Is Lihtc the Best Way Toi Deliver Affordable Housing

I.  Introduction

The Low Income Housing Tax Credit (LIHTC) program provides for the majority of new affordable housing units congenital in the United States and has resulted in the production of ane.5 million low‐income housing units since its inception in 1986. The LIHTC represents a radical departure from the structure of previous supply‐side housing programs, which accept mostly relied on direct provision or subsidization of low‐income housing. In add-on to being a critical federal housing program, the LIHTC is of involvement as an example of a novel type of tax expenditure program that is spreading to other policy domains. Nether the LIHTC plan, the government allocates taxation credits to developers of depression‐income housing, who and so sell the credits, often via intermediaries, to investors in exchange for equity financing. Credits are subsequently claimed by investors on their taxation returns. As a consequence, the revenue enhancement beneficiary is an investor rather than the provider or the targeted beneficiary of the subsidized service. We refer to this class of credits equally "investable tax credits."

Figure ane illustrates the rapid growth in tax expenditures for this program. Annual tax expenditures on the LIHTC program were estimated at $four.iii billion in 2004 and are projected to increase past almost 20% in real terms by 2011 (JCT, various years).i Expenditures on the LIHTC program relative to other federal housing programs are shown in effigy 2. While tax expenditures on the LIHTC are merely about a third of outlays on Section 8 vouchers, the program is large relative to other federal supply‐side programs. In improver, several of these other programs (such equally the Section 8 Project‐Based Program) are no longer agile (Rice and Sard 2007). The LIHTC is currently the largest and fastest‐growing federal plan for the production of affordable rental housing.

Fig. 1.

JCT estimates of LIHTC revenue enhancement expenditure. Source: JCT (various years). Figures for a given twelvemonth are taken from the written report immediately preceding that year. All figures are in 2004 dollars. Starting from 2008, estimates are based on a −0.3 annual alter in the consumer cost index, which reflects the average for 2005–7. The apparent 1‐year shift from individual to corporate claimants from 2000 to 2001 reflects a change in the method of estimation rather than a true change in the distribution of claimants.

Fig. 2.

Comparing of housing programs (2005). Sources: *Data for LIHTC tax expenditure are from JCT (various years). **Information on remaining programs come up from Rice and Sard (2007). Funding estimates for these programs are based on budget authority rather than expenditure. The Section eight Housing Choice Voucher program provides vouchers to low‐income tenants. The Section 8 Project‐Based Program subsidizes affordable rental housing. Section 515 and Section 521 provide low‐interest loans to encourage the production of affordable rental housing in rural areas. Department 202 and Department 811 provide subsidies to developers of affordable rental housing for the elderly and those with disabilities; Department 811 as well includes tenant‐based rental assistance for the disabled. The Domicile Investment Partnership provides a variety of subsidies (both project and tenant based) for rental and nonrental affordable housing. Homeless assistance refers to a number of programs that provide housing aid to the homeless. Meet Rice and Sard (2007) for further details.

Although the LIHTC has mostly been regarded with skepticism in the bookish literature (e.g., Weisbach 2006), the programme has remained politically pop. Reforms to the program since its inception accept all been in the direction of program expansion and have more often than not passed with overwhelming bipartisan support. In addition, other programs with a similar structure accept been introduced past the federal regime. Under the New Market Tax Credits program, for example, the federal government allocates tax credits to designated Community Development Entities (CDEs). The CDE then sells these credits to investors in substitution for equity finance, which is used by the CDE to provide investments in low‐income communities. The program was initiated in 2000 and has allocated $12.1 billion in taxation credits every bit of 2007 (GAO 2007). Recent legislation has likewise proposed the creation of a Homeowner's Tax Credit, modeled afterwards the LIHTC, for the construction or rehabilitation of nonrental affordable housing.ii

The LIHTC programme has also received a corking deal of contempo attention, as the market for credits has been severely afflicted past the financial crisis that began in 2007. Prices for LIH credits have fallen dramatically through 2008 and 2009. This strain in the credit market has come up precisely at a time when the need for affordable housing has arguably increased greatly. These recent events accept prompted the passage of new legislative initiatives and proposals for further policy reform. In particular, both the Housing and Recovery Act of 2008 (HERA) and the American Recovery and Reinvestment Deed of 2009 (ARRA) resulted in important changes to the LIHTC program. Many of these changes are temporary, but some permanently affect the structure of the program. Nosotros discuss these policies in detail below.

In this paper, we provide estimates of the magnitude of LIHTC tax expenditures and discuss efficiency costs of the programme. We likewise present empirical evidence on the characteristics of programme participants: developers, credit claimants, and beneficiaries. Finally, nosotros discuss recent events in the LIHTC market as well equally the potential implications of recent revenue enhancement legislation and reform proposals.

1Expenditure growth in the early years of the program primarily reflects lags betwixt credit allocation and claiming; afterwards expenditure growth reflects expansions in almanac credit allocations.

2In addition, fifteen states have enacted country LIHTC programs to supplement the federal LIHTC program. Nigh of these programs have a construction like to that of the federal program.

2.  An Overview of the LIHTC

The Taxation Reform Act of 1986 (TRA86) included several provisions that reduced the profitability of investment in rental housing. The LIHTC program was devised every bit office of TRA86 to preserve incentives for the provision of affordable rental housing. Nether this program, the Internal Acquirement Service (IRS) allocates nonrefundable tax credits to housing agencies run by the land governments, which then award the credits to selected housing projects proposed by developers. Federal law sets basic requirements for projects applying for LIHTC funds to ensure that they brand a strong commitment to provide low‐income housing. However, state housing agencies concord almost of the power in selecting program recipients, both through their individual plans for ranking programs and through a significant amount of additional discretion. Developers, in plough, sell the credits to investors in exchange for equity financing used to support the housing project. Investors may exist individuals, corporations, or financial institutions, and their return is limited to their taxation benefits. Intermediaries, also known as syndicators, create the market place in these tax credits. The flow of credits is depicted in simplified form in figure 3, based on GAO (1997).3

Fig. 3.

