Today we are building a new system of housing finance from the ashes of the system that crumbled in the last decade. Yet, we began this endeavor before we had a blueprint completed—we have laid pieces of the foundation before we know the design of the first floor, let alone the shape of the roofline. What is more, to over-extend the metaphor, the ground upon which we are building is shifting. As a consequence, I don’t think we will have a grand reform moment, but rather a series of discussions and incremental change.
As part of those discussions, we must continuously ask: in this new system, will homeownership continue to offer the advantages of physical security and psychological identity? Will it continue to offer economic opportunity—the opportunity to build wealth and assets that make so much else possible? And, if so, who among us will gain those benefits?
I fear we are expecting homeownership to accomplish too much: we’re asking it to close gaps in incomes and opportunities, and to address economic volatility—challenges that have deeper roots than housing. Homeownership cannot be the route to overcoming growing inequality and its persistence along racial lines. But if the new system cannot erase these inequalities, it is imperative to ask if it will exacerbate them—if it will calcify the divisions between the “asset haves” and “asset have nots,” and the “income haves” and “income have nots.”
I wish I could tell you what the future will bring, but you should be highly skeptical of anyone who has great confidence about how sweeping policy change and shifting socio-economic factors will combine to determine the future, as we are in a period of profound policy change. Underway now are the implementation of new Basel capital standards, which are changing financial institution behavior; the implementation of hundreds of Dodd-Frank provisions (most relevant QM and QRM, but inevitably others) that will also have consequences we cannot predict; efforts to restore the FHA Fund’s capital levels, which will either strengthen or hobble FHA; and the creation of a new “architecture”—including servicing standards and a new securities platform to segregate the MBS issuance and credit risk functions. The elephant in the room is the impact of what comes after Fannie and Freddie as they are (as the Administration likes to say) “unwound.”
As the principal focus of policymaking right now is on avoiding excess leverage and systemic risk, and not principally about ensuring affordability and access to credit, a key question is how access and affordability will be pursued in the new framework. Some mechanism will be politically necessary, but will they be sufficient? Will they be a fig leaf? Can access be done in a non-distorting way?
Beyond the future housing finance system, what will be the consequences of changes, if any, to the Mortgage Interest Deduction and broader tax reform? And what about likely further reductions in domestic discretionary spending?
The U.S. is also experiencing transformational socioeconomic and demographic shifts that will fundamentally affect housing demand and the effectiveness and desirability of housing as a wealth-building tool. As the Joint Center and some of my own colleagues at the Urban Institute and the Bipartisan Policy Center have so ably documented, household formation patterns are changing, our senior population is exploding, and our immigrant population is both increasing and suburbanizing.
For example, children are increasingly born into non-marital families: 40 percent of births are to unmarried moms, and these percentages vary dramatically by race. And while in some of those cases, a second adult is present, absent marriage, co-habitation doesn’t usually last. The result for homeownership could be profound, as single-earner households face a narrower set of housing choices.
We are also seeing significant generational differences in wealth accumulation. Just last month, Urban Institute scholars including Signe-Mary McKernan, Gene Steuerle, and others published work showing that even with the Great Recession, over the last forty years, people are wealthier than their parents’ generation—unless they were under forty. Today’s young people, in particular the cohort in their thirties, have accumulated far less wealth than their parents over a comparable period of time. Some of that stems from coming of age in the era that led to the housing collapse, but not all. Another challenge is that younger people without college degrees have less predictable earnings.
Sadly, that same economic volatility affects many of those with a college degree in the 21st century economy. Employment is far less secure. And deeper trends in the economy suggest that middle skill workers will not see any end to the wage stagnation of the last few decades, giving rise to questions about the ability of more of the workforce to sustain homeownership.
Minorities—who historically have had lower levels of educational attainment and thus incomes—make up a growing share of the U.S. population. While educational attainment is increasing, it is unclear whether these rising levels will overtake the effects of shifting population composition, or if the housing market demand will be shaped by lower-income consumers in the long term.
And an additional challenge is the aging of the population and its consequences for demand for different kinds of housing units. We just don’t know how all of these trends will intersect with policy change.
There are also three trends in the mortgage market – partly policy-driven, but not exclusively – that keep me up at night: underwriting rigidity, the cost of credit, and the role of down payments.
