Full text of DOI:10.1080/10511482.2000.9521377

Housing Policy Debate
ISSN: 1051-1482 (Print) 2152-050X (Online) Journal homepage: http://www.tandfonline.com/loi/rhpd20
Income tax concessions for owner‐occupied
housing
Steven C. Bourassa & William G. Grigsby
To cite this article: Steven C. Bourassa & William G. Grigsby (2000) Income tax
concessions for owner‐occupied housing, Housing Policy Debate, 11:3, 521-546, DOI:
10.1080/10511482.2000.9521377
To link to this article: http://dx.doi.org/10.1080/10511482.2000.9521377
Published online: 31 Mar 2010.
Submit your article to this journal
Article views: 121
View related articles
Citing articles: 21 View citing articles
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=rhpd20
Download by: [Simon Fraser University]
Date: 05 June 2016, At: 08:49

Housing Policy Debate
Volume 11, Issue 3
521
521
·
© Fannie Mae Foundation 2000. All Rights Reserved.
Income Tax Concessions for Owner-Occupied
Housing
Steven C. Bourassa
University of Louisville
William G. Grigsby
University of Pennsylvania
Abstract
This article considers arguments for and against the major tax concessions for
owner-occupied housing in the Internal Revenue Code—concessions that have a sig-
nificant effect on the federal budget. It recommends retaining two of the concessions—
nontaxation of net imputed income and exemption of capital gains—and abolishing
two others—the mortgage interest and real estate tax deductions.
After a review of the market impact of removing each deduction, the article con-
cludes that an appropriate phase-out period of 15 to 20 years would virtually elimi-
nate adverse effects on house prices and homeowners.
Keywords: Homeownership; Tax policy
Introduction
For more than a decade, Congress and three successive administrations
have tried to balance the budget, simplify the Internal Revenue Code,
reduce marginal tax rates, and flatten the rate schedule. Throughout
this series of exercises, the favoritism that has been accorded the hous-
ing sector through various tax concessions has been left largely un-
touched. The principal concessions all apply to owner-occupied housing,
which currently accounts for about two-thirds of the housing stock.1
1 There are six quantitatively less important housing tax concessions, only one of
which applies to owner-occupied housing: exclusion of interest on owner-occupied
mortgage subsidy bonds, exclusion of interest on state and local debt for rental hous-
ing, deferral of income from post-1987 installment sales, exemption from passive loss
rules for $25,000 of rental loss, credit for low-income housing investments, and accel-
erated depreciation on rental housing. The estimated total tax concession for these
privileges for fiscal 1998 is about $13 billion (Office of Management and Budget
[OMB] 1997). We reject the argument that certain normally deductible business
expenses should be considered tax concessions to landlords and, indirectly, to ten-
ants. Such an argument has been advanced, for example, by Litzenberger and Sosin
(1978) and Woodward and Weicher (1989).

522
Steven C. Bourassa and William G. Grigsby
The concessions are
1. Nontaxation of net imputed income from owner-occupied homes2
2. Nontaxation of the first $250,000 ($500,000 for a married couple)
of capital gains on the sale of an owner-occupied home3
3. Deductibility of mortgage interest on owner-occupied homes
4. Deductibility of local property taxes on owner-occupied homes
These concessions have a significant impact on the federal budget. It
has been estimated that in 1989, nontaxation of imputed income cost
the U.S. Treasury $109 billion, or about one-fourth of federal revenues
from individual income taxes in that year (Follain, Ling, and McGill
1993; Ling and McGill 1992).4 Although more recent estimates of this
concession have not been published, if this cost rose roughly in pro-
portion to the growth in individual income tax receipts, then it would
have been about $145 billion in 1998. For the same year (but prior to
the recent tax reform legislation that changed the tax treatment of
capital gains), it was estimated that nontaxation and deferral of capi-
tal gains on owner-occupied housing deprived the federal Treasury of
another $22 billion (Office of Management and Budget [OMB] 1997).
Mortgage interest and real estate tax deductions were expected to cost
the Treasury $52 billion and $17 billion, respectively.5 The actual cost
of the mortgage interest deduction is likely to be considerably less than
OMB’s estimate, because eliminating the deduction would increase the
cost of and reduce the demand for mortgage debt (see, for example,
Follain and Dunsky 1997). Follain and Melamed (1998) estimate that
the increase in revenue would be only about one-fourth of the amount
2 Because the federal government does not tax any type of imputed income, OMB
does not treat nontaxation of owner-occupants’ imputed rent as a tax expenditure. To
avoid confusion, we have chosen to use the term “tax concession” in this article since
OMB does not regard nontaxation of imputed income as a tax expenditure and also
since the tax expenditure concept itself is a bit fuzzy around the edges. Although not
part of the tax base, income from owner-occupied housing is imputed by the U.S.
Department of Commerce in estimating gross domestic product.
3 These provisions were changed in 1997. Previously, the capital gains tax could be
deferred indefinitely if the proceeds were used to buy another principal residence.
Also, taxpayers over 55 were eligible for a one-time exclusion of up to $125,000.
4 Follain, Ling, and McGill (1993) point out that the actual savings from taxing net
imputed rent would have been less than this estimate because of the general equilib-
rium effects of the tax change, such as reductions in the quantity of housing consumed
by owner-occupants.
5 The Joint Committee on Taxation (JCT) (1998a) estimated the revenue loss from the
mortgage interest deduction to be $43 billion for the same year. Inquiries to OMB and
JCT staffs failed to uncover the reason for the difference between these two estimates.

