As a result of their dissatisfaction with our rescission practices, in late 2009, Countrywide commenced litigation against us and Countrywide and its Bank of America affiliates stopped purchasing insurance from MGIC for the loans they originated. See our risk factor titled “We are subject to the risk of private litigation and regulatory proceedings” for more information about this litigation and the arbitration case we filed against Countrywide regarding rescissions. Bank of America recently informed us that it intends, at some point in the future, to implement procedures to enable them to cancel MGIC’s coverage on loans Bank of America purchases from correspondent lenders and substitute coverage from our competitors. In general, a correspondent lender funds loans and then sells them servicing-released to another lender who retains the servicing and either sells the loans to an investor or retains them in portfolio. Traditionally a correspondent lender’s selection of which mortgage insurer insures the loans it funds has not been changed by the lender to which those loans were sold. We estimate that during 2010 approximately 10% of our new insurance written was for loans purchased by Bank of America from correspondent lenders. The effect of Bank of America’s actions on us may depend on the reaction of correspondent lenders and any reaction from the GSEs, as well as other factors. While we will be taking various actions to seek to retain this business, we cannot predict the extent to which Bank of America’s actions will adversely affect us.
We continue to expect new insurance written in 2011 to increase modestly over the $12 billion written in 2010. Our level of new insurance written could also be affected by other items, including those noted in our Risk Factors.
From time to time, in response to market conditions, we change the types of loans that we insure and the guidelines under which we insure them. In addition, we make exceptions to our underwriting guidelines on a loan-by-loan basis and for certain customer programs. Together these exceptions accounted for fewer than 5% of the loans we insured in the second half of 2010 and fewer than 6% of the loans we insured in the first quarter of 2011. A large percentage of the exceptions were made for loans with debt-to-income ratios slightly above our guideline. Beginning in September 2009, we have made changes to our underwriting guidelines that have allowed certain loans to be eligible for insurance that were not eligible prior to those changes and we expect to continue to make changes in appropriate circumstances in the future. Our underwriting guidelines are available on our website at http://www.mgic.com/guides/underwriting.html.
Cancellations, insurance in force and risk in force
New insurance written and cancellations of primary insurance in force during the three months ended March 31, 2011 and 2010 were as follows:
| |
|
Three Months Ended
|
|
| |
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
(In billions)
|
|
| |
|
|
|
|
|
|
|
NIW
|
|
$ |
3.0 |
|
|
$ |
1.8 |
|
|
Cancellations
|
|
|
(7.4 |
) |
|
|
(6.9 |
) |
| |
|
|
|
|
|
|
|
|
|
Change in primary insurance in force
|
|
$ |
(4.4 |
) |
|
$ |
(5.1 |
) |
| |
|
|
|
|
|
|
|
|
|
Direct primary insurance in force as of March 31,
|
|
$ |
186.9 |
|
|
$ |
207.1 |
|
| |
|
|
|
|
|
|
|
|
|
Direct primary risk in force as of March 31,
|
|
$ |
47.9 |
|
|
$ |
53.0 |
|
Cancellation activity has historically been affected by the level of mortgage interest rates and the level of home price appreciation. Cancellations generally move inversely to the change in the direction of interest rates, although they generally lag a change in direction. Cancellations also include rescissions and policies cancelled due to claim payment. Since 2009, cancellations due to rescissions and claim payments have comprised a significant amount of our cancellations.
Our persistency rate was 83.7% at March 31, 2011 compared to 84.4% at December 31, 2010 and 85.6% at March 31, 2010. These persistency rates reflect the more restrictive credit policies of lenders (which make it more difficult for homeowners to refinance loans), as well as declines in housing values.
Bulk transactions
We ceased writing Wall Street bulk business in the fourth quarter of 2007. In addition, we wrote no new business through the bulk channel since the second quarter of 2008. We expect the volume of any future business written through the bulk channel will be insignificant. Wall Street bulk transactions, as of March 31, 2011, included approximately 86,700 loans with insurance in force of approximately $13.7 billion and risk in force of approximately $4.0 billion, which is approximately 63% of our bulk risk in force.
Pool insurance
We are currently not issuing new commitments for pool insurance and expect that the volume of any future pool business will be insignificant.
Our direct pool risk in force was $2.5 billion ($1.1 billion on pool policies with aggregate loss limits and $1.4 billion on pool policies without aggregate loss limits) at March 31, 2011 compared to $2.7 billion ($1.2 billion on pool policies with aggregate loss limits and $1.5 billion on pool policies without aggregate loss limits) at December 31, 2010.
Freddie Mac, one of our pool insurance insureds, is computing the aggregate loss limit under a pool insurance policy at a higher level than we are computing this limit because we believe the original aggregate limit decreases over time while they believe the limit remains constant. At March 31, 2011, the difference was approximately $535 million and under our interpretation this difference may increase by up to approximately $205 million in August 2011 depending on the development of losses under this pool policy. This difference has had no effect on our results of operations because the aggregate paid losses plus the portion of our loss reserves attributable to this policy have been below our interpretation of the loss limit. Based on our interpretation of the pool insurance policy, and our expected loss development, we believe it is possible that in the second quarter of 2011 the losses from delinquent loans under this policy will exceed our view of the aggregate loss limit, with the result that we will not recognize the excess portion of such losses as incurred losses.
Net premiums written and earned
Net premiums written and earned during the first quarter of 2011 increased when compared to the same period in 2010. The increase was due to lower levels of estimated premium refunds and lower ceded premiums due to captive terminations and run-offs, offset by lower average insurance in force. In a captive termination, the arrangement is cancelled, with no future premium ceded and funds for any incurred but unpaid losses transferred to us. In a run-off, no new loans are reinsured by the captive but loans previously reinsured continue to be covered, with premium and losses continuing to be ceded on those loans.
We expect our average insurance in force to continue to decline through 2011 because our expected new insurance written levels are not expected to exceed our cancellation activity. We expect our premium yields (net premiums written or earned, expressed on an annual basis, divided by the average insurance in force) for the remainder of 2011 to continue at approximately the level experienced during the first quarter of 2011.
Risk sharing arrangements
For the quarter ended March 31, 2011, approximately 5% of our flow new insurance written was subject to arrangements with captives which was comparable to the year ended December 31, 2010. We expect the percentage of new insurance written subject to risk sharing arrangements to continue to approximate 5% in 2011 for the reasons discussed below.
Effective January 1, 2009, we are no longer ceding new business under excess of loss reinsurance treaties with lender captive reinsurers. Loans reinsured through December 31, 2008 under excess of loss agreements will run off pursuant to the terms of the particular captive arrangement. New business will continue to be ceded under quota share reinsurance arrangements, limited to a 25% cede rate. Beginning in 2008, many of our captive arrangements have either been terminated or placed into run-off.
We anticipate that our ceded premiums related to risk sharing agreements will continue to decline in 2011 for the reasons discussed above.
See discussion under “-Losses—Losses incurred” regarding losses assumed by captives.
Investment income
Investment income in the first quarter of 2011 was lower when compared to the same period in 2010 due to a decrease in the average investment yield. The decrease in the average investment yield was caused both by decreases in prevailing interest rates and a decrease in the average maturity of our investments. The average maturity of our investments has continued to decrease as the proceeds from the April 2010 offerings have been invested in shorter term instruments. See further discussion under “Liquidity and Capital Resources” below. The portfolio’s average pre-tax investment yield was 2.7% at March 31, 2011 and 3.2% at March 31, 2010. The portfolio’s average pre-tax investment yield, excluding cash and cash equivalents, was 3.1% at March 31, 2011 and 3.5% at March 31, 2010.
We continue to expect a decline in investment income throughout 2011, compared to 2010, as the average amortized cost of invested assets decreases due to claim payments exceeding premiums received in future periods. See further discussion under “Liquidity and Capital Resources” below.
Realized gains and other-than-temporary impairments
Net realized gains for the first quarter of 2011 included $5.7 million in net realized gains on the sale of fixed income investments. Net realized gains for the first quarter of 2010 included $33.0 million in net realized gains on the sale of fixed income investments, offset by $6.1 million in OTTI losses.
Other revenue
Other revenue for the first quarter of 2011 decreased, when compared to the same period in 2010, due to a decrease in contract underwriting revenues.
Losses
As discussed in “Critical Accounting Policies” in our 10-K MD&A and consistent with industry practices, we establish loss reserves for future claims only for loans that are currently delinquent. The terms “delinquent” and “default” are used interchangeably by us and are defined as an insured loan with a mortgage payment that is 45 days or more past due. Loss reserves are established based on estimating the number of loans in our default inventory that will result in a claim payment, which is referred to as the claim rate, and further estimating the amount of the claim payment, which is referred to as claim severity. Historically, a substantial majority of borrowers have eventually cured their delinquent loans by making their overdue payments, but this percentage has decreased significantly in recent years.
Estimation of losses that we will pay in the future is inherently judgmental. The conditions that affect the claim rate and claim severity include the current and future state of the economy, including unemployment and local housing markets. Current conditions in the housing and mortgage industries make these assumptions more volatile than they would otherwise be. The actual amount of the claim payments may be substantially different than our loss reserve estimates. Our estimates could be adversely affected by several factors, including a further deterioration of regional or national economic conditions, including unemployment, leading to a reduction in borrowers’ income and thus their ability to make mortgage payments, and a further drop in housing values, which expose us to greater losses on resale of properties obtained through the claim settlement process and may affect borrower willingness to continue to make mortgage payments when the value of the home is below the mortgage balance. Our estimates are also affected by any agreements we enter into regarding claim payments, such as the settlement agreement discussed below under “Losses incurred.” Changes to our estimates could result in a material impact to our results of operations, even in a stable economic environment.
In addition, our loss reserving methodology incorporates the effects rescission activity is expected to have on the losses we will pay on our delinquent inventory. A variance between ultimate actual rescission rates and these estimates could materially affect our losses. See our risk factor titled “We may not continue to realize benefits from rescissions at the levels we have recently experienced and we may not prevail in proceedings challenging whether our rescissions were proper.”
Our estimates could also be positively affected by efforts to assist current borrowers in refinancing to new loans, assisting delinquent borrowers in reducing their mortgage payments, and forestalling foreclosures. If these benefits occur, we anticipate they will do so under non-HAMP programs. See discussion of HAMP under “Overview – Loan Modification and Other Similar Programs.”
Losses incurred
In the first quarter of 2011, net losses incurred were $310 million, which represented $347 million related to current year loss development offset by $37 million related to favorable prior years’ loss development. In the first quarter of 2010, net losses incurred were $455 million, which represented $513 million related to current year loss development offset by $58 million related to favorable prior years’ loss development. See Note 12 – “Loss Reserves” to our consolidated financial statements.
Current year losses incurred decreased in the first quarter of 2011 compared to the same period in 2010 primarily due to a decrease in the number of new notices received, from 53,393 in the first quarter of 2010 to 43,195 in the first quarter of 2011.
The amount of losses incurred relating to default notices received in prior years represents the actual claim rate and severity associated with those default notices resolved in the current period to the extent it differs from the estimated liability at the prior year-end, as well as a re-estimation of amounts to be ultimately paid on defaults remaining in inventory from the end of the prior year. This re-estimation of the claim rate and severity is the result of our review of current trends in default inventory, such as percentages of defaults that have resulted in a claim, the amount of the claims, changes in the relative level of defaults by geography and changes in average loan exposure. The $37 million decrease in losses incurred in the first quarter of 2011 was related to defaults that occurred in prior periods. This decrease in losses incurred primarily related to a slight decrease in severity on primary defaults which approximated $28 million as well as a slight decrease in the expected claim rate on primary defaults which accounted for a decrease of approximately $16 million. The decrease in the severity and claim rate was based on the resolution of approximately 22% of the prior year default inventory, as well as a re-estimation of amounts to be ultimately paid on defaults remaining in inventory from the end of the prior year. The offsetting increase in losses incurred related to prior years of approximately $7 million related to pool reserves, LAE reserves and reinsurance.
The $58 million decrease in losses incurred in the first quarter of 2010 was related to defaults that occurred in prior periods. This decrease in losses incurred primarily related to a decrease in the claim rate on primary and pool defaults which approximated $206 million. The decrease in the claim rate was based on the resolution of approximately 19% of the prior year default inventory, as well as a re-estimation of amounts to be ultimately paid on defaults remaining in inventory from the end of the prior year. The decrease in the claim rate was due to greater cures experienced during the first quarter of 2010, a portion of which resulted from loan modifications. The decrease in the claim rate on prior year defaults was offset by an increase in severity on primary and pool defaults which approximated $151 million. The increase in severity was based on the resolution of defaults that occurred in prior periods with higher claim amounts, as well as a re-estimation of amount to be ultimately paid on defaults remaining in inventory from the end of the prior year. The additional decrease in losses incurred related to prior years of approximately $3 million related to LAE reserves and reinsurance.
The decrease in the primary default inventory experienced during the first quarter of 2011 was generally across all markets and all book years. However the number of consecutive months a loan remains in the primary default inventory (the age of the item in default) has continued to increase, as shown in the table below. Historically as a default ages it becomes more likely to result in a claim.
Aging of the Primary Default Inventory
| |
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Consecutive months in the default inventory
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 months or less
|
|
|
27,744 |
|
|
|
14 |
% |
|
|
37,640 |
|
|
|
18 |
% |
|
|
36,256 |
|
|
|
15 |
% |
|
4 - 11 months
|
|
|
57,319 |
|
|
|
29 |
% |
|
|
58,701 |
|
|
|
27 |
% |
|
|
90,816 |
|
|
|
38 |
% |
|
12 months or more
|
|
|
110,822 |
|
|
|
57 |
% |
|
|
118,383 |
|
|
|
55 |
% |
|
|
114,172 |
|
|
|
47 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total primary default inventory
|
|
|
195,885 |
|
|
|
100 |
% |
|
|
214,724 |
|
|
|
100 |
% |
|
|
241,244 |
|
|
|
100 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Loans in our default inventory in our claims received inventory
|
|
|
17,686 |
|
|
|
9 |
% |
|
|
20,898 |
|
|
|
10 |
% |
|
|
17,384 |
|
|
|
7 |
% |
The length of time a loan is in the default inventory can differ from the number of payments that the borrower has not made or is considered delinquent. These differences typically result from a borrower making monthly payments that do not result in the loan becoming fully current. The number of payments that a borrower is delinquent is shown in the table below.
