Print

The Great Mortgage Refinancing, by the numbers

Deutsche Bank have returned with some numbers to back up their cost claims when it comes to the Great Mortgage ReFi Rumour of August 2010.

Analyst Steven Abrahams noted last week that “hitting that reset button” on US mortgages via a mass refinancing would be no easy, or costless, feat. Remember that the rumour kicked off when Morgan Stanley billed the possible move — which would involve extending refinancing for millions of Fannie Mae and Freddie Mac-backed mortgages — as the “slamdunk” and free, US “stimulus.”

Since then, we’ve had Fed officials shoot down the idea — but clearly it’s persisting in some corners of the MBS market. In his updated commentary, Abrahams has added some additional detail:

Almost all of the $5.7 trillion agency mortgage market trades at a premium to par today because of historically low mortgage rates and because of the clear difficulty that many borrowers face in refinancing. The average fixed rate agency MBS closed last Wednesday at $106-28 and the average ARM at $105-24. Fixed-rate mortgages constitute an estimated 93% of outstanding agency balances, adjustable-rate an estimated 7%.

If the government found a way to efficiently refinance the market down to par—an aggressive assumption given the effectiveness of HAMP, HARP and other programs, but useful for sizing the risk—the market would lose $390 billion of market value. That would clearly be a costless windfall to homeowners, who would gain that value in lower mortgage payments over time. But it would be far from costless for the banks, GSEs, mutual fund investors, sovereign portfolios and others institutions now holding agency MBS… They would be the providers of the windfall.

Unsurprisingly then, it’s the banks who are likely to be hit by any mass refinancing.

According to Deutsche’s estimates, banks, thrifts and credit unions together hold more than 20 per cent of outstanding agency MBS. In absolute terms, most of that sits at the biggest banks, but in relative terms (percentage of total assets), it’s the regional banks who are most exposed.

Under the crude assumption that these banks hold a representative portfolio of agency MBS, they would, by our estimates, lose 6.8% of economic value on those holdings in a ruthless refinancing. Since most banks account for MBS as available for sale and probably hold most of the asset on their books near par, the price impact may be muted. Nevertheless, the economic impact would show up over time in a smaller net interest margin and lower earnings. Given the substantial resources invested in rebuilding bank capital and bringing the sector back to full strength, the impact of ruthless refinancing might give policymakers reason to pause.

And as for non-bank impact:

Of course, other portfolios would feel the impact of ruthless refinancing as well. The GSEs would potentially lose substantial economic value that would show up as lower earnings over time. The Fed would be likely to send a smaller dividend from its portfolio back to the Treasury. These effectively would be losses to any U.S. taxpayer not financed by a Fannie Mae or Freddie Mac loan. Holders of mutual funds and private and public pensions, too, would be likely to lose economic value. And the central banks of China, Korea, Taiwan and other countries, and the commercial banks in those countries and Japan would probably see the impact as well. It is hard to know how policymakers would balance the interests of these constituencies against the windfall to U.S. homeowners.

And that’s not to mention the costs to future borrowers. By DB’s estimates, the “ruthless refinancing” could add as much as 40 basis points to borrowers’ mortgage rates in the years to come.

To sum it up, then:

Fortunately, the costs to all the constituents in any potential refinancing program are all roughly measurable— the benefits to current homeowners, the costs to current investors, the price paid by future borrowers. Traditionally, markets have allocated among those competing interests. Mortgage originators in the past have pilot tested loans where the note rate could not go up and the borrower had a continuing option to refinance to a lower rate at no cost. These loans usually had higher interest rates than traditional fixed-rate loans, and found almost no audience. Current and future borrowers accepted constrained refinancing in return for lower rates.

Under extraordinary circumstances, of course, rules change. Policy may become the arena for balancing competing interests. The surprise in the MBS market last week was the apparent quick discounting of the cost of refinancing to investors and future borrowers. Once fully considered, ruthless refinancing of existing agency mortgages looks more like a volatile zero-sum game than a costless windfall.

Hmm. Extraordinary circumstances versus traditional markets.

Surely MBS investors should be used to that by now.

Related links:
‘Free stimulus’ via refinancing, debate grows - WSJ Developments
The slow death of Hamp, the summer of delinquencies – FT Alphaville

Print