Will the U.S. federal government ever exit mortgage finance? Not any time soon. Let us explain why.
In September 2008, as investors shunned the debt of Fannie Mae and Freddie Mac, the U.S. Treasury put these government-sponsored enterprises (GSEs) into federal conservatorship, kicking off the most intense months of the financial crisis. Not long after, the CBO estimated the fair value of the GSEs' losses at $291 billion (or more than 5% of their end-2009 mortgage portfolios).
At the time, it was widely agreed that the GSEs should change, possibly even winding them down completely. In 2011, the Treasury proposed reforms to do just that. The plan (and hope) was that we could restore private involvement (and risk management) in the mortgage market, and protect taxpayers.
Well, as of end-2013, the fraction of outstanding residential mortgages accounted for by the two GSEs had reached a record 60.6% (see the red line in chart below). With the exception of the private securitization boom in the three years before the crisis, their market share has been rising for decades.
Residential Mortgages by Owner (Dollars in trillions, right, stacked) and Share of Mortgages Accounted for by GSEs (percent, left, line)
Source: Table L.218, flow of funds database, Board of Governors of the Federal Reserve System.
At the margin, the federal government’s role is even greater over the past few years. Collectively, federal agencies – including the GSEs, the Federal Housing Administration (FHA), and the Department of Veterans Affairs (VA) – backed 87% of mortgages issued in 2012, up from 35% in 2006 and 47% in 2000 (see report and handout of the Bipartisan Policy Center Housing Commission).
Last week, Treasury Secretary Lew pleaded for reform in a WSJ interview: “One shouldn’t wait until there’s a crisis to deal with this. We ought to deal with it now. […] the current system is one where the risk that taxpayers are bearing has never been properly priced and is open-ended.” (WSJ April 28, 2014) Yet, the Senate Banking Committee again delayed action on a limited reform bill. Even if the Committee acts, broader legislative gridlock casts doubt on serious change.
Of course, a long list of people benefit from the status quo. There are the GSEs’ own employees and managers; real estate brokers; mortgage originators and servicers; builders and construction workers; and last, but hardly least, the homeowners who can borrow up to $625,500 (the current limit for conforming loans in high-cost areas) at a below-market rate. Collectively, the CBO estimates that GSE backing of new mortgages over the period 2015-2023 will result in a subsidy averaging around $2.5 billion per year. The recipients of these benefits are so numerous that is hard to imagine how you would start to organize effective opposition.
One might hope (as policymakers did) that the long-term GSE subsidy boosts U.S. home ownership (recall President George W. Bush’s “ownership society”). Yet, despite the massive federal interventions, the pattern over the past two decades shows little sustained gain (see chart below) in the rate of U.S. home ownership, which dropped in the first quarter of 2014 to 65.0% -- the lowest level since 1995. Moreover, despite the relatively high U.S. mortgage debt/GDP ratio and the extent of direct government intervention in mortgage finance, the U.S. home ownership rate is unexceptional among advanced economies (see Figure 1 and Table 1 in this report). Switzerland and Germany, with homeownership rates of 44% and 53% stand out, as do Italy and Spain, at 74% and 79% respectively.
U.S. Home Ownership Rate
Sources: FRED and Census.
If boosting home ownership were truly the policy goal, focusing the federal subsidy on entry-level homeownership (or, at least on houses below the March 2014 median sales price of existing homes of $198,500) probably would be more effective. And, to make the subsidy less risky for the financial system as whole, policymakers could transform it from a debt subsidy to an equity subsidy of roughly equal size. Imagine, for example, if the federal government were to match $1 for $1 for low-income families that had saved 10% of the value of a starter home (call it a 21st century land grant), giving them enough capital to put 20% down. Keeping the annual cost at $2.5 billion, that would allow the government to subsidize 200,000 purchases of $125,000 homes every year. With a bigger equity cushion, the resulting mortgages would be less risky and easier to securitize. But then, this is a solution only an economist could love.
We all know about the dangers lurking in the current mortgage finance system. Long before the financial crisis, Fed officials warned about the systemic risk associated with GSEs who, by 2001, had assets equal to 40 times their equity. More recently, academics highlighted the GSE’s precariously leveraged balance sheets that made them Guaranteed to Fail. And other research highlights the role of household leverage in prompting the deepest economic slump in U.S. postwar experience.
Yet, it is hard to see how this awareness of systemic risks and economic fragilities could translate into serious reform anytime soon. The GSEs are now making profits again (as the Fed’s monetary stimulus boosts the incomes of mortgagees and the value of their homes), reducing the pressure on Congress to act. And, more generally, there has been little legislative progress on financial reform since the 2010 Dodd-Frank Act, which failed to address the GSEs (or the MMMFs, about which we wrote last week). Finally, any serious effort to wind down the debt subsidy is likely to widen the yield spread of mortgages over Treasury bonds, raising conventional mortgage rates. What politician wants that?
Perhaps governments like private debt as much as public debt, so long as it postpones problems for their successors to address. One would have thought that the financial crisis just past would have taught us the cost of such procrastination!