Tipping Points III: Debt-Financed Homeownership – Its Evolution, Impact, and Future

Center for Household Financial Stability, Federal Reserve Bank of St. Louis

On Friday, Oct. 12, 2018, the St. Louis Fed’s Center for Household Financial Stability and the Private Debt Project convened, in partnership with the Financial Security Program of The Aspen Institute, a one-day, invitation-only research symposium in Washington, D.C., entitled, “Debt-Financed Homeownership: Its Evolution, Impact, and Future”. The papers, discussions, main themes and key questions for future consideration are summarized here.

The symposium addressed, through commissioned papers from leading academics and discussion, four key issues:

• The evolution of homeownership from a relatively stable low-risk investment to a higher-risk financial asset;
• How highly leveraged homeownership came to be the norm in the U.S., and its effect on household balance sheets over time;
• The trade-off between the policy goal of expanding homeownership as a tool for reducing poverty and building wealth, and the risks entailed in navigating increasingly complex and cyclical housing and mortgage markets; and
• The policy reforms necessary for achieving affordable housing and financial stability in the future.

This symposium was the third in the “Tipping Points” series, which has explored when household debt “tips” from being productive and wealth-building for families and the economy to harmful to both. The summaries and papers from the first two symposia also may be found on the Private Debt Project site.

Finally, while not a formal part of the symposium, a public keynote conversation (video)—featuring Sarah Rosen Wartell of The Urban Institute, Paul Willen of the Boston Fed, and Michael Stegman of the St. Louis Fed’s Center for Household Financial Stability—was also convened at The Aspen Institute that day to explore some of our nation’s key issues around homeownership.

Framing Paper

Emmons et al. Kicking-off the symposium was a framing paper by William R. Emmons, Ana H. Kent and Lowell R. Ricketts of the Center for Household Financial Stability of the St. Louis Fed. The paper begins with the premises that debt-financed housing was central to the recent financial crisis and subsequent Great Recession; that millions of leveraged homeowners lost trillions of dollars of wealth, resulting in the near collapse of the global financial system; and that, accordingly, our system of highly leveraged homeownership deserves critical scrutiny. At the household level, the authors identify three key features of housing leading to the crisis: (1) a widespread preference and policy support for ownership over renting; (2) external financing exclusively in terms of debt (rather than equity); and (3) a relatively high amount of leverage, especially among ex ante financially vulnerable families. The collapse of the housing market proved, they demonstrate, to be profoundly damaging for many of the families concerned. They also show that, at the macro level, a boom-bust cycle has emerged as a result, at least in part, of government policies of three types: (1) tax preferences for owner occupation and debt financing; (2) an asymmetric monetary policy reaction to economic downturns known as the “Fed put,” which supported a private-sector debt buildup; and (3) financial liberalization, which contributed to large housing bubbles and financial instability by increasing typical loan-to-value ratios. Despite clear links between highly leveraged homeownership and severe financial crises, Emmons et al. believe that little serious discussion of changing our housing-finance system has taken place. As a first step towards reform, the authors discuss modifications to the tax code that could support alternatives to leveraged homeownership.

Panel One: Homeownership and the Household Balance Sheet

The first panel on homeownership and the household balance sheet featured two papers:

Ghent. The first paper, by Andra Ghent of the University of Wisconsin-Madison, documents patterns in U.S. mortgage debt and homeownership in recent decades and explores how well changes in mortgage credit constraints can explain these changes. In particular, she explores the effect of relaxing and tightening mortgage credit on homeownership and mortgage debt levels before and after the 2007 crisis. She finds that the age-adjusted homeownership rate actually declined slightly between 2000 and 2007, but the real stock of residential mortgage debt doubled. Since 2007, age-adjusted home ownership rates have trended down significantly as has the real stock of residential mortgage debt. From this, Ghent concludes that the relaxation of the loan-to-value (LTV) constraint cannot explain the housing boom but that the tightening of the LTV constraint does explain some of the decline following the crash. The main effect of the tightening of the LTV constraint has been to reduce the home ownership rate of young, high-income households. Based on these empirical observations, Ghent builds an equilibrium life-cycle model of tenure and mortgage choice that shows that the relaxation of the LTV constraint increases the homeownership for this group of households.There is a large and growing literature, including the paper presented by Mian, Sufi, and Verner for the 2017 Tipping Points symposium, exploring these questions. Also see Emmons et al. from this symposium for a review of the evidence on the question of how financial liberalization affected homeownership rates and/or house prices.

