Pre-payments are reducing value of mortgage-backed securities

If you read my recent post on How big banks earned so much money this quarter you would see that much of the income at Wells, JPMorgan, US Bank and others came from refinancing old mortgages. While this may be a boon to present income, it is very much a problem for the legacy mortgage-backed securities which contain the old mortgages. Let me explain.

When I was in business school, I took a course on debt markets with Professor Suresh Sundaresan, which was very helpful for me when I later joined a large bank in London. Professor Sundaresan worked at Lehman Brothers in their fixed income and derivatives area in the mid-1980s and was very familiar with all of the debt products we discussed in class.

One thing I found quite useful about the course was its explanation of the mortgage markets and mortgage-backed securities (MBS).  The mortgage market is a lot more complicated than other bond markets because MBS are not like other bonds. There is an embedded option contained in all mortgages, the pre-payment option, which makes valuation extremely tricky. In Chapter 9 of his book on “Securitization and Mortgage-Backed Securities,” which I have just consulted, Sundaresan talks about prepayment risk and what it means to the lower value of MBS.  He says the following in his book:

Mortgages permit the homeowners to prepay their loans. This prepayment provision introduces timing uncertainty into the originating bank’s cash flows from its loan portfolio. For example, if the bank originates a pool of mortgages with a weighted-average rate of 8% and six months later the mortgage rate drop significantly below 8%, say to 7%, then the loan portfolio is certain to experience significant prepayments as borrowers rush to refinance their mortgages with less-costly loans. The lender has a long position in the mortgage loan that entitles him to monthly scheduled payments, but also has sold an option to the homeowners that gives them the right [but not the obligation] to prepay the loan when the circumstances demand it. This means that the bank cannot predict the future cash flows from its loan portfolio with certainty. Clearly, the option to repay will be priced into the loan by the bank and the borrower will pay a higher interest rate on the loan as a consequence.

Here’s the deal. The bank does not want its customers to pre-pay because that means less income from interest payments for the bank. Less income lowers the value of the mortgage. So, pre-payments are bad for anyone holding debt or derivative instruments related to the expected cash flow of those mortgage loans.

As Sundaresan indicates, the originators and MBS packagers understand this and have tacked on a ‘fee’ in the form of a higher loan interest rate to cover their expected pre-payment risk.

Enter the Federal Reserve. To stave off a deflationary spiral, the Federal Reserve has lowered the effective short-term interest rate to zero and it is in the process of buying up shed loads of MBS paper and long-term bonds in order to artificially reduce long-term rates as well. The problem here lies in the term ‘expected’ from the previous paragraph, because the so-called toxic MBS paper now clogging balance sheets are now unexpectedly pre-paying at a record rate. This lowers the expected cash flow from those assets significantly, making the underlying assets worth even less.

Moreover, there is a certain perverse adverse selection at work here because not everyone can get a loan these days. That means that the individuals pre-paying are likely to be the most qualified borrowers. This leaves existing MBS borrower pools significantly worse off.

For example, say you have a mortgage-backed security collateralized by mortgage assets from prime borrowers. We enter a recession and a number of loans in that pool become distressed and a number of the borrowers in that pool default. This means that your MBS asset is worth less.

Simultaneously, interest rates drop unexpectedly and a number of the borrowers re-finance their mortgage meaning they pre-pay the mortgage in your pool. You are not going to get the interest payments that you had expected to receive. Again, this means that your MBS asset is worth less.

What’s more is that because of the recession, credit conditions are unusually tight and only the best qualified borrowers are refinancing. So the borrowers left in your asset pool are net lower-quality borrowers. Translation: you should now expect a higher default rate of those left in the pool. Again, this means that your MBS asset is worth less.

To sum up:

1. You just got the shaft because a recession has meant higher default rates generally.
2. You just got the shaft because of unexpected pre-payment and lower cash flows that result from this.
3. You just got the shaft because your pool of borrowers has been adversely impacted by the Fed’s interference in the MBS market

All of this is very bad for existing or so-called legacy assets.  Moreover, these assets must be marked-to-market to reflect asset value impairment.  So, this will mean massive writedowns going forward.

But, wait a minute, didn’t we just change the accounting rules? Enter new mark-to-market accounting a.k.a. mark-to-make-believe.  Because of the guidance on marked-to-market accounting in FAS 157-e, you can deem these changes to be temporary impairments.  There is no need to mark to market.  Problem solved.

Here’s what Wells Fargo had to say about its marking-to-market last quarter in their earnings release (PDF):

The net unrealized loss on securities available for sale declined to $4.7 billion at March 31, 2009, from $9.9 billion at December 31, 2008. Approximately $850 million of the improvement was due to declining interest rates and narrower credit spreads. The remainder was due to the early adoption of FAS FSP 157-4, which clarified the use of trading prices in determining fair value for distressed securities in illiquid markets, thus moderating the need to use excessively distressed prices in valuing these securities in illiquid markets as we had done in prior periods.

Nice.

And if you think people aren’t fooled by all of this, think again. Bank stocks are way, way up.