Mechanics of LIHTC allotment. Source: Adapted from GAO (1997)

At that place are two types of credits allocated under the program. The first blazon of credit ("nine% credit") is allocated to projects that are newly constructed and receive no other federal subsidies. The federal government allocates these credits to states in proportion to population. Initial allocations were $1.25 per capita. This amount was increased (in nominal terms) to $1.50 in 2002 and to $ane.75 in 2003 and was indexed for inflation thereafter. These allocations are subject to a small state minimum, which was $two million in 2003 and was also indexed for inflation. If a projection is financed with individual‐action tax‐exempt bonds, information technology is eligible for the second type of credit ("4% credit"); these credits are capped only indirectly through state private action bond caps.

One dollar of allocated credit entitles the claimant to a dollar tax credit each year for a ten‐twelvemonth period. The federal government also sets several guidelines with which projects must comply in gild to exist eligible for LIHTC funding. Near important, a certain share of units must be rent restricted and be occupied by depression‐income households. These conditions must, in well-nigh cases, be met for a minimum of 30 years.4 Conditional on coming together these guidelines, state housing agencies have broad discretionary powers in setting criteria for the credit resource allotment process. These criteria are mostly set out in state Qualified Activity Plans (QAPs). The normal allotment process can be overridden or bypassed; indeed, a General Accounting Role study found that 17 out of 20 studied QAPs contained such override provisions (GAO 1997). Olsen (2003) estimates that, on boilerplate, $3 is requested for every available $1, indicating periods of substantial excess demand for LIH credits. As a result, many states have forced developers to meet requirements that are stricter than the federal guidelines in order to be eligible for these credits.5

Tax credits are then awarded to called projects. The amount of tax credits a given project receives is determined by the qualified basis of the project. This is the product of two factors: the showtime reflects the development costs of the project, and the 2nd reflects the share of the project that is reserved as depression‐income.6 This method approximates a subsidy per low‐income unit, adjusted for development costs. Projects that are newly constructed and receive no federal subsidies other than the LIHTC receive an amount of credits such that the present value of credits over a 10‐year period will be equal to 70% of the qualified basis. All other projects receive credits such that the present value will be equal to 30% of the qualified basis.7 On average, ane‐third of the total financing of a typical project is provided through the proceeds of the sale of tax credits.

One time state housing agencies honour credits to projects, the taxation credits become available to the developer. The programmer may either utilize the tax credits to reduce its own tax bills, if it is a for‐profit developer, or sell the credits to investors. Investors who buy the credits provide the developers with equity, which is used to support the construction and creation of the depression‐income housing. The deal is often structured as a limited partnership betwixt investors and developers. Investors, the express partners in the partnership, mostly exercise not expect income from the equity but instead view the taxation credits as their return.

Note that the structure of the program creates taxation benefits for investors and intermediaries in the syndication process who accept no intrinsic interest in depression‐income housing or related issues. This creates a much wider constituency for the program than would be the case for an economically equivalent program of directly subsidies for low‐income housing.8 This may explain why developers, intermediaries, and housing advocates—not by whatever ways a natural political coalition—are fans of the program.ix

3For other overviews of the program, see Burman (2005) and U.South. Treasury (2008).

4In particular, projects must meet either a xx‐fifty rule (at least twenty% of the units in the development must be rent restricted and occupied by households with incomes at or below 50% of the area median income) or a 40‐60 rule (at least 40% of the units in the development must be hire restricted and occupied by households with incomes at or below 60% of the area median income).

5States may, e.thou., require developers to run across stricter rent requirements or remain in compliance for a longer period than mandated by the federal guidelines.

6More specifically, the qualified footing is the product of the eligible basis, the amount of all applicable depreciable development costs, and the applicable fraction. The eligible footing includes most depreciable costs but excludes such costs as the conquering costs of land and permanent financing costs. The project may exist eligible for a "basis boost" of 30% if it is located in a U.Due south. Housing and Urban Development–designated high‐cost surface area. The applicative fraction is either the fraction of units or the fraction of floor space (whichever is lower) reserved equally low‐income units. A potential business organization with this method is that the subsidy amount is set ex ante without a mechanism to adjust the subsidy ex post for changes in operating costs or area median income; see Usowski and Hollar (2008) for farther give-and-take.

7Technically, the eligible basis is multiplied by an "applicative pct" that is meant to produce a credit allocation equivalent to the present values given above. This leads to applicable percentages of approximately 9% and 4%, respectively. In practise, the percentages are set by the Treasury monthly and fluctuate with interest rates. If the amount of credits required to attract enough equity finance to fill the financial deficit is less than this calculated amount, the projection receives the lower corporeality. The 2008 HERA set up a flooring for the credit rate for new construction completed before 2022 at 9%.

8There is perhaps an analogy here with Ferejohn's (1986) account of the Food Stamp Programme, which argues that the program was enacted every bit the outcome of a (seemingly unnatural) coalition between rural legislators with farming constituencies and urban legislators representing the urban poor.

9The LIHTC too can create odd political coalitions in other policy dimensions. For example, LIHTC advocates suggested that the initial proposal for dividend exemption in 2003 would depress the market for tax credits and opposed the proposal on that basis.

3.  LIHTC Tax Expenditure Estimates

In this section, we consider two methods to calculate tax expenditures nether the LIHTC program and compare the resulting estimates to the Joint Committee on Taxation (JCT) estimates. First, nosotros summate implied taxation expenditures from credits allocated. As discussed above, the LIHTC differs from almost other tax expenditures in that a large share of full expenditure is capped and adamant by almanac per capita credit allocations. If the bulk of credits are claimed, it should be possible to approximate and projection tax expenditures under this program with a reasonably high caste of accurateness using only information on credits allocated. 2d, we summate tax expenditures using data from individual and corporate tax returns. As we discuss below, the information on the public‐use files is not ever sufficient to calculate actual credits claimed. We therefore compare the resulting revenue enhancement expenditure estimates to the estimates from JCT and credits allocated to assess the validity of using these data to examine the characteristics of credit claimants.

A.  Estimates from Credits Allocated

Figure 4 provides expenditure estimates based on credits allocated. All credits allocated by the federal authorities are assumed to be claimed by investors over the 10‐twelvemonth period immediately following project completion, with a 2‐year lag between credit allocation and project completion. Annual credits claimed are and so adapted to 2004 dollars. This method volition result in underestimates of full revenue enhancement expenditures since information technology accounts only for credits subject to the per capita cap.