Bartscher et al. The second paper, by Alina K. Bartscher, Moritz Kuhn, Moritz Schularick, and Ulrike I. Steins, all of the University of Bonn, was presented by Moritz Schularick. Using seven decades of household-level micro data, the paper makes the first systematic attempt to explain the household debt boom in postwar America. Overall, the authors find that, relative to income, household debt has risen by a factor of six. Yet empirically and theoretically, the drivers of rising debt levels remain poorly understood; their paper closes this knowledge gap. Based on their empirical work, the authors posit that a combination of rising house prices and individually rational life-cycle savings behavior explains the increase in U.S. housing debt. Specifically, they document (1) sharply higher debt-to-income ratios, driven by mortgage borrowing and increasingly concentrated among households in 50th to 80th percentiles of the income distribution; (2) substantial increases in aggregate LTV ratios for all income groups, with slightly higher increases in the middle and lower parts of the income distribution; and (3) stable home equity positions over time and across the distribution. In essence, they find that when house prices rise, prospective homeowners have to borrow larger amounts to buy more expensive houses.These questions have also been raised by Ghent and others in a growing body of literature: why and how do house prices rise faster than income, and how is mortgage borrowing related to that, as cause or consequence? In other words, Bartscher et al. argue that families start their savings path with expanded balance sheets for a given amount of housing equity; families will follow their life-cycle savings plan even if they enter the housing market at elevated price levels.

In terms of theory, the authors suggest that their findings provide new and important guiding principles for future theoretical research on household portfolio choices: how the interaction between house prices and debt appears to be central to gaining a complete and nuanced understanding of the surge in household debt since World War II. At the same time, their study speaks to and quantifies the financial stability risks in highly-leveraged economies where even comparatively small changes in asset prices are magnified and can quickly turn into substantial losses for households and financial intermediaries.

Panel Two: From the American Dream of Mortgage-Financed Homeownership to Risky Debt Accumulation

Like the first panel, the second panel included two papers:

Herndon. The first paper, by Thomas Herndon of Loyola Marymount University, explores the central institutional and regulatory changes in U.S. mortgage finance that contributed to the historic loss of wealth during the Great Recession. In the paper, he traces the developments that transformed mortgages from financial instruments that once largely shielded borrowers and savers from risk to ones that concentrated risk on those least able to bear it. Herndon finds that a consistent theme of the history of mortgage finance in the U.S. is that the risk-bearing capacity for private financial intermediaries is both costly and limited. When this capacity becomes stressed due to rapid growth, competitive pressures, or other economic conditions, there is a persistent tendency for these private financial institutions to redistribute risk towards end users of the system—borrowers and savers—often with severe consequences.

Herndon’s central argument is that stable mortgage finance on a widespread basis in the United States has always depended on direct public intervention, rather than occurring as the result of the unregulated market equilibrium. This stable form of lending was supported by a particular regulatory structure dating to the New Deal era, and heavily relied on the active participation of public institutions in the market.

Indeed, as he shows through international and historical comparisons, absent this regulatory structure, the unregulated equilibrium tends towards risky mortgages in the primary market and a boom-bust cycle in the secondary market driven by agency problems. In this light, Herndon argues that the 2007-2008 financial crisis and Great Recession can be reinterpreted as being caused by the problems that plagued 19th century mortgage finance. These problems re-emerged after deregulation and privatization eroded the New Deal institutional structure. Due to this historical continuity, the 2007-2008 financial crisis was, according to Herndon, both predictable and predicted.

