Tax Policy’s Role in Increasing Affordable Housing Supply for Working Families

While my affiliation today is with the Cato Institute, any views expressed are solely my own. In addition, I have no financial interests, other than as a taxpayer, homeowner and concerned citizen, in the issues being discussed at today’s hearing, and nor do I represent any interests that do.

Let me also clearly state that neither the Cato Institute nor its scholars either endorse or oppose specific pieces of legislation. Accordingly, nothing in my prepared testimony or oral remarks should be interpreted to either support or oppose any particular legislation.

The observations and statements made in my testimony are based not only on my years of public service, but also on my read of the relevant statistical series and academic research. Others may read the same research and reach different conclusions. I’ve attempted to limit any observations to those generally supported by multiple researchers and data series.

Let me first state that too many working families face significant housing costs burdens. While I have never been shy when it comes to disagreeing with some of the more conventional approaches to housing assistance, my disagreement has been based in a concern that too many programs are inefficient, poorly targeted, and even occasionally counter-productive. It is not a disagreement over the importance of housing affordability. At the risk of overgeneralization, I believe that too often, too much of our housing subsidies have been captured by providers, and too little of the economic benefit not ultimately received by the intended families. I also remain concerned that too often subsidies are not focused on those most in need, this is particularly the case with housing, where despite the positive tax law changes in 2017, the bunk of the benefit of the mortgage interest deduction, for instance, is captured by wealthier households.

Let me also emphasize that the best housing policy is a jobs policy. There is simply no greater force for housing affordability than broad-based economic growth. Efforts must be made to reduce the cost of housing, especially by reducing construction costs, but it is also critical we see strong inflation-adjusted income growth for working families. Future tax reform should be foremost focused on increasing wages.

We must also recognize that the rental housing programs created in our tax code are often used by developers in conjunction with other non-tax subsidies. Considerable expense is incurred coordinating and combining these various subsidies. I would encourage any changes to the Low-Income Housing Tax Credit (LIHTC) or related programs to be coordinated with programs changes in other programs. There is far too much complexity today in affordable housing development. We should work to reduce that complexity, not add to it.

I would also urge the Committee to consider where the multifamily rental market may be headed. To quote a recent Freddie Mac report¹, “Multifamily is at an inflection point.” The bulk of evidence suggests that national rental market has been softening, with some risk of actual oversupply in multifamily rental housing. While forecasting is always somewhat difficult, most indicators suggest that rents will decline over the next 12 to 18 months. The Committee should be cautious as to adding any additional stimulus to apartment construction when we are likely already passing the peak of this cycle.

America’s Housing Market

America contains just under 144 million housing units. Of these, 85 million are owner-occupied and 44 million are renter-occupied. As of year-end 2022, we also 14.5 million vacant housing units.

Of those renter households, around 2.2 million receive a federal housing voucher, usually under Section 8, another 840,000 live in federally assisted public housing, and 1.4 million live in a federally assisted, but privately-owned, unit, often a project-based Section 8. Just under a 1 million renters live in units covered by some sort of rent control or stabilization.

Since January 2021, America has experienced a dramatic decline in housing affordability. The Federal Reserve Bank of Atlanta’s national Home Ownership Affordability Monitor (HOAM) index², fell from 103.6 in January 2021 to 64.7 in October 2022. The index, where higher numbers indicate more affordability, has modestly recovered to 69.6 for year-end 2022. The most recent HOAM index numbers are at their lowest level since before the 2008 crisis. The rate of decline for 2022 is unparalleled.

While the dramatic increase in mortgage rates has been the primary driver of the decline in housing affordability, high home prices coupled with weak real income growth have also added to declines in affordability. Similarly negative trends have occurred in the rental market.

Rental housing often conjures up visions of urban, high density, apartment living. That vision fails to capture much of the character of renting. A third of rental housing is in the form of single-family units. Another fourth of rental units are in properties of under 10 units. In fact, only about 12 percent of rental units are in the higher density, 50 or more units, one generally finds utilizing the LIHTC. If we include single-family units, approximately half of all rental units are in properties of under 5 units.

The percent of renters earning below the poverty line is only modestly lower for renters living in single-family detached housing (20%) relative to renters overall, 22% of whom earn below the poverty line.

Lack of affordable rental housing is sometimes viewed as an urban problem, yet according to the U.S. Census Bureau’s Housing Vacancy Survey, suburban rental markets actually have tighter housing markets than those of central cities, 5.3 percent vacant compared to 5.9 percent vacant for central cities.

As Harvard’s Joint Center for Housing Studies has noted, in recent years, new construction has predominately added to the stock of higher unit rental properties, while the number of lower unit properties, particularly those in the 2‑to-4-unit range, have been on net leaving the housing stock.

A pressing policy question should be how do we leverage the large portion of small units that typically fall outside the footprint of the LIHTC? This is a particularly pressing issue in rural and suburban communities.

