Description

The market has thrown MTG and other mortgage insurance stocks in with the bond insurers.  However, while both industries are experiencing heavy losses (bottom line suffering), unlike monoline insurance, mortgage insurance is a viable business, as evidenced by the fact that mortgage insurers (MIs) are currently writing record amounts of new business (top line expanding).  They are generated this new business despite the fact that Q4 mortgage originations  were at a 6-year low in Q4.  So as max318 bravely did with respect to RDN in December, I am going to make the case for MTG.  Like RDN, MTG is trading at a fraction of intrinsic value using very conservative “through-cycle” assumptions.  However, MTG has some important characteristics that provide significant downside protection not afforded by an investment in RDN – it is a pure-play MI, is the industry leader, has an honest (albeit sometimes to a fault) management team and has reserved more aggressively than its peers.  More importantly, MTG has already raised capital, something RDN/PMI may struggle to do without selling assets at bargain prices.

While MTG does not have a strong immediate catalyst, it is better to be early than never.  More importantly, this value investment is also an arbitrage opportunity capitalizing on a severe dislocation between debt and equity markets.  This dislocation allows one to substantially limit the potential for permanent impairment on this investment.  As a bonus, for those of you like me, who are unfortunately stuck trying to reconcile mark-to-market with value (impossible as it is), this setup should also make the volatile holding period much more palatable; if the position is managed prudently, even poor performance will be prove profitable.  In summary, I believe that the potential to lose money in this investment/arbitrage is slim, while the probability of a two+ bagger is high.  The thesis is as follows.

MIs are massively levered companies.  Their leverage is not obvious from their balance sheets as they typically book a liability / reserve for losses only after mortgages are delinquent.  However, their maximum exposure is significantly higher.  For MTG, total potential insurance liabilities (risk in force – RIF) were ~$59bn at Q4 while total on-balance-sheet liabilities were only $5.1bn.  While MTG will not pay claims amounting to the entire $59bn (more on my expectations later), the uncertainty surrounding these mortgage liabilities in the current environment is great.

Subordinate to the insurance liabilities (RIF) is a slice of senior unsecured debt ($511mm net of cash) and another slice (thanks to recent stock sell-off) of common equity/market cap ($1.4bn).  (I’m going to leave out the $365mm wedge of subordinated converts raised last week for now, but more on these very attractive instruments later for those who care to read on.)  Furthermore, all of these securities are located at a shell holding company, while almost all of MTG’s assets and potential insurance liabilities are structurally senior at a regulated insurance subsidiary.  There are no upstream guarantees benefiting the unsecured debt so, like the common equity, it is dependant on dividends from the regulated insurance subsidiary for cash.

Given the extreme potential leverage (RIF), depressed market cap, and the structural subordination of the HoldCo obligations, there is virtually no difference between equity and unsecured risk.  It is binary.  Either both are wiped out or both are ok.  However, despite their similar risk profiles, MTG’s stocks enjoys the benefit of 100+% upside if the company survives the current cycle, while the upside in the unsecured debt is minimal with the CDS trading at 550bps. 

For this reason, I advocate buying MTG stock and shorting unsecured risk (buying CDS) at a hedge ratio of between 2- and 1-to-1 ($1-2 of CDS for every $1 of stock) to start.  In doing so, you are shorting uncertainty (e.g. duration and severity of economic downturn, etc.) and going long valuation.  Over a one to three year period when the dust settles, one of two things will happen.  If MTG survives (I believe they will), the stock will be worth 2-5x its current levels making the 6-11% annual cost-of-carry on the CDS immaterial.  If MTG goes belly up (I believe very unlikely), you will make a little money also.  Heads I win; tails I win too! 