McCargo. In slides prepared for the panel, Alanna McCargo of the Urban Institute makes a compelling case for debt-financed homeownership, though with some important recommendations for future reform. Her argument is as follows: (1) homeownership is an important source of wealth, and is a primary tool for building wealth; (2) there is no investment alternative that is as beneficial for wealth-building for middle-class Americans; (3) rental housing can be unstable, with rising rents and little control over maintenance; and (4) sustainable homeownership, though, is key—when borrowers are over-extended, in bad loans, over-leveraged, or not aware of the challenges of homeownership, problems can develop. McCargo further observes that a home is the only leveraged investment that most people have access to, and low-income people do not, in her view, have ready access to the one unleveraged investment that has proven to be successful: stocks. Furthermore, she observes that defined-benefit pension plans largely do not exist anymore, especially for new hires, and that defined-contribution plans are not broadly available to seasonal, part-time, low-wage, or many small-business workers. Accordingly, McCargo argues that—especially in an era of low interest rates, lack of good information on lower-risk financial investments, and limited access to pension plans—homeownership done right is by far the best alternative.

However, she observes that, unfortunately, it has become more difficult to be a homeowner because, among other things, credit has been tightening and home prices have been rising. But even if access to homeownership opens up, there are risks, especially for lower-income and minority households—a point also made by Emmons et al. and others. Accordingly, it is important to help borrowers engage in safe, sustainable homeownership—which McCargo defines as buying the right amount of home; buying when ready; getting the right kind of mortgage that helps build equity; not overleveraging; and having and accumulating back-up savings to manage unanticipated housing and non-housing expenses.

Panel Three: Restoring or Rethinking the American Dream

Like the first two panels, the third and final panel featured two papers:

Dokko and Dynan. The first paper, by Jane Dokko of the Chicago Fed and Karen Dynan of Harvard University, identifies the underlying risks associated with homeownership and offers six lessons from the financial crisis for reducing risks to homeowners. Those lessons are (1) the pro-cyclicality of mortgage credit can drive business cycles but it can also amplify business cycles; (2) the institutional design of the mortgage market affects the transmission of monetary policy to households; (3) negative equity not only imposed large direct costs on households, but also impeded the deployment of homeowner assistance; (4) a variety of other policy features excluded certain homeowners and hindered efforts to reach distressed homeowners with government assistance; (5) stabilizing a weak housing 5 market is not just about housing policy; and (6) the ongoing conservatorship of Fannie Mae and Freddie Mac makes every taxpayer a participant in the housing finance system.

With these lessons in mind, Dokko and Dynan identify the main economic objectives that should guide policymakers as they undertake mortgage finance reform and look to change other mortgage-related policies in the future. The principal goal of policy reform, they argue, should be to reduce the risks faced by homeowners and, to some extent, households more broadly (though, in discussion, others argued that low-risk itself is not necessarily the goal; improving the risk-return trade-off is). Another important goal would be to make the ex post renegotiation of mortgage contracts more streamlined and less costly.

Hockett. The panel’s final paper was from Cornell’s Robert Hockett, who first situates homeownership in the long history of ownership in the U.S. Recognizing the central role homeownership has played throughout American history in building a robust middle class, he argues that it is more important than ever to restore the institutional and regulatory foundation upon which successful homeownership rests. Yet today, ten years after failing and being rescued by our federal government, our nation’s principal secondary market makers in home mortgage loans—Fannie Mae and Freddie Mac—remain in federal receivership. The proximate reason for this is that neither Republicans nor Democrats in Congress have been able to find either interparty or intraparty consensus as to what should be done with our home mortgage GSEs post-crisis. The deeper, more positive reason for not returning Fannie and Freddie to private hands is that public—that is to say, citizen—ownership of secondary market-makers in home loans is in a certain sense “natural” in any republic, such as our own, where both middle-class standing and that standing’s primary indicator—owning a home—are deeply ingrained in the citizenry’s self-ascribed national identity. This truth is yet more compelling when home prices, given that homes are the dominant middle-class asset, become what Hockett calls “systemically significant”—that is, they become pervasive determinants both of other prices and of macroeconomic well-being. Hockett concludes that the only sustainable future for Fannie and Freddie—not to say for the American middle class and our other GSEs (including our student loan GSEs)—is to be found in their past. Fannie and Freddie should be forthrightly made citizen-owned once again as they were through our home markets’ healthiest decades.

