Fewer Restrictions on Lending Can Make Housing More Affordable

I appreciate the opportunity to submit a comment to the Department of Housing and Urban Development (HUD) regarding the elimination of regulatory barriers to affordable housing. I am a visiting fellow at the Mercatus Center at George Mason University, and this public interest comment reflects my own views based on my research. The Mercatus Center is dedicated to bridging the gap between academic ideas and real-world problems and to advancing knowledge about the effects of regulation on society. This comment, therefore, does not represent the views of any particular affected party or special interest group. Rather, it is designed to help HUD as it considers how to improve housing affordability for all Americans.

The mission of HUD regarding housing affordability can be divided into two broad categories:

• Rental affordability, especially for Americans of lesser means, to ensure broad access to acceptable shelter
• Access to home ownership for households able to commit to it financially

This request for comment appears to apply mostly to the first category. In many areas, regulatory issues make rents higher. For example, urban zoning laws and excessive barriers to new building drive up rents by limiting the supply of new housing. Many comments will rightly focus on these issues, which are especially important in a handful of metropolitan areas such as Boston, Los Angeles, New York City, and San Francisco. These cities have become excessively expensive because of a dearth of supply.

I will mainly address the second category, regarding home ownership, because homeowners are both suppliers and consumers of housing. So the second category of HUD’s mission is really a subset of the first. In fact, one could argue that programs meant to encourage home ownership first and foremost encourage residential investment that lowers costs for renters as much as or more than it lowers costs for the home buyers that utilize those public programs. Since the financial crisis of 2008, there has been an extreme retrenchment in public support of mortgage lending to borrowers with imperfect credit or low incomes. This has sharply reduced residential investment spurred by buyers of new homes, which has created a drag on new supply. As a result, one of the most important consequences of tight lending standards has been rising rents on American households with lower incomes.

This issue is especially important outside the handful of metropolitan areas where local supply obstructions have caused rents to rise to extremely high levels. Tight lending has made housing less affordable in cities and towns where ample affordable housing could be a local advantage. Tight lending standards are also an issue most directly managed by federal regulators, and in some ways, by HUD itself. This is an issue that HUD can take direct action on, through agencies like the Federal Housing Administration (FHA).

Preliminaries

I will introduce some evidence here for this claim and outline the many benefits that will accrue from reversing the limits on mortgage lending that have developed since the financial crisis. My analysis is based on the following premises, arising from my research:

• Most commenters on the financial crisis, including the Financial Crisis Inquiry Commission (FCIC), committed an important error in concluding that high prices during the precrisis housing boom had been the result of “irrational exuberance” and mortgage credit that was recklessly extended to unqualified borrowers. That conclusion was based on an important logical premise—that prices had become unmoored from the underlying rental value of homes. To the contrary, rising rents were the key factor leading to rising prices. This was especially true in cities such as Los Angeles and New York City, which the FCIC ironically identified as cities that were prime examples of prices rising in spite of rents. With each passing year, rents in the cities, where prices had risen to the highest levels, continue to press ever higher. Reckless lending and systemically destabilizing mortgage securities products clearly played an important role in how the financial crisis played out. But recognizing the primacy of rent as the fundamental driver of localized rising prices demands a wholesale reassessment of the policy decisions that followed the financial crisis.
• As late as the end of the year 2000, Ginnie Mae mortgages outstanding represented 9 percent of the total mortgage market. By the end of 2005, that number had collapsed to 3 percent. Ginnie Mae market share has risen back to about 13 percent today. This represents a tremendous level of countercyclical market influence, which HUD should be commended for. If anything, more countercyclical postcrisis mortgage lending from the FHA would have been broadly helpful. Certainly, one should be able to assert with confidence that high housing prices before 2007 and the collapse that followed had nothing to do with FHA lending activity. The previous bullet point should lead one to a wholesale reassessment of the retrenchment of postcrisis lending in general. That reassessment applies doubly to the FHA.

Some of my writing and research that supports this view, for reference, includes the following:

• “Housing Was Undersupplied during the Great Housing Bubble.” A short introduction to the thesis that a lack of supply was the key to rising home prices before the crisis.
• “Policymakers Should Reevaluate the Premises That Led to an Overhaul of Regulation Z.” A more detailed review of the effects of tightened lending on postcrisis housing markets, directed to the Consumer Financial Protection Bureau.
• “Housing Affordability Series.” A 16-part series of short posts published on The Bridge, a blog of the Mercatus Center, developing the fundamental importance of rent as the key to housing affordability, even for homeowners.
• Shut Out: How a Housing Shortage Caused the Great Recession and Crippled Our Economy. A book detailing the importance of housing supply in driving the housing bubble and general American economic malaise (copies of this can be made available on request).