Fig. 4.

Tax expenditure estimates from credits allocated. Figures are calculated from almanac per capita credits allocated and population figures and assume that all credits are claimed over the ten‐year flow following project completion. The figures are based on federal allocation guidelines and do non contain state‐specific minimum allocations or changes arising from 2008 and 2009 legislation. We permit a 2‐year lag between allocation and projection completion.

Unsaid expenditures increase quickly over the kickoff decade every bit the program reaches steady state. Expenditures then decline slightly equally the real value of credits allocated falls over time and increase as almanac per capita allocations are increased. Expenditure estimates are substantially college than JCT estimates in the early years of the programme simply conform closely to JCT estimates over the contempo period. This probable reflects the fact that many of the initially allocated taxation credits were left unused (GAO 1997). Lags between federal allocation and credit claiming may also have been longer than average in the early years of the program.

B.  Estimates from Revenue enhancement Return Data

A second measure of LIHTC tax expenditure comes from credits claimed by individuals and corporations on their revenue enhancement returns. There are at least 3 reasons why information on credits claimed from tax return information may non accurately reverberate bodily almanac tax expenditure on the LIH program.10 Kickoff, the line item for the current‐yr LIHTC is included every bit part of the tentative credit adding form for the general business tax credit (Form 3800). Taxpayers may not be eligible to merits the total corporeality of the tentative credit; for instance, credits are nonrefundable and cannot be used to offset culling minimum tax (AMT) liability.11 Although the total amount of general business organisation tax credit claimed is available on the tax return, it is difficult to decide the exact amount of credit claimed for these taxpayers. 2nd, LIH credits can in some cases exist carried frontward or carried back. In these cases, they are included as unmarried line items that include the sum of all general concern credit carryforwards or carrybacks. Finally, an individual does not need to file Form 3800 if she is non claiming any other general business organization credits and has no carryforwards or carrybacks. The allowable LIH credits are so entered in the "other credits" line detail on Form 1040. While the first factor may cause these estimates to represent an overestimate, the 2d two factors may effect in underestimates of total LIH credits claimed from the Form 3800 line item.

Given these caveats, estimates of LIHTC tax expenditure based on publicly available private and corporate tax return data are presented in figure v. The individual data come from public‐utilise samples of individual federal taxation returns.12 Information on LIH credit claiming is available from 1987–2002. The corporate data come up from table 21 of the Statistics of Income Corporation Complete Report (http://www.irs.gov/taxstats/commodity/0,,id=170734,00.html), which itemizes components of the general business tax credit.

Fig. 5.

Taxation expenditure estimates from revenue enhancement return data. Source: Individual information are from 1987–2002. Figures are taken from the tentative credit calculation of the full general business organization tax credit (Form 3800). Private data are taken from the public‐apply individual federal revenue enhancement render sample and corporate data are from the Statistics of Income Corporation Consummate Report.

The estimates in figure 5 are lower than estimates from credits allocated in the early years of the program. Over again, this is likely a result of initial credits remaining either unused or claimed with a substantial lag. In 2002, the last year for which we have individual information, implied tax expenditure on the program was near $four billion. Less than 10% of this expenditure was from claims by individual investors.

The gap between tentative and commanded LIH credits seems likely to be small for 2 reasons. First, investors must actively buy credits and would not exercise and then unless they expected to claim the full value. Second, while information technology is possible that investors could unexpectedly go ineligible to claim the full value, there is a secondary market for credits that allows investors to sell credits they cannot claim.13 A like argument applies to carryforwards and carrybacks: such claims would ascend if investors are not able to merits the full almanac allocation in a given year. I imperfect approximate of the magnitude of these biases is to examine what share of tentative credit dollars is "claimed" past investors with AMT liability.14 Using the individual public‐use files for 1987–2002, we find that less than x% of all individual credit dollars are from AMT payers. The possibility remains that individuals are ineligible to claim the total value of credits for other reasons, and the corporate AMT "claim" share may be quite different from the share for individuals. In addition, some credits claimed by individual investors may not appear on Form 3800. We unfortunately do non have the necessary data to quantify the magnitude of these biases.

Overall, the estimates from these tax data imply slightly higher revenue enhancement expenditures than the JCT estimates and estimates from credits allocated, consistent with tentative credits beingness an overestimate of concluding credits claimed. Nonetheless, the aggregate trends over fourth dimension besides as the shift from individual claims to corporate claims exercise track the estimates from the other information sources adequately well. We are therefore reasonably confident in using these figures as an approximation of credits claimed when nosotros examine the characteristics of credit claimants below.

10We are grateful to Tom Holtmann at JCT for several helpful discussions on these problems.

11The credits included in the full general business organization tax credit are then subject to a stacking guild that determines which credits are claimed first.

12See http://world wide web.irs.gov/pub/irs‐soi/weber.pdf for sampling details.

13There are some weather placed on investors who wish to sell their interests in LIHTC properties. In particular, until recently, the investor had to purchase a "recapture bond," which guarantees payment to the Treasury in the event of taxation credit recapture due to noncompliance. This was repealed nether HERA.

14Until 2008, LIH credits could not be used to offset AMT liability. Therefore, these tentative credits are presumably not actually claimed by investors.

Four.  Credit Pricing and Economic Efficiency

A.  Determinants of Credit Prices and Trends over Time

An interesting feature of the LIHTC program is that purchasers of credits practice not generally receive income from the property; the difference between the purchase cost and the value of the credit represents the return on the investment. However, credit prices sometimes deviate from the actuarially fair value. In the instance of the LIHTC, the price at which credits are sold has increased substantially over the life of the programme. Ernst & Young (2005) calculates the equity cost simply every bit housing credit equity divided by housing credits. By this measure out, median credit prices have increased from approximately $0.45 in the early years of the program to over $0.85 for projects placed into service in 2005.15 Anecdotal prove suggests that the very low initial pricing was primarily the result of dubiety about the rules of the system and about how long it would final.

