What role did the housing goals play in the demise of Freddie and Fannie? How risky were those targeted affordable loans, really?

In place for more than 15 years by the time the housing bubble popped in 2007, the GSE affordable housing measure how many units of affordable housing (based on different criteria and definitions) were financed by the GSEs in a given year, as a percentage of their overall business. The goals are set years in advance based on estimates of what the mortgage market will do in the future and are not easily adjusted if things don’t shape up as expected. So, even if the goals were unrealistic in a particular year, the GSEs would go to extraordinary lengths to meet them, lest they be publicly shamed…

To meet the tough regulatory requirement, Freddie needed to buy thousands and thousands of goal-qualifying mortgages year in and year out. Thanks to Fannie’s larger size, long history and influence, it was better at attracting those loans. Freddie’s traditional reputation as a tough cookie when it came to managing credit risk didn’t help us in the affordable department.

So Freddie struggled. Most years, Freddie managers would wait until near the end of the calendar year to see how far short the firm was from meeting the annual affordable housing goals–and then there would be an all-out hunt to purchase the loans needed to cover the shortfall.

If the firm could get them.

After a few years of this behavior, lenders realized that having the two GSEs competing for the same subset of mortgages would bid up the prices at which they could sell them. Lenders noticed that the GSEs started getting desperate for affordable loans by around October. Then they could start calling the shots. There were even suspicions that banks would withhold selling goal-rich loans earlier in the year so they could jack up the price in the fourth quarter when they knew Freddie would pay just about anything to keep from missing the goals. One year, I recall talking to a stressed out manager on New Years’ eve who was calling lenders looking for loans to buy.

But in 2003, there simply weren’t enough affordable loans to go around. The refinance boom was in full swing, which meant that most of the mortgages being originated were for refinances, which tended to go to higher-income borrowers whose loans didn’t count toward the goals. The goal levels had been concocted years before assuming an average refinance share of the market of 15 percent; in 2003, the refi share soared to over 50 percent.

There was no way we could make it.

Ironically Freddie Mac’s huge refinance business was allowing millions of borrowers to lower their mortgage rates, a welcome jolt to the sluggish post-9/11 economy. And yet Freddie was in danger of failing its “mission” by not meeting the affordable housing goals, as a share of our total business.

Goal failure was costly. Depending on the level of infraction, penalties ranged from having to develop and submit a business plan for reaching the goals in the following year to paying monetary and civil penalties to HUD. However, the real cost of missing a goal was not financial but political. If the goals were, as Fannie Mae had shrewdly foreseen, the price of admission in terms of maintaining congressional support for the GSE charter, then missing a goal threatened to chip away that support…

As dysfunctional as the goals came to be, however, they were not the sine qua non cause of the housing crisis, in contrast to what some have argued. Many of the very worst loans purchased by Freddie Mac did not even qualify for the goals, but were loans to higher income borrowers in a bid to compete with Wall Street, which was under no such regulatory directive.

Mortgages made to borrowers of modest means did not, by in large, bring down the U.S. housing finance system. That said, the affordable housing goals did contribute to a sea change in GSEs’ and others’ views of what constitutes good underwriting, and eased the transition to even greater levels of risk posed by non-traditional (e.g., low or no documentation) and subprime mortgages. 

While no one in Washington wants to repeat the failed social experiment of numerical housing goals, there remains a stubborn refusal to learn from the past. To call for a return to traditional underwriting standards is out of favor once again. Bankers are against that. Home builders and real estate agents are against that. Consumer groups are against that, as are an increasing number of lawmakers.

The ritual drubbing of anyone who tries to bring up the need for underwriting restraint indicates that disparities of income and wealth, as well as the explosive issues of race and immigration, lie simmering beneath the surface unaddressed.