A Few Important Misunderstood Facts about Mortgage Insurance Companies

Misunderstood Fact #1 – Record Amounts of Quality Business Being Written Now

Mortgage insurance companies are NOT bond insurers.  MIs insure mortgages (assets people will always need to purchase homes) and not ABS (assets that are rapidly disappearing from the face of the planet).  This key fact has been ignored by that much of the market, but unlikely by VIC readers.  What be more surprising to VIC readers is that MIs are currently writing record amounts of business despite the current turmoil.  I will pause on this and let it sink in.  How is this possible when home sales and mortgage originations have plummeted?  To quote Curt Culver (CEO): “MI penetration is at an all time high at somewhere between 17 and 20%. This is up from 10% a year ago, and somewhere around 8.5% 18 months ago. As evidence of the significance of the penetration increase, our flow NIW in the fourth quarter was up 108% year over year and 75% for the year. Even with the underwriting changes our industry has introduced, I think penetration will still run in excess of 15%.”

During the peak of the housing bubble, 2^nd lien loans served as MIs main competition (not other MIs or FHA).  When credit is cheap, originators have no problem writing 100% LTV loans through 2^nd liens and HELOCs.  However, during and coming out of a downturn, mortgage insurance is a borrowers only option for high LTV loans as cash-strapped lenders pull back.  Mortgage insurance penetration rates peaked following each of the past recessions with the highest penetration rates following the severe early 80’s housing recession.

As an important added bonus, the business the MIs are writing now is also significantly better quality than during the “bull market” – the subprime market is dead.

Misunderstood Fact #2 – Defaults are Cleansing

It should be no surprise that the bursting housing bubble is taking its toll on MTG and the other MIs.  While barriers are relatively high for a new entrant (due to capital requirements, infrastructure and state-by-state regulatory approval), mortgage insurance is a commoditized product.  Given the fierce competition from 2^nd liens over the last several years, MIs had little choice but to underwrite a lot of garbage in order to maintain share.  “Data-dump” (which is exactly what Countrywide would do with respect to the MIs) is a much better term for the risk taking process of the MIs than is “underwriting”.  In fact, industry loss ratios have been steadily increasing since 2000 despite the bull market.

But there is a light as the end of the tunnel for the surviving MIs.  While unprecedented losses will depress earnings over the next several quarters and cashflow for the quarters that follow (surprisingly, MIs are currently still generating boatloads of FCF due to a lag between defaults and paid claims), the losses will also cleanse the MIs’ books in the process.  As claims are paid on bad loans, the remaining book becomes skewed towards better loans.

Misunderstood Fact #1 PLUS Misunderstood Fact #2

The point is that the best time for an MI is not when home prices are soaring, credit is cheap and Mozillo is especially tan.  Rather, Mis perform best coming out of down-turn when consumers are recovering.  The combination of (i) more industry penetration, (ii) less competition, (ii) higher insurance persistency rates and, most importantly (iv) less non-traditional competition from cheap credit creates an especially strong revenue stream for surviving MIs.  In addition, the “cleansing process” described above provides for a relatively lean cost structure.  With strong revenue and low costs, MIs bottom lines will bounce early in the housing rebound.  It’s not a coincidence that the 1990 recession was followed by a wave of IPOs for MIs (MGIC 8/91, RDN 10/92, PMI 4/95). 

Valuation

This entire arbitrage is predicated on a surviving MTG being severely undervalued, so I will try to make my case below.  But before I do, I provide the following disclaimer.  MTG is not a great “Buffet-style” business.  MIs compete fiercely with each other, cheap credit and FHA (who’s role is expanding as I write).  In addition, MIs (including MTG) are old-school and have done a horrible job reacting to the evolving mortgage market with new risk management techniques (I believe that improving hedging capabilities through interest rate and credit default swaps is a big opportunity for MTG).  Despite those factors, MTG represents a chance to own (i) the leading company (ii) in an industry that has a reason to exist (iii) at an inexplicably cheap valuation (especially given inherent leverage of the investment).

While it is difficult to value MTG given its cyclical nature and the fact that none of the MIs were public during the last severe housing downturn, below I will take a stab at putting some brackets around intrinsic valuation.

