NYSE Symbol MTG. We don t make home loans. We make home loans possible.

Transcription

1 NYSE Symbol MTG We don t make home loans. We make home loans possible.

2 Safe Harbor Statement During the course of this presentation, I may make comments about expectations for the future. Actual results could differ materially from those contained in these forward-looking statements. By making these statements the company is not undertaking an obligation to update these statements in the future should subsequent events cause these expectations to change. Additional information about the factors that could cause actual results to differ materially from statements made during this presentation are contained in our press release dated July 14, 2004, which has been filed with the SEC, and are also presented at the end of this presentation. 2

3 Financial Highlights MGIC Investment Corporation(1) Q Diluted Earnings Per Share $1.56 % Change Q H 2004 $2.87 Net Income (millions) Total Revenue (millions) Incurred Losses (millions) $154.5 $403.1 $ % -3.0% -19.2% $284.6 $818.6 $344.7 Paid Losses (millions) $ % $282.0 Expenses (millions) New Insurance Written (billions) $73.6 $ % 24.8% $141.8 $29.0 Insurance In Force (billions) Persistency $ % -2.6% 51.0% Average ROE (2) 15.6% 14.7% (1) Financial Highlights are presented on a consolidated basis and include subsidiaries (2) Average ROE is the result of dividing Net Income for the reporting period(s) divided by average Shareholders Equity for the same reporting period(s) as publicly reported 3

4 Origination Market Current MBA Origination Estimates stands at $2.4 trillion A 35% reduction from 2003 Approximately $1.3 trillion already booked as of 6/30/2004 Current MBA Forecast for 2005 and 2006 is estimated at $1.8 trillion MGIC New Insurance Written 1 st half of 2004: Bulk $5 billion Flow $24 billion 4

5 Persistency and Insurance in Force 85 $ $200 Annual Persistency Longer Term Persistency Expectations Flow/Total IIF (Billions) $175 $150 $125 $100 $75 45 $50 40 $ Q Q $ Q Q Annual Persistency Bulk Flow Total The Bulk channel has contributed 15 30% of the New Insurance Written over the last several years and is now 22% of Insurance in Force SOURCE: MGIC Press Releases

6 In Force Mix 90% Total Risk In Force 13% 58% Bulk Risk in Force 35% Prime (FICO 620 & >) 85% 80% 75% 70% 65% 60% 55% 11% 8% 6% Non-Prime (FICO < 620) Prime (FICO 620 & >) 55% 53% 30% 25% 20% 15% Non-Prime (FICO < 620) 50% Q1 02 Q2 02 Q3 02 Q4 02 Q1 03 Q2 03 Q3 03 Q4 03 Q1 04 Q2 04 3% 50% Q1 02 Q2 02 Q3 02 Q4 02 Q1 03 Q2 03 Q3 03 Q4 03 Q1 04 Q % 620 and > < and > < 575 Bulk Business began in 1998 Bulk Currently Comprises 22% of IIF Expanded GSE credit guidelines began in 1998/1999 SOURCE: MGIC Press Releases

7 Credit Environment Jobs are the Key Quarterly Change in Non-Farm Payroll And Delinquent Inventory ,000 Chanhe In Non-Farm Payroll (Thousands) Q Q Q Q Q Q Q Q Q Q Q Q Q Q ,000 39,000 34,000 29,000 24,000 19,000 14,000 Delinquent Inventory -300 Change in Payroll Flow Delq Bulk Delq 9,000 Total Delinquent Inventory increased 34.3% from Q2 02 to Q2 03 Total Delinquent Inventory increased 2.3% from Q2 03 to Q2 04 Source: Bureau of Labor Statistics Establishment Survey, Bloomberg, Company Press Releases 7

8 MGIC Strengths Market Share Leader Broad Geographic/Risk Distribution Productivity Financial Strength

9 Market Share Leader As of June 30, 2004 Sorted By Flow Share FLOW BULK MGIC 21.5% 29.6% PMI 19.4% 23.8% Radian 16.8% 12.6% UGI 15.2% 9.0% RMIC 11.7% 0.0% GE 10.6% 13.0% TRIAD 4.8% 11.8% SOURCE: Inside Mortgage Finance 8/13/04