Number of Payments Delinquent
| |
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 payments or less
|
|
|
40,680 |
|
|
|
21 |
% |
|
|
51,003 |
|
|
|
24 |
% |
|
|
50,045 |
|
|
|
21 |
% |
|
4 - 11 payments
|
|
|
61,060 |
|
|
|
31 |
% |
|
|
65,797 |
|
|
|
31 |
% |
|
|
98,753 |
|
|
|
41 |
% |
|
12 payments or more
|
|
|
94,145 |
|
|
|
48 |
% |
|
|
97,924 |
|
|
|
45 |
% |
|
|
92,446 |
|
|
|
38 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total primary default inventory
|
|
|
195,885 |
|
|
|
100 |
% |
|
|
214,724 |
|
|
|
100 |
% |
|
|
241,244 |
|
|
|
100 |
% |
Before paying a claim, we can review the loan file to determine whether we are required, under the applicable insurance policy, to pay the claim or whether we are entitled to reduce the amount of the claim. For example, all of our insurance policies provide that we can reduce or deny a claim if the servicer did not comply with its obligation to mitigate our loss by performing reasonable loss mitigation efforts or diligently pursuing a foreclosure or bankruptcy relief in a timely manner. We also do not cover losses resulting from property damage that has not been repaired. We are currently reviewing the loan files for the majority of the claims submitted to us.
In addition, subject to rescission caps in certain of our Wall Street bulk transactions, all of our insurance policies allow us to rescind coverage under certain circumstances. Because we can review the loan origination documents and information as part of our normal processing when a claim is submitted to us, rescissions occur on a loan by loan basis most often after we have received a claim. Historically, claim rescissions and denials, which we collectively refer to as rescissions, were not a material portion of our claims resolved during a year. However, beginning in 2008 our rescissions of policies have materially mitigated our paid and incurred losses. While we have a substantial pipeline of claims investigations that we expect will eventually result in future rescissions, we expect that rescissions will not continue to mitigate paid and incurred losses at the same level we have recently experienced. In addition, if an insured disputes our right to rescind coverage, the outcome of the dispute ultimately would be determined by legal proceedings. In each of 2009 and 2010, rescissions mitigated our paid losses by approximately $1.2 billion, and in the first quarter of 2011, rescissions mitigated our paid losses by approximately $0.2 billion. These figures include amounts that would have resulted in either a claim payment or been charged to a deductible or aggregate loss limit under a bulk or pool policy, and may have been charged to a captive reinsurer. The amounts that would have been applied to a deductible do not take into account previous rescissions that may have been applied to a deductible.
Our loss reserving methodology incorporates the effect that rescission activity is expected to have on the losses we will pay on our delinquent inventory. We do not utilize an explicit rescission rate in our reserving methodology, but rather our reserving methodology incorporates the effects rescission activity has had on our historical claim rate and claim severities. A variance between ultimate actual rescission rates and these estimates could materially affect our losses incurred. Our estimation process does not include a direct correlation between claim rates and severities to projected rescission activity or other economic conditions such as changes in unemployment rates, interest rates or housing values. Our experience is that analysis of that nature would not produce reliable results, as the change in one condition cannot be isolated to determine its sole effect on our ultimate paid losses as our ultimate paid losses are also influenced at the same time by other economic conditions. The estimation of the impact of rescissions on losses incurred, as shown in the table below, must be considered together with the various other factors impacting losses incurred and not in isolation.
The table below represents our estimate of the impact rescissions have had on reducing our loss reserves, paid losses and losses incurred.
| |
|
Three Months Ended
|
|
| |
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
(In billions)
|
|
| |
|
|
|
|
|
|
|
Estimated rescission reduction - beginning reserve
|
|
$ |
1.3 |
|
|
$ |
2.1 |
|
| |
|
|
|
|
|
|
|
|
|
Estimated rescission reduction - losses incurred
|
|
|
- |
|
|
|
0.6 |
|
| |
|
|
|
|
|
|
|
|
|
Rescission reduction - paid claims
|
|
|
0.2 |
|
|
|
0.4 |
|
|
Amounts that may have been applied to a deductible
|
|
|
- |
|
|
|
(0.1 |
) |
|
Net rescission reduction - paid claims
|
|
|
0.2 |
|
|
|
0.3 |
|
| |
|
|
|
|
|
|
|
|
|
Estimated rescission reduction - ending reserve
|
|
$ |
1.1 |
|
|
$ |
2.4 |
|
The decrease in the estimated rescission reduction to losses incurred in the first quarter of 2011 compared to the same period in 2010 is due to a decline in the expected rescission rate for loans in our default inventory, compared to an increasing expected rescission rate in the first quarter of 2010.
At March 31, 2011, our loss reserves continued to be significantly impacted by expected rescission activity. We expect that the reduction of our loss reserves due to rescissions will continue to decline because our recent experience indicates new notices in our default inventory have a lower likelihood of being rescinded than those already in the inventory due to their product mix, geographic location and vintage.
The liability associated with our estimate of premiums to be refunded on expected future rescissions is accrued for separately. At March 31, 2011 and December 31, 2010 the estimate of this liability totaled $89 million and $101 million, respectively. Separate components of this liability are included in “Other liabilities” and “Premium deficiency reserve” on our consolidated balance sheet. Changes in the liability affect premiums written and earned and change in premium deficiency reserve.
If the insured disputes our right to rescind coverage, the outcome of the dispute ultimately would be determined by legal proceedings. Legal proceedings disputing our right to rescind coverage may be brought up to three years after the lender has obtained title to the property (typically through a foreclosure) or the property was sold in a sale that we approved, whichever is applicable, although in a few jurisdictions there is a longer time to bring such an action. For nearly all of our rescissions that are not subject to a settlement agreement, the period in which a dispute may be brought has not ended. We consider a rescission resolved for reporting purposes even though legal proceedings have been initiated and are ongoing. Although it is reasonably possible that, when the proceedings are completed, there will be a determination that we were not entitled to rescind in all cases, we are unable to make a reasonable estimate or range of estimates of the potential liability. Under ASC 450-20, an estimated loss from such proceedings is accrued for only if we determine that the loss is probable and can be reasonably estimated. Therefore, when establishing our loss reserves, we do not include additional loss reserves that would reflect an adverse outcome from ongoing legal proceedings, including those with Countrywide. Countrywide has filed a lawsuit against MGIC alleging that MGIC has denied, and continues to deny, valid mortgage insurance claims. MGIC has filed an arbitration case against Countrywide regarding rescissions and Countrywide has responded seeking damages, including exemplary damages. For more information about this lawsuit and arbitration case, see Note 5 – “Litigation and contingencies” to our consolidated financial statements.
In 2010, we entered into a settlement agreement with a lender-customer regarding our rescission practices and we may enter into additional settlement agreements with other lenders in the future. See our risk factor titled “Changes in the business practices of the GSEs, federal legislation that changes their charters or a restructuring of the GSEs could reduce our revenues or increase our losses” for a discussion of GSE restrictions regarding future settlement agreements.
We continue to discuss with other lenders their objections to material rescissions. In addition to the proceedings involving Countrywide, we are involved in legal proceedings with respect to rescissions that we do not consider to be collectively material in amount. Although it is reasonably possible that, when these discussions or proceedings are completed, there will be a conclusion or determination that we were not entitled to rescind in all cases, we are unable to make a reasonable estimate or range of estimates of the potential liability.
Information regarding the ever-to-date rescission rates by the quarter in which the claim was received appears in the table below. No information is presented for claims received in the most recent two quarters to allow sufficient time for a substantial percentage of the claims received in those two quarters to reach resolution.
As of March 31, 2011
Ever to Date Rescission Rates on Primary Claims Received
(based on count)
|
Quarter in Which the
Claim was Received
|
|
ETD Rescission
Rate (1)
|
|
ETD Claims Resolution
Percentage (2)
|
| |
|
|
|
|
|
Q2 2009
|
|
27.8%
|
|
100.0%
|
|
Q3 2009
|
|
27.4%
|
|
100.0%
|
|
Q4 2009
|
|
24.0%
|
|
99.9%
|
|
Q1 2010
|
|
21.1%
|
|
99.6%
|
|
Q2 2010
|
|
20.0%
|
|
98.7%
|
|
Q3 2010
|
|
17.2%
|
|
93.2%
|
(1) This percentage is claims received during the quarter shown that have been rescinded as of our most recently completed quarter divided by the total claims received during the quarter shown. In certain cases we rescind coverage before a claim is received. Such rescissions, which have not been material, are not included in the statistics in this table.
(2) This percentage is claims received during the quarter shown that have been resolved as of our most recently completed quarter divided by the total claims received during the quarter shown. Claims resolved principally consist of claims paid plus claims for which we have informed the insured of our decision not to pay the claim. Although our decision to not pay a claim is made after we have given the insured an opportunity to dispute the facts underlying our decision to not pay the claim, these decisions are sometimes reversed after further discussion with the insured. The number of rescission reversals has been immaterial.
We anticipate that the ever-to-date rescission rate on the more recent quarters will increase, to a greater or lesser degree, as the ever-to-date resolution percentage moves closer to 100%.
As discussed under “–Risk sharing arrangements,” approximately 5% of our flow new insurance written is subject to reinsurance arrangements with lender captives. Captive agreements are written on an annual book of business and the captives are required to maintain a separate trust account to support the combined reinsured risk on all annual books. MGIC is the sole beneficiary of the trust, and the trust account is made up of capital deposits by the lender captive, premium deposits by MGIC, and investment income earned. These amounts are held in the trust account and are available to pay reinsured losses. The reinsurance recoverable on loss reserves related to captive agreements was approximately $214 million at March 31, 2011 which was supported by $460 million of trust assets, while at December 31, 2010 the reinsurance recoverable on loss reserves related to captives was $248 million which was supported by $484 million of trust assets. As of March 31, 2011 and December 31, 2010 there was an additional $26 million of trust assets in captive agreements where there was no related reinsurance recoverable on loss reserves. For additional discussion regarding our captive arrangements see “Losses—Losses incurred” in our 10-K MD&A.
In the first quarter of 2011 the captive arrangements reduced our losses incurred by approximately $14 million, compared to a $25 million captive reduction in the first quarter of 2010. We anticipate that the reduction in losses incurred will continue to be lower in 2011, as some of our captive arrangements were terminated in 2010.
A rollforward of our primary default inventory for the three months ended March 31, 2011 and 2010 appears in the table below. The information concerning new notices and cures is compiled from monthly reports received from loan servicers. The level of new notice and cure activity reported in a particular month can be influenced by, among other things, the date on which a servicer generates its report and by transfers of servicing between loan servicers.
| |
|
Three Months Ended
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
Default inventory at beginning of period
|
|
|
214,724 |
|
|
|
250,440 |
|
|
Plus: New Notices
|
|
|
43,195 |
|
|
|
53,393 |
|
|
Less: Cures
|
|
|
(45,639 |
) |
|
|
(49,210 |
) |
|
Less: Paids (including those charged to a deductible or captive)
|
|
|
(13,466 |
) |
|
|
(9,194 |
) |
|
Less: Rescissions and denials
|
|
|
(2,929 |
) |
|
|
(4,185 |
) |
|
Default inventory at end of period
|
|
|
195,885 |
|
|
|
241,244 |
|
Information about the composition of the primary default inventory at March 31, 2011, December 31, 2010 and March 31, 2010 appears in the table below.
| |
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
Total loans delinquent (1)
|
|
|
195,885 |
|
|
|
214,724 |
|
|
|
241,244 |
|
|
Percentage of loans delinquent (default rate)
|
|
|
16.35 |
% |
|
|
17.48 |
% |
|
|
18.14 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
Prime loans delinquent (2)
|
|
|
123,046 |
|
|
|
134,787 |
|
|
|
148,101 |
|
|
Percentage of prime loans delinquent (default rate)
|
|
|
12.24 |
% |
|
|
13.11 |
% |
|
|
13.33 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
A-minus loans delinquent (2)
|
|
|
28,073 |
|
|
|
31,566 |
|
|
|
34,821 |
|
|
Percent of A-minus loans delinquent (default rate)
|
|
|
33.80 |
% |
|
|
36.69 |
% |
|
|
38.63 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
Subprime credit loans delinquent (2)
|
|
|
10,053 |
|
|
|
11,132 |
|
|
|
12,536 |
|
|
Percentage of subprime credit loans delinquent (default rate)
|
|
|
42.51 |
% |
|
|
45.66 |
% |
|
|
47.80 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
Reduced documentation loans delinquent (3)
|
|
|
34,713 |
|
|
|
37,239 |
|
|
|
45,786 |
|
|
Percentage of reduced documentation loans delinquent (default rate)
|
|
|
40.37 |
% |
|
|
41.66 |
% |
|
|
44.58 |
% |
General Notes: (a) For the information presented for March 31, 2011 and December 31, 2010, the FICO credit score for a loan with multiple borrowers is the lowest of the borrowers’ “decision FICO scores.” For the information presented for March 31, 2010, the FICO score for a loan with multiple borrowers was the income weighted average of the “decision FICO scores” for each borrower. A borrower’s “decision FICO score” is determined as follows: if there are three FICO scores available, the middle FICO score is used; if two FICO scores are available, the lower of the two is used; if only one FICO score is available, it is used.
(b) Servicers continue to pay our premiums for nearly all of the loans in our default inventory, but in some cases, servicers stop paying our premiums. In those cases, even though the loans continue to be included in our default inventory, the applicable loans are removed from our insurance in force and risk in force. Loans where servicers have stopped paying premiums include 13,098 defaults with a risk of $636.2 million as of March 31, 2011.
(1) At March 31, 2011, December 31, 2010 and March 31, 2010 33,421, 36,066 and 41,727 loans in default, respectively, related to Wall Street bulk transactions.
(2) We define prime loans as those having FICO credit scores of 620 or greater, A-minus loans as those having FICO credit scores of 575-619, and subprime credit loans as those having FICO credit scores of less than 575, all as reported to us at the time a commitment to insure is issued. Most A-minus and subprime credit loans were written through the bulk channel. However, we classify all loans without complete documentation as “reduced documentation” loans regardless of FICO score rather than as a prime, “A-minus” or “subprime” loan; in the table above, such loans appear only in the reduced documentation category and they do not appear in any of the other categories.
(3) In accordance with industry practice, loans approved by GSE and other automated underwriting (AU) systems under "doc waiver" programs that do not require verification of borrower income are classified by MGIC as "full documentation." Based in part on information provided by the GSEs, we estimate full documentation loans of this type were approximately 4% of 2007 NIW. Information for other periods is not available. We understand these AU systems grant such doc waivers for loans they judge to have higher credit quality. We also understand that the GSEs terminated their “doc waiver” programs, with respect to new commitments, in the second half of 2008.