Computing the price in this way implicitly assumes that investors are realizing the full nominal value of credits received. In practice, however, investors realize the credits over a 10‐year menstruum. To obtain a "true" price, which reflects the equity per effective credit dollar, the stream of tax credits should be discounted. Discounting credits accordingly essentially increases the implied disinterestedness price: Cummings and DiPasquale (1999) judge that the average credit toll over their sample menstruum (1987–96) increases from $0.52 to more than $0.70 if the present value of the stream of credits (using a discount rate of half-dozen.7%) is used in the toll calculations.

Historical prices below actuarially fair values may have reflected additional compensation for the risk of default: projects may neglect to remain in compliance over the total lifetime of the credits. If a property is institute to be in noncompliance, the investor forfeits hereafter tax credit claims and must repay one‐third of previously claimed credits with interest. The monitoring costs associated with ensuring that the developers are in compliance and the discounting of the risk of developer noncompliance could result in a risk premium. Empirically, this hazard premium appears negligible: just before the contempo financial crisis, credits were trading at or above their actuarially off-white value, consistent with evidence indicating that the ex post probability of punishment appears low. Ernst & Young (2005) finds that only 0.4% of surveyed properties had been audited and faced a loss of tax benefits in the grade of recapture or disallowance of future claims.xvi While the average run a risk of noncompliance may exist minor, risk may assistance to explain cross‐sectional variation in credit prices. Using data from LIHTC transactions in California, Eriksen (2009) finds that projects with characteristics correlated with an increased risk of noncompliance are associated with lower taxation credit prices. This suggests that states may exist able to use credit prices every bit an effective ex dues screening device when choosing how to allocate credits across proposed projects.

In many respects, the puzzling characteristic of credit prices (until the very recent menstruum) was not why they are too depression, but rather why they appear too loftier. In recent years, credit prices approached or exceeded one when the stream of tax credits is discounted appropriately.17 A possible explanation for this anomaly is that for certain fiscal institutions, investment in low‐income housing tin can serve to satisfy some of their obligations under the Community Reinvestment Deed (CRA). The CRA requires banks and other depository institutions to provide credit throughout their local communities, including in low‐income areas. CRA evaluations are subjective and are based on a diffuse set of metrics that are deliberately not quantified. The record of the financial establishment is reviewed and taken into account when regulators evaluate applications for deposit facilities or during mergers. As such, there are no specific, quantifiable linkages betwixt LIHTC purchases and CRA ratings, though information technology is widely acknowledged that LIHTC tin can count toward the hardest standard for banks, the investment of resources into their low‐income communities. A 2003 report estimated that 43% of depression‐income housing investors were financial institutions subject to CRA requirements (Ernst & Young 2003). The interaction with the CRA opens up the possibility that entities may be willing to bid the price of taxation credits in a higher place their actuarially fair value every bit they tin jointly realize revenue enhancement advantages and fulfill CRA obligations.xviii Such mechanisms for increasing the funding for the supply of low‐income housing should, still, be balanced against possible inefficiencies stemming from any expansion of CRA obligations.

Credit prices may also experience brusk‐term fluctuations as a upshot of supply and demand shocks. About notably, the puncturing of the bubble in the U.Southward. housing marketplace in 2007 and the subsequent financial crisis of 2008 have had profound consequences for the LIHTC program. Prior to the crisis, $1 of taxation credits traded at an undiscounted cost of nearly $0.xc;19 by early 2009, the corresponding toll had fallen beneath $0.seventy (run into, e.grand., Federal Reserve Bank of Dallas 2009). This price decline reflects decreased investor need for LIH credits, due principally to two factors. First, the large losses incurred past banks and other financial institutions rendered tax credits significantly less valuable. A 2d major evolution was the exit from this market of two major buyers of LIH credits, Fannie Mae and Freddie Mac, which entered government conservatorship in September 2008.

15Cummings and DiPasquale (1999) find like patterns in their surveyed properties.

16The low probability of punishment may ascend from low noncompliance or ineffective monitoring. The IRS is largely dependent on monitoring by land housing agencies, and a GAO review constitute a number of potential problems in state oversight procedures (GAO 1997).

17Many participants utilize an after‐revenue enhancement yield to measure out pricing more accurately to grab the temporal aspects of the credits. Participants so use comparisons between these yields and municipal bonds to benchmark appropriate pricing. When yields are lower than municipal bonds, most participants believe that CRA incentives, discussed beneath, are operative.

18Marquis and Guthrie (2007) talk over the interactions between the CRA and the LIHTC at length. They cite Federal Reserve analyses of bank mergers that highlight LIHTC investments equally ways of satisfying CRA requirements. Marquis and Guthrie get on to argue that states design their LIHTC programs, in part, to help banks address CRA guidelines in order to foster merger activity.

19As discussed earlier, this cost corresponds to one that may accept exceeded $one when tax credits are accordingly discounted; this point does not, nevertheless, touch on the magnitude of the postcrisis decline in prices.

B.  Efficiency of the Program

A notable feature of the LIHTC is that a big share of credits (specifically, the ix% credits) are capped. This means that there is no monetary doubt most the amount of resource committed by the authorities to low‐income housing, which may be politically advantageous. Withal, there is dubiety over the actual provision of housing; in particular, lower prices of credits may dictate that fewer housing units are constructed. For case, the decreased market value of LIH credits during the contempo crunch has had a severe negative bear upon on the ability of developers to obtain equity financing through the auction of credits.20 At the same time, a number of developments accept potentially increased the demand for depression‐income rental housing. For instance, would‐be homeowners at present face greater difficulty in obtaining credit to purchase houses. At that place has likewise been an increase in unemployment associated with the recession. Demand has also increased from quondam homeowners whose houses have been foreclosed on or who accept decided to abandon houses in which they had negative disinterestedness.21 This extreme case illustrates a more general issue with the construction of the LIHTC programme: credit prices and the supply of affordable housing through this program may fluctuate with aggregate economic conditions.

In addition to variability in the prices of credits, information technology is important to notation that not all of the equity finance received through the sale of credits goes into housing projects; there are a number of transactions costs associated with nigh LIHTC projects. Indeed, the GAO (1997) estimates, for example, that syndication costs may consume 10%–27% of equity invested in LIH credit projects. Consequent with this finding, Cummings and DiPasquale (1999) find that the average ratio of net equity to gross disinterestedness in their data is 0.71. Syndication costs announced to be declining over time as the market has get more competitive. It is also important to note that many projects constructed with LIH credits also benefit from additional supply‐side subsidies, such every bit taxation‐exempt bail financing and Department 515 rural housing loans. In improver, a substantial share of residents in these housing projects are recipients of Department 8 vouchers or other forms of rental assistance.22 Therefore, while the LIHTC program is credited with great success in the production of affordable housing, it is not clear that the program would be successful in meeting project goals were it to be in isolation.