10 Broad Geographic/Risk Distribution Total: 17.1% Total: 17.1% Flow: 17.3% Total: 13.8% Flow: 17.3% Total: 13.8% Flow: 14.7% Flow: 14.7% Total: 25.8% Total: 25.8% Flow: 27.4% Total: 20.3% Flow: 27.4% Total: 20.3% Flow: 16.6% Flow: 16.6% Total: 23.0% Total: 23.0% Flow: 23.9% Flow: 23.9% Risk in Force as of June 30, 2004

11 Home Price Appreciation: 2003Q1-2004Q1 United States (AK & HI Inset) by Yr/Yr Chg 0.0% - 2.4% (2) 2.5% - 4.9% (20) 5.0% - 9.9% (18) 10% or more (11) Source: OFHEO 2004Q1 HPI 11

12 30% 28% 26% Productivity Expense Ratio Analysis Public MIs * Statutory Expense Ratio was 27.2% for % 22% 20% 19.2% 19.4% 18 % 16 % 14 % 16.1% 16.8% 14.6% 13.5% 13.5% 12 % 10 % YTD 06/ YTD 06/04 *RDN, PMI, TGIC Source: Statutory filings

13 Financial Strength Q Q Asset - Size (billions) $ Investment Portfolio (billions) $5.4 $5.0 Shareholder Equity (billions) $4.0 $3.6 Net Loss Reserves (millions) $1,106 $842 Debt-to-Equity Ratio 15.1% 16.6% Risk-to-Capital (consolidated) 8.3:1 9.6:1 13

14 Summary Short Term 30-35% reduction in NIW MI Avoidance Increasing Persistency Improved Credit Performance as Economy Continues to Recover Long Term Growth in Homeownership Steady Home Price Appreciation MGIC s Franchise and Financial Strength Clear Primary/Secondary Market Definitions

15 NYSE Symbol MTG We don t make home loans. We make home loans possible.

16 Safe Harbor Statem ent Forw ard-looking Statem ents and Risk Factors: The Com pany s revenues and expenses could be affected by the risk factors discussed below. These factors m ay also cause actual results to differ m aterially from the results contem plated by forw ard looking statem ents that the C om pany m ay m ake. Forw ard looking statem ents consist of statem ents w hich relate to m atters other than historical fact. A m ong others, statem ents that include w ords such as the Company believes, anticipates or expects, or words of similar import, are forward looking statem ents. T he C om pany is not undertaking any obligation to update any forw ard looking statem ents it may make. As the domestic economy deteriorates, more homeowners may default and the Company s losses may increase. Losses result from events that reduce a borrow er s ability to continue to m ake m ortgage payments, such as unemployment, and whether the home of a borrower who defaults on his mortgage can be sold for an amount that will cover unpaid principal and interest and the expenses of the sale. Favorable economic conditions generally reduce the likelihood that borrow ers w ill lack sufficient incom e to pay their m ortgages and also favorably affect the value of homes, thereby reducing and in some cases even eliminating a loss from a mortgage default. A deterioration in econom ic conditions generally increases the likelihood that borrow ers w ill not have sufficient income to pay their mortgages and can also adversely affect housing values. The m ix of business the Com pany w rites also affects the likelihood of losses occurring. In recent years, a greater percentage of the Company s volum e than in the past has included segments that the Company views as having a higher probability of claim, including loans with LTV ratios over 95%, FICO credit scores below 620 or limited underwriting, including limited borrower documentation. In response to lower national origination volume in 2004 compared to 2003, m ortgage lenders m ay seek to m aintain their ow n volum e through a greater focus on lending to borrowers in segments that the Com pany views as having a higher probability of claim. A pproxim ately 8% of the C om pany s risk in force w ritten through the flow channel, and m ore than half of the C om pany s risk in force w ritten through the bulk channel, consists of A R M s. T he Company believes that during a prolonged period of rising interest rates, claims on ARM s would be substantially higher than for fixed rate loans, although the performance of ARM s has not been tested in such an environment.