Pool insurance notice inventory decreased from 43,329 at December 31, 2010 to 40,769 at March 31, 2011. The pool insurance notice inventory was 42,664 at March 31, 2010. We expect that the trend of increased pool claim payments shown below in the table below titled “Net paid claims” will continue.
The primary and pool loss reserves at March 31, 2011 and December 31, 2010 appear in the table below.
|
Gross Reserves
|
|
March 31,
2011
|
|
|
December 31,
2010
|
|
| |
|
|
|
|
|
|
|
Primary
|
|
|
|
|
|
|
|
Direct loss reserves (in millions)
|
|
$ |
4,766 |
|
|
$ |
5,146 |
|
|
Default inventory
|
|
|
195,885 |
|
|
|
214,724 |
|
|
Average direct reserve per default
|
|
$ |
24,331 |
|
|
$ |
23,966 |
|
| |
|
|
|
|
|
|
|
|
|
Pool
|
|
|
|
|
|
|
|
|
|
Direct loss reserves (in millions)
|
|
$ |
697 |
|
|
$ |
730 |
|
|
Default inventory
|
|
|
40,769 |
|
|
|
43,329 |
|
| |
|
|
|
|
|
|
|
|
|
Other gross reserves (in millions)
|
|
$ |
8 |
|
|
$ |
8 |
|
Note: Since a number of our pool policies include aggregate loss limits and/or deductibles, we do not disclose an average direct reserve per default for our pool business.
The primary default inventory and primary loss reserves by region at March 31, 2011, December 31, 2010 and March 31, 2010 appears in the table below.
|
Primary Default Inventory
|
|
|
|
|
|
|
|
|
|
| |
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
|
Region
|
|
2011
|
|
|
2010
|
|
|
2010
|
|
|
Great Lakes
|
|
|
24,595 |
|
|
|
27,663 |
|
|
|
30,762 |
|
|
Mid-Atlantic
|
|
|
8,901 |
|
|
|
9,660 |
|
|
|
11,004 |
|
|
New England
|
|
|
7,356 |
|
|
|
7,702 |
|
|
|
8,600 |
|
|
North Central
|
|
|
22,348 |
|
|
|
24,192 |
|
|
|
26,708 |
|
|
Northeast
|
|
|
17,822 |
|
|
|
19,056 |
|
|
|
20,032 |
|
|
Pacific
|
|
|
23,336 |
|
|
|
25,438 |
|
|
|
31,089 |
|
|
Plains
|
|
|
6,235 |
|
|
|
7,045 |
|
|
|
7,570 |
|
|
South Central
|
|
|
25,419 |
|
|
|
28,984 |
|
|
|
33,126 |
|
|
Southeast
|
|
|
59,873 |
|
|
|
64,984 |
|
|
|
72,353 |
|
|
Total
|
|
|
195,885 |
|
|
|
214,724 |
|
|
|
241,244 |
|
|
Primary Loss Reserves
|
|
|
|
|
|
|
|
|
|
|
|
|
|
(In millions)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Region
|
|
March 31,
2011
|
|
|
December 31,
2010
|
|
|
March 31,
2010
|
|
|
Great Lakes
|
|
$ |
388 |
|
|
$ |
426 |
|
|
$ |
516 |
|
|
Mid-Atlantic
|
|
|
213 |
|
|
|
231 |
|
|
|
238 |
|
|
New England
|
|
|
145 |
|
|
|
174 |
|
|
|
205 |
|
|
North Central
|
|
|
456 |
|
|
|
495 |
|
|
|
534 |
|
|
Northeast
|
|
|
330 |
|
|
|
374 |
|
|
|
461 |
|
|
Pacific
|
|
|
847 |
|
|
|
886 |
|
|
|
1,005 |
|
|
Plains
|
|
|
96 |
|
|
|
107 |
|
|
|
122 |
|
|
South Central
|
|
|
508 |
|
|
|
555 |
|
|
|
588 |
|
|
Southeast
|
|
|
1,295 |
|
|
|
1,395 |
|
|
|
1,660 |
|
|
Total before IBNR and LAE
|
|
$ |
4,278 |
|
|
$ |
4,643 |
|
|
$ |
5,329 |
|
|
IBNR and LAE
|
|
|
488 |
|
|
|
503 |
|
|
|
648 |
|
|
Total
|
|
$ |
4,766 |
|
|
$ |
5,146 |
|
|
$ |
5,977 |
|
|
Regions contain the states as follows:
|
|
Great Lakes: IN, KY, MI, OH
|
|
Mid-Atlantic: DC, DE, MD, VA, WV
|
|
New England: CT, MA, ME, NH, RI, VT
|
|
North Central: IL, MN, MO, WI
|
|
Northeast: NJ, NY, PA
|
|
Pacific: CA, HI, NV, OR, WA
|
|
Plains: IA, ID, KS, MT, ND, NE, SD, WY
|
|
South Central: AK, AZ, CO, LA, NM, OK, TX, UT
|
|
Southeast: AL, AR, FL, GA, MS, NC, SC, TN
|
The primary loss reserves at March 31, 2011, December 31, 2010 and March 31, 2010 separated between our flow and bulk business appears in the table below.
|
Primary loss reserves
|
|
|
|
|
|
|
|
(In millions)
|
|
|
|
|
December 31,
2010
|
|
|
|
|
|
Flow
|
|
$ |
3,047 |
|
|
$ |
3,329 |
|
|
$ |
3,634 |
|
|
Bulk
|
|
|
1,231 |
|
|
|
1,314 |
|
|
|
1,695 |
|
|
Total primary reserves
|
|
$ |
4,278 |
|
|
$ |
4,643 |
|
|
$ |
5,329 |
|
The average claim paid, as shown in the table below, can vary materially from period to period based upon a variety of factors, on both a national and state basis, including the geographic mix, average loan amount and average coverage percentage of loans for which claims are paid.
The primary average claim paid for the top 5 states (based on 2011 paid claims) for the three months ended March 31, 2011 and 2010 appears in the table below.
|
Primary average claim paid
|
|
|
|
| |
|
Three Months Ended
|
|
| |
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
|
Florida
|
|
$ |
57,540 |
|
|
$ |
66,315 |
|
|
California
|
|
|
80,609 |
|
|
|
95,061 |
|
|
Arizona
|
|
|
55,696 |
|
|
|
63,830 |
|
|
Georgia
|
|
|
41,059 |
|
|
|
43,483 |
|
|
Michigan
|
|
|
33,750 |
|
|
|
36,076 |
|
|
All other states
|
|
|
44,062 |
|
|
|
47,243 |
|
| |
|
|
|
|
|
|
|
|
|
All states
|
|
$ |
47,921 |
|
|
$ |
53,070 |
|
The primary average loan size of our insurance in force at March 31, 2011, December 31, 2010 and March 31, 2010 appears in the table below.
|
Primary average loan size
|
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
|
2010
|
|
|
Total insurance in force
|
|
$ |
156,010 |
|
|
$ |
155,700 |
|
|
$ |
155,730 |
|
|
Prime (FICO 620 & >)
|
|
|
155,550 |
|
|
|
155,050 |
|
|
|
154,430 |
|
|
A-Minus (FICO 575-619)
|
|
|
129,970 |
|
|
|
130,360 |
|
|
|
129,950 |
|
|
Subprime (FICO < 575)
|
|
|
117,090 |
|
|
|
117,410 |
|
|
|
118,040 |
|
|
Reduced doc (All FICOs)
|
|
|
197,270 |
|
|
|
198,000 |
|
|
|
202,120 |
|
The primary average loan size of our insurance in force at March 31, 2011, December 31, 2010 and March 31, 2010 for the top 5 states (based on 2011 paid claims) appears in the table below.
|
Primary average loan size
|
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
|
2010
|
|
|
Florida
|
|
$ |
173,888 |
|
|
$ |
174,203 |
|
|
$ |
177,318 |
|
|
California
|
|
|
282,745 |
|
|
|
283,459 |
|
|
|
287,154 |
|
|
Arizona
|
|
|
184,116 |
|
|
|
184,508 |
|
|
|
187,735 |
|
|
Georgia
|
|
|
148,174 |
|
|
|
148,002 |
|
|
|
148,468 |
|
|
Michigan
|
|
|
121,461 |
|
|
|
121,282 |
|
|
|
121,288 |
|
|
All other states
|
|
|
149,619 |
|
|
|
149,182 |
|
|
|
148,537 |
|
Information about net paid claims during the three months ended March 31, 2011 and 2010 appears in the table below.
|
Net paid claims (In millions)
|
|
|
|
|
|
|
| |
|
Three Months Ended
|
|
| |
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
|
Prime (FICO 620 & >)
|
|
$ |
451 |
|
|
$ |
288 |
|
|
A-Minus (FICO 575-619)
|
|
|
76 |
|
|
|
62 |
|
|
Subprime (FICO < 575)
|
|
|
19 |
|
|
|
21 |
|
|
Reduced doc (All FICOs)
|
|
|
100 |
|
|
|
113 |
|
|
Pool
|
|
|
72 |
|
|
|
34 |
|
|
Other
|
|
|
1 |
|
|
|
1 |
|
|
Direct losses paid
|
|
|
719 |
|
|
|
519 |
|
|
Reinsurance
|
|
|
(48 |
) |
|
|
(17 |
) |
|
Net losses paid
|
|
|
671 |
|
|
|
502 |
|
|
LAE
|
|
|
16 |
|
|
|
17 |
|
|
Net losses and LAE paid before terminations
|
|
|
687 |
|
|
|
519 |
|
|
Reinsurance terminations
|
|
|
(1 |
) |
|
|
- |
|
|
Net losses and LAE paid
|
|
$ |
686 |
|
|
$ |
519 |
|
Primary claims paid for the top 15 states (based on 2011 paid claims) and all other states for the three months ended March 31, 2011 and 2010 appears in the table below.
|
Paid Claims by state (In millions)
|
|
|
|
|
| |
|
Three Months Ended
|
|
| |
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
|
|
|
|
|
|
Florida
|
|
$ |
84 |
|
|
$ |
71 |
|
|
California
|
|
|
80 |
|
|
|
66 |
|
|
Arizona
|
|
|
46 |
|
|
|
37 |
|
|
Georgia
|
|
|
37 |
|
|
|
19 |
|
|
Michigan
|
|
|
35 |
|
|
|
28 |
|
|
Texas
|
|
|
32 |
|
|
|
19 |
|
|
Nevada
|
|
|
31 |
|
|
|
24 |
|
|
Illinois
|
|
|
26 |
|
|
|
21 |
|
|
Ohio
|
|
|
23 |
|
|
|
16 |
|
|
Virginia
|
|
|
18 |
|
|
|
16 |
|
|
Maryland
|
|
|
16 |
|
|
|
13 |
|
|
Washington
|
|
|
16 |
|
|
|
10 |
|
|
Colorado
|
|
|
14 |
|
|
|
7 |
|
|
Minnesota
|
|
|
13 |
|
|
|
11 |
|
|
North Carolina
|
|
|
13 |
|
|
|
6 |
|
|
All other states
|
|
|
162 |
|
|
|
120 |
|
| |
|
$ |
646 |
|
|
$ |
484 |
|
|
Other (Pool, LAE, Reinsurance)
|
|
|
40 |
|
|
|
35 |
|
| |
|
$ |
686 |
|
|
$ |
519 |
|
The primary default inventory in those same states at March 31, 2011, December 31, 2010 and March 31, 2010 appears in the table below.
| |
|
|
|
|
|
|
| |
March 31,
|
|
|
December 31,
|
March 31,
|
| |
2011
|
|
|
2010
|
|
2010
|
|
Florida
|
30,883
|
|
|
32,788
|
|
37,619
|
|
California
|
12,712
|
|
|
14,070
|
|
18,490
|
|
Arizona
|
6,006
|
|
|
6,781
|
|
8,320
|
|
Georgia
|
7,929
|
|
|
9,117
|
|
10,700
|
|
Michigan
|
9,015
|
|
|
10,278
|
|
12,098
|
|
Texas
|
9,870
|
|
|
11,602
|
|
13,055
|
|
Nevada
|
4,268
|
|
|
4,729
|
|
5,640
|
|
Illinois
|
11,748
|
|
|
12,548
|
|
13,685
|
|
Ohio
|
8,779
|
|
|
9,850
|
|
10,442
|
|
Virginia
|
3,177
|
|
|
3,627
|
|
4,237
|
|
Maryland
|
4,051
|
|
|
4,264
|
|
4,863
|
|
Washington
|
3,735
|
|
|
3,888
|
|
3,917
|
|
Colorado
|
2,575
|
|
|
2,917
|
|
3,365
|
|
Minnesota
|
3,346
|
|
|
3,672
|
|
4,462
|
|
North Carolina
|
5,093
|
|
|
5,641
|
|
6,356
|
|
All other states
|
72,698
|
|
|
78,952
|
|
83,995
|
| |
195,885
|
|
|
214,724
|
|
241,244
|
The primary default inventory at March 31, 2011, December 31, 2010 and March 31, 2010 separated between our flow and bulk business appears in the table below.
| |
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
|
2010
|
|
|
Flow
|
|
|
147,267 |
|
|
|
162,621 |
|
|
|
180,898 |
|
|
Bulk
|
|
|
48,618 |
|
|
|
52,103 |
|
|
|
60,346 |
|
| |
|
|
195,885 |
|
|
|
214,724 |
|
|
|
241,244 |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
The flow default inventory by policy year at March 31, 2011, December 31, 2010 and March 31, 2010 appears in the table below.
Flow default inventory by policy year
| |
|
March 31,
|
|
|
December 31,
|
|
|
March 31,
|
|
|
Policy year:
|
|
2011
|
|
|
2010
|
|
|
2010
|
|
|
2002 and prior
|
|
|
13,343 |
|
|
|
14,914 |
|
|
|
16,679 |
|
|
2003
|
|
|
8,121 |
|
|
|
9,069 |
|
|
|
10,062 |
|
|
2004
|
|
|
10,917 |
|
|
|
12,077 |
|
|
|
13,420 |
|
|
2005
|
|
|
17,092 |
|
|
|
18,789 |
|
|
|
20,784 |
|
|
2006
|
|
|
25,624 |
|
|
|
28,284 |
|
|
|
31,911 |
|
|
2007
|
|
|
56,577 |
|
|
|
62,855 |
|
|
|
71,545 |
|
|
2008
|
|
|
14,984 |
|
|
|
16,059 |
|
|
|
16,261 |
|
|
2009
|
|
|
572 |
|
|
|
546 |
|
|
|
236 |
|
|
2010
|
|
|
37 |
|
|
|
28 |
|
|
|
- |
|
| |
|
|
147,267 |
|
|
|
162,621 |
|
|
|
180,898 |
|
Beginning in 2008, the rate at which claims are received and paid slowed for a combination of reasons, including foreclosure moratoriums, servicing delays, court delays, loan modifications and our claims investigations. Although these factors continue to affect our paid claims, we believe that paid claims for 2011 will be higher than 2010 given the large number of loans that are 12 months or more past due and the approximately 18,000 claims that have been received but not yet paid. Within this increase, we also expect the portion of our paid claims that are pool paid claims to continue to increase, in part, as a result of deductibles being exhausted.
