Opinions on GSE Culpability Conflict at Housing Conference

In mid-November thernFederal Reserve Bank of St. Louis sponsored a conference on the “Past,rnPresent, and Future of the Government Sponsored Enterprises (GSEs)”.  The invitation-only event featured sixrnpresentations on the title topic provided by participants from colleges andrnuniversities and one conservative and one liberal “think tank.”  The papers broke down into two categoriesrnwhich can be broadly described as “how did we get into this mess?”rnand “what can we do to get out of it?”  </p

Bank President JimrnBullard in his opening remarks said that the U.S. government has often modifiedrnthe structure of housing finance after a crisis, pointing to the expansion of mortgagesrnfrom the five to seven year non-amortizing balloons that predominated beforernthe Depression to 20 to 30 year, fixed-rate loans afterward.  It was the Depression that also led to therncreation of Fannie Mae.  </p

Bullard said thern”extent of Congressional meddling in this market has been astonishing tornthe point where one can barely identify what the private sector outcomes wouldrnbe in the absence of intervention,” and urged the government to let thernprivate sector provide the bulk of housing finance going forward.  It was “incentive-distorting”rngovernment programs and taxpayer guarantees that caused the current system torncollapse; programs that meant well, he said, but ended up costing everyone dearly.  “It makes little sense to try to designrnprograms that subsidize everyone.  Ifrneveryone is subsidized, then no one is subsidized.</p

Bullard laid out hisrnprinciples for reforming the process:</p<ul class="unIndentedList"<liGovernment subsidies to lower-income and firstrntime buyers should be disentangled from more broadly defined housing financernand the function regulated alongside FHA and Ginnie Mae.</li<liThe flow of credit to specific sectors of therneconomy should be done so as to shelter taxpayers from the risk of insolvency.</li<liExcess leveraging of home values must berncompensated with other factors such as default insurance.</li<liRecourse regulation should be considered similarrnto Europe's which does not permit discharge of mortgage debt.</li<liMortgage pooling should be constrained to loansrnwith similar characteristics and the financial intermediaries required to hedgernMBS portfolios through insurance or other means.</li</ul

In evaluating therncauses behind the failure of the GSEs, some conference presentations did notrnsupport Bullard’s theory about the “incentive-distorting” nature ofrngovernment intervention.  Jason Thomas,rnof the Department of Finance and Robert Van Order, holder of the Oliver Carr Chairrnin Finance and Real Estate, both of George Washington University presentedrntheir paper on Housing Policy, Subprime Markets and Fannie Mae and FreddiernMac:  What we Know, What we Think wernKnow and What We Don’t Know.  The twornsaid that the current narrative about the financial crisis tends to putrnparticular emphasis on the affordable housing goals that the GSEs were requiredrnto meet; the theory being that this pushed them into excessive risk taking andrninto making the market in subprime loans. rn”That Fannie and Freddie required an expensive taxpayer-financedrnrescue just years after they were identified as posing a systemic risk isrnconsistent with this line of reasoning.” rnThis, however, the pair said is not supported by their data.  </p

Their study foundrnthat the growth of the subprime market was largely a non-GSE phenomenon occurringrnoutside of normal mortgage origination channels and funded by non-agency orrnprivate label securities (PLS). rn”The GSEs did build a large portfolio of AAA PLS, probably inrnresponse to affordable housing goals, but such investments were unlikely tornhave had much of an impact on subprime mortgage origination, and were not arnlarge share of their credit risk.”</p

Where the GSEs didrnincrease their risk-taking it was not in pursuit of housing goals.  They did not purchase non-traditionalrnmortgages in any quantity until the market had peaked in 2005 and then thesernwere predominantly Alt-A mortgages made to borrowers with high credit scoresrnand substantial equity.  Subsequent losesrncan be attributed largely to the acquisition of these Alt-A and otherrn”prime-like” paper at the height of the bubble.  While they do not have the data to supportrntheir conclusion, they feel that a large share of the GSE losses were due tornproperty value declines.</p