The Effect of Lending in Boom and Bust

As mentioned above, there are some metropolitan areas where political limits to housing supply are the key contributor to an affordability problem. The worst metropolitan areas in this regard are Boston, Los Angeles, New York City, San Diego, and San Francisco. I refer to these as Closed Access cities because political obstruction to housing there really does limit the mobility of American families. Each year, thousands of families must move out of those metropolitan areas because there isn’t enough housing to meet demand. The families that move away tend to be the families with less financial means, who can’t afford the rising rents. A lack of housing in those cities has created a national process of segregation by income. The modest rise in residential investment and home building from around 2000 to 2005 led to an acceleration of that segregation. In fact, the precrisis housing boom was mostly facilitating a way for Americans to move out of those Closed Access cities and into other, more affordable cities. The hundreds of thousands of households streaming out of the Closed Access cities created a sort of refugee crisis in places like Arizona, inland California, Florida, and Nevada. Those areas had brief bubbles in their housing markets, in part, because they were flooded with families seeking more affordable places to live. I call cities in those areas Contagion cities. They represent a special case, and they experienced something different than the rest of the country.

For our purposes here, I will compare Atlanta and Los Angeles. Los Angeles represents the housing-constrained Closed Access cities. Atlanta is a strong example of a metropolitan area that allowed high rates of new building to accommodate growth before the crisis. Figure 1 compares rent affordability (the percentage of the median household’s income required to rent the median house) and mortgage affordability (the percentage of the median household’s income required to make a conventional mortgage payment on the median house) for both metropolitan areas over time. A higher number means that the rent or mortgage is less affordable.

The teal lines are mortgage affordability, the orange lines are rent affordability, and the red and dark teal lines represent the US national median for each of those measures, respectively.

Rent has been relatively high in Los Angeles throughout the period shown and low in Atlanta. In the late 1970s and 1980s, mortgages tended to be less affordable than rents in all areas because of high inflation, which made interest rates high. However, by the 1990s, interest rates were stabilizing, and by the 2000s, they were declining to levels not seen since the decades before the inflationary 1970s.

Declining interest rates meant that mortgages remained affordable in most of the country. Understandably, lower interest rates were associated with moderate increases in home prices in most areas. This meant that for the United States in general, including Atlanta, mortgage affordability was similar to or slightly better than rent affordability throughout the 1990s and the early 2000s. In other words, the monthly payment on a conventional mortgage was about the same as the rental payment would be for the same property. In most of the country, that comparison was relatively stable from the early 1990s until the top of the housing boom in 2005.

This, however, had not been true for Los Angeles. Although as late as 1999 mortgages in Los Angeles were more affordable than rents, this changed at the turn of the century: even with declining interest rates, home prices in Los Angeles were rising at such a clip that mortgages became much less affordable than rents.

This pattern was true of all the Closed Access cities. This is because the lack of housing in a few key urban centers has become a defining characteristic of the postindustrial economy. By the late 1990s, segregation by migration was starting to accelerate. Rents in Los Angeles and the other Closed Access cities had been high already, and rent inflation in those cities continued to rise. Rent affordability was reaching a point where it was beyond the means of many households. By the late 1990s, rent affordability (rent divided by income) was being moderated by rising local incomes rather than by declining rents. And one way local incomes were rising was through the relocation of households with lower incomes out of those metropolitan areas to make room for households with higher incomes who could afford the rents. By the height of the housing boom, households with incomes below the national median were moving out of the Closed Access cities, on net, at a clip of about 2 percent annually.

This migration was happening systematically across the country. The cities where prices jumped so high that mortgage affordability suddenly looked like the late 1970s again were generally cities where rents had been high and were continuing to rise. In Atlanta and in the rest of the country, the typical homeowners in 2006 did not have an unusual amount of trouble affording the payments on their mortgage.