The liability associated with our estimate of premiums to be refunded on expected claim payments is accrued for separately at March 31, 2011 and December 31, 2010 and approximated $112 million and $113 million, respectively. Separate components of this liability are included in “Other liabilities” and “Premium deficiency reserve” on our consolidated balance sheet. Changes in the liability affect premiums written and earned and change in premium deficiency reserve, respectively.
As of March 31, 2011, 59% of our primary insurance in force was written subsequent to December 31, 2006. On our flow business, the highest claim frequency years have typically been the third and fourth year after the year of loan origination. On our bulk business, the period of highest claims frequency has generally occurred earlier than in the historical pattern on our flow business. However, the pattern of claims frequency can be affected by many factors, including persistency and deteriorating economic conditions. Low persistency can have the effect of accelerating the period in the life of a book during which the highest claim frequency occurs. Deteriorating economic conditions can result in increasing claims following a period of declining claims.
Premium deficiency
During the first quarter of 2011 the premium deficiency reserve on Wall Street bulk transactions declined by $9 million from $179 million, as of December 31, 2010, to $170 million as of March 31, 2011. The $170 million premium deficiency reserve as of March 31, 2011 reflects the present value of expected future losses and expenses that exceeded the present value of expected future premium and already established loss reserves. The discount rate used in the calculation of the premium deficiency reserve at March 31, 2011 was 2.7%.
The components of the premium deficiency reserve at March 31, 2011 and December 31, 2010 appear in the table below.
| |
|
March 31,
|
|
|
December 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
(In millions)
|
|
|
Present value of expected future paid losses and expenses, net of expected future premium
|
|
$ |
(1,170 |
) |
|
$ |
(1,254 |
) |
| |
|
|
|
|
|
|
|
|
|
Established loss reserves
|
|
|
1,000 |
|
|
|
1,075 |
|
| |
|
|
|
|
|
|
|
|
|
Net deficiency
|
|
$ |
(170 |
) |
|
$ |
(179 |
) |
The decrease in the premium deficiency reserve for the three months ended March 31, 2011 was $9 million, as shown in the table below, which represents the net result of actual premiums, losses and expenses as well as a net change in assumptions for these periods. The net change in assumptions for the first quarter of 2011 is primarily related to higher estimated ultimate premiums.
| |
|
Three Months Ended
|
|
| |
|
March 31, 2011
|
|
| |
|
(In millions)
|
|
| |
|
|
|
|
|
|
|
Premium Deficiency Reserve at beginning of period
|
|
|
|
|
$ |
(179 |
) |
| |
|
|
|
|
|
|
|
|
Paid claims and loss adjustment expenses
|
|
$ |
75 |
|
|
|
|
|
|
Decrease in loss reserves
|
|
|
(75 |
) |
|
|
|
|
|
Premium earned
|
|
|
(33 |
) |
|
|
|
|
|
Effects of present valuing on future premiums, losses and expenses
|
|
|
(11 |
) |
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
Change in premium deficiency reserve to reflect actual premium, losses and expenses recognized
|
|
|
|
|
|
|
(44 |
) |
| |
|
|
|
|
|
|
|
|
Change in premium deficiency reserve to reflect change in assumptions relating to future premiums, losses, expenses and discount rate (1)
|
|
|
|
|
|
|
53 |
|
| |
|
|
|
|
|
|
|
|
|
Premium Deficiency Reserve at end of period
|
|
|
|
|
|
$ |
(170 |
) |
(1) A positive number for changes in assumptions relating to premiums, losses, expenses and discount rate indicates a redundancy of prior premium deficiency reserves.
Each quarter we perform a premium deficiency analysis on the portion of our book of business not covered by the premium deficiency described above. As of March 31, 2011, the analysis concluded that there was no premium deficiency on such portion of our book of business. For the reasons discussed below, our analysis of any potential deficiency reserve is subject to inherent uncertainty and requires significant judgment by management. To the extent, in a future period, expected losses are higher or expected premiums are lower than the assumptions we used in our analysis, we could be required to record a premium deficiency reserve on this portion of our book of business in such period.
The calculation of premium deficiency reserves requires the use of significant judgments and estimates to determine the present value of future premium and present value of expected losses and expenses on our business. The present value of future premiums relies on, among other things, assumptions about persistency and repayment patterns on underlying loans. The present value of expected losses and expenses depends on assumptions relating to severity of claims and claim rates on current defaults, and expected defaults in future periods. These assumptions also include an estimate of expected rescission activity. Similar to our loss reserve estimates, our estimates for premium deficiency reserves could be adversely affected by several factors, including a deterioration of regional or economic conditions leading to a reduction in borrowers’ income and thus their ability to make mortgage payments, and a drop in housing values that could expose us to greater losses. Assumptions used in calculating the deficiency reserves can also be affected by volatility in the current housing and mortgage lending industries. To the extent premium patterns and actual loss experience differ from the assumptions used in calculating the premium deficiency reserves, the differences between the actual results and our estimates will affect future period earnings and could be material.
Underwriting and other expenses
Underwriting and other expenses for the first quarter of 2011 decreased when compared to the same period in 2010. The decrease reflects our lower contract underwriting volume as well as reductions in headcount.
Ratios
The table below presents our loss, expense and combined ratios for our combined insurance operations for the three months ended March 31, 2011 and 2010.
| |
|
Three Months Ended
|
|
| |
|
March 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
|
|
|
|
|
|
Loss ratio
|
|
|
107.6 |
% |
|
|
167.1 |
% |
|
Expense ratio
|
|
|
16.2 |
% |
|
|
18.4 |
% |
|
Combined ratio
|
|
|
123.8 |
% |
|
|
185.5 |
% |
The loss ratio is the ratio, expressed as a percentage, of the sum of incurred losses and loss adjustment expenses to net premiums earned. The loss ratio does not reflect any effects due to premium deficiency. The decrease in the loss ratio in the first quarter of 2011, compared to the same period in 2010, was due to a decrease in losses incurred, as well as a slight increase in premiums earned. The expense ratio is the ratio, expressed as a percentage, of underwriting expenses to net premiums written. The decrease in the expense ratio in the first quarter of 2011, compared to the same period in 2010, was due to a decrease in underwriting and other expenses of the combined insurance operations, as well as a slight increase in premiums written. The combined ratio is the sum of the loss ratio and the expense ratio.
Interest expense
Interest expense for the first quarter of 2011 increased when compared to the same period in 2010. The increase is primarily due to the issuance of our 5% Convertible Senior Notes in April 2010 as well as an increase in amortization on our junior debentures.
Income taxes
The effective tax rate provision on our pre-tax loss was 5.5% in the first quarter of 2011, compared to the effective tax rate (benefit) of (0.7%) on our pre-tax loss in the first quarter of 2010. During those periods, the benefit from income taxes was reduced by the establishment of a valuation allowance.
We review the need to establish a deferred tax asset valuation allowance on a quarterly basis. We analyze several factors, among which are the severity and frequency of operating losses, our capacity for the carryback or carryforward of any losses, the expected occurrence of future income or loss and available tax planning alternatives. Based on our analysis and the level of cumulative operating losses, we have reduced our benefit from income tax by establishing a valuation allowance.
For the three months ended March 31, 2010, our deferred tax valuation allowance was reduced by the change in the deferred tax liability related to $9.2 million of unrealized gains on investments that were recorded to other comprehensive income. For the three months ended March 31, 2011, our deferred tax valuation allowance was increased by the change in the deferred tax liability related to $26.2 million of unrealized losses on investments that were recorded to other comprehensive income. In the event of future operating losses, it is likely that the valuation allowance will be adjusted by any taxes recorded to equity for changes in unrealized gains or losses or other items in other comprehensive income.
| |
Three Months Ended
|
|
| |
March 31,
|
|
| |
2011
|
|
2010
|
|
| |
(In millions)
|
| |
|
|
|
|
|
|
|
Benefit from income taxes
|
|
$ |
(19.2 |
) |
|
$ |
(60.7 |
) |
|
Change in valuation allowance
|
|
|
21.0 |
|
|
|
59.7 |
|
| |
|
|
|
|
|
|
|
|
|
Tax provision (benefit)
|
|
$ |
1.8 |
|
|
$ |
(1.0 |
) |
The increase in the valuation allowance that was included in other comprehensive income was $9.2 million for the three months ended March 31, 2011. There was no valuation allowance within other comprehensive income for the first three months of 2010. The total valuation allowance as of March 31, 2011 and December 31, 2010 was $440.5 million and $410.3 million, respectively.
Legislation enacted in 2009 expanded the carryback period for certain net operating losses from 2 years to 5 years. A total benefit for income taxes of $282.0 million was recorded during 2009 in the consolidated statement of operations for the carryback of 2009 losses. The refund related to these benefits was received in the second quarter of 2010.
Giving full effect to the carryback of net operating losses for federal income tax purposes, we have approximately $1,331 million of net operating loss carryforwards on a regular tax basis and $505 million of net operating loss carryforwards for computing the alternative minimum tax as of March 31, 2011. Any unutilized carryforwards are scheduled to expire at the end of tax years 2029 through 2031.
Financial Condition
At March 31, 2011, based on fair value, approximately 95% of our fixed income securities and cash and cash equivalents were invested in ‘A’ rated and above, readily marketable securities, concentrated in maturities of less than 15 years. The composition of ratings at March 31, 2011, December 31, 2010 and March 31, 2010 are shown in the table below.
Investment Portfolio Ratings
| |
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
|
AAA
|
|
|
50 |
% |
|
|
51 |
% |
|
|
45 |
% |
|
AA
|
|
|
25 |
% |
|
|
25 |
% |
|
|
30 |
% |
|
A
|
|
|
20 |
% |
|
|
20 |
% |
|
|
19 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
A or better
|
|
|
95 |
% |
|
|
96 |
% |
|
|
94 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
BBB and below
|
|
|
5 |
% |
|
|
4 |
% |
|
|
6 |
% |
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
Total
|
|
|
100 |
% |
|
|
100 |
% |
|
|
100 |
% |
Approximately 13% of our investment portfolio, excluding cash and cash equivalents, is guaranteed by financial guarantors. We evaluate the credit risk of securities through analysis of the underlying fundamentals. The extent of our analysis depends on a variety of factors, including the issuer’s sector, scale, profitability, debt cover, ratings and the tenor of the investment. A breakdown of the portion of our investment portfolio covered by a financial guarantor by credit rating, including the rating without the guarantee is shown below. The ratings are provided by one or more of the following major rating agencies: Moody’s, Standard & Poor’s and Fitch Ratings.
|
At March 31, 2011
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
(In millions)
|
|
Guarantor Rating
|
|
| |
|
AA-
|
|
|
BBB
|
|
|
NR
|
|
|
|
R |
|
|
All
|
|
|
Underlying Rating:
|
|
|
|
|
AAA
|
|
$ |
- |
|
|
$ |
- |
|
|
$ |
- |
|
|
$ |
19 |
|
|
$ |
19 |
|
|
AA
|
|
|
107 |
|
|
|
241 |
|
|
|
- |
|
|
|
111 |
|
|
|
459 |
|
|
A
|
|
|
82 |
|
|
|
167 |
|
|
|
- |
|
|
|
136 |
|
|
|
385 |
|
|
BBB
|
|
|
- |
|
|
|
31 |
|
|
|
10 |
|
|
|
25 |
|
|
|
66 |
|
| |
|
$ |
189 |
|
|
$ |
439 |
|
|
$ |
10 |
|
|
$ |
291 |
|
|
$ |
929 |
|
NR – not rated
R – in regulatory receivership
At March 31, 2011, there were no fixed income securities relying on financial guaranty insurance to elevate their rating to ‘A’ and above. Any future downgrades of these financial guarantor ratings would leave the percentage of fixed income securities ‘A’ and above effectively unchanged.
We primarily place our investments in instruments that meet high credit quality standards, as specified in our investment policy guidelines. The policy guidelines also limit the amount of our credit exposure to any one issue, issuer and type of instrument. At March 31, 2011, the modified duration of our fixed income investment portfolio, including cash and cash equivalents, was 3.1 years, which means that an instantaneous parallel shift in the yield curve of 100 basis points would result in a change of 3.1% in the fair value of our fixed income portfolio. For an upward shift in the yield curve, the fair value of our portfolio would decrease and for a downward shift in the yield curve, the fair value would increase.
We held $334.3 million in auction rate securities (“ARS”) backed by student loans at March 31, 2011. ARS are intended to behave like short-term debt instruments because their interest rates are reset periodically through an auction process, most commonly at intervals of 7, 28 and 35 days. The same auction process has historically provided a means by which we may rollover the investment or sell these securities at par in order to provide us with liquidity as needed. The ARS we hold are collateralized by portfolios of student loans, substantially all of which are ultimately 97% guaranteed by the United States Department of Education. At March 31, 2011, approximately 89% of our ARS portfolio was rated AAA/Aaa by one or more of the following major rating agencies: Moody’s, Standard & Poor’s and Fitch Ratings.
In mid-February 2008, auctions began to fail due to insufficient buyers, as the amount of securities submitted for sale in auctions exceeded the aggregate amount of the bids. For each failed auction, the interest rate on the security moves to a maximum rate specified for each security, and generally resets at a level higher than specified short-term interest rate benchmarks. At March 31, 2011, our entire ARS portfolio, consisting of 32 investments, was subject to failed auctions; however, from the period when the auctions began to fail through March 31, 2011, $190.4 million in par value of ARS was either sold or called, with the average amount we received being approximately 98% of par which approximated the aggregate fair value prior to redemption. To date, we have collected all interest due on our ARS.