Income Tax Concessions for Owner-Occupied Housing
523
calculated by OMB. Households would likely shift taxable investments
into nontaxed housing equity to reduce the use of more costly debt.6
Moreover, the concessions are not additive for at least two reasons.
First, imposing a tax on net imputed rent logically entails eliminating
the concessions for mortgage interest and property taxes. Although
both would remain deductible expenses, they would no longer be tax
concessions. This means, in effect, that the cost of not taxing imputed
income under current tax law also includes the cost of the two deduc-
tions. Second, due to the standard deduction, the mortgage interest and
property tax deductions are not additive. Taken together, the two deduc-
tions make it advantageous for about 45 percent of homeowners to
itemize, but if either were to be removed, the value of the remaining
one would be reduced or eliminated for many of these homeowners.
Although the concessions are not additive, they are nevertheless equiv-
alent to substantial proportions of expected individual income tax re-
ceipts (about $700 billion for 1998). The mortgage interest deduction
alone is exceeded by only two of the tax concessions calculated by OMB:
exclusion of employer contributions for medical insurance and care at
$77 billion and exclusion of employer pension plan contributions and
earnings at $56 billion (1997).
The four major concessions are not only quite large, they are extraor-
dinarily generous to individual homeowners. If net imputed income
were taxed, then mortgage interest payments and local real estate
taxes on the homeowner’s income tax return would be deducted when
calculating net imputed income. However, since the federal government
does not tax imputed income (on owner-occupied housing or any other
asset), the deductibility of mortgage interest and real estate taxes
simply adds another layer of tax concessions to the nontaxation of
imputed income.
For a number of years, various people have argued that eliminating
these tax concessions should be done on grounds of simple equity and
allocative efficiency.7 According to this argument, these concessions
favor the well-to-do and also cause resources to be allocated to the
housing sector that could be used more productively elsewhere in the
economy. It has also been argued that, by encouraging families to
purchase more housing than they would otherwise, these concessions
contribute to urban sprawl (Voith 1999). Perhaps most important, it
is argued that they do little to increase the rate of owner-occupancy
among the income groups Congress wishes to assist.
6 This is consistent with the fact that countries such as Australia, which do not allow
the mortgage interest deduction, have much lower average loan-to-value ratios than
the United States (Bourassa, Greig, and Troy 1994).
7 Examples are cited in the detailed discussions of each tax concession.

524
Steven C. Bourassa and William G. Grigsby
Despite these arguments, Congress has only nibbled at these conces-
sions.8 Mortgage interest cannot be deducted on home loan balances
in excess of $1 million, a restriction that affects very few taxpayers.
Also, to prevent the diversion of the proceeds from mortgage loans
into the financing of consumer durables, there is a cap on the amount
of interest that can be deducted in connection with mortgage refinanc-
ing. This gesture, however, is not aimed at housing but rather at the
attempt by some homeowners to circumvent the prohibition of interest
deduction on consumer loans. Indicative of the extreme reluctance of
Congress and the administration to touch owner-occupied housing itself
has been their refusal to eliminate even the tax concessions enjoyed
by owners of second homes that are not rented out.
The federal income tax code appears to treat owner-occupied housing
as a sacred cow for four reasons: First, it is widely felt that the tax
concessions encourage owner-occupancy, a highly valued national pri-
ority that has been promoted not only through the tax code, but also
through direct subsidies to lower-income households. Second, removing
these tax concessions without causing a significant reduction in the
market value of owner-occupied homes is thought to be difficult and
problematic. Third, some analysts argue that eliminating the conces-
sions would increase rather than reduce distortions in the tax code
(e.g., Woodward and Weicher 1989). Finally, several influential housing
trade associations are against eliminating the concessions because of
concern that such action would adversely affect their members. Whether
members of Congress themselves seriously subscribe to these arguments
is unclear. They may feel only that an old tax concession is a good tax
concession, unless the arguments for eliminating it are especially
compelling.
Without endorsing or rejecting the various arguments that have been
put forward over the years about the benefits of homeownership, this
article explores the question of whether the four major tax concessions
for owner-occupied housing are indeed good ones.9 The article examines
difficulties involved in eliminating each of them, as well as some of the
8 Indirectly, the value of the tax concessions has been reduced for higher-income tax-
payers. Cuts in marginal tax rates in 1981 and 1986 reduced the value of nontaxation
of imputed rents and deductible expenses. More recently, however, effective marginal
tax rates have increased for higher-income households, but limitations on deductions
have made the concessions less valuable for households with incomes above $124,500.
In addition, the Alternative Minimum Tax, which has been in effect in its current gen-
eral form since 1978, removes the deductibility of real estate taxes for higher-income
households (JCT 1998c). It does not affect the mortgage interest deduction itself, how-
ever, provided the mortgage was used to buy, build, or substantially improve the tax-
payer’s principal or second home.
9 Our evaluations of the concessions would, in any case, be unaffected even if we did
endorse such arguments.

Income Tax Concessions for Owner-Occupied Housing
525
consequences of their removal on the housing market, the tax system,
and the economy. It then suggests ways in which certain of the conces-
sions that are recommended for removal can be eliminated without
hurting existing homeowners.
The analysis is divided into eight parts. The next section briefly puts
the tax concessions into a historical and international context. The
following four sections discuss each of the concessions separately and
conclude that both the mortgage interest and real estate tax deduc-
tions should be removed. The sixth section discusses the potential
adverse impact of their removal on existing homeowners. The seventh
section deals with the critical question of how to phase in the removal
painlessly for these owners. The final section summarizes our conclu-
sions. The pros and cons of retaining current tax concessions are
summarized in table 1.
Historical and international context
The most important fact about these tax concessions is that their cre-
ation was largely accidental (Baer 1975). The mortgage interest and
real estate tax deductions became embedded in the first Internal Rev-
enue Code in 1913 without any thought to their effect on housing pur-
chases, and for nearly 30 years they had little effect because tax rates
were very low and the proportion of households who paid no tax at all
was fairly high.10 Only with the advent of high tax rates during World
War II did their impact on home purchases become significant. They
came to be defended as a revered part of the code only after the fact.
Also, nontaxation of imputed income from owner-occupied dwellings
and taxation of capital gains on home sales were not issues until after
the war.11
Like the United States, most developed countries give favored status
to owner-occupied housing. However, 10 Organisation for Economic
Cooperation and Development (OECD) countries—Belgium, Denmark,
France, Greece, Italy, Luxembourg, The Netherlands, Norway, Spain,
and Switzerland—currently tax at least some imputed income. Five
other countries—Austria, Finland, Germany, Sweden, and the United
Kingdom—have imposed such a tax in recent decades (Ministry of
Housing, Physical Planning, and the Environment, The Netherlands
10 The deductions first appeared as part of the income tax legislation passed to
finance the Civil War (Baer 1975).
11 The exclusion of owner-occupants’ imputed income from taxation was, however,
mentioned explicitly in the legislation to finance the Civil War (Baer 1975).