Since the primary problem all along in the Closed Access cities such as Los Angeles was a lack of supply and high rents, the collapse in home-building markets after 2005 was not helpful at all. Rents have remained exceedingly high in Los Angeles. Initially, mortgage affordability dropped back to pre-2000 levels, and it has since moved back to a level similar to rent affordability.

In some ways, the effect of relative lending access on home prices in Los Angeles is irrelevant. Higher prices cannot trigger significant new supply in Los Angeles because building is largely blocked through local political opposition. In fact, higher prices only stir up more animosity, leading locals to complain that only “luxury” units can be built. These are the cities where reforming local regulations is the key to bringing down rents. Increasing access to lending in the Closed Access cities can create more housing units for the families that have access to those mortgages, but those units don’t become available through added new supply. They become available because the demand triggered by that new lending pushes up prices and rents until existing owners sell and move away from the metropolitan area with windfall capital gains or existing renters are forced to move away from the metropolitan area by rising rents.

In the Closed Access cities, the limit on supply is a hard political limit. The basic market forces that create positive externalities when programs such as FHA lending make home ownership available and affordable can’t come about in that context. In the aftermath of the financial crisis, lending was widely blamed for the distress that Closed Access housing markets caused. But that distress is fundamentally the result of deprivation and denial. Closed Access housing makes all good things seem bad. It is even common now to see politicians and local activists complain about the development of good-paying local jobs because it will only lead to rising rents. The blame on lending was similar to these complaints about jobs. Closed Access just makes everything seem bad. However, even though lending took too much blame for the precrisis housing boom, there is little that generous lending can do to help with housing affordability in the Closed Access cities. All it can do is to increase the rate of segregation that is pushing Americans out of the Closed Access cities to more affordable markets.

In other cities where obstacles to building aren’t so binding, however, generous and prudent lending can produce positive externalities by increasing the supply of housing, removing barriers to home ownership for families that value it, and putting downward pressure on rents. The change in lending standards since the financial crisis has been extreme. As the Obama administration noted, the average FICO score on originated loans has increased by more than 40 points. Home ownership rates and the sales of new affordable homes have plummeted. The lack of positive externalities from generous lending has had deleterious effects in most of the country.

In figure 1, Atlanta demonstrates what has happened in those areas. When lending was scaled back, home prices collapsed, even in places like Atlanta where prices had never been particularly high and mortgage affordability had never been far from the norm. In fact, after the crisis, mortgages became affordable in Atlanta and most of the rest of the United States to an unprecedented degree. They still are. Low prices made building new affordable homes unprofitable. The lack of demand, blocked through newly tight lending standards, became the key obstacle to new supply. As seen in figure 1, this led to rising rents in Atlanta. This is an important factor to keep in mind. There isn’t a generalized lack of income or demand for housing in cities like Atlanta. That demand is expressed through rent, and rents have been relentlessly rising. There is, specifically, a lack of demand for buying homes, in spite of such strong demand for living in them.

Tight lending and the associated low prices have caused a housing affordability problem in all of the cities where it shouldn’t be a problem. Figure 2 compares the rates of housing permits in the Los Angeles and Atlanta metropolitan areas to the US average. In Los Angeles, rents have been higher for some time, and the rate of building has been low. This was one of the cities with extremely high prices, which were blamed on loose lending. Those high prices have been generally associated with overbuilding also, yet clearly Los Angeles hasn’t come close to building a sustainable quantity of new homes in decades.

Yet Los Angeles is back to normal, such as it is. Its building rates have recovered to those of the boom period. Rent inflation is high there, but rent inflation was high there in 2005 too. Those rising rents, and the growing realization that they weren’t going to reverse, were the primary cause of high prices.

Atlanta, on the other hand, was willing and able to build enough to accommodate significant growth. Prices increased moderately during the boom. Building rates were moderately higher in 2005 than they had been in the late 1990s. Atlanta was creating the positive externalities of a healthy building market. There was no rent inflation in Atlanta from 2002 to 2006. Now rates of building in Atlanta are not much higher than they are in Los Angeles, and rent inflation has been running at about 5 percent annually in Atlanta for the past few years—similar to rent inflation in Los Angeles.