--------------------------

As of Q4 2007, MTG had $212bn of insurance in force (this represents the face amount of the mortgages they insure and not their maximum risk, which is ~25% of that).  MTG earns recurring cash revenue on this book equal to approximately 67bps annually or $1.4bn.  In terms of costs, MTG has about $300mm labor and other underwriting costs.  The real costs are losses/reserves.  MTG is currently reserved for $3.9bn and expects another ~1bn of losses for a total of $5bn, which is a slightly more optimistic than my own modeling, so let’s assume ~$6bn for conservatism.  Therefore, if MTG doesn’t grow their book at all (keep in mind the record growth in Q4), and all of the loss reserves are all taken over the next 4 Qs, MTG’s 2008 income statement would look something like this:

 

Revenues           1.4bn

Overhead:           (300)mm

Reserves:            (2.0)bn

Interest               (76)mm

Net Income         $(957)mm

EPS                    $(6.31) per share

Note: I have not included any income from (i) investment income on MTG’s $6.7bn of invested assets (MTG earned $260mm of investment income in 2007), or (ii) MTG’s 24% minority interest in Sherman (worth $64mm of 2007 EBT).   MGIC is currently in negotiations with Sherman management to sell its interest for $242.5mm. 

Margin of Safety

MIs are misvalued by nearly every equity analyst, who simply slap a multiple on a misleading book value.  What I have tried to do above is formulate some conservative inputs so that we can complete the earnings cycle necessary to value MTG in our survival case.  Looking at historicals, from Q1 1995 through Q2 2007, MTG has averaged annual EPS of $3.11.  Rolling forward those numbers to complete the cycle by including 2H 2007 actual results and our 2008 projections from above (front-loading all estimated losses in 2008 for conservatism), we have cycle-average EPS of $1.45, which is 7.6x MTG’s current trading levels.  Keep in mind that MTG has grown revenues from $600mm to $1.6bn during this time period.

Upside

To give people a sense for the upside, the cleansing process described above will leave the Company with a spick-and-span book of business that should generate (assuming 30% loss ratio based on historical performance) $4+ EPS once we are through the cycle in 2009/10.  This implies a PE ratio of less than 2.8x for a business that has historically traded between 7x and 13x.

While the duration of this downturn (and therefore this investment) is uncertain, continued aggressive reserving should put MTG in a position to return to profitability in 2009 or 2010. 

(The share count used in the figures above are pro forma for new equity and convert assuming conversion.)

The Case for the Convert

The recently issued subordinated converts offer an even more attractive investment opportunity than the stock when paired with CDS.  The 9% cash coupon provides very good financing for the CDS (even at a 2-to-1 hedge ratio the negative carry is only 2% annually).  More importantly, the coupon provides protection in the unlikely case of a strategic acquisition at a depressed valuation.  In this case, the CDS could be left worthless and the acquisition may not make up for the loss via the stock gain (depending on timing and take-out price)   However, the perpetual 9% coupon will likely more than cover the CDS loss in this case if one chooses to remain outstanding especially if the strategic is of high credit quality (as I expect) – e.g. AIG, GNW. 

Relative to the stock, the converts offer very similar upside in survival case, more upside in blow-up case, and significant downside protection in the unlikely case of a distressed sale to a strategic.

Questions

Why not just go long equity/convert – is the CDS hedge worth it?

While I do think the equity and convert are attractive on their own, there is enough uncertainty here to warrant the CDS hedge.  MTG is an extremely levered company and its performance/survival is tied to very macro trends, which are difficult to accurately assess.  Overall, the cost of the CDS hedge is insignificant relative to upside in the stock/convert.

As a bonus, the CDS also helps manage volatility (which I unfortunately cannot ignore in my current seat).

What about the rating agencies?