Table 1. Pros and Cons of Retaining Current Tax Concessions
Concession
Pros
Cons
Overall Assessment
Nontaxation
of imputed
net income
A tax on wealth would effectively be created
Problems would arise with setting a single tax
rate due to variations in rates of return
across space and time
Assessment standards vary across jurisdictions
Additional complexity of the tax system would
be onerous
Inefficiencies in allocation of capital
investments result
Pros outweigh cons:
Administrative infeasi-
bility of accurately tax-
ing net imputed income
makes it undesirable
Nontaxation
of capital
gains
Much of the nominal gain is due to general
inflation
If gains are taxed, then losses should be
deductible, thus reducing revenues
Additional complexity of the tax system would
be onerous
Inefficiencies in allocation of capital
investments may result
Pros outweigh cons:
Efficiency gains do not
appear to justify high
administrative costs and
small revenue gains
Mortgage
interest
deduction
The homeownership rate rises
The amount of housing constructed increases,
thus preventing a housing shortage
The capital value of existing homes is
maintained
At best, it has only a marginal impact on
the rate of homeownership (and may ben-
efit only those who would become home-
owners anyway, but at a later date)
It is unlikely to have a big impact on the
number of new starts
Cons outweigh pros:
Benefits of the deduction
are negligible
Capital values could be protected by phas-
ing in tax reform
Higher-income households receive undue
benefits
Renters are treated inequitably
It complicates the tax system unnecessarily
It encourages urban sprawl
Real estate tax The real estate tax is analogous to a very high Housing deductions are themselves regres- Cons outweigh pros:
deduction
sales tax, which calls for offsetting treatment
in the federal tax code
It is often unrelated to ability to pay
It is imperfectly assessed
It is regressive with respect to household income
sive, meaning that they do not offset the
regressivity of the real estate tax
The real estate tax deduction would
become more regressive after the mort-
gage interest deduction is eliminated
Benefits of the deduction
appear to be the opposite
of what is intended

Income Tax Concessions for Owner-Occupied Housing
527
1992; Robinson 1988; Scholten 1999; Werczberger 1997).12 Most of
these countries—plus Japan, Portugal, Turkey, and the United States—
allow deduction of mortgage interest payments. Only two OECD mem-
bers—The Netherlands and Sweden—tax capital gains. Australia,
Canada, and New Zealand do not allow mortgage interest or real estate
tax deductions. Nor, however, do they tax capital gains or imputed rent.
Nontaxation of net imputed income from
owner-occupied housing
As already noted, nontaxation of net imputed income is the single
largest tax concession for owner-occupants. Using time series data,
Rosen and Rosen (1980) concluded that about one-quarter (four percent-
age points) of the post–World War II growth in the homeownership rate
was due to this tax concession combined with the concessions for mort-
gage interest and property taxes. Using cross-sectional data for 1970,
Rosen (1979) reports a similar percentage. Subsequent estimates of the
impact of the tax deductions by Rosen (1989) suggest that nontaxation
of net imputed rent probably accounts for at least two percentage
points of the homeownership rate.
The nontaxation of imputed income on owner-occupied homes was orig-
inally criticized on grounds of simple fairness (e.g., Goode 1960; White
and White 1965). A family that invests $10,000 in a savings bond, for
example, is taxed on the interest earned. An identical family that in-
vests the same amount in a home is not taxed, even though the home
also yields a return, albeit one that cannot be precisely measured.
The nontaxation of net imputed rent is usually criticized on the grounds
of allocative inefficiency. Nontaxation lowers the cost of investing in
owner-occupied housing below the cost of investing in other forms of
physical capital, causing the economy to invest too much in housing
and not enough in other productive sectors, such as plant and machin-
ery (Aaron 1970; Hendershott 1987; Ott and Ott 1973). One solution
to this problem would be to liberalize the tax treatment of income
from investment in nonhousing capital (Hendershott and Hu 1980).
Another more often recommended solution, and one that would also
increase equity and allocative efficiency across income and tenure
groups, is to tax investment returns on owner-occupied housing. For
12 Wisconsin taxed imputed rental income from owner-occupied housing between
1911 and 1917 (Goode 1960). Australia may have been the first country to tax net
imputed rent. The tax was created in 1915 when the federal income tax was intro-
duced, but it was abolished eight years later, in part because it appeared to discour-
age homeownership (Reece 1975).