As a result of the persistent failed auctions, and the uncertainty of when these investments could be liquidated at par, the investment principal associated with failed auctions will not be accessible until successful auctions occur, a buyer is found outside of the auction process, the issuers establish a different form of financing to replace these securities, or final payments come due according to the contractual maturities of the debt issues. However, we continue to believe we will have liquidity to our ARS portfolio by December 31, 2014.
At March 31, 2011, our total assets included $1.1 billion of cash and cash equivalents as shown on our consolidated balance sheet.
At March 31, 2011, we had $77.4 million, 5.625% Senior Notes due in September 2011 and $300 million, 5.375% Senior Notes due in November 2015, with a combined fair value of $361.3 million, outstanding. At March 31, 2011, we also had $345 million principal amount of 5% Convertible Senior Notes outstanding due in 2017, with a fair value of $377.8 million and $389.5 million principal amount of 9% Convertible Junior Subordinated Debentures due in 2063 outstanding, which at March 31, 2011 are reflected as a liability on our consolidated balance sheet at the current amortized value of $322.3 million, with the unamortized discount reflected in equity. The fair value of the convertible debentures was approximately $404.1 million at March 31, 2011.
The Internal Revenue Service ("IRS") completed separate examinations of our federal income tax returns for the years 2000 through 2004 and 2005 through 2007 and issued assessments for unpaid taxes, interest and penalties. The primary adjustment in both examinations related to our treatment of the flow-through income and loss from an investment in a portfolio of residual interests of Real Estate Mortgage Investment Conduits ("REMICs"). This portfolio has been managed and maintained during years prior to, during and subsequent to the examination period. The IRS indicated that it did not believe that, for various reasons, we had established sufficient tax basis in the REMIC residual interests to deduct the losses from taxable income. We appealed those adjustments and, in August 2010, we reached a tentative settlement agreement with the IRS. The settlement agreement is subject to review by the Joint Committee on Taxation of Congress because net operating losses incurred in 2009 were carried back to taxable years that were included in the agreement. A final agreement is expected to be entered into when the review is complete, although we do not expect there will be any substantive change in the terms of a final agreement from those in the tentative agreement. We adjusted our tax provision and liabilities for the effects of this agreement in 2010 and believe that they accurately reflect our exposure in regard to this issue.
The IRS is currently conducting an examination of our federal income tax returns for the years 2008 and 2009, which is scheduled to be completed in 2011.
The total amount of unrecognized tax benefits as of March 31, 2011 is $109.5 million. The total amount of the unrecognized tax benefits that would affect our effective tax rate is $96.9 million. We recognize interest accrued and penalties related to unrecognized tax benefits in income taxes. We have accrued $26.1 million for the payment of interest as of March 31, 2011. Based on our tentative agreement with the IRS, we expect our total amount of unrecognized tax benefits to be reduced by $103.6 million during 2011, while after taking into account prior payments and the effect of available net operating loss carrybacks, we expect net cash outflows to equal approximately $22 million.
Our principal exposure to loss is our obligation to pay claims under MGIC’s mortgage guaranty insurance policies. At March 31, 2011, MGIC’s direct (before any reinsurance) primary and pool risk in force, which is the unpaid principal balance of insured loans as reflected in our records multiplied by the coverage percentage, and taking account of any loss limit, was approximately $50.4 billion. In addition, as part of our contract underwriting activities, we are responsible for the quality of our underwriting decisions in accordance with the terms of the contract underwriting agreements with customers. We may be required to provide certain remedies to our customers if certain standards relating to the quality of our underwriting work are not met, and we have an established reserve for such obligations. Through March 31, 2011, the cost of remedies provided by us to customers for failing to meet the standards of the contracts has not been material. However, a generally positive economic environment for residential real estate that continued until approximately 2007 may have mitigated the effect of some of these costs, and claims for remedies may be made a number of years after the underwriting work was performed. A material portion of our new insurance written through the flow channel in recent years, including for 2006 and 2007, has involved loans for which we provided contract underwriting services. We believe the rescission of mortgage insurance coverage on loans for which we provided contract underwriting services may make a claim for a contract underwriting remedy more likely to occur. Beginning in the second half of 2009, we experienced an increase in claims for contract underwriting remedies, which continued into the first quarter of 2011. Hence, there can be no assurance that contract underwriting remedies will not be material in the future.
Liquidity and Capital Resources
Overview
At March 31, 2011, our sources of funds consisted primarily of:
|
|
·
|
our investment portfolio (which is discussed in “Financial Condition” above), and interest income on the portfolio,
|
|
|
·
|
net premiums that we will receive from our existing insurance in force as well as policies that we write in the future and
|
|
|
·
|
amounts that we expect to recover from captives (which is discussed in “Results of Consolidated Operations – Risk sharing arrangements” and “Results of Consolidated Operations – Losses – Losses incurred” above).
|
At March 31, 2011, our obligations consisted primarily of:
|
|
·
|
claim payments under MGIC’s mortgage guaranty insurance policies,
|
|
|
·
|
$77.4 million of 5.625% Senior Notes due in September 2011,
|
|
|
·
|
$300 million of 5.375% Senior Notes due in November 2015,
|
|
|
·
|
$345 million of Convertible Senior Notes due in 2017,
|
|
|
·
|
$389.5 million of Convertible Junior Debentures due in 2063,
|
|
|
·
|
interest on the foregoing debt instruments, and
|
|
|
·
|
the other costs and operating expenses of our business.
|
Holders of both of the convertible issues may convert their notes into shares of our common stock at their option prior to certain dates prescribed under the terms of their issuance, in which case our corresponding obligation will be eliminated.
For the first time in many years, beginning in 2009, claim payments exceeded premiums received. We expect that this trend will continue. Due to the uncertainty regarding how certain factors, such as foreclosure moratoriums, servicing and court delays, failures by servicers to follow proper procedures in foreclosure proceedings, loan modifications and claims investigations and rescissions, will affect our future paid claims it has become even more difficult to estimate the amount and timing of future claim payments. When we experience cash shortfalls, we can fund them through sales of short-term investments and other investment portfolio securities, subject to insurance regulatory requirements regarding the payment of dividends to the extent funds were required by an entity other than the seller. In addition, we align the maturities of our investment portfolio with our estimate of future obligations. A significant portion of our investment portfolio securities are held by our insurance subsidiaries. As long as the trends discussed above continue, we expect to experience significant declines in our investment portfolio.
Debt at Our Holding Company and Holding Company Capital Resources
The senior notes, convertible senior notes and convertible debentures are obligations of MGIC Investment Corporation and not of its subsidiaries. We are a holding company and the payment of dividends from our insurance subsidiaries, which prior to raising capital in the public markets in 2008 and 2010 had been the principal source of our holding company cash inflow, is restricted by insurance regulation. MGIC is the principal source of dividend-paying capacity. In 2009, 2010 and the first quarter of 2011, MGIC has not paid any dividends to our holding company. Through 2011, MGIC cannot pay any dividends to our holding company without approval from the OCI.
At March 31, 2011, we had $887 million in cash and investments at our holding company. As of March 31, 2011, our holding company’s obligations included $77.4 million of debt which is scheduled to mature in September 2011, $300 million of Senior Notes due in November 2015 and $345 million in Convertible Senior Notes due in 2017, all of which must be serviced pending scheduled maturity. On an annual basis, as of March 31, 2011 our use of funds at the holding company for interest payments on our Senior Notes and Convertible Senior Notes approximated $38 million. As of March 31, 2011, our holding company’s obligations also include $389.5 million in Convertible Junior Debentures due in 2063 and interest on these debentures. See Note 3 – “Debt” to our consolidated financial statements for additional information about this indebtedness, including our right to defer interest on our Convertible Junior Debentures.
We may from time to time seek to acquire our debt obligations through cash purchases and/or exchanges for other securities. We may do this in open market purchases, privately negotiated acquisitions or other transactions. The amounts involved may be material.
Risk-to-Capital
We compute our risk-to-capital ratio on a separate company statutory basis, as well as for our combined insurance operations and is our net risk in force divided by our policyholders’ position. Our net risk in force includes both primary and pool risk in force, and excludes risk on policies that are currently in default and for which loss reserves have been established. The risk amount includes pools of loans or bulk deals with contractual aggregate loss limits and in some cases without these limits. Policyholders’ position consists primarily of statutory policyholders’ surplus (which increases as a result of statutory net income and decreases as a result of statutory net loss and dividends paid), plus the statutory contingency reserve. The statutory contingency reserve is reported as a liability on the statutory balance sheet. A mortgage insurance company is required to make annual contributions to the contingency reserve of approximately 50% of net earned premiums. These contributions must generally be maintained for a period of ten years. However, with regulatory approval a mortgage insurance company may make early withdrawals from the contingency reserve when incurred losses exceed 35% of net earned premium in a calendar year.
The premium deficiency reserve discussed under “Results of Consolidated Operations – Losses – Premium deficiency” above is not recorded as a liability on the statutory balance sheet and is not a component of statutory net income. The present value of expected future premiums and already established loss reserves and statutory contingency reserves, exceeds the present value of expected future losses and expenses on our total in force book, so no deficiency is recorded on a statutory basis. On a GAAP basis, contingency loss reserves are not established and thus not considered when calculating premium deficiency reserve, additionally policies are grouped based on how they are acquired, serviced and measured.
MGIC’s separate company risk-to-capital calculation appears in the table below.
| |
|
March 31,
|
|
|
December 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
(In millions, except ratio)
|
|
| |
|
|
|
|
|
|
|
Risk in force - net (1)
|
|
$ |
33,577 |
|
|
$ |
33,817 |
|
| |
|
|
|
|
|
|
|
|
|
Statutory policyholders' surplus
|
|
$ |
1,703 |
|
|
$ |
1,709 |
|
|
Statutory contingency reserve
|
|
|
- |
|
|
|
- |
|
| |
|
|
|
|
|
|
|
|
|
Statutory policyholders' position
|
|
$ |
1,703 |
|
|
$ |
1,709 |
|
| |
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
Risk-to-capital
|
|
19.7:1
|
|
|
19.8:1
|
|
(1) Risk in force – net, as shown in the table above, is net of reinsurance and exposure on policies currently in default and for which loss reserves have been established.
Our combined insurance companies’ risk-to-capital calculation appears in the table below.
| |
|
March 31,
|
|
|
December 31,
|
|
| |
|
2011
|
|
|
2010
|
|
| |
|
(In millions, except ratio)
|
|
| |
|
|
|
|
|
|
|
Risk in force - net (1)
|
|
$ |
38,976 |
|
|
$ |
39,369 |
|
| |
|
|
|
|
|
|
|
|
|
Statutory policyholders' surplus
|
|
$ |
1,684 |
|
|
$ |
1,692 |
|
|
Statutory contingency reserve
|
|
|
8 |
|
|
|
5 |
|
| |
|
|
|
|
|
|
|
|
|
Statutory policyholders' position
|
|
$ |
1,692 |
|
|
$ |
1,697 |
|
| |
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
Risk-to-capital
|
|
23.0:1
|
|
|
23.2:1
|
|
(1) Risk in force – net, as shown in the table above, is net of reinsurance and exposure on policies currently in default ($10.1 billion at March 31, 2011 and $11.0 billion at December 31, 2010) and for which loss reserves have been established.
If our statutory policyholders’ position decreases at a greater rate than our risk in force, then our risk-to-capital ratio will increase.
For additional information regarding regulatory capital see “Overview-Capital” above as well as our Risk Factor titled “Even though our plan to write new insurance in MGIC Indemnity Corporation has received approval from the Office of the Commissioner of Insurance of the State of Wisconsin (“OCI”) and the GSEs, because MGIC is not expected to meet statutory risk-to-capital requirements to write new business in various states, we cannot guarantee that the implementation of our plan will allow us to continue to write new insurance on an uninterrupted basis.”
Financial Strength Ratings
The financial strength of MGIC, our principal mortgage insurance subsidiary, is rated Ba3 by Moody’s Investors Service with a positive outlook. Standard & Poor’s Rating Services’ insurer financial strength rating of MGIC is B+ and the outlook for this rating is negative.
For further information about the importance of MGIC’s ratings, see our Risk Factor titled “MGIC may not continue to meet the GSEs’ mortgage insurer eligibility requirements”.
Contractual Obligations
At March 31, 2011, the approximate future payments under our contractual obligations of the type described in the table below are as follows:
| |
|
Payments due by period
|
|
|
Contractual Obligations (In millions):
|
|
|
|
|
Less than
|
|
|
|
|
|
|
|
|
More than
|
|
| |
|
Total
|
|
|
1 year
|
|
|
1-3 years
|
|
|
3-5 years
|
|
|
5 years
|
|
|
Long-term debt obligations
|
|
$ |
3,147 |
|
|
$ |
148 |
|
|
$ |
137 |
|
|
$ |
437 |
|
|
$ |
2,425 |
|
|
Operating lease obligations
|
|
|
6 |
|
|
|
3 |
|
|
|
2 |
|
|
|
1 |
|
|
|
- |
|
|
Tax obligations
|
|
|
17 |
|
|
|
17 |
|
|
|
- |
|
|
|
- |
|
|
|
- |
|
|
Purchase obligations
|
|
|
1 |
|
|
|
1 |
|
|
|
- |
|
|
|
- |
|
|
|
- |
|
|
Pension, SERP and other post-retirement benefit plans
|
|
|
169 |
|
|
|
10 |
|
|
|
25 |
|
|
|
32 |
|
|
|
102 |
|
|
Other long-term liabilities
|
|
|
5,471 |
|
|
|
2,626 |
|
|
|
2,462 |
|
|
|
383 |
|
|
|
- |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Total
|
|
$ |
8,811 |
|
|
$ |
2,805 |
|
|
$ |
2,626 |
|
|
$ |
853 |
|
|
$ |
2,527 |
|
Our long-term debt obligations at March 31, 2011 include our $77.4 million of 5.625% Senior Notes due in September 2011, $300 million of 5.375% Senior Notes due in November 2015, $345 million of 5% Convertible Senior Notes due in 2017 and $389.5 million in convertible debentures due in 2063, including related interest, as discussed in Note 3 – “Debt” to our consolidated financial statements and under “Liquidity and Capital Resources” above. Our operating lease obligations include operating leases on certain office space, data processing equipment and autos, as discussed in Note 19 – “Leases” to our consolidated financial statements in our Annual Report on Form 10-K for the year ended December 31, 2010. Purchase obligations consist primarily of agreements to purchase data processing hardware or services made in the normal course of business. See Note 13 - “Benefit plans” to our consolidated financial statements in our Annual Report on Form 10-K for the year ended December 31, 2010 for discussion of expected benefit payments under our benefit plans.
