The clamoring for an increase in GSE loan limits is getting deafening, with the National Association of Homebuilders, the Housing Policy Council of the Financial Services Roundtable, and the National Association of Realtors issuing press statements in the last few days that call for legislators to raise the conforming lending limit. The logic between all three groups is pretty much the same -- borrowers need access to more credit, we're staring at a recession because of credit problems, more access to credit helps borrowers and the economy, and the GSEs are in a position to provide that credit. (Feel free to read the press releases linked above to get similar sounding quotes, if you'd like.) HW readers may recall last week's coverage of the Office of Federal Housing Enterprise and Oversight's position on the issue of raising the conforming loan limit, which suggested that credit availability may actually suffer as a result:
OFHEO said that “newly eligible mortgages originated in high-cost areas might pose greater credit risk on average than loans now purchased by Fannie Mae and Freddie Mac,� which would lead to even further hikes in GSE guarantee fees. Both GSEs have hiked their so-called g-fees in recent months in response to deteriorating industry conditions. Further increases would likely have the effect of reducing available credit to borrowers during a time when more credit availability is what is ultimately needed, according to the report.
There's much more to OFHEO's argument, including some technical points regarding how secondary markets work. Linda Lowell, who writes a column over at Market News International, highlights one such technical point (no link available):
... OFHEO wonders whether the newly eligible jumbo loans can be placed in pools that trade TBA and can be advantageously financed in the "dollar-roll" market (MBS dollar-rolls are a lesson for another day). This issue is very important, because only the very largest, most standardized programs trade TBA or to-be-announced. TBA pools are considered to be fungible in the sense that the underlying loans have similar characteristics and hence similar expected prepayment behavior. Large amounts of these pools outstanding provide liquidity. Based on my experience, MBS investors and traders would refuse to allow pools containing these loans to trade TBA and insist the agency sequester them in their own pool type. The result would be a very small security universe consisting of only one vintage year. Investors would rightly expect it to trade "way behind" TBA. The resultant wider yields translate into higher borrower rates.
I'm not patently against raising the conforming lending limit, although my hunch is that doing so is probably a bad idea; I'm guessing that higher-limit loans in this case are synonymous with higher-risk loans, at least until the housing market can rebalance itself. The point here is that any discussion of raising the conforming lending limit needs to be framed within a larger analysis of risk -- to borrowers, to lenders, to investors -- and within an strong understanding of how the primary and secondary mortgage markets interrelate.