To prevent an FHA bailout, lower the conforming limit on GSE loans
In a recent post, I noted that FHA mortgage delinquencies skyrocketed more than 25%. Since most FHA borrowers only put 3.5% down, when factoring in a 6% commission, 2% closing costs, and a declining market, nearly all FHA borrowers over the last five years are effectively underwater. When these borrowers sell or quit paying, the losses will be huge because the capital recovery will be far less than the original loan balance. With a 9.4% serious delinquency rate, the FHA is facing 713,104 future foreclosures. This rate has been rising steadily, and many more delinquencies are coming because many borrowers will want to move before prices regain peak values. Many borrowers may opt for a strategic short sale or strategic default, after all, they don’t have much skin in the game, so it’s easy for them to walk away. The FHA insurance fund has far less in reserve than its federal mandate, and many observers are concerned a bailout of the FHA is inevitable.
To avoid a federal bailout, the fund must be replenished quickly with user fees. To this end, the FHA recently raised the fee it charges at origination and the fee for ongoing insurance. Even these measures may not be enough. To raise more money, the FHA must abandon its mandate to provide loans to low and moderate income borrowers and tap into the market for high wage earners. After the fund stabilizes, the conforming limit for FHA loans can be lowered again and the FHA can return to exclusively making loans for low to moderate income borrowers.
In 2008, the conforming limit for both GSE and FHA loans was raised from $417,000 to $729,750 as non-insured private financing evaporated. In late 2011, the conforming loan limit for high income areas like Orange County or the San Francisco Bay area was lowered on GSE loans from $729,750 to $625,000. What perplexed many at the time was that the conforming limit for FHA loans was not similarly reduced. Why did the government do this? They too recognized the need to get high wage earners into the fund.
Ever since the change, many of Shevy’s clients faced the cost of financing dilemma this change creates. If the borrower uses a $625,000 GSE-backed loan, even with less than 20% down and private mortgage insurance, the cost of ownership is significantly less (PMI is half of the FHA insurance fee). The cost of the incremental dollars when the borrower goes from $625,000 to $729,750 with less than 20% down is significant (nearly $500 per month) because the FHA insurance is much higher. The borrower pays the extra money into the FHA insurance fund. Anywhere high wage earners are using FHA loans — and many are — the FHA is getting a considerable revenue boost.
Why not do more? If we really wanted to get more high wage earners to pay FHA insurance, why not lower the GSE conforming limit further?
Think about what would happen if the GSE conforming limit were reduced back to $417,000 and the FHA limit were maintained at $729,750. Anyone putting less than 20% down on a house between $450,000 and $750,000 — which is a huge portion of the coastal markets in California and New England — would be pushed into FHA loans. If these markets really have bottomed — and that’s a big “if” — these new high wage earning borrowers would not cause large losses, and the increase in FHA insurance payments would be dramatic.
Further, if the conforming loan limit on GSE loans were reduced, it would also help reduce the size of their operations and make it easier to someday dismantle them. As long as the GSEs are insuring more than half the housing market, there is no way we can reform or eliminate them.
Lowering the GSE conforming loan limit will significantly increase FHA insurance fund revenues while simultaneously reducing the footprint of the GSEs. Don’t be surprised if you see this happen sometime soon. If politicians don’t do something like this, they will have to bail the FHA out, and none of them want to face that problem.
[gview file=”http://ochousingnews.g.corvida.com/wp-content/PDF/huddoc.pdf” height=”675″ width=”550″ save=”1″]