Our other long-term liabilities represent the loss reserves established to recognize the liability for losses and loss adjustment expenses related to defaults on insured mortgage loans. The timing of the future claim payments associated with the established loss reserves was determined primarily based on two key assumptions: the length of time it takes for a notice of default to develop into a received claim and the length of time it takes for a received claim to be ultimately paid. The future claim payment periods are estimated based on historical experience, and could emerge significantly different than this estimate. Due to the uncertainty regarding how certain factors, such as foreclosure moratoriums, servicing and court delays, failures by servicers to follow proper procedures in foreclosure proceedings, loan modifications, claims investigations and claim rescissions, will affect our future paid claims it has become even more difficult to estimate the amount and timing of future claim payments. Current conditions in the housing and mortgage industries make all of the assumptions discussed in this paragraph more volatile than they would otherwise be. See Note 12 – “Loss reserves” to our consolidated financial statements and “-Critical Accounting Policies” in our 10-K MD&A. In accordance with GAAP for the mortgage insurance industry, we establish loss reserves only for loans in default. Because our reserving method does not take account of the impact of future losses that could occur from loans that are not delinquent, our obligation for ultimate losses that we expect to occur under our policies in force at any period end is not reflected in our financial statements or in the table above.
Forward Looking Statements and Risk Factors
General: Our revenues and losses could be affected by the risk factors referred to under “Location of Risk Factors” below. These risk factors are an integral part of Management’s Discussion and Analysis.
These factors may also cause actual results to differ materially from the results contemplated by forward looking statements that we may make. Forward looking statements consist of statements which relate to matters other than historical fact. Among others, statements that include words such as we “believe,” “anticipate” or “expect,” or words of similar import, are forward looking statements. We are not undertaking any obligation to update any forward looking statements we may make even though these statements may be affected by events or circumstances occurring after the forward looking statements were made. Therefore no reader of this document should rely on these statements being current as of any time other than the time at which this document was filed with the Securities and Exchange Commission.
Location of Risk Factors: The risk factors are in Item 1 A of our Annual Report on Form 10-K for the year ended December 31, 2010, as supplemented by Part II, Item 1 A of this Quarterly Report on Form 10-Q. The risk factors in the 10-K, as supplemented by this 10-Q and through updating of various statistical and other information, are reproduced in Exhibit 99 to this Quarterly Report on Form 10-Q.
Item 3. Quantitative and Qualitative Disclosures about Market Risk
At March 31, 2011, the derivative financial instruments in our investment portfolio were immaterial. We place our investments in instruments that meet high credit quality standards, as specified in our investment policy guidelines; the policy also limits the amount of credit exposure to any one issue, issuer and type of instrument. At March 31, 2011, the modified duration of our fixed income investment portfolio was 3.1 years, which means that an instantaneous parallel shift in the yield curve of 100 basis points would result in a change of 3.1% in the market value of our fixed income portfolio. For an upward shift in the yield curve, the market value of our portfolio would decrease and for a downward shift in the yield curve, the market value would increase.
Item 4. Controls and Procedures
Our management, with the participation of our principal executive officer and principal financial officer, has evaluated our disclosure controls and procedures (as defined in Rule 13a-15(e) under the Securities Exchange Act of 1934, as amended), as of the end of the period covered by this Quarterly Report on Form 10-Q. Based on such evaluation, our principal executive officer and principal financial officer concluded that such controls and procedures were effective as of the end of such period. There was no change in our internal control over financial reporting that occurred during the first quarter of 2011 that materially affected, or is reasonably likely to materially affect, our internal control over financial reporting.
PART II. OTHER INFORMATION
Item 1. Legal Proceedings
Consumers are bringing a growing number of lawsuits against home mortgage lenders and settlement service providers. Mortgage insurers, including MGIC, have been involved in litigation alleging violations of the anti-referral fee provisions of the Real Estate Settlement Procedures Act, which is commonly known as RESPA, and the notice provisions of the Fair Credit Reporting Act, which is commonly known as FCRA. MGIC settled class action litigation against it under RESPA in October 2003. MGIC settled the named plaintiffs’ claims in litigation against it under FCRA in December 2004 following denial of class certification in June 2004. Since December 2006, class action litigation has been brought against a number of large lenders alleging that their captive mortgage reinsurance arrangements violated RESPA. On November 29, 2010, six mortgage insurers (including MGIC) and a large mortgage lender (which was the named plaintiffs’ lender) were named as defendants in a complaint, alleged to be a class action, filed in Federal District Court for the District of Columbia. The complaint alleges various causes of action related to the captive mortgage reinsurance arrangements of this mortgage lender, including that the defendants violated RESPA by paying the lender’s captive reinsurer excessive premiums in relation to the risk assumed by that captive. The named plaintiffs’ loan was not insured by MGIC and it is our understanding that it was not reinsured by this mortgage lender’s captive reinsurance affiliates. In March 2011, the complaint was voluntarily dismissed by the plaintiffs as to MGIC and all of the other mortgage insurers. There can be no assurance that we will not be subject to future litigation under RESPA (or FCRA) or that the outcome of any such litigation would not have a material adverse effect on us.
In addition to the above litigation, we face other litigation and regulatory risks. For additional information about such other litigation and regulatory risks you should review our Risk Factor titled “We are subject to the risk of private litigation and regulatory proceedings.”
Item 1 A. Risk Factors
With the exception of the changes described and set forth below, there have been no material changes in our risk factors from the risk factors disclosed in the Company’s Annual Report on Form 10-K for the year ended December 31, 2010. The risk factors in the 10-K, as supplemented by this 10-Q and through updating of various statistical and other information, are reproduced in their entirety in Exhibit 99 to this Quarterly Report on Form 10-Q.
The amount of insurance we write could be adversely affected if the definition of Qualified Residential Mortgage results in a reduction of the number of low down payment loans available to be insured or if lenders and investors select alternatives to private mortgage insurance.
The financial reform legislation that was passed in July 2010 (the “Dodd-Frank Act” or “Dodd-Frank”) requires a securitizer to retain 5% of the risk associated with mortgage loans that are securitized, and in some cases the retained risk may be allocated between the securitizer and the lender that originated the loan, all as specified by regulations to be adopted under Dodd-Frank by various federal financial institutions regulators. This risk retention requirement does not apply to mortgage loans that are Qualified Residential Mortgages (“QRMs”) or that are insured by the FHA or another federal agency. In March 2011, federal regulators issued the proposed risk retention rule that includes a definition of QRM. The proposed definition of QRM allows a maximum loan-to-value ratio (“LTV”) of 80% on a home purchase transaction. The LTV is to be calculated without including mortgage insurance. Public comments to the proposed rule are due June 10, 2011. The regulators requested that the public comments include information that may be used to assess whether loans with mortgage insurance are less likely to default than other loans and they requested public comments regarding the possibility of expanding the QRM definition to include loans with 90% LTVs that have mortgage insurance. Under the proposed rule, because of the capital support provided by the U.S. Government, the GSEs satisfy the Dodd-Frank risk-retention requirements while they are in conservatorship. Therefore, lenders that originate loans that are sold to the GSEs while they are in conservatorship will not be required to retain risk associated with those loans. Depending on the maximum LTV allowed in the final definition of QRM, to what extent, if any, the presence of mortgage insurance would allow for a higher LTV in the definition of QRM, and whether lenders choose mortgage insurance for non-QRM loans, the amount of new insurance that we write may be materially adversely affected. See also “— If the volume of low down payment home mortgage originations declines, the amount of insurance that we write could decline, which would reduce our revenues.” The following table shows the percentage of our new risk written by LTV for the first quarter of 2011 and for the year ended December 31, 2010.
| |
|
Percentage of new risk written
|
|
| |
|
1Q
|
|
|
Full Year
|
|
| |
|
2011
|
|
|
2010
|
|
|
LTV:
|
|
|
|
|
|
|
|
85% and under
|
|
7%
|
|
|
7%
|
|
|
85.1% - 90%
|
|
44%
|
|
|
48%
|
|
|
90.1% - 95%
|
|
48%
|
|
|
44%
|
|
|
95.1% - 97%
|
|
1%
|
|
|
1%
|
|
|
> 97%
|
|
0%
|
|
|
0%
|
|
Alternatives to private mortgage insurance include:
|
|
·
|
lenders using government mortgage insurance programs, including those of the Federal Housing Administration, or FHA, and the Veterans Administration,
|
|
|
·
|
lenders and other investors holding mortgages in portfolio and self-insuring,
|
|
|
·
|
investors using credit enhancements other than private mortgage insurance, using other credit enhancements in conjunction with reduced levels of private mortgage insurance coverage, or accepting credit risk without credit enhancement, and
|
|
|
·
|
lenders originating mortgages using piggyback structures to avoid private mortgage insurance, such as a first mortgage with an 80% loan-to-value ratio and a second mortgage with a 10%, 15% or 20% loan-to-value ratio (referred to as 80-10-10, 80-15-5 or 80-20 loans, respectively) rather than a first mortgage with a 90%, 95% or 100% loan-to-value ratio that has private mortgage insurance.
|
The FHA substantially increased its market share beginning in 2008. We believe that the FHA’s market share increased, in part, because private mortgage insurers tightened their underwriting guidelines (which led to increased utilization of the FHA’s programs) and because of increases in the amount of loan level delivery fees that the GSEs assess on loans (which result in higher costs to borrowers). In addition, federal legislation and programs provided the FHA with greater flexibility in establishing new products and increased the FHA’s competitive position against private mortgage insurers. However, the FHA’s 2010 and 2011 pricing, when compared to our credit-tiered pricing introduced in 2010 (and considering the effects of GSE pricing changes), may allow us to be more competitive with the FHA than in the recent past for loans with high FICO credit scores. We cannot predict, however, what impact these premium changes will have on new insurance written in the future.
Changes in the business practices of the GSEs, federal legislation that changes their charters or a restructuring of the GSEs could reduce our revenues or increase our losses.
The majority of our insurance written is for loans sold to Fannie Mae and Freddie Mac. The business practices of the GSEs affect the entire relationship between them, lenders and mortgage insurers and include:
|
|
·
|
the level of private mortgage insurance coverage, subject to the limitations of the GSEs’ charters (which may be changed by federal legislation), when private mortgage insurance is used as the required credit enhancement on low down payment mortgages,
|
|
|
·
|
the amount of loan level delivery fees (which result in higher costs to borrowers) that the GSEs assess on loans that require mortgage insurance,
|
|
|
·
|
whether the GSEs influence the mortgage lender’s selection of the mortgage insurer providing coverage and, if so, any transactions that are related to that selection,
|
|
|
·
|
the underwriting standards that determine what loans are eligible for purchase by the GSEs, which can affect the quality of the risk insured by the mortgage insurer and the availability of mortgage loans,
|
|
|
·
|
the terms on which mortgage insurance coverage can be canceled before reaching the cancellation thresholds established by law,
|
|
|
·
|
the programs established by the GSEs intended to avoid or mitigate loss on insured mortgages and the circumstances in which mortgage servicers must implement such programs, and
|
|
|
·
|
the extent to which the GSEs intervene in mortgage insurers’ rescission practices or rescission settlement practices with lenders. Freddie Mac recently advised its servicers that they must obtain its prior approval for related settlements and Fannie Mae recently advised its servicers that they are prohibited from entering into such settlements. In addition, Fannie Mae recently notified us that we must obtain its prior approval to enter into certain settlements.
|
In September 2008, the Federal Housing Finance Agency (“FHFA”) was appointed as the conservator of the GSEs. As their conservator, FHFA controls and directs the operations of the GSEs. The appointment of FHFA as conservator, the increasing role that the federal government has assumed in the residential mortgage market, our industry’s inability, due to capital constraints, to write sufficient business to meet the needs of the GSEs or other factors may increase the likelihood that the business practices of the GSEs change in ways that may have a material adverse effect on us. In addition, these factors may increase the likelihood that the charters of the GSEs are changed by new federal legislation. The Dodd-Frank Act required the U.S. Department of the Treasury to report its recommendations regarding options for ending the conservatorship of the GSEs. This report was released on February 11, 2011 and while it does not provide any definitive timeline for GSE reform, it does recommend using a combination of federal housing policy changes to wind down the GSEs, shrink the government’s footprint in housing finance, and help bring private capital back to the mortgage market. Members of the House of Representatives and the Senate have since introduced several bills intended to scale back the GSEs. The bills include proposals to abolish the GSEs’ affordable housing goals, reduce the conforming loan limits, increase guarantee fees and set annual limits on the size of each GSE’s retained portfolio. As a result of the matters referred to above, it is uncertain what role the GSEs, FHA and private capital, including private mortgage insurance, will play in the domestic residential housing finance system in the future or the impact of any such changes on our business. In addition, the timing of the impact on our business is uncertain. Any changes would require Congressional action to implement and it is difficult to estimate when Congressional action would be final and how long any associated phase-in period may last.
The GSEs have different loan purchase programs that allow different levels of mortgage insurance coverage. Under the “charter coverage” program, on certain loans lenders may choose a mortgage insurance coverage percentage that is less than the GSEs’ “standard coverage” and only the minimum required by the GSEs’ charters, with the GSEs paying a lower price for such loans. During 2010 and the first quarter of 2011, more than 90% of our volume was on loans with GSE standard coverage. We charge higher premium rates for higher coverage percentages. To the extent lenders selling loans to GSEs in the future choose charter coverage for loans that we insure, our revenues would be reduced and we could experience other adverse effects. The pricing changes we implemented in 2010 (see “— The premiums we charge may not be adequate to compensate us for our liabilities for losses and as a result any inadequacy could materially affect our financial condition and results of operations”) may eliminate a lender’s incentive to use GSE charter coverage in place of standard coverage.
Both of the GSEs have guidelines on terms under which they can conduct business with mortgage insurers, such as MGIC, with financial strength ratings below Aa3/AA-. (MGIC’s financial strength rating from Moody’s is Ba3, with a positive outlook and from Standard & Poor’s is B+, with a negative outlook.) For information about how these guidelines could affect us, see “— MGIC may not continue to meet the GSEs’ mortgage insurer eligibility requirements.”
We may not continue to realize benefits from rescissions at the rates we have recently experienced and we may not prevail in proceedings challenging whether our rescissions were proper.