This evidence leaves open the question of whether new construction is the about efficient means of getting housing assistance to low‐income households or whether subsidies should instead be provided in the form of demand‐side programs, such every bit vouchers. The relative claim of supply‐side programs and vouchers are a subject of ongoing debate, and some (e.g., Weisbach 2006) have advocated replacing the LIHTC with a voucher program. A potential disadvantage of vouchers is their effect on market rents: Susin (2002) and Gibbons and Manning (2006), for example, find large effects on hire subsidies on rents. Since rent voucher programs cover only a relatively small fraction of people with low incomes, recipients of vouchers may exist ameliorate off on balance despite the hire increases, but at that place is a larger population of low‐income nonrecipients who face college rents without receiving any regime assistance. In contrast, supply‐side housing policies not but benefit those low‐income families that obtain the new housing just besides may potentially do good the wider low‐income population through lower marketplace rents (Coate, Johnson, and Zeckhauser 1994).

Even so, these benefits depend on whether housing provided nether this plan increases the net supply of low‐income housing or but crowds out other low‐income housing. Estimates suggest that crowd‐out is likely to exist substantial: Sinai and Waldfogel (2005) estimate crowd‐out of upward to two‐thirds from government‐subsidized housing mostly, Malpezzi and Vandell (2002) estimate full crowd‐out from the LIHTC program, and Eriksen and Rosenthal (2008) guess crowd‐out of between 1‐half and one. There is, however, some show that the LIHTC programme increases supply at very local levels. Baum‐Snowfall and Marion (2009) exploit a discontinuity in the LIHTC program. Specifically, a census tract qualifies for discontinuously higher funding (a 30% increase in the revenue enhancement credit) if more than than 50% of its households have incomes beneath a certain threshold (sixty% of surface area median gross income). They find a large bear on of this discontinuity on the supply of LIHTC housing. They also observe that LIHTC‐funded housing increases possessor turnover, reduces neighborhood income in gentrifying areas, and increases property values in declining areas.

On rest, the available empirical evidence suggests that hire vouchers are probable to exist more cost effective than the LIHTC plan. Deng (2006) computes the costs associated with LIHTC‐funded developments and vouchers in six U.S. metropolitan areas and finds that vouchers are generally less costly. However, it is important to note that in that location is considerable heterogeneity in the relative costs, depending on housing market weather condition in the metropolitan surface area, the type of housing, and various other factors. Supply‐side remedies may thus exist preferred for item targeted groups or in sure geographic areas. Certain groups, such as large households, single nonelderly persons, and the elderly, statistically have a lower take chances of success in using their vouchers (Finkel and Buron 2001). There may also be markets in which the LIHTC is less costly than vouchers (DiPasquale, Fricke, and Garcia‐Diaz 2003). Finally, LIHTC‐funded developments may achieve better outcomes forth other dimensions, such as neighborhood integration and school quality (Deng 2007).

There are 3 other considerations that are relevant when thinking about the efficiency consequences of the structure of the LIHTC program specifically. Starting time, the fact that producers need not be the claimers of credits may be beneficial for productive efficiency. If credits are not refundable and not transferable, and so participation in the program volition be tied to a provider'southward taxation liability. Investable tax credits neutralize the bias toward for‐profit providers inherent in a nonrefundable tax credit; this characteristic is especially critical if the ascendant organizational grade for delivering the product is nonprofit. Such a market place can also improve productive efficiency if at that place is heterogeneity among for‐profit providers and more than efficient providers practice non have sufficient tax liability to utilize the full value of the credits.

In this sense, the provision of investable taxation credits under the LIHTC is closely analogous to the widespread employ of leasing every bit a means of transferring tax benefits (such as depreciation allowances) amidst firms. The leasing market place allows firms facing relatively depression marginal tax rates to benefit from investment incentives by allowing a firm facing a high marginal tax charge per unit to exist the legal owner of a piece of equipment. Specifically, Congress'south experiment with "safe harbor" leasing removed virtually obstacles for the transfer of tax benefits through leasing arrangements as function of the Economic Recovery Taxation Human action of 1981 (see Warren and Auerbach 1982). This provision proved to be short‐lived because of perceived abuses, and these concerns have limited more than expansive efforts in the leasing domain. Nevertheless, the leasing market continues to serve this basic function today. Investable tax credits achieve the same separation of the legal taxation beneficiary from the agent undertaking the targeted activity. This analogy also makes clear that transferability of taxation benefits is not a unique characteristic of the LIHTC and has important precedents elsewhere in the tax system.

Second, investors claim credits over a 10‐year horizon. This provides a mechanism for the developer to receive up‐forepart financing for the project while maintaining the ability of the authorities to enforce program guidelines later the projection is built through the threat of recapture of credits from investors. This structure may therefore be preferable to direct subsidies or refundable credits that allocate the unabridged subsidy to the developer prior to project completion.

Finally, the investable nature of the program allows for the creation of delegated monitors to ensure compliance with housing agency and IRS requirements. Without the investable feature of the program, housing agencies and the IRS would be exclusively monitoring compliance of the many projects. With investable taxation credits, monitoring is largely undertaken past the investors and their agents (oftentimes accounting firms) since their entire economic return for their investment is contingent on compliance. If at that place are differential monitoring capabilities, this delegation of monitoring can exist an important dimension to the desirability of the LIHTC.23

Notation that, in practise, the incentives for investors to monitor depend on the regime oversight of the program. The IRS requires land housing agencies to develop compliance monitoring programs that run into certain federal guidelines and report cases of noncompliance. Notwithstanding, the IRS has no specific dominance to evaluate or inspect state housing agencies to ensure that they are actually meeting these federal guidelines, and some evidence suggests that the agencies may not in fact e'er be fully compliant with their oversight responsibilities (GAO 1997). As discussed, the IRS rarely recaptures credits in practice, and it is not clear whether this is a result of constructive monitoring and high project compliance or merely poor government oversight.