Shawn Moulton,rnDepartment of Economics, University of Notre Dame also debunked a GSE-housingrncollapse narrative, saying there is little econometric evidence linking thern1992 GSE Act which established low-income housing goals to relaxed GSE orrnlender standards.  Moulton uses loanrnapplication level data to examine whether the affordable housing goals alteredrnthe probability that 1) a loan application is originated, 2) an originated loanrnis purchased by a GSE, or 3) an originated loan is a sub-prime orrn”high-price” loan.  He alsornused census-tract level data to examine the effect of the GSE Act onrnforeclosures, vacancies, high price loans, and other housing outcomes.</p

The three goals setrnforward by the GSE Act define the percentage of GSE purchases that must be fromrn</p

 a. very low-income borrowers and low-incomernborrowers living in low-income areas – the Special Affordable Goal (SAG);</p

b.  lower incomernborrowers – the Low and Moderate Income Goal (LMIG);</p

c.  low income andrnminority neighborhoods – the Underserved Areas Goal (UAG).</p

Moulton found thatrnthe SAG increased GSE purchases from very low-income borrowers by four percentrnbut had no effect on mortgage lending and found no evidence that the LMIG orrnUAG altered purchase or mortgage lending decisions.  Finally, using the census tract-level data hernfound no relationship between the GSE Act’s goals and increased foreclosures,rnvacancies, or other housing outcomes.  </p

Moulton concludes,rn”As I find no evidence that the GSE Act increase loan originations, Irnconclude that the GSE Act could not have caused the recent housing crisis.  Hence, attention should be moved away fromrnthe GSE Act to other potential causes of the crisis.”</p

Dwight M. Jaffee ofrnthe University of California, Berkeley, however, does blame the GSEs and theirrnhousing goals for a portion of the problem. rnHis research found that what he defined as high-risk mortgages made uprnabout 43 percent of the total mortgage acquisitions by the GSEs in 2007,  This domination of the market for high-riskrnmortgages is, he said, just the opposite of what would be expected ofrngovernment sponsored enterprises with a public mission to stabilize thernmortgage market.  “In fact, the GSEsrnpiled onto an already highly overheated market, thus creating terrible consequencesrnfor themselves, the overall housing and mortgage markets, and therneconomy.”</p

Jaffe said that thernGSEs took these risks, even at a time when the extremes of the housing marketrnwere already apparent, because subprime mortgages offered above averagerninterest rates and the GSEs felt they could evaluate and tolerate the risk andrninvestors in GSE debt and MBS felt protected by the implicit guarantee.  In addition, the expansion of the subprimernmarket drained the prime market and the GSEs were unable to maintain theirrnusual portfolios which caused a decline in their stock price.  The aggressive acquisition of sub-primernmortgages was an attempt to regain market share.  Finally, the acquisition of high -risk mortgagesrncould help the GSEs reach their housing goals.</p

This risk-takingrnbehavior was fully predictable, Jaffee maintains, because of the incentivesrnavailable to the GSEs.  The implicitrnguarantee allowed them to issue almost unlimited amounts of debt and MBS atrninterest rates that exceeded comparable Treasury debt by small spreads.  In this situation, enterprise profits andrnmanagement bonuses could be maximized by expanding the volume and degree ofrnrisk-taking.  </p

At the same time,rngovernment studies vastly underestimated the cost to the taxpayers of the riskyrnbehavior with most studies, even those as late at 2008, projecting that futurernsubsidies to the GSEs had only a minuscule chance of exceeding $100rnbillion.  The losses are now expected to exceedrn$200 billion.  </p

Jaffe and threernother speakers presented conference attendees with a variety of suggestions forrnrestructuring the GSEs.  Those will bernsummarized in a future article.

All Content Copyright © 2003 – 2009 Brown House Media, Inc. All Rights Reserved.nReproduction in any form without permission of MortgageNewsDaily.com is prohibited.