While the rating agencies are important, unlike for the bond insurers (who are in the business of renting out a “AAA” rating), mortgage insurers’ viability is ultimately determined by the their customers (the GSEs).  Fannie and Freddie have a vested interest in the MIs’ survival as they are holding billions of dollars of insured loans on their books.  For example, the GSEs recently altered their guidelines to support the MIs during this downturn by suspending “Type II Insurer requirements otherwise automatically applicable to mortgage insurers that are downgraded below AA- or Aa3 by the rating agencies provided the mortgage insurer commits to submitting a complete remediation plan for our review and approval within 90 days of the downgrade.”  This was big news largely ignored by the markets. 

Also, while the rating agencies are impossible to judge right now (and downgrades are likely), the only hard data point they have given is that the MIs must maintain a risk-to-capital ratio of less than 15x.  Even using my conservative, front-loaded 2008 projections, MTG’s risk-to-capital should be better.  A likely sale of their stake in Sherman and / or additional reinsurance will further boost MTG’s capital base.

Why not other mortgage insurers?

While this arbitrage may work for other MIs, I believe MTG is safest way to play this theme.  MTG has already put the dilutive capital raise behind it and in doing so has chosen to remain independent and see the cycle through.  In addition (as previously mentioned), MTG is (i) a pure play MI, (ii) the industry leader and (iii) has an honest (although not always rational) management.  While RDN may have more upside given different pockets of value (financial guarantee business) the CDS hedge is much more expensive and survival independently much more in question.  For PMI, there is likely less upside and less safety. 

Also, I believe MTG has reserved more conservatively than its competitors due to a special reserve taken in Q4.  MTG now has $3.9bn of reserves posted.  To put this in perspective, MTG has $1.4bn of subprime and $8.2bn of Al-A exposure.  Therefore, MTG has reserved for a 40% default rate with 100% severity on their subprime and Alt-A exposure.  Importantly, 30% of MTG’s exposure was originated from pre-2005 and should perform much better.

Comparatively, MTG has booked reserves representing 28% of its mid/sub prime and Alt-A exposure, more than double that of RDN/PMI.  MTG’s more conservative reserving may allow it to outperform peers nearer term and shorten the duration of the investment.

Is the industry viable?

The industry is in for a rough ride well into 2008-9, and will likely undergo significant changes (acquisitions, government intervention through FHA, and the potential failure of a smaller player like TGIC), but it has staying power.  Heavy losses, ratings downgrades, and negative research coverage will be the theme in the near to medium term.  However, unlike other mortgage-related industries, the MIs have liquidity in the form of cash/investments.  They also have sticky revenue as they earn monthly premiums on their policies.  For the purer-play MIs (MTG is only large player without exposure to structured credit), the jump-to-default risk is minimal.  

The current mortgage/housing crisis was caused by lack of accountability and a poor incentive structure in the mortgage lending chain.  The originators took advantage of homeowners by giving them loans they couldn’t afford – the larger (high LTVs) and more complicated (Option ARMs, HELOC) the loan, the more accretive to their bottom lines.  The originators then quickly sold off these loans to Wall Street, who gladly accepted them (the more the merrier), without any thought for quality, as they skimmed fees before repackaging and shoving the loans into the market with investment grade ratings (ABS).  The fundamental issue is that neither (i) the originator, (ii) the Wall Street firm, or (iii) the rating agencies had any skin in the game.  Losses didn’t hit their bottom line (until now), so execs clipped healthy bonuses. 

During this “bull market”, MIs struggled to maintain market share as they were competing against cheap credit.  MIs were at a huge pricing disadvantage.  Because they have skin the game, they can’t give away free money while others, who pass it off can (for a while anyway).  This is one of the main reason the MI industry survives this shakeout.  MIs facilitate home buying for people, who couldn’t otherwise afford homes, while also having a vested interest in the performance of the loan.

Catalyst

(i) Arbitrage opportunity, which will likely make investment profitable even during a selloff, (ii) outperformance vs. competitors due to more conservative reserving, (iii) capital infusion may eventually force rating agencies to downgrade other MIs (RDN/PMI) and not (/ more than) MTG, (iv) cheapness.