17 T he perform ance of the servicing function on a m ortgage loan, particularly a subprim e loan, can affect the likelihood that the loan w ill default as w ell as the loss resulting from a default. T he C om pany believes S elect P ortfolio S ervicing f/k/a F airbanks C apital C orp. ( F airbanks ) is the servicer of approxim ately 1.2% of the loans insured by the Com pany and approximately 5.7% of the loans insured by the Com pany written through the bulk channel (a substantial num ber of which are subprim e). The servicer ratings assigned to Fairbanks by M oody s and S& P were downgraded to "below average" during the second quarter of 2003 due in part to concerns expressed by those rating agencies about Fairbanks regulatory compliance and operational controls. In the second quarter of 2004, these rating agencies raised Fairbanks' service ratings to "average." C om petition or changes in the C om pany s relationships w ith its custom ers could reduce the C om pany s revenues or increase its losses. Com petition for private m ortgage insurance prem ium s occurs not only am ong private m ortgage insurers but also w ith m ortgage lenders through captive m ortgage reinsurance transactions. In th ese transaction s, a lend er s affiliate rein sures a p ortio n of th e in suran ce w ritten b y a private m ortgage insurer on m ortgages originated or serviced by the lender. T he level of com petition w ithin the private m ortgage insurance industry has also increased as m any large mortgage lenders have reduced the num ber of private m ortgage insurers with whom they do business. At the same tim e, consolidation am ong m ortgage lenders has increased the share of the m ortgage lending m arket held by large lenders. Our private mortgage insurance competitors include: PM I M ortgage Insurance Company Genworth Financial f/k/a/ GE Capital M ortgage Insurance Corporation United Guaranty Residential Insurance Company Radian Guaranty Inc. Republic M ortgage Insurance Company Triad Guaranty Insurance Corporation CM G M ortgage Insurance Company Assured Guaranty Limited f/k/a/ AGC Holdings Limited, a financial guaranty company whose mortgage insurance business is primarily reinsurance, has announced that it intends to write investment grade mortgage guaranty insurance on a direct basis.

18 If interest rates decline, house prices appreciate or m ortgage insurance cancellation requirem ents change, the length of tim e that our policies rem ain in force could decline and result in declines in our revenue. In each year, m ost of the C om pany s prem ium s are from insurance that has been w ritten in prior years. A s a resu lt, th e len g th o f tim e in su ran ce rem ain s in fo rce (w h ich is also gen erally referred to a s p e rsisten c y ) is a n im p o rta n t d eterm in a n t o f re v e n u e s. T h e fac to rs a ffe c tin g th e le n g th o f tim e the C om pany s insurance rem ains in force include: the level of current m ortgage interest rates com pared to the m ortgage coupon rates on the insurance in force, w hich affects the vulnerability of the insurance in force to refinancings, and m ortgage insurance cancellation policies of m ortgage investors along w ith the rate of hom e price appreciation experienced by the hom es underlying the m ortgages in the in suran ce in force. D uring the 1990s, the C om pany s year-end persistency ranged from a high of 87.4% at D ecem ber 31, 1990 to a low of 68.1% at D ecem ber 31, A t June 30, 2004 persistency w as at 53.8%, w hich w as an im provem ent over the record low of 44.9% at Septem ber 30, O ver the past several years, refinancing has becom e easier to accom plish and less costly for m any consum ers. H e n c e, e v e n in a n in te re st ra te e n v iro n m e n t fa v o ra b le to p ersiste n c y im p ro v e m e n t, th e C o m p a n y does not expect persistency w ill approach its D ecem ber 31,1990 level. If the volum e of low dow n paym ent hom e m ortgage originations declines, the am ount of insurance that the C om pany w rites could decline w hich w ould reduce the C om pany s revenues. T h e fac to rs th at a ffec t th e v o lu m e o f lo w d o w n p a y m e n t m o rtg a g e o rig in a tio n s in c lu d e : th e le v e l o f h o m e m o rtg a g e in te rest rate s, the health of the dom estic econom y as w ell as conditions in regional and local econom ies, housing affordability, population trends, including the rate of household form ation, the rate of hom e price appreciation, w hich in tim es of heavy refinancing can affect w hether refinance loans have loan-to-value ratios that require private m ortgage insurance, and governm ent housing policy encouraging loans to first-tim e hom ebuyers.