Panel Discussions on Debt-Financed Homeownership

Contributing significantly to the symposium’s rich discussions were the symposium’s discussants, who not only offered insightful and constructive comments on the symposium’s papers but also enriched our understanding of the symposium’s key questions around the evolution, impact and future of debtfinanced homeownership in the U.S. Discussants were Eric Belsky of the Board of Governors of the Federal Reserve, Clinton Key of the Pew Charitable Trusts, Dan Alpert of Westwood Capital, and Michael Stegman of the St. Louis Fed’s Center for Household Financial Stability.

Evolution. As stated already, as the importance of homeownership to wealth-building has increased over time, the risk associated with debt-financed homeownership has grown as well. Housing has become a financial asset whose value is now subject to larger asset fluctuations and credit cycles. In fact, as Eric Belsky pointed out, debt-fueled home buying—under unusually lax underwriting standards and including to investors that in earlier downturns would have relied far more on equity—itself was involved in the rise in prices and subsequent collapse, while Bartscher et al. found that when house prices rise, prospective homeowners have to borrow larger amounts to buy more expensive houses.

Moreover, Belsky observed, it was not just relaxed lending to consumers to buy homes but also to businesses to buy other businesses, and investors to purchase other assets, that drove asset values higher across much of the economy. Indeed, he noted, stock market valuations also soared and became increasingly unmoored from earnings.

Much of this evolution is, in many ways, devolution, as Herndon points out: the deregulation and privatization associated with the financial crisis and Great Recession recalls the 19th century mortgagefinance problems the New Deal then set out to correct, which raises the policy questions: Do we need another set of New Deal type of reforms? Did Dodd-Frank go far enough? Must, as Herndon, Hockett, and Stegman and others asked, the public sector once again play a more prominent role?

Impact. As McCargo, Belsky, Bartscher et al. and others observed, owning a home is popular in part because it can result in wealth accumulation and serve as a hedge against rent inflation. Indeed, as Belsky pointed out, even after controlling for income and past propensity to save and invest, there are papers that show that over time the average homeowner accumulates significantly more wealth than does the average renter. This reflects the fact that over longer periods, home prices tend to grow faster than general price inflation, and mortgage payments result in forced saving and building equity even if prices remain flat.

Yet, as Emmons et al. demonstrated, those benefits were and remain highly uneven, with benefits largely accruing to better-off families and costs and risks largely borne by less-educated and non-white families. Moreover, they note that characteristics over which you have no control—the moment in economic history you were born, your race, your parents’ education and income, etc.—have played an outsized role in determining who did or did not build wealth through homeownership. Finally, highly leveraged homeownership—while making wealth building via homeownership possible for many—also greatly amplified the risks that homeownership would lead to wealth destruction, as the most recent financial crisis painfully demonstrated.

Yet, as McCargo and others argued, a lack of wealth-building alternatives keeps homeownership as the best choice for many, and that it may be, she argues, the only leveraged investment that most Americans have access to. Clearly a critical agenda for the future is to develop those alternatives.

Another impact noted by Hockett and others is that home prices are “systemically significant”—in other words, they critically influence other prices. Along these lines, as Daniel Alpert observed, mortgage equity withdrawals were more than just a danger to households; they distorted the entire macroeconomic picture as well. Equity extraction ran at nearly 8 percent of aggregate disposable personal income per annum from 2003–2007 and was already elevated from the turn of the century. In just 2005, 2006 and 2007, Alpert observed that mortgage equity withdrawals totaled $917 billion—the equivalent of the entire growth of U.S. GDP in the final year, 2007. This raises the question: Is the U.S. economy still too dependent on housing at the expense of other investments? What’s the optimal level of investment in homeownership?