528
Steven C. Bourassa and William G. Grigsby
example, Nobel laureate economist William Vickrey, in his 1993 presi-
dential address to the American Economic Association, argued for
the inclusion in taxable income of the rental value of owner-occupied
residences. This would not only improve the equity and progressiv-
ity of the income tax but also go a substantial way toward making
more units available for rental and, to a modest extent, promoting
the construction of additional affordable rental housing and abat-
ing the problem of homelessness. (4)13
Congress has consistently ignored these recommendations, and perhaps
for good reason. Proposals to bring the taxation of owner-occupied hous-
ing into line with the taxation of other forms of capital investment run
into four major difficulties. The first is inherent in the concept of imput-
ed income. Because there is no measurable money income stream to
serve as the tax base, rental income must be calculated from an estimat-
ed market value and an assumed gross rental rate of return for each
property.14 The tax, therefore, is really a levy on wealth, even though
gross rent is adjusted downward to reflect occupancy costs, financing
expenses, and depreciation. Indeed, any tax on imputed income could
be considered a tax on wealth, and vice versa. Except in the case of
estate taxes, the federal government does not tax wealth, because that
is a state and local prerogative in the present federal system.
The second difficulty is that the tax bears little relationship to capaci-
ty to pay, weighing most heavily on lower-income, elderly homeowners,
the group for which most state governments now provide some form of
property tax relief for this very reason.15 Substantial exclusions would
be required to protect retired homeowners from being taxed out of
their own homes.16 Such exclusions would blunt the efficiency gains
and reduce tax revenues.
13 For other examples, see Goode (1960), Laidler (1969), and Simons (1938).
14 This exercise is made more difficult by the fact that there is no active single-family
rental market in the middle and upper price ranges that could provide comparative
data on prices and rents.
15 This also appears to be the primary reason that Follain, Ling, and McGill (1993)
found that taxation of net imputed rent, accompanied by elimination of mortgage
interest and property tax concessions, would have a disproportionately adverse
impact on lower-income households (Crowe 1993).
16 In a study of the impact of a hypothetical tax on net imputed rent proposed for
Australia, Bourassa and Hendershott (1994) recommended exempting retired house-
holds or taxing them at a reduced rate. For working-age households, the proposed tax
was found to be progressive when net imputed rent was included in household income.
Because net imputed rent is not cash income, the finding of progressivity did not
imply anything about ability to pay.

Income Tax Concessions for Owner-Occupied Housing
529
A third set of difficulties relates to the fact that both current and long-
term net rates of real return vary substantially from one property to
another, depending on location and time of purchase. This variation
makes it grossly inaccurate to apply a single imputed rate of return
across all properties. For example, properties in a depressed inner-city
neighborhood may experience declining values and high rental returns
relative to values, while properties in a more prosperous neighborhood
or city may have rapidly appreciating values and low rental returns
relative to values. To assume a constant rate of net rental return
across properties would ignore the wide disparity in rates of appreci-
ation or depreciation.
A final difficulty in trying to tax imputed income is administrative.
The Internal Revenue Service (IRS) does not have the resources to
treat homeowners as landlords and process 65 million Schedule C in-
come and expense statements. To tax imputed income, some sort of
simplification would be necessary. In the OECD countries that tax
imputed income, the taxable rent is set at a specified fraction of the
dwelling’s assessed value minus mortgage interest payments and (in
two of the countries) imputed maintenance (Scholten 1999). For all
practical purposes, even this streamlined approach is too cumbersome
to adopt here. The large number of local property-taxing jurisdictions
in the United States, with their wide range of valuation standards
and methods, makes it extremely difficult and expensive to develop a
common basis for assessment across jurisdictions. The federal govern-
ment would have to create a large tax equalization facility or rely on
taxpayers to accurately report the market value of their homes. Nei-
ther approach seems feasible.
Nontaxation of capital gains
If the investment returns that homeowners earn on the equity in their
homes cannot feasibly be taxed, then surely, one might argue, the cap-
ital gains these owners receive should be taxed. The taxation of capi-
tal gains, however, also raises difficult policy and administrative issues.
Three are especially important. First, a large part of the capital gain
on a particular home sale, or for that matter any asset, may be due to
general inflation and hence be fictitious.
Second, contrary to what is popularly believed, in the absence of mod-
ernization, a large proportion of the owner-occupied stock depreciates
over time. If capital gains from home sales are taxed, capital losses
should also be deductible. If that were the case, the net revenue that
would accrue to the Treasury from taxing capital gains from home
sales would be considerably less than the figure the OMB currently
estimates would be received.

530
Steven C. Bourassa and William G. Grigsby
Third, an accurate accounting of capital gains and losses requires ac-
curate records of improvement expenditures by owner-occupants and
careful auditing by the IRS. If all of the 4 to 5 million homeowners
who sell their homes each year reported their capital gains and losses,
the burden on them and on the IRS would be unacceptably large.
In exempting all but very large capital gains from taxation, Congress
has probably resolved these issues as satisfactorily as possible.
Deductibility of mortgage interest
Since the Internal Revenue Code does not treat homeowners as their
own landlords and require them to report imputed rental income,
logic would seem to dictate that the code also should not allow home-
owners to take deductions from this unreported income (Goode 1960).
Yet, despite a continuous stream of analyses calling for its elimination,
the mortgage interest deduction survives. Defenders argue that (1) it
increases the rate of homeownership, helping families achieve an es-
sential element of the American dream (e.g., Tretzger 1998); (2) it in-
creases the volume of housing constructed, enabling the home building
industry to accommodate an expanding population without recourse to
direct subsidies;17 and (3) eliminating it would seriously depress house
values, causing unfair capital losses to homeowners whose purchase
decisions were based in part on the availability of the deduction (e.g.,
Capozza, Green, and Hendershott 1997). Opponents dispute these
arguments and argue further that the deduction (1) provides undue
benefits to higher-income households at the expense of lower-income
households (e.g., Baer 1975; Dreier 1997; Pierce 1989), (2) unfairly
favors homeowners over renters (e.g., Goode 1960; White and White
1977), and (3) causes excessive consumption of housing because it in-
creases housing expenditures among higher-income groups that do not
need any subsidy to acquire adequate accommodation (e.g., Baer 1975;
Hendershott and Hu 1981). We will examine five of these points of con-
tention in this section, saving the question of potential capital losses
for the section on market impact of tax reform.
Impact on the rate of homeownership
The primary purpose of the mortgage interest deduction is allegedly
to increase the homeownership rate. Through owner-occupancy, fami-
lies are presumably able to acquire better housing, increased finan-
17 For example, the National Association of Home Builders has lobbied Congress to
preserve the mortgage interest deduction, evidently because of its perceived impact
on housing starts (see Downey 1989).