The “good news” is that after years of under-building, both single-family and multifamily construction is strong. The “bad new” is that we are almost certainly overbuilding in the multifamily market and are due for a sharp correction in the apartment market. Over the past year, multifamily housing starts have been at levels not seen since the early 1980s.

Primary causes of rental affordability burdens

The primary drivers of housing costs nationally are land costs, labor costs/​shortages, and materials costs. To a lesser extent, issues related to construction finance are also adversely impacting housing affordability.

Despite being a country rich in land, only around 2 percent is currently used for urban purposes, we, especially state and local governments, have made it increasingly difficult to use our vast land resources for housing. According to the National Association of Home Builders, regulation, mostly state and local, adds 40% to the cost of multifamily development.³

There is perhaps little that Congress can really do to substantially change local land use rules, but we must at least recognize that local supply constraints can render useless many federal attempts at affordable housing, and in some instances actually make them worse. When supply is relatively fixed or inelastic, do not apply demand subsidies, as such only drives up prices.

One area where Congress can make considerable progress, at least for a few states, is to convert federally held land into land that can be used to develop housing. The model used by Nevada, which is 80 percent federal land, to convert federal land to developable uses should be a model for other states. If not for this mechanism, Las Vegas would be considerably more expensive. While few states have the degree of federal ownership as Nevada, one of the least affordable housing markets, California, does have a federal ownership of 45 percent. Even a modest level of converting federal lands in California to housing would make a massive difference in housing affordability. Colorado, Arizona, Idaho, Oregon, Utah, New Mexico, and Washington State are prime examples of housing markets were much of the affordability problem could be solved by a federal land to housing conversion. Of course, a large amount of that land would not be suitable for housing, but enough of it would be.

In relation to materials costs, we can make a significant contribution to reducing construction costs by resolving many of our outstanding trade disputes. This is especially the case with Canadian lumber and aluminum. While we have witnessed dramatic declines in lumber prices relative to those of 2021 and 2022, lumber prices still remain above their long-run averages. Similar, aluminum is down significantly from its summer 2022 highs but still remain highly elevated relative to its long run average.

When thinking in terms of directing subsidies, I urge the committee to think clearly about which constraints are the most binding. If subsidies are not directed at those constraints, then one risks simply driving up the cost of the input in short supply. For instance, in a housing market with limited land availability, or as the economists would say an “inelastic” supply of land, then subsidies not directed at easing the supply of land will largely bid up the cost of land without increasing total units (housing) produced. For this reason, the typical use of demand side tax credits, such as the home buyer tax credit enacted in 2008 as part of the Housing and Economic Recovery Act, simply caused a short-term spurt in housing prices, that later reverted. While it was popular, it was a subsidy that was largely wasted.⁴ As economist Dean Baker noted, the 2008 buyer credit “delayed the deflation of the bubble, but did not stop it.”⁵ We should avoid similar gimmicks this time around.

Excess Complexity in Affordable Rental Development

The most prominent rental housing program in our tax code is the Low-Income Housing Tax Credit (LIHTC). According to the U.S. Census Bureau’s Rental Housing Finance Survey (RHFS), there are around 2.6 million rental units currently in low-income housing tax credit program (about 5% of the total U.S. rental stock), representing over a little less than 60,000 properties, with an average of 40 units per property.

LIHTC properties often receive additional subsidies. For instance, at least a third of LIHTC properties have at least five tenants receiving a section 8 housing voucher. A full fifth of tax credit properties have a third of their tenants on Section 8 housing vouchers. A considerable amount of the “affordability” of these properties is mostly, if not exclusively, provided by the Section 8 voucher program. Other subsidies include: 22% of tax credit properties have subsidized mortgage rates; a third of LITHC developments get either HOME or CDBG; About half get state low-income housing tax credit, and another third are getting some sort of local tax relief.

If we could better streamline subsidies, my read of the literature is that we could reduce development costs somewhere between 2 and 10 percent, depending upon locality. So not a silver bullet, but worth doing, as it would require no additional funding. Not surprising, legal fees are a big cost, as often each subsidy stream performs a separate legal review. There is little market discipline among these fees, often cost-plus. Developer fees in LIHTC are usually around 10–15%. This is much higher than for-profit, unsubsidized development. While some higher development fees are to be expected, given the complexity, I would urge the Committee to either cap development fees or have Treasury promulgate a rulemaking on “reasonable” development fees for LIHTC developments.

Another cost of subsidy-layering is that different subsidy application cycles can add time and delay. One approach would be to limit the use of other federal subsidies on LIHTC developments, else the Committee should explore avenues to better align or consolidate subsidy application cycles.