This is the pattern throughout the country. Of the 16 largest metropolitan areas in which the Bureau of Labor Statistics tracks a Consumer Price Index (CPI) measure for rent inflation, 6 cities approved at least 35 housing permits per 1,000 residents for the period 2001 to 2005: Atlanta, Dallas, Houston, Minneapolis, Seattle, and Tampa. (For comparison, Los Angeles approved 13 units per 1,000 residents over those five years.) Annual rent inflation ranged from 1.2 percent to 3.4 percent in those cities over those five years. American households were moving to those cities in large numbers, and healthy building was keeping rents moderate.

In the five years leading up to 2018, rates of building in those six cities were still, on average, 40 percent below what they were from 2001 to 2005. And annual rent inflation during this period ranged from 3.3 percent to 5.4 percent. Now all of those cities have rates of building and rent inflation similar to Los Angeles. This is not because those cities have all suddenly developed the excessive local political obstructions to building that Los Angeles has.

Tight lending has mainly shut down demand for home ownership (and for the new supply that would bring down rents) in the cities where the housing boom was facilitating the production of affordable homes. Only the Closed Access cities have fully recovered to boom-era rates of building. Those were the places where unconventional lending had been blamed for pushing home prices into the stratosphere. Fundamentally, rising rents are what caused Closed Access prices to rise so high. Loose lending in cities like Atlanta had nothing to do with it. Imposing new tight lending standards in cities like Atlanta was not the appropriate response to the housing boom and the financial crisis. HUD can play an important role in reversing this error.

Figure 3 shows how important incomes have been as a factor in collapsing housing supply. In cities with per capita incomes of $50,000 or less, building rates are generally still less than half what they were in 2005. This isn’t for lack of demand for shelter—rent inflation is now similarly high in most of these metropolitan areas. It is lack of demand for purchasing homes, even though mortgage affordability is now much, much better in those cities than rent affordability, compared to decades of experience from before the financial crisis.

The Mechanism for Tight Lending to Dampen Entry-Level Demand

The question of exactly how newly tight lending standards are preventing reasonable home buying decisions in entry-level and affordable markets is complicated. That is an analysis that I must leave to the underwriters at the FHA. Yet the shift in the character of home buyers after 2007 is so extreme that there can be little doubt that some combination of new regulations and rules, such as stringent rules about income recognition on applications, new mandated costly underwriting requirements, limits on fees and spreads, etc., has had severe dampening effects.

Sales of new homes under $200,000 are down 85 percent since 2005. In inflation-adjusted dollars, mortgage origination volume to borrowers with FICO scores greater than 760 is higher than it was in 2005. For all FICO scores below that level, mortgage originations are down more than 65 percent. Home ownership rates for all age groups of working-age households remain well below any precrisis levels going back at least 40 years. Rates of building, prices vs. rent affordability, borrower characteristics, new home sales—all of these indicators are screaming that there is an extreme lack of capital funding, especially for affordable homes.

Residential investment had almost always run at more than 4 percent of GDP annually before the financial crisis. In the decade since, it has failed to reach 4 percent once. Until residential investment sustainably reaches levels well above 4 percent, there is little reason to question spreading broad access to mortgage credit. This is especially true for rates of home building at lower price points. If critics claim that more mortgage access through the FHA is recklessly pushing up home prices, one must ask why higher prices have not triggered more building. The lack of building in the face of persistently high rent inflation suggests that the supply that would maintain rent affordability requires higher prices.

Since tight lending standards were created based on the false premise that rising prices were unrelated to rising rents, I would suggest that the proper starting point would be to return to some pre-2008 lending standard and recalibrate from there. That would be simpler than trying to ascertain exactly which features of the new lending apparatus are blocking funding in affordable markets. And it would be justifiable, especially for an institution such as the FHA that demonstrated highly countercyclical trends during the housing boom.

Free Lunches All Over

This leaves two clear areas for improvement in American housing markets: (1) the reduction of barriers to new supply in cities with ample job opportunities so that Americans can move there affordably, and (2) the reduction of barriers to mortgage lending that have cut off building in the cities where Americans can live affordably. Both of these barriers have prevented American households from engaging in reasonable economic activity. The removal of these barriers can produce unmitigated positives.

Home Ownership

Home ownership rates are well below 20th century norms. For the 35-to-44-year-old group, for instance, the rate of home ownership ranged between about 65 percent and 70 percent from 1982 to 2010. It is now 60 percent. Much of the sharp decline in ownership rates happened when originations to less-than-pristine borrowers and the construction of affordable homes collapsed. Millions of potential American homeowners could be reintroduced to the benefits of home ownership before the American housing market would even start to tickle the lower end of what was once considered normal.