20For examples of projects that have lost financing as a result, meet Terry Pristin, "Shovel Set, simply Investor‐Deprived," New York Times, May 6, 2009.

21However, the end of the housing bubble has increased housing affordability, as business firm prices have declined from their previously excessive levels toward their central values.

22See Cummings and DiPasquale (1999) for estimates of the total development costs of housing produced nether the LIHTC programme and Deng (2006) for comparisons with the voucher program.

23Thanks to Michael Novogradac for highlighting this signal to us.

Five.  Distribution of LIHTC Tax Expenditure

Determining the truthful economical incidence of the LIHTC program is quite challenging for several reasons. Credit allocation formulas reflect a range of factors, including development costs and other received subsidies. In addition, states oftentimes allocate credits to developers willing to serve detail target populations or run into affordability guidelines that are stricter than federal requirements. Finally, virtually depression‐income housing projects receive additional federal supply‐ and need‐side subsidies, making it hard to isolate the effects of the LIHTC. We focus instead on examining the distribution of three groups affected by the program: providers of depression‐income housing, investors in LIH credits, and low‐income households that benefit from the plan.

A.  Distribution of Providers

Every bit discussed to a higher place, an advantage of the "investable" feature of the LIHTC program is that it levels the playing field between for‐profit and nonprofit developers. Federal police requires states to reserve 10% of LIHTC funds for projects with nonprofit developers, but this requirement does not appear to exist binding in practice. Figure 6 illustrates the share of LIHTC‐funded projects developed by nonprofit developers past yr placed into service. This share increased rapidly in the early years of the program, peaked at over 35% in 1998, and has since declined slightly to 25% in 2003. Some states practise have additional nonprofit gear up‐asides or favor nonprofits in their QAPs (Gustafson and Walker 2002), but information technology is unlikely that these provisions alone can explain the observed levels of participation of nonprofits in this market place.

Fig. half dozen.

Share of projects completed past a not‐for‐profit developer. Source: U.Due south. Housing and Urban Evolution LIHTC database. Year refers to twelvemonth placed in service.

B.  Distribution of Credit Claimants

1.  Individuals

A surprising characteristic of individual LIHTC investors is that they are small: betwixt 1987 and 2002, the average credit among claimers was slightly more than than $3,000. The number of claimants has remained fairly constant over fourth dimension, only the credit claimed per return has declined substantially, as shown in figure seven. Among claimers, the boilerplate credit per return nearly halved betwixt 1995 and 2002, failing from $4,100 to $2,100 in real terms. This implies both an overall shift toward corporate investors and a shift toward smaller investments amidst remaining individual investors.

Fig. seven.

LIHTC claimed by private investors. Source: Public‐utilize sample of individual revenue enhancement returns (1987–2002). LIHTC claimed is calculated from the LIHTC line item in the tentative general business organisation tax credit calculation.

To examine the distribution of private claimants, an approximation of cash income tin can exist constructed from the public‐utilise data files.24Figure 8 plots the share of the annual value of credits claimed past individuals by income category between 1999 and 2002. There are two features to annotation. Kickoff, the distribution is hump shaped in each twelvemonth. This is non surprising: low‐income individuals may non have revenue enhancement liability to offset, and college‐income individuals (except those at the very top of the distribution) are more likely to be subject area to the AMT. 2nd, the distribution of credits claimed shifts noticeably down the income distribution betwixt 2000 and 2001. In particular, the share of credit value claimed by those in the $100,000–$250,000 income category declines and appears to shift to the 2 income categories that are immediately below.25 These findings suggest that the AMT may have significant effects on the distribution of credit claimants and may help to explain both the overall shift toward corporate investors and the move toward lower‐income, smaller individual investors.26

Fig. viii.

Distribution of claimants by income. Source: Public‐utilise sample of private tax returns (1987–2002). LIHTC claimed is calculated from the LIHTC line item in the tentative general business tax credit calculation. Income refers to an approximation of cash income (run into the text for details).

24In item, we use the following definition for income: cash income = adjusted gross income − saving and loan tax refunds + individual retirement account deduction + student loan involvement deduction + alimony paid deduction + tuition and fees deduction + health savings business relationship deduction + 1/2 of self‐employment tax + punishment on early withdrawal of saving + self‐employed health insurance deduction + medical savings business relationship deduction + Keogh deduction + tax‐exempt interest + nontaxable social security benefits − min (other income, 0). Non all components are available in all years, then we construct a measure every bit shut every bit possible to the above definition in each year.

25Those with very high incomes are likely to accept regular tax liability that exceeds the AMT. This may be one explanation for why the claim share for those in the highest income category ($i million and to a higher place) remains fairly stable over this menstruation. The change in distribution of individual investors could be more precisely attributed to the AMT by comparing changes in the distribution beyond states with high vs. low shares of AMT taxpayers. This is unfortunately not possible in the public‐use files since the state of residence is not available for those with adjusted gross incomes greater than $200,000.

26Individual investments in LIH credits are also limited by passive loss limitations. Nosotros give thanks Rob Dietz for highlighting this betoken to us.

2.  Corporations

Tabulations by income category are non available in the published corporate reports. The reports do tabulate claims by sector, which is perhaps a more interesting categorization for corporate claimants. Table 1 illustrates the share of annual credit value claimed past diverse sectors in 2000 and 2006. We include data from the five sectors in the Statistics of Income classification that account for the largest shares of credit value claimed as well as information for the real estate and rental and leasing sectors.

Table 1

Distribution of Corporate Claimants by Selected Sectors

	Share of Annual Value of Corporate Credits Claimed
	2000	2006
Finance and insurance	33.four	42.nine
Management of companies (holding companies)	31.5	45.6
Utilities	9.61	NA
Manufacturing	14.7	4.xi
Information	5.43	3.56
Real estate and rental and leasing	one.02	.06

The majority of corporate credits are claimed past corporations in two sectors: finance and insurance and management of companies (property companies). These ii sectors have also deemed for a larger share of LIH credit dollars claimed over time. Together, corporations in these sectors claimed 65% of corporate credit dollars in 2000 and 89% in 2006. In that location is nothing in the construction of the LIHTC program alone that suggests that challenge of credits should exist then concentrated across sectors. These findings strongly suggest that corporations in these sectors derive additional benefits from investments in depression‐income housing.