Historically, rescissions of policies for which claims have been submitted to us were not a material portion of our claims resolved during a year. However, beginning in 2008, our rescissions of policies have materially mitigated our paid losses. In each of 2009 and 2010, rescissions mitigated our paid losses by approximately $1.2 billion and in the first quarter of 2011, rescissions mitigated our paid losses by approximately $0.2 billion (in each case, the figure includes amounts that would have either resulted in a claim payment or been charged to a deductible under a bulk or pool policy, and may have been charged to a captive reinsurer). While we have a substantial pipeline of claims investigations that we expect will eventually result in future rescissions, we expect that rescissions will not continue at the same rates (as a percentage of claims received) we have previously experienced.
In addition, our loss reserving methodology incorporates the effects we expect rescission activity to have on the losses we will pay on our delinquent inventory. A variance between ultimate actual rescission rates and these estimates, as a result of the outcome of claims investigations, litigation, settlements or other factors, could materially affect our losses. See “—Because loss reserve estimates are subject to uncertainties and are based on assumptions that are currently very volatile, paid claims may be substantially different than our loss reserves.” We estimate rescissions mitigated our incurred losses by approximately $2.5 billion in 2009 and $0.2 billion in 2010. For the first quarter of 2011, we estimate that rescissions had no material impact on our losses incurred. All of these figures include the benefit of claims not paid in the period as well as the impact of changes in our estimated expected rescission activity on our loss reserves in the period. In recent quarters, between 20% and 28% of claims received in a quarter have been resolved by rescissions. At March 31, 2011, we had 195,885 loans in our primary delinquency inventory; the resolution of a significant portion of these loans will not involve paid claims.
If the insured disputes our right to rescind coverage, the outcome of the dispute ultimately would be determined by legal proceedings. Legal proceedings disputing our right to rescind coverage may be brought up to three years after the lender has obtained title to the property (typically through a foreclosure) or the property was sold in a sale that we approved, whichever is applicable, although in a few jurisdictions there is a longer time to bring such an action. For nearly all of our rescissions that are not subject to a settlement agreement, the period in which a dispute may be brought has not ended. We consider a rescission resolved for reporting purposes even though legal proceedings have been initiated and are ongoing. Although it is reasonably possible that, when the proceedings are completed, there will be a determination that we were not entitled to rescind in all cases, we are unable to make a reasonable estimate or range of estimates of the potential liability. Under Accounting Standards Codification (“ASC”) 450-20, an estimated loss from such proceedings is accrued for only if we determine that the loss is probable and can be reasonably estimated. Therefore, when establishing our loss reserves, we do not include additional loss reserves that would reflect an adverse outcome from ongoing legal proceedings, including those with Countrywide. Countrywide has filed a lawsuit against MGIC alleging that MGIC has denied, and continues to deny, valid mortgage insurance claims. MGIC has filed an arbitration case against Countrywide regarding rescissions and Countrywide has responded seeking damages, including exemplary damages. For more information about this lawsuit and arbitration case, see “— We are subject to the risk of private litigation and regulatory proceedings” as well as Item 3, “Legal Proceedings,” in our Annual Report on Form 10-K for the year ended December 31, 2010.
In 2010, we entered into a settlement agreement with a lender-customer regarding our rescission practices and we may enter into additional settlement agreements with other lenders in the future. For information regarding GSE restrictions on such settlement agreements, see “— Changes in the business practices of the GSEs, federal legislation that changes their charters or a restructuring of the GSEs could reduce our revenues or increase our losses.”
We continue to discuss with other lenders their objections to material rescissions. In addition to the proceedings involving Countrywide, we are involved in legal proceedings with respect to rescissions that we do not consider to be collectively material in amount. Although it is reasonably possible that, when these discussions or proceedings are completed, there will be a conclusion or determination that we were not entitled to rescind in all cases, we are unable to make a reasonable estimate or range of estimates of the potential liability.
We are subject to the risk of private litigation and regulatory proceedings.
Consumers are bringing a growing number of lawsuits against home mortgage lenders and settlement service providers. Mortgage insurers, including MGIC, have been involved in litigation alleging violations of the anti-referral fee provisions of the Real Estate Settlement Procedures Act, which is commonly known as RESPA, and the notice provisions of the Fair Credit Reporting Act, which is commonly known as FCRA. MGIC settled class action litigation against it under RESPA in October 2003. MGIC settled the named plaintiffs’ claims in litigation against it under FCRA in December 2004 following denial of class certification in June 2004. Since December 2006, class action litigation has been brought against a number of large lenders alleging that their captive mortgage reinsurance arrangements violated RESPA. On November 29, 2010, six mortgage insurers (including MGIC) and a large mortgage lender (which was the named plaintiffs’ lender) were named as defendants in a complaint, alleged to be a class action, filed in Federal District Court for the District of Columbia. The complaint alleges various causes of action related to the captive mortgage reinsurance arrangements of this mortgage lender, including that the defendants violated RESPA by paying the lender’s captive reinsurer excessive premiums in relation to the risk assumed by that captive. The named plaintiffs’ loan was not insured by MGIC and it is our understanding that it was not reinsured by this mortgage lender’s captive reinsurance affiliates. In March 2011, the complaint was voluntarily dismissed by the plaintiffs as to MGIC and all of the other mortgage insurers. There can be no assurance that we will not be subject to future litigation under RESPA (or FCRA) or that the outcome of any such litigation would not have a material adverse effect on us.
We are subject to comprehensive, detailed regulation by state insurance departments. These regulations are principally designed for the protection of our insured policyholders, rather than for the benefit of investors. Although their scope varies, state insurance laws generally grant broad supervisory powers to agencies or officials to examine insurance companies and enforce rules or exercise discretion affecting almost every significant aspect of the insurance business. Given the recent significant losses incurred by many insurers in the mortgage and financial guaranty industries, our insurance subsidiaries have been subject to heightened scrutiny by insurance regulators. State insurance regulatory authorities could take actions, including changes in capital requirements or termination of waivers of capital requirements, that could have a material adverse effect on us. In addition, the Dodd-Frank Act establishes the Bureau of Consumer Financial Protection to regulate the offering and provision of consumer financial products or services under federal law. We are uncertain whether this Bureau will issue any rules or regulations that affect our business. Such rules and regulations could have a material adverse effect on us.
In June 2005, in response to a letter from the New York Insurance Department, we provided information regarding captive mortgage reinsurance arrangements and other types of arrangements in which lenders receive compensation. In February 2006, the New York Insurance Department requested MGIC to review its premium rates in New York and to file adjusted rates based on recent years’ experience or to explain why such experience would not alter rates. In March 2006, MGIC advised the New York Insurance Department that it believes its premium rates are reasonable and that, given the nature of mortgage insurance risk, premium rates should not be determined only by the experience of recent years. In February 2006, in response to an administrative subpoena from the Minnesota Department of Commerce (the “MN Department”), which regulates insurance, we provided the MN Department with information about captive mortgage reinsurance and certain other matters. We subsequently provided additional information to the MN Department, and beginning in March 2008 the MN Department has sought additional information as well as answers to questions regarding captive mortgage reinsurance on several occasions. In addition, beginning in June 2008, we have received subpoenas from the Department of Housing and Urban Development, commonly referred to as HUD, seeking information about captive mortgage reinsurance similar to that requested by the MN Department, but not limited in scope to the state of Minnesota. Other insurance departments or other officials, including attorneys general, may also seek information about or investigate captive mortgage reinsurance.
The anti-referral fee provisions of RESPA provide that HUD as well as the insurance commissioner or attorney general of any state may bring an action to enjoin violations of these provisions of RESPA. The insurance law provisions of many states prohibit paying for the referral of insurance business and provide various mechanisms to enforce this prohibition. While we believe our captive reinsurance arrangements are in conformity with applicable laws and regulations, it is not possible to predict the outcome of any such reviews or investigations nor is it possible to predict their effect on us or the mortgage insurance industry.
Five previously-filed purported class action complaints filed against us and several of our executive officers were consolidated in March 2009 in the United States District Court for the Eastern District of Wisconsin and Fulton County Employees’ Retirement System was appointed as the lead plaintiff. The lead plaintiff filed a Consolidated Class Action Complaint (the “Complaint”) on June 22, 2009. Due in part to its length and structure, it is difficult to summarize briefly the allegations in the Complaint but it appears the allegations are that we and our officers named in the Complaint violated the federal securities laws by misrepresenting or failing to disclose material information about (i) loss development in our insurance in force, and (ii) C-BASS, including its liquidity. Our motion to dismiss the Complaint was granted on February 18, 2010. On March 18, 2010, plaintiffs filed a motion for leave to file an amended complaint. Attached to this motion was a proposed Amended Complaint (the “Amended Complaint”). The Amended Complaint alleged that we and two of our officers named in the Amended Complaint violated the federal securities laws by misrepresenting or failing to disclose material information about C-BASS, including its liquidity, and by failing to properly account for our investment in C-BASS. The Amended Complaint also named two officers of C-BASS with respect to the Amended Complaint’s allegations regarding C-BASS. The purported class period covered by the Amended Complaint began on February 6, 2007 and ended on August 13, 2007. The Amended Complaint sought damages based on purchases of our stock during this time period at prices that were allegedly inflated as a result of the purported violations of federal securities laws. On December 8, 2010, the plaintiffs’ motion to file an amended complaint was denied and the Complaint was dismissed with prejudice. On January 6, 2011, the plaintiffs appealed the February 18, 2010 and December 8, 2010 decisions to the United States Court of Appeals for the Seventh Circuit. We are unable to predict the outcome of these consolidated cases or estimate our associated expenses or possible losses. Other lawsuits alleging violations of the securities laws could be brought against us.
Several law firms have issued press releases to the effect that they are investigating us, including whether the fiduciaries of our 401(k) plan breached their fiduciary duties regarding the plan’s investment in or holding of our common stock or whether we breached other legal or fiduciary obligations to our shareholders. We intend to defend vigorously any proceedings that may result from these investigations.
With limited exceptions, our bylaws provide that our officers and 401(k) plan fiduciaries are entitled to indemnification from us for claims against them.
On December 17, 2009, Countrywide filed a complaint for declaratory relief in the Superior Court of the State of California in San Francisco (the “California State Court”) against MGIC. This complaint alleges that MGIC has denied, and continues to deny, valid mortgage insurance claims submitted by Countrywide and says it seeks declaratory relief regarding the proper interpretation of the insurance policies at issue. On January 19, 2010, we removed this case to the United States District Court for the Northern District of California (the “District Court”). On March 30, 2010, the District Court ordered the case remanded to the California State Court. We have appealed this decision to the United States Court of Appeals for the Ninth Circuit (the “Court of Appeals”) and asked the Court of Appeals to vacate the remand and stay proceedings in the District Court. On May 17, 2010, the Court of Appeals denied a stay of the District Court’s remand order. On May 28, 2010, Countrywide filed an amended complaint substantially similar to the original complaint in the California State Court. On July 2, 2010, we filed a petition in the California State Court to compel arbitration and stay the litigation in that court. On August 26, 2010, Countrywide filed an opposition to our petition. Countrywide’s opposition states that there are thousands of loans for which it disputes MGIC’s interpretation of the flow insurance policies at issue. On September 16, 2010, we filed a reply to Countrywide’s opposition. On October 1, 2010, the California State Court stayed the litigation in that court pending a final ruling on our appeal.
In connection with the Countrywide dispute discussed above, on February 24, 2010, we commenced an arbitration action against Countrywide seeking a determination that MGIC was entitled to deny and/or rescind coverage on the loans involved in the arbitration action, which were insured through the flow channel and numbered more than 1,400 loans as of the filing of the action. On March 16, 2010, Countrywide filed a response to our arbitration action objecting to the arbitrator’s jurisdiction in view of the case initiated by Countrywide in the California State Court and asserting various defenses to the relief sought by MGIC in the arbitration. On December 20, 2010, we filed an amended demand in the arbitration proceeding. This amended demand increased the number of loans for which we denied and/or rescinded coverage and which were insured through the flow channel to more than 3,300. We continue to rescind insurance coverage on additional Countrywide loans. On December 20, 2010 Countrywide filed an amended response. In the amended response, Countrywide is seeking relief for rescissions on loans insured by MGIC through the flow channel and more than 30 bulk insurance policies. In April 2011, Countrywide indicated that it believes MGIC has improperly rescinded coverage on more than 5,000 loans. The amended response also seeks damages as a result of purported breaches of insurance policies issued by MGIC and additional damages, including exemplary damages, on account of MGIC’s purported breach of an implied covenant of good faith and fair dealing. The amended response states that Countrywide seeks damages “well-exceeding” $150 million; the original response sought damages of at least $150 million. On January 17, 2011, Countrywide filed an answer to MGIC’s amended demand and MGIC filed an answer to Countrywide’s amended response. Countrywide and MGIC have each selected 12 loans for which a three-member arbitration panel will determine coverage. While the panel’s determination will not be binding on the other loans at issue, the panel will identify the issues for these 24 “bellwether” loans and strive to set forth findings of fact and conclusions of law in such a way as to aid the parties to apply them to the other loans at issue. The hearing before the panel on the bellwether loans that had previously been scheduled to begin in October 2011 has been postponed to May 2012.
From January 1, 2008 through March 31, 2011, rescissions of Countrywide-related loans mitigated our paid losses on the order of $360 million. This amount is the amount we estimate we would have paid had the loans not been rescinded. On a per loan basis, the average amount that we would have paid had the loans not been rescinded was approximately $72 thousand. At March 31, 2011, 41,696 loans in our primary delinquency inventory were Countrywide-related loans (approximately 21% of our primary delinquency inventory). Of these 41,696 loans, some will cure their delinquency and the remainder will either become paid claims or will be rescinded. From January 1, 2008 through March 31, 2011, of the claims on Countrywide-related loans that were resolved (a claim is resolved when it is paid or rescinded; claims that are submitted but which are under review are not resolved until one of these two outcomes occurs), approximately 72% were paid and the remaining 28% were rescinded.
The flow policies at issue with Countrywide are in the same form as the flow policies that we use with all of our customers, and the bulk policies at issue vary from one another, but are generally similar to those used in the majority of our Wall Street bulk transactions. Because our rescission practices with Countrywide do not differ from our practices with other servicers with which we have not entered into settlement agreements, an adverse result in the Countrywide proceeding may adversely affect the ultimate result of rescissions involving other servicers and lenders. From January 1, 2008 through March 31, 2011, we estimate that total rescissions mitigated our incurred losses by approximately $3.1 billion, which included approximately $2.2 billion of mitigation on paid losses, excluding amounts that would have been applied to a deductible. At March 31, 2011, we estimate that our total loss reserves were benefited from rescissions by approximately $1.1 billion.
