US Housing Finance: Let’s Put Quality Before Quantity

Douglas J. Peebles

The US government’s housing finance policies in recent decades can be summarized by one simple phrase: quantity over quality. The implicit goal was to increase the quantity of housing finance by keeping mortgage rates low and promoting wider home ownership. For several decades, the system worked. But if we view the long-term stability of home prices as one measure of the quality of housing finance, those policies now don’t look so successful.

In our view, this prioritization of quantity over quality has had unacceptable consequences. Taxpayers have been saddled with losses at Fannie Mae and Freddie Mac, and the price of a house—the biggest store of value for the vast majority of American families—has fallen nationwide for the first time since the Great Depression. 

So what’s the answer? As argued in my last post on this topic, a housing market without any government involvement at all is not a realistic option. Nonetheless, it is clear that the private sector needs to take the reins, going forward. My colleagues Michael Canter and Matthew Bass argue in their recently released white paper, Increasing the Role of Private Capital in the Mortgage Market, that there are two key principals that should govern the transition.

First, to protect taxpayers, private capital needs to provide a significant buffer against losses before a government guarantee kicks in. Second, to reduce volatility in housing prices, this capital should be unleveraged. Let’s look at each of these principles in turn.

Historically, the US government underwrote 100% of the risk in agency mortgage-backed securities issued by Fannie Mae, Freddie Mac and Ginnie Mae. Although this government backstop may have helped keep mortgage rates low, it subjected taxpayers to risk, since the government had no way of pricing this “insurance” correctly.

In our view, a better approach would be to require that private investors price and invest in the “first loss” risk in mortgage securitizations—the first piece of the transaction that is at risk of loss when homeowners default on loans in a mortgage pool. The government guarantee would come into play for the rest of the securitization only after private investors’ first-loss capital is completely wiped out. If the first-loss piece is appropriately sized relative to the underlying quality of the mortgage loans, the first-loss capital should be wiped out only in extreme stress scenarios.

Transparent, market-based pricing of the first-loss capital provided by private investors would help the government price its catastrophic insurance appropriately, reducing the risk to taxpayers.

But perhaps most importantly, in order to minimize volatility in the housing market, the private capital should come from a stable base of long-term investors, such as pension funds and sovereign-wealth funds, as opposed to leveraged investors, which tend to be more flighty because their investment decisions are largely driven by the availability of financing. A strict prohibition on investors’ ability to leverage the first-loss piece would help facilitate this change.

This would also effectively end “credit tranching” of the first-loss risk in mortgage securitizations. In the years leading up to the recent financial crisis, nonagency (or private-label) securitizations were commonly tranched, or sliced up into different classes of securities that had a differing order in the priority of claims on interest and principal repayments. Often, some of the “subordinated” tranches, which are inherently leveraged, were further repackaged into collateralized debt obligations (CDOs), magnifying the leverage even further. The end result of this system was that when home prices declined and mortgage losses escalated, the availability of credit became volatile and leveraged institutions became fragile. The ultimate impact was extreme housing price volatility.

Although low mortgage rates are an admirable goal, we believe that price instability for housing is too high a price to pay—for individual homeowners as well as taxpayers. Under the privately led system that we propose, mortgage rates are likely to rise somewhat, but in our view, this would be an acceptable tradeoff for a more stable housing market. How could the transition to a privately-led system actually happen? I’ll have more to say about that in a future post.

The views expressed herein do not constitute research, investment advice or trade recommendations and do not necessarily represent the views of all AllianceBernstein portfolio-management teams

Douglas J. Peebles is Chief Investment Officer and Head of Fixed Income, Michael S. Canter is Director of Structured Asset Research and Portfolio Management and Matthew D. Bass is VP— Structured Assets, all at AllianceBernstein.



  1. Thomas Boyle

    The decline in house prices is a debateable loss of real value to American families: it”s effectively a wealth transfer from families who already own homes to the families who will buy the homes from them in the future. It”s a zero-sum change: the houses still exist, and no actual physical assets or utility have been lost.

    Let”s stipulate that the government has a role in enforcing contracts and punishing fraud. Beyond that, it”s not remotely clear it has any useful or even legitimate role in providing financing for home buyers.

    Indeed, it”s arguable that taxing third parties to provide government guarantees that reduce the mortgage costs for home buyers, is neither desirable, nor equitable. It”s government support of rent-seeking by homeowners and lenders, yes, but hardly good public policy.

    • Matthew Bass, at AllianceBernstein

      I would argue that the decline in home prices is not a zero-sum game.
      Every asset has a corresponding liability and the increase in home (an asset) prices during the housing bubble was largely financed with increasing amounts of debt (a liability). Since the housing bubble burst, homeowner equity—or the difference between home values and the mortgage debt —has decreased by nearly $7 trillion. This is not a transfer of wealth, but a destruction of household wealth that may have only existed on paper, but that still has has had knock-on effects on overall economic activity—most notably through consumer spending, which represents approximately 70% of GDP.
      This destruction of household wealth has also constrained credit availability: banks are less apt to lend given mortgage debt writedowns. Furthermore, many homeowners are unable to refinance and take advantage of today””s very low mortgage rates, because they do not have any home equity.
      This is why we need to put quality before quantity in rebuilding housing finance. It wouldl build a stronger foundation for the market to grow with less price volatility.

      • Thomas Boyle

        This gets us into a discussion of the difference between paper wealth and “real” wealth. Real wealth is resources you can eat, sleep in, stay warm with, be entertained by, etc. In the end, it is real wealth that matters; you can print paper wealth all day, and it makes no-one any better off, as we all agree, I think.

        When house prices collapse, no real wealth is lost; there is no loss of economic capacity or functionality. This is very different to the effect of, say, an earthquake, or crushing viable used cars, which actually destroy real resources and are not zero sum. Of course, the creation of real wealth – by creating a new service or building a house, say, is also not zero sum. But a change in the price of houses is neither wealth creation nor destruction; it is a re-pricing of houses relative to everything else.

        When house prices climb, homeowners wind up with the ability to convert their homes into a greater amount of other types of real resources (food, for example, or air travel), and non-homeowners need more other resources to trade for a home. When they fall, the reverse happens. Claims on resources shift between homeowners and non-homeowners, but the resources are not themselves created or destroyed. It”s economically zero sum to society, even if, in dollar terms, it”s not.

        Turning to spending effects, people who sell their homes at lower prices in the future will spend less afterward, yes, but those who buy the same homes from them at lower prices will subsequently spend more than they would have if they were paying off a more expensive home (truly zero sum, even in dollar terms). People who don”t sell their homes, but experience a loss of paper wealth, may also spend less as they try to rebuild their “savings,” but those who have not yet bought homes will be able to spend more, as they will not need to save as much in anticipation of buying a home (again zero sum, even in dollar terms). Future consumption of real resources has shifted between homeowners and non-homeowners, but it”s zero sum; no real resources have been created or lost.

        What about lenders? To the extent that a lender, rather than a homeowner, put money into the property, the homeowner”s loss becomes the lender”s loss. So instead of the change in house prices becoming a real wealth transfer between homeowners and non-homeowners, it becomes a real wealth transfer between lenders and non-homeowners.

        Now, I recognize that there are frictions, and that sudden shifts in people”s relative real wealth requires a period of adaptation, during which time some real opportunities will be lost. But that”s got to be pretty second-order compared to the “loss of wealth” we”ve all been hearing so much about.

        Thanks for the discussion!

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>