Future. In the discussion, Belsky argued that it is time to return to what was a common view back in the late 1990s and early 2000s: to focus on reducing the risks of bad outcomes from leveraged lending for homes, recognizing that few can buy homes outright. In other words, it is time to return to what both McCargo and Belsky call sustainable homeownership. In Belsky's view, this means (1) creating initial conditions to increase the chances of riding out storms in housing and labor markets through counseling and helping borrowers make prudent financial and loan choices; (2) doing what we can to help borrowers save for rainy days and help them avoid making choices after their purchase that expose them to higher costs or risks, and (3) helping people who do get behind on their payments to work through loan modifications with counselors and lenders.

It is worth remembering here that one cannot build meaningful wealth through homeownership, or any other means for that matter, without taking on some risks. So the key is to ensure borrowers understand those risks and undertake reasonable efforts to mitigate them. Belsky remarked that the Community Advantage Program, funded by the Ford Foundation and Fannie Mae, demonstrated that even low-income families in the midst of a financial crisis could build wealth through homeownership. We could and should learn from that experience.

It’s clear that a fundamental question going forward revolves around the proper role of government in housing, housing finance and homeownership. As Hockett and Herndon concluded, the best if not only sustainable future for leveraged homeownership may lie in our past, when public policy played a far more prominent role in subsidizing and sharing the risks of homeownership and homeownership markets than recently. If that’s true, what does that mean for the GSEs, and for public policy more generally? Should we, as Emmons et al. argue, discuss modifications to the tax code that could support alternatives to leveraged homeownership? What might those alternatives be? Or, more fundamentally, do we need to rethink the long-standing emphasis on debt-financed home ownership itself?

As Stegman emphasized, we are still learning. Ten years after the financial crisis, he notes, the empirically-based scholarly literature is still in the early stages of the peer-review publication process. This long-term process of learning will undoubtedly change what is viewed as conventional wisdom about the causes and policy responses to the financial crisis. In this spirit, Bartscher et al. offered, as mentioned, findings that they believe are providing new and important guiding principles for future theoretical research on household portfolio choices. And Dokko and Dynan, as discussed, offered six important lessons that will inform future research and public policy efforts as well.

In the end, our own long history, as well as experiences in other countries, can be a rich source of insight into the risks and rewards of leveraged homeownership. We hope that this symposium has meaningfully contributed to that critical inquiry.

Five Closing Observations and Ten Key Questions

Five Observations. After the papers were presented, Ray Boshara of the St. Louis Fed’s Center for Household Financial Stability and Sherle Schwenninger of the Private Debt Project offered a few final observations, drawing out some of the symposium’s major themes:

1. An unwavering belief in homeownership and rising prices. Despite the housing crisis of the past decade, there remains a strong and enduring faith in the power of homeownership—as a bedrock principle of our middle-class society, as a source of social stability, and as the principal means of building wealth. Part of this is an assumption, though not a correct one, that home prices are always rising.
2. Homeownership’s prominent role on family balance sheets. Home ownership is, by far, the largest component of wealth for the majority of U.S. families but also the principal source of debt. And, as Moritz Schularick pointed out, one of the disadvantages of relying on homeownership for wealth building is that the only way of accessing that wealth is to take on debt.
3. Outsized risks to families and the economy. Precisely because of its prominence, debt-financed homeownership poses large risks to both household and macro-level financial stability—and thus to economic growth. Today, policymakers face this dilemma: even as the importance of home ownership to wealth-building has increased, the risk of debt-financed homeownership has grown. Once a relatively low-risk investment, housing has become a financial asset whose value is subject to a larger movement in asset prices and to the “boom and busts” of credit cycles.
4. Homeownership’s uneven costs, risks and benefits. Homeownership also results in an uneven distribution of benefits, costs, and risks—with benefits largely accruing to wealthier families who can better manage the risks of debt-financed homeownership, while costs and risks are disproportionately borne by less-educated, younger and non-white families who cannot. That said, Paul Willen’s and Andra Ghent’s observations are worth recalling here: that the crisis was not solely concentrated among low-income homeowners or in subprime mortgages but also hit higherincome groups who bought larger homes than they should have, and who extracted too much wealth via home equity loans. In other words, the risk of homeownership can impact households in nearly all income groups.
5. The role that sheer luck, in the form of timing, has played in determining who has built, maintained or lost wealth from homeownership. In their paper, Emmons et al. found that (1) older Americans have reaped the most benefits from homeownership because they bought well before the housing bubble; (2) middle-aged Americans, or those approaching middle age, especially Gen Xers, have barely gained if at all from homeownership because they tended to buy homes during the housing bubble—which meant that many of them suffered severe housing wealth losses, some even to this day; and (3) younger Americans—many of whom started their careers in or following the Great Recession, and many of whom are saddled with student loans— have weak homeownership rates and have generally found homeownership elusive because of tighter credit standards and other factors.