We intend to defend MGIC against Countrywide’s complaint and arbitration response, and to pursue MGIC’s claims in the arbitration, vigorously. However, we are unable to predict the outcome of these proceedings or their effect on us. Also, although it is reasonably possible that, when the proceedings are completed, there will be a determination that we were not entitled to rescind in all cases, we are unable to make a reasonable estimate or range of estimates of the potential liability. Under ASC 450-20, an estimated loss is accrued for only if we determine that the loss is probable and can be reasonably estimated. Therefore, we have not accrued any reserves that would reflect an adverse outcome in this proceeding. An accrual for an adverse outcome in this (or any other) proceeding would be a reduction to our capital. In this regard, see “— Even though our plan to write new insurance in MGIC Indemnity Corporation (“MIC”) has received approval from the Office of the Commissioner of Insurance of the State of Wisconsin (“OCI”) and the GSEs, we cannot guarantee that the implementation of our plan will allow us to continue to write new insurance on an uninterrupted basis.”
In addition to the rescissions at issue with Countrywide, we have a substantial pipeline of claims investigations (including investigations involving loans related to Countrywide) that we expect will eventually result in future rescissions. In 2010, we entered into a settlement agreement with a lender-customer regarding our rescission practices. We continue to discuss with other lenders their objections to material rescissions. In addition to the proceedings involving Countrywide, we are involved in legal proceedings with respect to rescissions that we do not consider to be collectively material in amount. For additional information about rescissions as well as the settlement referred to above, see “— We may not continue to realize benefits from rescissions at the rates we have recently experienced and we may not prevail in proceedings challenging whether our rescissions were proper.”
In addition to the matters described above, we are involved in other legal proceedings in the ordinary course of business. In our opinion, based on the facts known at this time, the ultimate resolution of these ordinary course legal proceedings will not have a material adverse effect on our financial position or results of operations.
Loan modification and other similar programs may not continue to provide material benefits to us and our losses on loans that re-default can be higher than what we would have paid had the loan not been modified.
Beginning in the fourth quarter of 2008, the federal government, including through the Federal Deposit Insurance Corporation (the “FDIC”) and the GSEs, and several lenders have adopted programs to modify loans to make them more affordable to borrowers with the goal of reducing the number of foreclosures. During 2010 and the first quarter of 2011, we were notified of modifications that cured delinquencies that had they become paid claims would have resulted in approximately $3.2 billion and $0.5 billion, respectively, of estimated claim payments. As noted below, we cannot predict with a high degree of confidence what the ultimate re-default rate will be. For internal reporting purposes, we assume approximately 50% of those modifications will ultimately re-default, and those re-defaults may result in future claim payments. Because modifications cure the defaults with respect to the previously defaulted loans, our loss reserves do not account for potential re-defaults unless at the time the reserve is established, the re-default has already occurred. Based on information that is provided to us, most of the modifications resulted in reduced payments from interest rate and/or amortization period adjustments; less than 5% resulted in principal forgiveness.
One loan modification program is the Home Affordable Modification Program (“HAMP”). Some of HAMP’s eligibility criteria relate to the borrower’s current income and non-mortgage debt payments. Because the GSEs and servicers do not share such information with us, we cannot determine with certainty the number of loans in our delinquent inventory that are eligible to participate in HAMP. We believe that it could take several months from the time a borrower has made all of the payments during HAMP’s three month “trial modification” period for the loan to be reported to us as a cured delinquency.
We rely on information provided to us by the GSEs and servicers. We do not receive all of the information from such sources that is required to determine with certainty the number of loans that are participating in, or have successfully completed, HAMP. We are aware of approximately 15,400 loans in our primary delinquent inventory at March 31, 2011 for which the HAMP trial period has begun and which trial periods have not been reported to us as completed or cancelled. Through March 31, 2011 approximately 27,700 delinquent primary loans have cured their delinquency after entering HAMP and are not in default. We believe that we have realized the majority of the benefits from HAMP because the number of loans insured by us that we are aware are entering HAMP trial modification periods has decreased significantly over time.
The effect on us of loan modifications depends on how many modified loans subsequently re-default, which in turn can be affected by changes in housing values. Re-defaults can result in losses for us that could be greater than we would have paid had the loan not been modified. At this point, we cannot predict with a high degree of confidence what the ultimate re-default rate will be. In addition, because we do not have information in our database for all of the parameters used to determine which loans are eligible for modification programs, our estimates of the number of loans qualifying for modification programs are inherently uncertain. If legislation is enacted to permit a portion of a borrower’s mortgage loan balance to be reduced in bankruptcy and if the borrower re-defaults after such reduction, then the amount we would be responsible to cover would be calculated after adding back the reduction. Unless a lender has obtained our prior approval, if a borrower’s mortgage loan balance is reduced outside the bankruptcy context, including in association with a loan modification, and if the borrower re-defaults after such reduction, then under the terms of our policy the amount we would be responsible to cover would be calculated net of the reduction.
Eligibility under loan modification programs can also adversely affect us by creating an incentive for borrowers who are able to make their mortgage payments to become delinquent in an attempt to obtain the benefits of a modification. New notices of delinquency increase our incurred losses.
Competition or changes in our relationships with our customers could reduce our revenues or increase our losses.
In recent years, the level of competition within the private mortgage insurance industry has been intense as many large mortgage lenders reduced the number of private mortgage insurers with whom they do business. At the same time, consolidation among mortgage lenders has increased the share of the mortgage lending market held by large lenders. During 2010 and the first quarter of 2011, approximately 11% and 9%, respectively, of our new insurance written was for loans for which one lender was the original insured, although revenue from such loans was significantly less than 10% of our revenues during each of those periods. Our private mortgage insurance competitors include:
|
|
·
|
PMI Mortgage Insurance Company,
|
|
|
·
|
Genworth Mortgage Insurance Corporation,
|
|
|
·
|
United Guaranty Residential Insurance Company,
|
|
|
·
|
Republic Mortgage Insurance Company, whose parent, based on information filed with the SEC through May 9, 2011, is our largest shareholder,
|
|
|
·
|
CMG Mortgage Insurance Company, and
|
Until recently, the mortgage insurance industry had not had new entrants in many years. Recently, Essent Guaranty, Inc. announced that it began writing new mortgage insurance. Essent has publicly reported that one of its investors is JPMorgan Chase which is one of our customers. The perceived increase in credit quality of loans that are being insured today combined with the deterioration of the financial strength ratings of the existing mortgage insurance companies could encourage new entrants. We understand that one potential new entrant has advertised for employees. The FHA, which in recent years was not viewed by us as a significant competitor, substantially increased its market share beginning in 2008.
Our relationships with our customers could be adversely affected by a variety of factors, including tightening of and adherence to our underwriting guidelines, which have resulted in our declining to insure some of the loans originated by our customers and rescission of loans that affect the customer. We have ongoing discussions with lenders who are significant customers regarding their objections to our rescissions. In the fourth quarter of 2009, Countrywide commenced litigation against us as a result of its dissatisfaction with our rescission practices shortly after Countrywide ceased doing business with us. See “— We are subject to the risk of private litigation and regulatory proceedings” for more information about this litigation and the arbitration case we filed against Countrywide regarding rescissions. Countrywide and its Bank of America affiliates were the original insured for 12.0% of our flow new insurance written in 2008 and 8.3% of our new insurance written in the first three quarters of 2009. Bank of America recently informed us that it intends, at some point in the future, to implement procedures to enable them to cancel MGIC’s coverage on loans Bank of America purchases from correspondent lenders and substitute coverage from our competitors. In general, a correspondent lender funds loans and then sells them servicing-released to another lender who retains the servicing and either sells the loans to an investor or retains them in portfolio. Traditionally a correspondent lender’s selection of which mortgage insurer insures the loans it funds has not been changed by the lender to which those loans were sold. We estimate that during 2010 approximately 10% of our new insurance written was for loans purchased by Bank of America from correspondent lenders. The effect of Bank of America’s actions on MGIC may depend on the reaction of correspondent lenders and any reaction from the GSEs, as well as other factors. While we will be taking various actions to seek to retain this business, we cannot predict the extent to which Bank of America’s actions will adversely affect us.
We believe some lenders assess a mortgage insurer’s financial strength rating as an important element of the process through which they select mortgage insurers. MGIC’s financial strength rating from Moody’s is Ba3 with a positive outlook and from Standard & Poor’s is B+ with a negative outlook. It is possible that MGIC’s financial strength ratings could decline from these levels. As a result of MGIC’s less than investment grade financial strength rating, MGIC may be competitively disadvantaged with these lenders.
The mix of business we write also affects the likelihood of losses occurring.
Even when housing values are stable or rising, mortgages with certain characteristics have higher probabilities of claims. These characteristics include loans with loan-to-value ratios over 95% (or in certain markets that have experienced declining housing values, over 90%), FICO credit scores below 620, limited underwriting, including limited borrower documentation, or higher total debt-to-income ratios, as well as loans having combinations of higher risk factors. As of March 31, 2011, approximately 26.8% of our primary risk in force consisted of loans with loan-to-value ratios greater than 95%, 8.6% had FICO credit scores below 620, and 11.1% had limited underwriting, including limited borrower documentation, each attribute as determined at the time of loan origination. A material portion of these loans were written in 2005 — 2007 or the first quarter of 2008. In accordance with industry practice, loans approved by GSEs and other automated underwriting systems under “doc waiver” programs that do not require verification of borrower income are classified by us as “full documentation.” For additional information about such loans, see footnote (3) to the composition of primary default inventory table under "Results of Consolidated Operations-Losses-Losses incurred in Management’s Discussion and Analysis of Financial Condition and Results of Operations."
Beginning in the fourth quarter of 2007 we made a series of changes to our underwriting guidelines in an effort to improve the risk profile of our new business. From time to time, in response to market conditions, we change the types of loans that we insure and the guidelines under which we insure them. In addition, we make exceptions to our underwriting guidelines on a loan-by-loan basis and for certain customer programs. Together these exceptions accounted for fewer than 5% of the loans we insured in the second half of 2010 and fewer than 6% of the loans we insured in the first quarter of 2011. A large percentage of the exceptions were made for loans with debt-to-income ratios slightly above our guideline. Beginning in September 2009, we have made changes to our underwriting guidelines that have allowed certain loans to be eligible for insurance that were not eligible prior to those changes and we expect to continue to make changes in appropriate circumstances in the future. Our underwriting guidelines are available on our website at http://www.mgic.com/guides/underwriting.html.
As of March 31, 2011, approximately 2.8% of our primary risk in force written through the flow channel, and 35.7% of our primary risk in force written through the bulk channel, consisted of adjustable rate mortgages in which the initial interest rate may be adjusted during the five years after the mortgage closing (“ARMs”). We classify as fixed rate loans adjustable rate mortgages in which the initial interest rate is fixed during the five years after the mortgage closing. We believe that when the reset interest rate significantly exceeds the interest rate at loan origination, claims on ARMs would be substantially higher than for fixed rate loans. Moreover, even if interest rates remain unchanged, claims on ARMs with a “teaser rate” (an initial interest rate that does not fully reflect the index which determines subsequent rates) may also be substantially higher because of the increase in the mortgage payment that will occur when the fully indexed rate becomes effective. In addition, we have insured “interest-only” loans, which may also be ARMs, and loans with negative amortization features, such as pay option ARMs. We believe claim rates on these loans will be substantially higher than on loans without scheduled payment increases that are made to borrowers of comparable credit quality.
Although we attempt to incorporate these higher expected claim rates into our underwriting and pricing models, there can be no assurance that the premiums earned and the associated investment income will be adequate to compensate for actual losses even under our current underwriting guidelines. We do, however, believe that given the various changes in our underwriting guidelines that were effective beginning in the first quarter of 2008, our insurance written beginning in the second quarter of 2008 will generate underwriting profits.
Item 6. Exhibits
The accompanying Index to Exhibits is incorporated by reference in answer to this portion of this Item, and except as otherwise indicated in the next sentence, the Exhibits listed in such Index are filed as part of this Form 10-Q. Exhibit 32 is not filed as part of this Form 10-Q but accompanies this Form 10-Q.
Pursuant to the requirements of the Securities Exchange Act of 1934, the registrant has duly caused this report to be signed on its behalf by the undersigned thereunto duly authorized, on May 10, 2011.
| |
|
MGIC INVESTMENT CORPORATION
|
|
|
|
|
\s\ J. Michael Lauer |
|
| |
|
J. Michael Lauer |
|
| |
|
Executive Vice President and |
|
| |
|
Chief Financial Officer |
|
| |
|
|
|
|
|
|
\s\ Timothy J. Mattke |
|
| |
|
Timothy J. Mattke |
|
| |
|
Vice President, Controller and Chief Accounting Officer |
|
| |
|
|
|
INDEX TO EXHIBITS
(Part II, Item 6)
|
Exhibit
|
|
|
Number
|
Description of Exhibit
|
| |
|
|
3.1
|
Articles of Incorporation, as amended
|
| |
|
|
10.3.2
|
2011 Omnibus Incentive Plan (incorporated by reference to Appendix B to the Company’s Proxy Statement for its May 5, 2011 Annual Meeting of Shareholders)
|
| |
|
|
|
Certification of CEO under Section 302 of Sarbanes-Oxley Act of 2002
|
| |
|
|
|
Certification of CFO under Section 302 of Sarbanes-Oxley Act of 2002
|
| |
|
|
|
Certification of CEO and CFO under Section 906 of Sarbanes-Oxley Act of 2002 (as indicated in Item 6 of Part II, this Exhibit is not being "filed")
|
| |
|
|
|
Risk Factors included in Item 1 A of our Annual Report on Form 10-K for the year ended December 31, 2010, as supplemented by Part II, Item 1A of our Quarterly Report on Form 10-Q for the quarter ended March 31, 2011, and through updating of various statistical and other information
|
| |
|
|
101
|
The following financial information from MGIC Investment Corporation’s Quarterly Report on Form 10-Q for the quarter ended March 31, 2011, formatted in XBRL (eXtensible Business Reporting Language): (i) Consolidated Balance Sheets as of March 31, 2011 and December 31, 2010, (ii) Consolidated Statements of Operations for the three months ended March 31, 2011 and 2010, (iii) Consolidated Statements of Shareholders’ Equity for the year ended December 31, 2010 and the three months ended March 31, 2011, (iv) Consolidated Statements of Cash Flows for the three months ended March 31, 2011 and 2010, and (v) the Notes to Consolidated Financial Statements.
|
90