In terms of underwriting, we have gone backwards. We now have limited flexibility with little consideration for compensating factors. Rigid rules have replaced rules of reason. This change, which falls disproportionately on lower income households, recent immigrants, retired Americans and the self-employed, comes from a confluence of regulatory policy and private sector incentives. Well-intentioned individuals who committed to avoiding the mistakes of the bubble years are overshooting. One statistic stands out: I am told the average loan that is turned away by GSE automated underwriting systems has an average credit scores well north of 700 and average LTV below 80. Manual underwriting is a shallow promise in the era of rep and warranty concerns. So we see overlays based on rigid ratios not only on GSE loans, but also FHA.
A second issue is that we haven’t really absorbed the extent to which the cost of credit is going to rise in the years ahead. When we fully price the value of the guarantee – especially as there will be pressure to price it for another “100-year flood” or, in this case, financial catastrophe, the real cost of credit is going to rise A LOT.
There is also pressure in Congress to change the rules of how we estimate what credit costs taxpayers. “Fair value accounting” will make the accounting for government credit risk more expensive, which will lead to higher premiums and guarantee fees. Even today, before these changes, despite low interest rates, the cost of credit is not as inexpensive as it seems. Today, the primary-secondary market spread is 50 to 100 bps higher than its historic levels.
And then there are risk-based premiums. We’ve always had pooling of risk, but we can expect much less in the future. While we have traditionally had price banding—pooling of risks but not a single pool, we are moving toward more granular pricing, some at the expense of credit-impaired borrowers.
A real challenge for those who care about affordability is managing in a world of greater risk-based pricing. We have built a financial system that prices credit based on risk, yet pools or average prices prepayment risk. Low-wealth individuals are the losers on both sides of these trades. We need more research to provide greater transparency on the consequences of risk based pricing.
I am also concerned about learning the wrong lessons about consumer leverage and downpayments. I accept that loss severity is greater with lower equity, but am less convinced that proclivity to default is affected by LTV. Further, there is a difference between what drove the crisis and what may have exacerbated the losses suffered: when unemployment rises so dramatically and prices fall 50 percent, does it matter whether current LTV is 190 or 180?
Going forward, we need to understand how combinations of factors influence performance and create prudent regimes that allow for other factors to compensate for lower equity. I also am attracted to testing the concepts discussed yesterday that build in savings, first in the lead up to ownership with a promise of mortgage at the end, and throughout ownership to provide the necessary cushion against unexpected costs of loss of income.
We also need a way to lower the cost of capital for innovation, to provide additional credit enhancement for innovative products and services. The Center for American Progress’ and Mortgage Finance Working Group’s reform paper has a concept known as the “Market Access Fund” which we would fund out of an assessment on all MBS – guaranteed or not—and that can be delivered through a combination of high performing state and local HFAs and competitive allocations to innovative products by MIs, CDFIs, or other nonprofit and private capital providers. Some will say, isn’t that FHA’s role? I have long hoped it would be so. But I think we need an additional strategy – a nimbler and more flexible tool – as I have little confidence Congress will ever stop writing FHA underwriting standards in statute.
Finally, we face a pressing need for better information – particularly, more analysis and quick learning that connects the seemingly disparate macro and micro economic forces that determine a household’s ability to purchase a home and save. In addition, much of the impact on families and communities is determined by capital markets, secondary markets, and financial institution regulatory policy, and not the retail consumer and mortgage issues that advocates for affordable housing typically focus on. Yet the gulf between the two worlds is so great, and the language of analysis is so different. If we could manage to better translate between the two, and create a body of shared knowledge, I am confident we would be in a much better position to reimagine and build our housing system of the future.
In closing, we have only just begun to reconstruct our national housing finance system, and we face tremendous uncertainty around the forces that will affect it. Will homeownership continue to serve the critical asset-building role it has for 75 years? And if so, how do we create a system that enables access to it in the face of volatile economic and social dynamics? We need to show how we can do access without increasing risk to the system.
If anyone can determine the best way forward, it is the minds assembled here today. You all have not only the brainpower, but the commitment to affordability and access. Thanks again to Eric, the Joint Center, the Ford Foundation, NeighborWorks America, and Bank of America for bringing us together today.