19 In general, the m ajority of the underw riting profit (prem ium revenue m inus losses) that a book of m o rtg a g e in su ra n ce g e n e rate s o c c u rs in th e e a rly y e a rs o f th e b o o k, w ith th e la rg e st p o rtio n o f th e underw riting profit realized in the first year. Subsequent years of a book generally result in m odest underw riting profit or underw riting losses. This pattern of results occurs because re la tiv e ly fe w o f th e c la im s th a t a b o o k w ill u ltim a te ly e x p e rie n c e o c c u r in th e first fe w y e a rs o f the book, w hen prem ium revenue is highest, w hile subsequent years are affected by declining prem ium revenues, as persistency decreases due to loan prepaym ents, and higher losses. If a ll o th e r th in g s w e re eq u al, a d e c lin e in n e w in su ra n c e w ritte n in a y e a r th at fo llo w e d a n u m b e r of years of higher volum e could result in a low er contribution to the m ortgage insurer s overall results. T his effect m ay occur because the older books w ill be experiencing declines in revenue and increases in losses w ith a low er am ount of underw riting profit on the new book available to o ffset th ese resu lts. W hether such a low er contribution w ould in fact occur depends in part on the extent of the v o lu m e d e c lin e. E v e n w ith a su b sta n tial d e c lin e in v o lu m e, th e re m a y b e o ffse ttin g fa c to rs th at could increase the contribution in the current year. T hese offsetting factors include higher persistency and a m ix of business w ith higher average prem ium s, w hich could have the effect of increasing revenues, and im provem ents in the econom y, w hich could have the effect of reducing losses. In addition, the effect on the insurer s overall results from such a low er contribution m ay b e o ffse t b y d e c re ase s in th e m o rtg a g e in su re r s e x p e n se s th at are u n rela te d to cla im o r d e fa u lt a c tiv ity, in clu d in g th o se re late d to lo w e r v o lu m e. T he C om pany s new insurance w ritten during w as $86.1 billion, $92.5 billion and $96.8 billion, respectively and w as $49.5 billion and $29.1 billion in the first half of 2003 and 2004, respectively. C onsistent w ith a m id-june 2004 m ortgage finance forecast of the M ortgage B ankers A ssociation, w hich projects that quarterly m ortgage originations in the U nited States are expected to decline m aterially in 2004 com pared to 2003, the C om pany expects new insurance w ritten for the last tw o quarters of 2004 w ill be m aterially low er than for the com parable period in T he am ount of insurance the C om pany w rites could be adversely affected if lenders and investors select alternatives to private m ortgage insurance. T hese alternatives to private m ortgage insurance include: lenders structuring m ortgage originations to avoid private m ortgage insurance, such as a first m ortgage w ith an 80% loan-to-value ratio and a second m ortgage w ith a 10%, 15% or 20% loan-to-value ratio (referred to as an , or loans, re sp e c tiv e ly ) rath e r th a n a first m o rtg a g e w ith a 9 0 %, 9 5 % o r % lo a n -to -v a lu e ra tio, in v e sto rs h o ld in g m o rtg a g e s in p o rtfo lio a n d se lf-in su rin g, investors using credit enhancem ents other than private m ortgage insurance or using other credit enhancem ents in conjunction w ith reduced levels of private m ortgage insurance coverage, and lenders using governm ent m ortgage insurance program s, including those of the Federal H ousing A dm inistration and the V eterans A dm inistration.