Ten Key Questions. Considering all the papers and discussion, the symposium raised ten larger questions and challenges for future consideration:

1. Balancing the risks and rewards of homeownership? Could we develop products and public policies to better share the risks and rewards of homeownership and family wealth? Could social insurance schemes be enhanced along these lines? Could risk-pooling products be created, scaled and profitable?
2. Transitioning to sustainable homeownership? What would it take to make homeownership more sustainable, especially for lower-income and minority households? How do we resolve the dilemma of expanding homeownership among low-income and minority households without exposing them to an unacceptable level of risk—a dilemma made worse by a severe lack of 9 reasonably priced starter homes that offer good value for the first-time home buyer? Along these lines, should we more seriously consider the potential of manufactured housing?
3. Government’s best role in promoting and protecting homeownership? Key questions include: With one in eight American families experiencing eviction, should government be more involved in affordable housing and less involved in homeownership? Is government, overall, too involved, or not involved enough? Can we, should we, go beyond a homeownership rate of 64 percent? How could and should the FHA and GSEs be reformed—should they be entirely publically owned? How could and should tax subsidies for homeownership be restructured? Could the government, as in Germany and other nations, reduce the need for private wealth accumulation in the first place?
4. Reducing family and macroeconomic risk within the “American” mortgage? Is it possible to update and strengthen the standard mortgage that Thomas Herndon calls the unique (by international standards) “American” mortgage—the long-term, fixed rate, fully amortizing mortgage that includes a universal ability to repay—with the aim of reducing the risk of debtfinanced homeownership? And is homeownership subject to too many larger macroeconomic risks related to credit cycles and asset inflation?
5. Optimal housing stock? To what extent does debt-financed homeownership produce an optimal housing stock? To what extent is the problem not the risk of debt-financed homeownership but the lack of affordable housing?
6. Alternatives to homeownership? Do we rely too much on homeownership to build wealth in the United States? Should and can we develop alternatives to homeownership to build wealth and promote asset ownership? How can we diversify the household balance sheet?
7. U.S. economy still too dependent on housing as a source of growth? Similarly, do we emphasize housing investment at the expense of other investments? What’s the optimal level of investment in homeownership?
8. Relationship between housing and credit cycles? As a robust academic debate is exploring, did a larger credit bubble derived from financial deregulation and global macroeconomic conditions lead to a rise in housing prices that led to the crisis? Or did overly optimistic expectations of the economic returns to housing (rising prices and accumulated wealth) lead to the credit bubble?
9. Role of banking and financial reform? Relatedly, should our reform efforts focus on large macro-prudential regulation, or should it focus more specifically on mortgage market reform? What reforms would make banks more effective and efficient lending institutions for homeownership?
10. Housing, financial fragility, and growth? What is the relationship between debt-financed housing and financial fragility? Have banks become too dependent on real estate and housing lending? Does the preponderance of debt for mortgages affect our vulnerability to interest rate changes and tie the hands of central banks in making monetary policy? Has propping up private debt—and the housing it supports—become a default goal of monetary policy?