Finally, it is interesting to annotation that the real estate sector accounts for a negligible share of credits claimed. In office, this suggests that the separation of the provision of the service from the tax beneficiary immune by investable taxation credits has been important.

C.  Distribution of Beneficiaries

Overall, it appears that the program is successful in providing affordable housing to households with beneath‐average incomes, but it may not do good those with the very lowest incomes. This is not surprising since the income limits for rent‐restricted units are 50%–threescore% of area median income, although the additional use of vouchers or other rent subsidies may aid to make the units affordable for lower‐income households. A survey of properties placed into service betwixt 1992 and 1994 indicates that three‐quarters of households in LIHTC properties had income below 50% of the surface area median and 40% had income below 30% of the area median income (GAO 1997). Thirty‐nine percentage of resident households received straight rental assist, and their average income was 25% of the area median. Similar patterns have been observed in other surveys of LIHTC properties (Ernst & Immature 1997). Nosotros practice not attempt to examine the effects of the program on other (nonresident) low‐income households.

Half-dozen.  Recent Tax Reforms

Congress has responded to the contempo bug in the LIHTC market by adopting various measures to encourage the supply of low‐income housing, within the context of its overall strategy of "economic stimulus." The HERA, enacted in July 2008, included provisions designed both to make LIH credits more than attractive to investors and to expand the scale of the LIHTC program (see Keightley 2009). In detail, HERA established a flooring for the annual credit rate (as a percentage of a project's eligible basis) obtained past investors in LIH credits. The deed also increased LIHTC allocations to the states past 10% for 2008 and 2009.

In addition, HERA allowed LIH credits to be used to offset AMT liability. A number of housing advocates had argued strongly for enabling credits to be used against AMT liability in order to make credits attractive to investors who are field of study to the AMT or who are concerned about facing the AMT over the lifetime of the credits. Since credit allocations are fixed, this provision should not substantially affect tax expenditure on the program.27 In theory, this modify may shift credit challenge back to private investors (especially those with higher incomes), since this grouping is more probable to be constrained past the AMT. This could shift the distribution of investors, which could affect the effectiveness of the programme in producing low‐income housing as well as the types of projects built under the program. Projects with corporate investors tend to accept lower syndication costs (Ernst & Young 1997) and are probable to be more effective at ex mail service monitoring than a diffuse group of small individual investors, resulting in a reduction of risk and higher credit prices. A shift back to individual investors could therefore decrease the amount of equity financing available for low‐income housing. The changing distribution of investors may as well influence the types of projects that are financed under the programme. A report by the National Association of Home Builders (2005), for example, argued that individual investors tend to prefer smaller projects and may be more likely to adopt rural projects besides as projects catering toward special‐needs populations. However, the application of credits against the AMT may have a small consequence relative to the growing use of LIH credits by corporations against CRA requirements. To the extent that this effect dominates, credit claiming may continue to shift increasingly toward corporations.

More than recently, the ARRA, signed into law by President Obama on February 17, 2009, marked a significant deviation for the LIHTC through the introduction of a credit commutation program. Specifically, ARRA authorized the Treasury to make cash grants to the states in lieu of role of their LIHTC allocations (Keightley 2009). For example, a state was permitted to elect to receive equally cash grants upward to 40% of its 2009 LIH credit allocations (as well every bit to receive grants in exchange for unused credits for 2008). The federal government will pay $0.85 for each $1 of LIH credits given up by the state. The idea underlying this provision is that states would award these funds to developers to pursue projects that conform to LIHTC requirements. These projects demand not take any other LIHTC funding, although developers are required to demonstrate that they have made a skillful‐religion try to obtain disinterestedness investment. The funds awarded in this style are not taxable income to the recipients and practice not reduce the project's eligible basis for LIHTC purposes. This LIHTC exchange program volition essentially bypass the current depressed market for LIH credits. The provision is anticipated to have a substantial budgetary touch, relative to the overall size of the LIHTC program.28

27In theory, allowing credits to be used to offset the AMT could increase tax expenditures nether the program if some investors were previously unable to merits purchased credits after being hitting by the AMT. This result seems likely to exist small since investors were well enlightened of the AMT provisions and made purchasing decisions taking into business relationship the likelihood that they would exist subject to the AMT over the life of the credits. In addition, investors could brand apply of carryforward allowances and the secondary market for credits, making it likely that most credits allocated were somewhen claimed.

28The JCT estimates a cost of $419 1000000 over the 10‐year upkeep horizon (Keightley 2009).

Vii.  Decision

The LIHTC plan has get the principal federal program subsidizing the development of low‐income housing and appears to accept broad support amid policy makers, low‐income housing advocates, developers, and institutional investors. Contempo and proposed reforms to the plan accept been in the direction of further expansion, and the structure of the LIHTC has been replicated in other related federal and state programs.

Several features of the LIHTC program distinguish information technology from traditional supply‐side provision or subsidization. The unbundling of the service provider and the tax beneficiary has potential advantages for competition and productive efficiency. This unbundling also creates better incentives for ex post monitoring and compliance in theory, although information technology is non articulate whether these benefits are being realized in do. Withal, the LIHTC structure too means that the success of the plan in delivering affordable housing thus depends on the incentives of a diversity of market participants.

Taxation expenditures under the plan are quite anticipated, but the supply of housing provided for a given level of tax expenditure (even abstracting from full general equilibrium effects) tin can vary. Corporations are at present the primary investors in LIH credits, and their incentives may reflect features of not only the LIHTC program specifically but likewise related programs, such as CRA requirements.

The contempo crisis has illustrated the potential vulnerability of the program to aggregate market conditions. Several provisions of contempo policy reforms have loosened the supply of credits or made the terms of their use more favorable merely take not altered the program'due south central construction. However, the replacement of some LIHTC allocations past cash grants represents a significant (albeit fractional and perhaps temporary) shift in program blueprint from the "investable revenue enhancement credit" model discussed earlier toward the direct subsidy approach. In view of this, analysis of the relative merits of the LIHTC structure and a traditional subsidy appears more pertinent than e'er.