Income Tax Concessions for Owner-Occupied Housing
531
cial security, higher social status, a greater stake in society, and more
control over and satisfaction with their living environments. The de-
duction does not appear, however, to have had a significant impact on
the country’s overall rate of homeownership.
A weak measure of possible impact is provided by a comparison with
Canada and Australia, two countries that are strikingly similar to the
United States except for their much smaller populations. Although
both of these countries provide less favorable tax treatment for home-
ownership than the United States, they have similar rates of owner-
occupancy.18 Rosen (1989) has provided a more rigorously derived esti-
mate. According to his calculations, without the mortgage interest de-
duction, the homeownership rate in the United States in 1988 would
have been about 62 percent instead of the 63.7 percent actually ob-
tained in that year. One cannot, unfortunately, tell from such a com-
parison whether eliminating the deduction would have locked most of
the affected families permanently in the rental market or only post-
poned their purchase for a few years. Accelerated purchases raise the
rate of homeownership without reducing the proportion of households
that will never become owner-occupants.19 Also, we simply do not know
enough about the characteristics and tenure considerations of potential
itemizers at the margins between owning and renting to determine who
is influenced by the mortgage interest deduction. We suspect, however,
that only a tiny fraction of potential home buyers get to the point of
making a pro forma tax calculation on Form 1040 to decide whether
to buy or not.
Even more troubling than the questionable impact of the mortgage
interest deduction on the overall rate of owner-occupancy is the fact
that low- and moderate-income families, which Congress wants to help
become homeowners, benefit least from the deduction. Fully 90 percent
of the benefit goes to homeowners with incomes over $50,000 a year
(table 2, line 4b) because most low- and moderate-income homeowners
find that taking the standard deduction is more advantageous than
itemizing expenses. Whereas over 85 percent of homeowners earning
more than $75,000 a year itemize, less than one-sixth of those earning
under $30,000 do so (table 2, line 3). If the standard deduction were
raised, as is often suggested, even fewer low- and moderate-income
18 The ownership rates for 1996 were 69.8 percent in Australia, 64.0 percent in Cana-
da, and 65.4 percent in the United States (Australian Bureau of Statistics 1999; Sta-
tistics Canada 1999; U.S. Bureau of the Census 1998).
19 Bourassa et al. (1994) concluded that removing Australia’s substantial subsidies to
first-time home buyers in 1990 merely delayed the time to first ownership by two
years. Heavy subsidies to households that would soon become homeowners in any
case are not justified by the resulting small gains in aggregate homeownership rates.

Table 2. Benefits from the Mortgage Interest Deduction by Income Category, 1998a
Under
$20,000
$20,000–
$29,999
$30,000–
$39,999
$40,000–
$49,999
$50,000–
$74,999
$75,000–
$99,999
$100,000–
$199,999 $200,000+
1. a. Taxable returns, owners and renters
(thousands)
b. Percentage of total returns
2. a. Mortgage interest itemizers
(thousands)
b. Percentage of returns within income
group (line 2a divided by line 1a)
3. Percentage of owner-occupants who
itemize
4. a. Mortgage interest tax concessions
(millions of $)
9,915
12,446
12,860
11,269
19,167
9,976
8,367
2,125
12
359
14
1,134
15
2,375
13
3,080
22
8,201
12
6,358
10
6,306
2
1,554
4
3
9
16
18
34
27
45
43
67
64
86b
75
86b
73
86b
131
466
1,238
2.6
2,270
4.8
7,667
16.3
10,029
21.4
15,739
33.5
9,438
20.1
b. Percentage of total mortgage interest
tax concessions
5. a. Tax liability, all taxable returns
(millions of $)
0.3
1.0
–11,587
14,101
29,025
38,022
99,413
94,567
164,726
272,265
b. Percentage of total taxes paid
6. Average tax liability for all taxable
returns (line 5a divided by line 1a, $)
7. Average tax liability if mortgage
interest tax concession is abolished ($)
8. a. Average mortgage interest tax
concession for itemizers (line 4a
divided by line 2a, $)
–1.6
–1,169
2.0
1,133
4.1
2,257
5.4
3,374
14.2
5,187
13.5
9,480
23.5
19,688
38.9
128,125
–1,155
365
1,170
410
1,965
521
3,573
737
5,587
935
10,485
1,577
21,570
2,496
132,567
6,073
b. As a percentage of average tax liabilityc
—
35.0
26.5
20.6
16.7
14.6
11.6
4.6
Source: JCT (1998a) and authors’ calculations.
a The JCT defines income categories according to adjusted gross income plus tax-favored income excluded from the adjusted gross income.
b The average across the highest three categories is listed.
c This figure is based on the assumption that within income groups, average tax liability for itemizers and nonitemizers would have been the same in the
absence of the mortgage deduction.

