Can the US mortgage market be privately funded? October 6, 2013 at 9:13 am
Naked Capitalism links to testimony given to the Senate Banking Committee testimony by Georgetown law professor Adam Levitin. Levitin makes an interesting claim that private sources of residential mortgage funding
will be able to support no more than $500 billion of annual housing finance; [while] the US housing finance market needs anywhere between $1.5 trillion and $4 trillion in annual financing, depending on interest rate conditions.
Part of Levitin’s claim is based on the unusually high quality nature of post-crisis US RMBS deals:
The average mortgage size in these deals was nearly $825,000, over four times the national average. These deals were backed by ultra-prime collateral: the average loan-to-value (LTV) ratio on these deals was 65%, and only 0.5% of the mortgages had an LTV of above 80%.
I don’t find this particularly persuasive as one could make the claim that there is crowding out by the GSEs. Levitin does make the interesting point though that
The fully prepayable 30-year fixed-rate mortgage is a uniquely American and uniquely consumer friendly product that furthers economic stability and monetary policy. The 30-year FRM is the crown jewel of the American housing finance system. Its long amortization period lowers mandatory monthly payments. The fixed rate shields households from inflation and facilitates stabile household budgeting. The ability to prepay enables consumers to take advantage of improved rate environments and to pay down the mortgage faster if they have excess funds. And the prepayment feature greatly facilitates Federal Reserve monetary policy by enabling lower interest rates to easily translate into greater disposable income for consumers and increased consumer spending in the real economy. 30-year FRMs underwritten with full documentation did not blow up in the housing bubble. Any restructuring of the system should start with the question of how to ensure the widespread availability of the 30-year FRM.
History indicates that the private market will not produce 30-year FRMs in any volume.
Now, even if you believe Levitin that the private label market’s dislike* of 30 year fixed rate deals means that they will not supply sufficient funding for them, it is not obvious that the answer to the preserve the 30 year FRM at all costs. No other country has a residential mortgage market based on such long term fixed rate loans. Perhaps the US could look to the experience of others and possibly conclude that the 30 year FRM shares risk inequitably between borrower and lender. Five year FRMs are much more manageable for the lender and scarcely† less useful for the borrower. Rather than sanctifying a mortgage structure that can only survive with taxpayer subsidy, perhaps the key to US mortgage market reform is finding something that works for borrowers and lenders?
*This dislike is based on the large, not-fully-hedgeable prepayment risk of these deals. Mortgage prepayment models are not accurate, so you don’t know exactly how the loans will prepay, especially given that prepayment rates depend on default rates. This imprecision means that you don’t know whether a prepayment hedge of a private label RMBS will work.
†The WAL of a typical 80% LTV 30 year FRM is around 7 years in normal markets, so asking the borrower to take on average 2 years worth of rate risk does not seem unreasonable.