References

• Baum‐Snow, N., and J. Marion. 2009. "Effects of Low‐Income Housing Developments on Neighborhoods." Periodical of Public Economics 93:654–66.

• Burman, L. 2005. "Low‐Income Housing Credit." In NTA Encyclopedia of Revenue enhancement and Tax Policy, ed. J. J. Cordes, R. D. Ebel, and J. Thousand. Gravelle. Washington, DC: Urban Institute Press.

• Coate, South., S. Johnson, and R. Zeckhauser. 1994. "Pecuniary Redistribution through In‐Kind Programs." Journal of Public Economics 55:1–xl.

• Cummings, J. L., and D. DiPasquale. 1999. "The Low‐Income Housing Tax Credit: An Assay of the Get-go Ten Years." Housing Policy Debate 10, no. ii:251–307.

• Deng, L. 2006. "The Cost‐Effectiveness of the Low‐Income Housing Taxation Credit Relative to Vouchers: Evidence from Six Metropolitan Areas." Housing Policy Fence 16:469–511.

• ———. 2007. "Comparing the Furnishings of Housing Vouchers and Depression‐Income Housing Tax Credits on Neighborhood Integration and Schoolhouse Quality." Journal of Planning Education and Research 27:xx–35.

• DiPasquale, D., D. Fricke, and D. Garcia‐Diaz. 2003. "Comparing the Costs of Federal Housing Assistance Programs." Federal Reserve Banking concern of New York Economical Policy Review ix, no. 2:147–66.

• Eriksen, M. 2009. "The Market Price of Depression‐Income Housing Taxation Credits." Journal of Urban Economics 66, no. 2:141–49.

• Eriksen, M., and S. Rosenthal. 2008. "Crowd Out Furnishings of Place‐Based Subsidized Rental Housing: New Evidence from the LIHTC Program." Unpublished manuscript, University of Georgia.

• Ernst & Young. 1997. "The Depression Income Housing Revenue enhancement Credit: The First Decade." Unpublished manuscript, Ernst & Young.

• ———. 2003. "The Touch on of the Dividend Exclusion Proposal on the Production of Affordable Housing." Unpublished manuscript, Ernst & Young.

• ———. 2005. "Understanding the Dynamics Three: Housing Taxation Credit Investment Functioning." Unpublished manuscript, Ernst & Young.

• Federal Reserve Bank of Dallas. 2009. "Low‐Income Housing Tax Credits in Texas: Achievements and Challenges." Banking and Customs Perspectives, no. 2 (May).

• Ferejohn, J. A. 1986. "Logrolling in an Institutional Context: A Example Study of Food Stamps Legislation." In Congress and Policy Change, ed. G. Wright, L. Rieselbach, and L. Dodd. New York: Agathon.

• Finkel, M., and L. Buron. 2001. Study on Section 8 Voucher Success Rates. Vol. 1: Quantitative Study of Success Rates in Metropolitan Areas. Cambridge, MA: Abt Associates.

• GAO (General Accounting Function). 1997. Tax Credits: Opportunities to Amend Oversight of the Depression Income Housing Program. GAO/GGD/RCED‐97‐55. Washington, DC: GAO.

• ———. 2007. Tax Policy: New Markets Tax Credit Appears to Increase Investment by Investors in Depression‐Income Communities, but Opportunities Exist to Better Monitor Compliance. GAO‐07‐296. Washington, DC: GAO.

• Gibbons, Due south., and A. Manning. 2006. "The Incidence of U.k. Housing Benefit: Bear witness from the 1990's Reforms." Journal of Public Economics 90:799–822.

• Gustafson, J., and J. C. Walker. 2002. "Analysis of State Qualified Allocation Plans for the Low‐Income Housing Taxation Credit Program." Unpublished manuscript, Urban Institute, Washington, DC.

• JCT (Joint Commission on Tax). Diverse years. "Estimates of Federal Tax Expenditures." Study, JCT, Washington, DC.

• Keightley, Mark P. 2009. An Introduction to the Blueprint of the Low‐Income Housing Tax Credit. Report no. RS22389. Washington, DC: Congressional Research Service.

• Malpezzi, S., and K. Vandell. 2002. "Does the Low‐Income Housing Tax Credit Increase the Supply of Housing?" Journal of Housing Economics eleven:360–80.

• Marquis, C., and D. Guthrie. 2007. "State Activism and the Subconscious Incentives behind Bank Acquisitions." Unpublished manuscript, Harvard Business School.

• National Clan of Dwelling house Builders. 2005. "Increasing Individual Investor Need for the LIHTC." White Newspaper, National Association of Dwelling Builders, Washington, DC.

• Olsen, E. O. 2003. "Housing Programs for Low‐Income Households." In Means‐Tested Transfer Programs in the U.S., ed. Robert Moffitt. Chicago: University of Chicago Press (for NBER).

• Rice, D., and B. Sard. 2007. "The Effects of the Federal Budget Clasp on Low‐Income Housing Aid." Unpublished manuscript, Center on Budget and Policy Priorities, Washington, DC.

• Sinai, T., and J. Waldfogel. 2005. "Do Low‐Income Housing Subsidies Increment the Occupied Housing Stock?" Journal of Public Economics 89:2137–64.

• Susin, S. 2002. "Rent Vouchers and the Toll of Low‐Income Housing." Journal of Public Economics 83:109–52.

• Usowski, K., and M. Hollar. 2008. "Social Policy and the U.S. Tax Lawmaking: The Curious Instance of the Low‐Income Housing Tax Credit." Unpublished manuscript, U.S. Department of Housing and Urban Development, Washington, DC.

• U.Southward. Treasury. 2008. "Depression‐Income Housing Tax Credits: Affordable Housing Investment Opportunities for Banks." Community Evolution Insights, February, U.Due south. Treasury, Washington, DC.

• Warren, A. C., Jr., and A. J. Auerbach. 1982. "Transferability of Tax Incentives and the Fiction of Safe Harbor Leasing." Harvard Law Review 95:1752–86.

• Weisbach, D. A. 2006. "Tax Expenditures, Main Amanuensis Problems, and Redundancy." Washington University Law Review 84:1823–lx.

cartwrightarage1945.blogspot.com

Source: https://www.journals.uchicago.edu/doi/full/10.1086/649832

Share this post