Income Tax Concessions for Owner-Occupied Housing
533
taxpayers would itemize, and the mortgage interest deduction would
become even more a benefit targeted at upper-income families.20
The small proportion of lower-income owner-occupants who itemize
receive substantial tax benefits from the mortgage interest deduction.
For owners with annual incomes under $30,000, the deduction lowers
their federal income tax liability by 35 percent (table 2, line 8b). The
conjunction of a relatively small proportion of itemizers in this
income group and relatively large tax benefits for those who itemize
may seem incongruous. One would think that the large benefits to
the few would be sought after by the many. No doubt, however, most
of these itemizers have either temporarily low incomes or temporarily
large medical deductions in combination with other deductions.
Regardless of how these data are interpreted, it does seem likely that
because homeownership has become so much a part of the American
dream, the mortgage interest deduction has had only a marginal
impact on tenure choice overall. The deduction is too poorly targeted
to achieve its objective.
Impact on the volume of new housing construction
For the past half-century, federal policy toward the residential con-
struction industry has been influenced by the belief that because the
price of new housing is beyond the reach of such a large segment of
the population, home builders are unable to sell enough unsubsidized
dwelling units each year to accommodate household growth (Grigsby
and Baratz 1986). Many housing subsidies over the years, including
the mortgage interest deduction, have been justified in part on this
assumption. In this connection, it is worth noting that over the past
three decades, there has been a substantial increase in vacancies in
the low-rent sector of the market in most metropolitan areas, sug-
gesting that while the subsidies may have improved housing quality,
all of them may not have been needed. Also, a portion of the assisted
housing starts simply substituted for homes that would have been
built anyway had the subsidies not been in place.
With respect to the question of whether the mortgage interest deduc-
tion acts as a stimulus to home building, it is clear that to the extent
that the deduction is not capitalized into home values—a possibility
to be discussed later in the article—it enables itemizers to buy more
expensive housing than would otherwise be possible. The deduction
20 Follain and Ling (1991) point out that the Tax Reform Act of 1986 reduced the value
of the mortgage interest deduction for many low- and moderate-income households by
increasing the standard deduction and by making certain formerly deductible expenses
nondeductible.

534
Steven C. Bourassa and William G. Grigsby
has the potential, therefore, of moving a certain number of buyers
upward from the market for inexpensive used homes into the market
for slightly more expensive, but still modestly priced, new units.
The size of this group, hence the potential impact of the deduction on
the volume of new construction, varies across market areas, depending
on the affordability of the least expensive newly constructed homes in
these markets. If the minimum-price new home in a particular market
is already affordable to moderate-income families taking the standard
deduction, then eliminating the mortgage deduction would have little,
if any, effect on the amount of new construction. There would be mod-
est price and cost adjustments on the supply side of the market and,
perhaps, spending adjustments among a few families on the demand
side without affecting the number of units built. This is likely to hap-
pen in market areas where home builders are able to reach far enough
down the income scale to include families normally taking the standard
deduction or those whose itemized deductions exceed the standard de-
duction by only a small amount. In a few market areas, however, where
housing is quite expensive and new housing is apparently out of reach
for a large majority of households, eliminating the deduction might
have an adverse effect on the volume of housing starts. Whether the
number of starts would drop enough to create a housing shortage, how-
ever defined, is a question that requires more study.
Undue benefits to higher-income households
The mortgage interest deduction is widely criticized because it reduces
the progressivity of the federal income tax. For example, a family with
a taxable income of $40,000 and a marginal tax rate of 15 percent
would receive only $1,500 in tax benefits from mortgage interest pay-
ments of $10,000 (assuming for the purpose of simplicity, no standard
deduction), while a family with a taxable income of $60,000 and a mar-
ginal tax rate of 28 percent would gain $2,800 in benefits from the same
deduction. The regressivity is made more pronounced by the standard
deduction, which, as discussed earlier, makes it disadvantageous for
most low- and moderate-income homeowners to itemize their interest
expenses (table 2, line 3).
It is interesting to note that the $200,000+ income group contributes
proportionately much more in taxes than it receives in concessions.21
21 This is undoubtedly due to the limitations on deductions applicable to higher-
income households. The JCT (1998b) estimated that 4.5 million households would be
affected by the limitation in 1998, including 95 percent of households with incomes
above $200,000 (defined by the JCT to include adjusted gross income plus tax-favored
income excluded from adjusted gross income). Only about a third of households with
incomes between $100,000 and $200,000 and a negligible number of households with
lower incomes would be affected.

Income Tax Concessions for Owner-Occupied Housing
535
Households in the highest income group receive only about 20 percent
of the mortgage interest tax concessions, even though they account for
almost 40 percent of taxes paid.
Contrary to popular opinion, then, the mortgage interest deduction is
almost entirely a tax concession from the wealthy ($200,000+) to the
well-to-do ($50,000 to $200,000) and from well-to-do nonitemizers to
well-to-do itemizers. Low- and moderate-income homeowners (below
$50,000) are, on average, only slightly worse off because of the deduc-
tions, and it is likely that many of the nonitemizing homeowners in
this income group benefited from the deductions when their mortgage
loans were larger.
On balance, the very modest regressivity of the mortgage interest de-
duction does not by itself warrant eliminating it, especially given the
fact that, even with this tax concession in place, the overall tax struc-
ture remains progressive. Since, however, the deduction contributes
only marginally, at best, to achieving its primary alleged objective—an
increase in the rate of homeownership among households that could
not otherwise afford to become homeowners—its regressivity adds still
another reason for eliminating it.
Some researchers have emphasized the fact that mortgage interest de-
ductibility achieves neutrality between debt and equity financing of
owner-occupied housing (Bourassa and Hendershott 1992; Ling and
McGill 1992; Woodward and Weicher 1989). Even so, providing well-to-
do homeowners with large tax concessions is a clumsy way to achieve
this small gain in efficiency.
Inequity to renters
Since renters cannot take the mortgage interest deduction at all, they
appear to be even more disadvantaged by the Internal Revenue Code
than lower-income owner-occupants. However, comparisons across ten-
ure involve a different set of considerations, and we must look at the
treatment of renters separately. The question is perhaps most appro-
priately phrased as follows: To the extent that tax deductions confer
benefits on the owner-occupancy market, are the beneficiary home-
owners favored over renters, or are there circumstances that might
make renters either more or less disadvantaged?
The short answer to this question is that inequities do exist but not
nearly to the degree assumed. If households are unrestricted in their
tenure choice, they are obviously not penalized, in the strict sense of
the word, if they choose to rent. For a variety of reasons, however, some
households are confined, as it were, to the rental sector. For example,
they may require only two or three rooms, and nothing that small is