20 Changes in the business practices of Fannie M ae and Freddie M ac could reduce the Company s revenues or increase its losses. The business practices of Fannie M ae and Freddie M ac affect the entire relationship between them and m ortgage insurers and include: the level of private m ortgage insurance coverage, subject to the lim itations of Fannie M ae and Freddie M ac s charters, when private mortgage insurance is used as the required credit enhancement on low down payment mortgages, w hether Fannie M ae or Freddie M ac influence the m ortgage lender s selection of the m ortgage insurer providing coverage and, if so, any transactions that are related to that selection, whether Fannie M ae or Freddie M ac will give mortgage lenders an incentive, such as a reduced guaranty fee, to select a m ortgage insurer that has a "A A A " claim s-paying ability rating to benefit from the low er capital requirem ents for Fannie M ae and Freddie M ac when a mortgage is insured by a company with that rating, the underw riting standards that determ ine w hat loans are eligible for purchase by Fannie M ae or Freddie M ac, which thereby affect the quality of the risk insured by the mortgage insurer and the availability of m ortgage loans, the terms on which mortgage insurance coverage can be canceled before reaching the cancellation thresholds established by law, and the circum stances in w hich m ortgage servicers m ust perform activities intended to avoid or m itigate loss on insured m ortgages that are delinquent. T he m ortgage insurance industry is subject to litigation risk. Consumers are bringing a growing number of lawsuits against home mortgage lenders and settlement service providers. In recent years, seven mortgage insurers, including the Company s M GIC subsidiary, have been involved in litigation alleging violations of the Real Estate Settlement Procedures Act, which is commonly known as RESPA. M GIC s settlement of litigation against it under RESPA became final in October There can be no assurance that M G IC w ill not be subject to future litigation under R ESPA.

21 In March 2003 an action against MGIC was filed in Federal District Court in Orlando, Florida seeking certification of a nationwide class of consumers who were required to pay for private mortgage insurance written by MGIC and whose loans were insured at less than MGIC s best available rate based on credit scores obtained by MGIC. (A portion of MGIC s A minus and subprime premium rates are based in part on the credit score of the borrower.) The action alleges that the Federal Fair Credit Reporting Act ( FCRA ) requires a notice to borrowers of such adverse action and that M GIC has violated FCRA by failing to give such notice. The action seeks statutory damages (which in the case of willful violations, in addition to punitive damages, may be awarded in an amount of $100 to $1,000 per class member) and/or actual damages of the persons in the class, and attorneys fees, as well as declaratory and injunctive relief. The action also alleges that the failure to give notice to borrowers in Florida in the circumstances alleged is a violation of Florida s Unfair and Deceptive Acts and Practices Act and seeks declaratory and injunctive relief for such violation. In December 2003, the Court denied MGIC s motion seeking dismissal of the portion of the case covering damages under FCRA but dismissed the remainder of the case. In late June 2004, the Court denied the plaintiffs motion to certify the class. There can be no assurance that the outcome of the litigation will not materially affect the Company s financial position or results of operations. Similar actions have been filed against six other mortgage insurers. N et premiums written could be adversely affected if the Department of Housing and Urban Development reproposes and adopts a regulation under the Real Estate Settlement Procedures Act that is equivalent to a proposed regulation that was recently withdrawn. The regulations of the Department of Housing and Urban Development under the Real Estate Settlement Procedures Act prohibit paying lenders for the referral of settlement services, including mortgage insurance, and prohibit lenders from receiving such payments. In July 2002, the Department of Housing and Urban Development proposed a regulation that would exclude from these anti-referral fee provisions settlement services included in a package of settlement services offered to a borrower at a guaranteed price. HUD withdrew this proposed regulation in March Under the proposed regulation, if mortgage insurance was required on a loan, the package must include any mortgage insurance premium paid at settlement. Although certain state insurance regulations prohibit an insurer s payment of referral fees, had this regulation been adopted in this form, the Company s revenues could have been adversely affected to the extent that lenders offered such packages and received value from the Company in excess of what they could have received were the anti-referral fee provisions of the Real Estate Settlement Procedures Act to apply and if such state regulations were not applied to prohibit such payments.