Share |

The GSE Black Hole

Christopher Papagianis | 11/19/2009 |

Click here for a printer-friendly version of this article

In February, the Obama Administration is expected to offer a reform proposal for the future of Fannie Mae and Freddie Mac (a.k.a Government Sponsored Enterprises or GSEs).  That has prompted interested parties to begin rolling out suggestions for what they want to see included in the reforms.  As you read these plans this winter – like the recent one by the influential Mortgage Bankers Association (MBA) – it’s important to remember why Fannie and Freddie went bust and the government had to take them over.  It’s also essential to keep a running tab of their recent losses, if for no other reason than to remind us how the black hole is still growing.


Let’s first take a look at the accounting numbers.  A couple of weeks ago, Fannie Mae posted a quarterly loss of $19.8 billion.  When you factor in the their recent calls for more government aid, the tab to prop up both of these enterprises since they were put into conservatorship in September 2008 comes in at a little over $110 billion.

Important to remember is that legislation passed last year allows the Treasury Department to spend whatever amount it wants to support Fannie and Freddie.  When the Treasury committed to back-stop these enterprises, it said that at least $400 billion would be available, if needed.  That large number was chosen because it made clear to investors around the world that the U.S. Government would not let Fannie and Freddie fail.  At the time, nobody thought spending all of that money would be necessary.  Amazingly, the situation at the enterprises has deteriorated to such a degree that spending the entire $400 billion is now a distinct possibility. 

Why are things still getting worse at Fannie and Freddie when the economy appears to be rebounding and house prices are starting to stabilize?  Because the Obama Administration – in an effort to subsidize more mortgage modifications – is okay with the GSE’s losing taxpayer dollars while they are in conservatorship.  For example, Fannie Mae recently documented $22 billion of their recent losses as foreclosure costs with close to $8 billion directly going to support the Obama Administration’s mortgage modification program.  (When Fannie or Freddie buy individual mortgages out of securities and then modify the terms, they generally book a loss.)

The following excerpt from a recent WSJ editorial explains why this isn’t getting a lot of attention:


Through this program, taxpayers are directly subsidizing homeowners who borrowed more than they could afford, or more  than their house is now worth, or both. The government is doing this under the cover of losses at Fannie and Freddie because Congress and the White House know these programs are both expensive and unpopular with the poor saps still paying their mortgages on time.


Unfortunately, there is not a lot Congress can do right now given that it already signed-off on this blank check for Treasury.  In many ways, one of the big benefits of passing new legislation that maps out how to resolve the GSE problem, once and for all, is that Congress will be able to take back some of its “lost” authority. 

It’s therefore surprising that Senator Dodd does not address the future of Fannie or Freddie in his recent proposal to reform financial regulation.  (See “First Impressions: A Legislative Dud from Dodd” for more information.)  As the WSJ put it in a different piece, “From a macroeconomic perspective, it is unhealthy when the government faces few checks when pouring billions of dollars into one sector of the economy, in this case, the housing market.”

Why exactly did Fannie and Freddie go bust?

Until 2007, analysts generally believed that annual credit losses at Fannie and Freddie would average just over 10 basis points (0.10%) of their total mortgage business.  The calculation was derived from the expectation that about one-half of one percent of all mortgages would ever default, and that the losses on those bad mortgages would be about 20%.  To compensate for these “expected” losses, Fannie and Freddie charged mortgage originators a guarantee fee of about 20 basis points per year.  When the housing market was healthy, this gave them a nice profit and helped them build a capital cushion. 

However, the risks posed by Fannie and Freddie came not from this basic insurance business, but from their retained portfolio and the issuance of debt.  Instead of just issuing mortgage-backed securities to fund more mortgage originations, Fannie and Freddie increasingly issued direct obligations (so-called “Agency debt”).  This debt issuance then allowed Fannie and Freddie to earn the difference between the mortgage interest rate and the lower coupon from their “implicitly” guaranteed Agency debt. 

Here is a side-by-side comparison of these two funding mechanisms:

20 basis point example: If Fannie purchased $100 million of 30-year mortgages with 6.5% interest  rates – by issuing mortgage-backed securities – they could expect to earn $200,000 annually (or 0.2 %). 

Agency debt example: If those same mortgages were financed through the issuance of 2-year debt with 5% coupons, Fannie would earn $1.5 million annually.  They would earn $6.5 million in mortgage interest and only pay out $5 million to the Agency debt holders.

The topline point here is that by issuing Agency debt to acquire mortgages, Fannie and Freddie assumed interest rate risk that could leave them insolvent.  Consider the following scenario that builds on the agency debt example above: if mortgage rates fell to 4% (from 6.5%) and the individual homeowners refinanced, the new mortgages would only generate $4 million in annual interest.  Conversely, if, after two years, rates on two-year Agency debt rose to 8%, the interest expense would be $8 million even though the mortgages would still only generate $6.5 million. 

In both cases, the annual loss would likely wipeout the tangible equity capital buffer maintained by Fannie and Freddie.  This interest rate risk drew the attention of a lot of people in Washington, including the Federal Reserve, the Bush Administration, and the Senate Banking Committee

This back story is important because, in the end, it was not this interest rate risk that brought down Fannie and Freddie, but rather a dramatic rise in credit losses.  Very few, if anybody, anticipated this.

Instead of the “expected” default rate of one-half of one percent, Fannie’s reached nearly 4% this summer.  The losses on these defaults increased from 20 to 50%.  Overall, the expected credit losses of 10 basis points ballooned to roughly 200 basis points.  This translated to Fannie and Freddie losing more than $100 billion combined in 2008, with an additional $100 billion in losses expected in 2009.  This is also why the Federal government had to put them into conservatorship in September 2008.  
  
So, what’s the lesson here for policymakers?  While the scale of these losses might be the result of reckless business decisions, it’s not clear that Fannie and Freddie would ever have survived a 30% decline in national housing prices.  In retrospect, it’s therefore difficult to surgically point to any one feature of their business model as the one thing to fix moving forward.  What is clear now is that the very existence of these federally subsidized financial guarantors with a combined $5 trillion exposure to nothing but mortgages must be called into question.   New plans for the GSEs that keep in tact Fannie and Freddie’s monoline status, where all they do is buy and/or guarantee mortgages, ignores one of the obvious lessons of the GSE collapse: the value of diversification.

Christopher Papagianis is the Managing Director of e21 and was Special Assistant for Domestic Policy to President George W. Bush.


e21 Projects & Partnerships