536
Steven C. Bourassa and William G. Grigsby
available in the owner-occupancy market in their community. Or they
must live in a location where few or no single-family homes are avail-
able in their price range. Or they change residence so frequently that
the annualized costs associated with purchase and sale are excessive.
Or they are barred from owner-occupancy by unnecessarily conservative
mortgage lending practices. In all of these situations, renters cannot
compete for housing of as high quality as would be within their reach
if the benefits of tax concessions in the rental and owner-occupancy
markets were more or less equal.
To the degree that tax concessions increase housing starts, renters
may benefit indirectly, but the horizontal inequity relative to owner-
occupants remains. Still, for several reasons, the extent of the inequity
may be less than it seems. As explained later, some of the concessions
to owner-occupants are capitalized into house prices, meaning that
landowners and developers receive part of the benefit.22 Also, a large
percentage of current renters have been or will be owners at some
point during their lifetime, so any tax inequity they now suffer as
renters will be partially or fully offset by their favored position later
when they are owners. In addition, many renters do not pay income
taxes, so they cannot be said to be treated inequitably in the Internal
Revenue Code with respect to households that do pay taxes. Moreover,
most renters (as is already true of a majority of homeowners) would
not itemize deductions on their tax returns, even if they did become
owner-occupants. Further, the rise of condominiums has made small
dwellings more broadly available for owner-occupancy. And as already
suggested, some renters have incomes high enough to suggest that
they have ample freedom to choose the tenure status they wish. Final-
ly, the tax treatment of investment in rental housing mitigates some
of the apparent disadvantage to renters.23
Excessive consumption of housing
As noted earlier, analysts have argued that failure to tax imputed in-
come from owner-occupied housing results in overinvestment in and
overconsumption of housing, particularly for upper-income house-
holds. Because owner-occupied housing is favored by the Internal Rev-
enue Code, it is argued that owners invest too much in housing and
not enough in other assets that would be more productive for the U.S.
economy. If imputed income were taxed, then mortgage interest would,
of course, be deductible. However, given that imputed income is not now
22 However, White and White (1977) conclude that to the extent that capitalization of
benefits occurs in the owner-occupied sector, rents would increase in the rental sec-
tor, so that renters are themselves disadvantaged by higher house prices.
23 See the list of tax concessions in footnote 1.

Income Tax Concessions for Owner-Occupied Housing
537
part of the federal government’s tax base and is not likely to become
so, the mortgage interest deduction itself is subject to the criticism
that it encourages overconsumption of housing, particularly among
higher-income households. This is a problem only to the extent that the
spending diverted to housing would have contributed to the efficiency
and productivity of the U.S. economy. If, instead, the spending would
have been mainly on other luxuries, such as imported cars or interna-
tional vacations, then the misallocation of resources to housing con-
sumption and investment may not matter or may even be desirable.24
Recent empirical research by Capozza, Green, and Hendershott (1997)
concludes that the basic premise of the overconsumption argument is
incorrect. According to their research findings, housing tax concessions
do not affect housing consumption and investment at all, because they
are fully capitalized into residential land prices. If Capozza and col-
leagues are correct, then it is also true that housing tax incentives do
not produce any of the other benefits or problems attributed to them:
that is, increased owner-occupancy, larger volumes of new construction,
favoritism to owners over renters, or undue benefits to higher-income
households.
However, Sinai (1997), in research critical of the Capozza et al. analy-
sis, concludes that there is a much lower rate of capitalization, on the
order of 14 percent. This conclusion seems more plausible. Full capi-
talization of the housing deductions implies an inelastic long-run sup-
ply curve for urban residential land, an argument that both economic
theory and urban sprawl call into question. Whether the added con-
sumption that the housing deductions encourage is in some sense ex-
cessive is, of course, a value judgment. Since the deductions benefit the
income groups least in need of assistance, we find that it is excessive.
On balance, our analysis leads us to the conclusion that the arguments
for eliminating the mortgage interest deduction considerably outweigh
those in favor of retaining it. Although encouraging homeownership
may be desirable, the deduction does little to achieve that goal. Rather,
it primarily provides large and unnecessary tax relief to upper-income
households. Equally, it does little to increase the number of housing
starts, except possibly in a few high-cost market areas. Members of
Congress may realize that the mortgage interest deduction is, indeed,
a clumsy policy tool, but they may also share the widespread fear that
removing it would hurt their home-owning constituents by causing
capital losses. To address this concern, we will show later how the
deduction can be painlessly removed.
24 There is also the difficult question, which we will not attempt to address here, of
whether conspicuous consumption by the wealthy has a negative impact on those
who are less wealthy.