Working-Class Net Worth

In more affordable markets, mortgage affordability has diverged from rent affordability. If mortgage funding again became more broadly available those indicators would reconverge. That convergence would come from both declining rents (because of new supply in the market for shelter) and rising prices (because of new access to capital). The time since the financial crisis has seen an increase in wealth inequality, and one important reason is that the lack of mortgage access has pushed down prices in affordable markets. Clearly, whatever the effects of labor costs, land costs, regulation, etc. are, prices that aren’t inducing new supply can’t be anywhere close to unsustainably high. There is plenty of room for more healing in working-class American net worth before concerns about potential overvaluation should become salient. There is nothing wrong with moderately higher home prices resulting from reasonably generous lending associated with potential home ownership rates in the ranges common before 2010. Here the risk of imposing one’s own mental model on the economy is evident. If low-tier home prices could harmlessly recover some more, the moral implications of limiting credit access to prevent that from happening are not attractive.

Real Incomes

The lower rents that would result from new residential investment would lower the cost of living, especially for Americans with lower incomes. Lower rents would mean less income being transferred to landlords and more income available to spend on other goods and services.

Easier Monetary Policy

The Federal Reserve continues to walk the tightrope between keeping inflation in check and allowing an aging economic recovery to continue. Lower rents would make that job much easier. In fact, rent inflation has been a major driver in rising costs since the mid-1990s. Since 2012, core CPI inflation has remained around 2 percent. However, much of that is owing to measured rent inflation, which is a result of low residential investment and low housing starts. Rent inflation has been above 3 percent for several years, and core CPI inflation with the rent components removed has rarely moved above 1.5 percent. Residential investment triggered by a reinvigorated mortgage market would lead to stabilizing inflation and growth in real incomes. The destabilizing and potentially inflationary borrowing that happened before the financial crisis largely came from the local markets where home prices had skyrocketed, leaving existing owners with ample home equity to use for current spending. The purpose of more generous lending today would be to help housing markets recover in the cities where prices did not rise so rapidly. There simply isn’t a danger that cities like Atlanta today pose the potential for inflationary spending of the type that developed before the financial crisis from harvested home equity. Worrying that those developments might eventually reoccur is not a good reason to keep fetters on new housing supply and middle-class home ownership there today.

Construction Employment

Millions of construction workers were thrown into long-term unemployment after the financial crisis. This is a sector where growing employment would improve the lives of many households with a variety of skills and educational backgrounds. Growing construction employment would be a boon to working-class Americans. Builders today sometimes complain that labor is in short supply. This problem is partly the result of selling a product that is being systematically underfunded with new limits on borrowing in low-tier markets. Increased demand, fueled by a return to earlier lending standards, would make those cost issues less of a problem. What would be the downside? That construction-sector wages would increase at the cost of moderately higher entry-level home prices? May the country hope to have such problems! Furthermore, moderately higher home prices may be associated with those rising wages and rising employment, but the increase in building would bring down rents, which are the appropriate measure of the cost of shelter. There would be no wage-driven inflation in that context. To the contrary, while workers’ wages would be rising, rent inflation would be declining.

Conclusion

There are certainly many areas where regulatory barriers to building need to be eliminated in order to keep housing affordable throughout the United States. That should certainly be the priority of government at all levels. Yet today there are many areas in this country where those barriers aren’t the binding constraint that is blocking supply and pushing up rents. Those cities do lack adequate supply today, but it is because they lack suppliers. They lack potential home buyers. Potential home buyers frequently need mortgages. The most direct and immediate boost to housing supply that HUD could create today would be to increase suppliers, to broaden the availability of mortgages to households that have been locked out in one way or another from today’s market. Trends in prices, building, and borrowing suggest that many of those potential buyers would buy more affordable and more modest homes than the homes that are bought by buyers who can qualify today. The most important task for HUD today is to figure out what is preventing the construction of homes that would sell for less than $200,000. The answer to that puzzle is surely a bit counterintuitive, because it is clear that more broad-based lending and more residential investment will be required for that to happen. The families that would use that funding, for the most part, aren’t living under a bridge or in a car today. They are stacked into the existing housing stock, where they frequently spend much more on rent than they would need to spend on a mortgage to buy that very same house. Spending less on rent must begin with spending more on residential investment.