A new disclosure "Magically" appears

In August – which in this market seems like years ago – I did two posts on MGIC  – the largest mortgage insurer and a company colloquially known as Magic. 

These posts are here and here…

I concluded that I did not believe in Magic – but I outlined the bull case (which was that the companies had something close to positive cash flow) and had a readers who tried to convince me that the bull case was right.  (I was and remain short.) 

In the second post I made an observation about statutory capital and possible shortages thereof. 

My further objection is that statutory capital deficiency happens way earlier than this - and stat capital deficiency will close the business as per TGIC.

It is an innocuous line – but two readers (both of whom I respect) followed me up on it – and one was sure that statutory capital was going to be adequate (he was long).  The other modelled it with brackets for his own uncertainty – and we thought that statutory capital might run short.

Anyway MGIC’s results contained a new disclosure on this issue:

Some states that regulate us have provisions that limit the risk-to-capital ratio of a mortgage guaranty insurance company to 25:1. If an insurance company’s risk-to-capital ratio exceeds the limit applicable in a state, it may be prohibited from writing new business in that state until its risk-to-capital ratio falls below the limit.

We believe that our 2006 and 2007 books of business will continue to generate material incurred and paid losses for a number of years. The ultimate amount of these losses will depend in part on the direction of home prices in California, Florida and other distressed markets, which in turn will be influenced by general economic conditions and other factors. Because we cannot predict future home prices or general economic conditions with confidence, we cannot predict with confidence what our ultimate losses will be on our 2006 and 2007 books. Depending on the extent of future losses, MGIC’s policy holders position could decline and its risk-to-capital could increase beyond the levels necessary to meet these regulatory requirements and this could occur before the end of 2009. As a result, we are considering options to obtain additional capital, which could occur through the sale of equity or debt securities and/or reinsurance.

That is slightly worse than the worst end of the models my readers created for me in August.

In other words it is bad out there.  And if you look its bad not in Magic’s subprime book (where the number of delinquent loans are up only a little in year – because the bad loans have been mostly foreclosed).  It is bad in the prime book (which is really a near-prime book).  Delinquent so called “prime loans” are the problem as they were when I wrote the original posts.  In that category the delinquency (by which they mean edge-of-default delinquency) has nearlydoubled and the loss per claim has also nearly doubled in the last 5 quarters…  [For those interested in actual loans – a year ago they had 11,700 subprime near default loans.  Now they have 12,700.  For “prime loans” the near default pool has risen from 36,700 to 71,100 loans.]

Anyway the disclosure about needing more capital is super-bad.  If the company is forced to stop writing business it will end nastily because the holding company is levered.  And I doubt very much it can raise capital now. 

The stock is down less than 50% in the past two months – so I can’t crow about the short.  (Plenty that is that levered is down more.)  But if the long investors can explain to me why they shouldn’t run for the exits right now I would appreciate it. 

There is also some disclosure that appeared far more prominently.  I looked for it before and it was in footnotes.  I know I should read the footnotes but…

Anyway it says more about Fannie and Freddie than it does about Magic.  It is a definition of “full documentation” that could only exist in the twenty first century mortgage industry:

In accordance with industry practice, loans approved by Government Sponsored Enterprise and other automated underwriting systems under “doc waiver” programs that do not require verification of borrower income are classified by us as “full documentation.”