Although complexity, and its additional costs, are one concern, I am also concerned that the extent of “double-dipping” in LIHTC development reduces the amount of subsidy available to non-LIHTC developments. One avenue to minimize the current incentives for double-dipping is the approach of Section 306 of last Congress’ Cantwell-Young bill (S.1136) prohibition on local contribution requirements.

In addition to the complexities of affordable housing development, the Committee should also consider the tendency of LIHTC properties to, as author Richard Rothstein observed in his book Color of Law, reinforce existing patterns of segregation.⁶ In part this effect is driven by the urban bias in site location, which is, in part, influenced by the more difficult economics of multifamily development in rural and suburban areas, due to the typically lower project density.

Housing Tax Credit Reform

While I would mostly recommend caution at this point in the real estate cycle, if the Committee were to revisit the legislative framework for the LIHTC, I would offer the following recommendations. First, make the LIHTC look at lot more like HUD’s HOME program. HOME allows a wider range of uses, such as direct tenant-based assistance, so that the particular local market circumstances can be taken into consideration. Despite some convergence, housing conditions are still largely locally determined. I would particularly urge the Committee to allow LIHTC funds to be converted into short-term tenant-based rental assistance.⁷ There are many families who do not need permanent or even long-term assistance, but rather only a short-term bridge during periods of financial distress.

I commend those state housing finance agencies which add tighter income targeting requirements to their awarding of tax credits. That said, I believe the current federal level income targeting requirements are insufficient to result in the prioritization of those families most in need. I am also concerned that recent efforts at “income averaging”, while reducing administrative burdens, will result in weaker targeting. For most areas, the median renter income is around 60 percent of the overall median income, implying that full income averaging would result in almost no actual targeting. The Committee should consider the addition of a subgoal of a certain percent of units going to households under 30 percent of area median income. The Committee may also want to consider capping the percent of units that can be occupied by households over 120 percent of area median income.

As one of the Banking Committee staff responsible for the drafting and negotiation of the 2005 reauthorization of the Violence Against Women Act (VAWA) and its housing title, I remain frustrated at the slow movement on the part of Treasury in meeting their obligations under VAWA, in respect to LIHTC properties.

One of the strongest, and most transparent, features of the LIHTC is its per-capita allocation formula. I would strongly encourage the Committee to resist any efforts to move to a different formula. We have repeatedly seen the perverse outcome in programs like CDBG that the allocation formula results in a disproportionate percent of funding going to the richest areas of our country, often rewarding local areas for choosing high housing cost policies.

Progress has been made to broaden the ultimate investors in the LIHTC, but the program is still overly reliant on the banking industry. I will remind the Committee that during the 2008 financial crisis, the market for tax credits largely disappeared when the banking industry as a whole become unprofitable. If we see a continued softening of the economy, it is highly likely that bank profitability will decline and the demand for tax credits will again fall.

Broader rental tax reform

As mentioned above, there are a large number of rental units in lower density properties that will likely never be attractive candidates for the LIHTC, at least not under current rules. Also mentioned was the over 14 million vacant housing units in the United States. How do we better leverage the millions of lower density units, while also attracting vacant unit onto the rental market? I would suggest the Committee directly examine the tax treatment of rental income. A modest proposal would be to make the first $500 in monthly rental income tax-free. In many markets, that savings would be passed along to renters. It would also encourage vacant units to enter the rental stock. As I am concerned as to our long-term fiscal situation, which I believe is unsustainable, this change could be paid-for by reducing the tax expensing of mortgage interest payments of rental properties.

Other Committee issues

While somewhat tangent, I do want to take this opportunity to alert the Committee to significant vulnerabilities in the REIT sector that were witnessed in March 2020. In particular, so-called mortgage REITs were a major source of fragility in our financial markets in March 2020, resulting in both assistance from FHFA and the Federal Reserve. A small handful of REITs have moved beyond acting as passive investment vehicles and have adopted active trading strategies, not in substance different from those that contributed to the failure of Bear Stearns. As I believe these activities represent a threat to financial stability and run counter to Congressional intent, I would encourage the Committee to investigate this issue. If interest, I would be happy to provide the Committee with additional detail.

Conclusions

The housing market has been softening and will continue to do so for at least another 18 months. Rents and prices have weakened and will continue to do so. In fact, I believe we are closer to the beginning of this housing correction than to the middle or end. Accordingly, Congress should move slowly and cautiously when it comes to providing any additional stimulus to the housing market. Such will only delay the inevitable.

Legislating is, of course, an often-slow process and should focus primarily on longer term issues. Despite the short-term outlook for an oversupply in multifamily housing, we still have longer run cost pressures. We also have significant housing resources that are under-utilized.

As the most effective housing program is a jobs and incomes program, I would encourage Congress to approach changes to our housing tax provisions within the broader context of overall tax and housing reform. I would also encourage the Committee to favor simplicity over complexity, and to expand the focus of the LIHTC beyond higher density, urban projects.