538
Steven C. Bourassa and William G. Grigsby
Deductibility of local real estate taxes
From the perspective of federal tax policy, the deductibility of local
real estate taxes, like the deductibility of mortgage interest, should be
eliminated. From the perspective of the federal-state-local tax system,
however, there are several arguments for retaining the property tax
concession as an offset to a not very efficient or equitable local tax sys-
tem. The first argument has to do with the size of the real estate tax.
Although the tax is a small percentage of the value of the home on
which it is levied, averaging slightly more than 1 percent across U.S.
cities (Ling and McGill 1992), it is quite a large proportion of annual
housing expenditures. The real estate tax typically accounts for about
15 percent of a homeowner’s annual cash costs for housing.25 Excluding
financing costs, property taxes rise to as much as 40 percent of out-of-
pocket expenditures. The property tax is analogous to a very high sales
tax and, as such, is thought to repress both housing consumption and
housing investment.26 Because the property tax is so high, most states
have enacted various types of tax relief, such as homestead exemptions
and tax caps, to lessen its impact on either selected groups of owners
or owners as a whole. Deductibility of the tax on income tax returns
provides similar relief, but to a more well-to-do group of households.
The second problem with the local real estate tax is that it is imper-
fectly related to income and, therefore, often does not reflect ability to
pay. This is an especially serious problem for elderly homeowners and
is one of the reasons for the tax caps just mentioned.
A third problem relates to tax administration. In many communities,
real estate assessment practices are so poor that the taxes homeown-
ers pay do not accurately reflect the value of their properties. Although
improvements in local assessment procedures have begun to ameliorate
this problem, it is still widespread.
The final problem has to do with the regressivity of the local real estate
tax. Data reported by Ling and McGill (1992), based on the 1989 Amer-
ican Housing Survey, indicate that property taxes averaged across the
United States are regressive (table 3).27 Deductibility of property taxes
25 This is based on median housing costs from the 1989 American Housing Survey
(U.S. Department of Commerce and U.S. Department of Housing and Urban Develop-
ment 1991, table 3–13). Netzer (1966, table 2–8) reported a figure of 17 percent,
based on the 1960 Census of Housing.
26 Some analysts see the tax simply as a payment for local public services, but more
than 50 percent of the tax in a large proportion of communities funds public educa-
tion, although most households do not have school-age children. The cost of housing
is expensive in part because public education is expensive.
27 As table 3 shows, however, the taxes are progressive with respect to owners’ equity.

Income Tax Concessions for Owner-Occupied Housing
539
Table 3. Regressivity of the Real Estate Tax, 1989
Average
Adjusted
Gross
Income
($)
Tax as a
Percent of
Adjusted
Gross
House
Value
($)
Average
Equity
($)
Tax as a
Percent
of Equity
Adjusted Gross
Income Range ($)
Average
Tax ($)
Income
0–5,000
5,000–10,000
10,000–15,000
15,000–20,000
20,000–25,000
25,000–30,000
30,000–35,000
35,000–40,000
40,000–45,000
45,000–50,000
50,000–60,000
60,000–75,000
75,000–100,000
100,000–120,000
120,000+
1,575
7,598
60,876
66,839
66,790
74,435
72,261
76,244
78,089
90,650
99,175
107,776
115,110
128,633
170,067
196,363
260,522
51,927
56,679
52,029
57,538
48,054
43,688
45,760
53,846
55,736
61,648
64,116
76,923
101,020
109,178
131,824
593
682
37.7
9.0
5.5
4.5
3.5
3.0
2.6
2.5
2.6
2.4
2.3
2.1
2.3
2.2
2.0
1.1
1.2
1.4
1.4
1.7
1.9
1.9
1.8
2.0
1.9
2.0
1.8
2.0
2.2
2.3
12,826
18,235
23,163
27,788
32,927
38,023
42,822
48,017
55,283
67,130
87,398
108,486
152,856
711
820
802
842
868
967
1,112
1,156
1,260
1,414
1,991
2,435
3,081
All
35,855
94,675
63,622
1,042
2.9
1.6
Source: Derived from Ling and McGill (1992), table 1.
could, in theory, help offset this regressivity, but as discussed earlier,
the regressivity of the housing deductions themselves means that this
does not happen in practice.
Although, collectively, the reasons for leaving the deductibility of local
real estate taxes intact seem compelling, they are not. The deductions
do nothing to offset the inequities just described. Moreover, if the de-
ductibility of mortgage interest is eliminated, then the number of
households finding it to their advantage to itemize deductions would
drop dramatically to a very low level, and the few households that con-
tinued to deduct real estate taxes would be those with high nonhous-
ing deductions and high incomes. An already regressive real estate tax
deduction would become more regressive.
Market impact of removing mortgage interest
and real estate tax deductions
If Congress removed the mortgage interest and real estate tax de-
ductions for owner-occupied housing, existing owner-occupants with
outstanding mortgage loans could and, in our view, should be held
harmless with respect to these deductions. Even with this protection,
however, many homeowners would be adversely affected if the deduc-
tions were eliminated overnight rather than phased out gradually.

540
Steven C. Bourassa and William G. Grigsby
Buyers could no longer afford homes as expensive as those they had
previously been able to purchase and would move down-market, leav-
ing an overhang of used housing in the upper reaches of the market.
Even if home building in the higher price ranges stopped immediately,
the overhang would persist, as homeowners had to move for one rea-
son or another in the normal course of events. While the price declines
would encourage some buyers to spend more for housing than they
would have otherwise, the net effect in the short run would still be a
drop in home values.
Various analysts have estimated that these capital losses might range
from 10 to 20 percent or more, depending on market conditions and
other factors (Brimer, Lasky, and Wyss 1995; Buchanan 1996; Capoz-
za, Green, and Hendershott 1996, 1997; Cecchetti and Rupert 1996).
The estimated reductions are greatest in high-price cities, with losses
of more than 15 percent in Boston due to elimination of the property
tax deduction and of over 20 percent in Honolulu, San Francisco, and
San Jose (CA) due to elimination of the mortgage interest deduction
(Capozza, Green, and Hendershott 1997). Even though, as argued
earlier, it seems unlikely that the deductions would be fully capital-
ized into home values over the long term, these percentages may be
reasonable estimates of the probable average impact on house prices
in various metropolitan areas in the short run. Follain and Melamed
(1998) suggest, however, that “a substantial portion of the impact will
be absorbed by the mortgage market because of portfolio reshuffling
on the part of households, especially higher-income households” (197).
The amount of these price declines would be different in different seg-
ments of local housing markets, depending on the proportion of buyers
in a particular submarket who had previously been able to take advan-
tage of the deductions. Prices at the top end would drop the most, be-
cause despite portfolio reshuffling, most households in this price range
would opt for less housing, resulting in a general shift in demand from
higher-price to somewhat lower-price homes. Demand for homes in the
middle price ranges would stay approximately the same, as buyers who
were forced to shift downward and into these markets replace other
buyers who had to shift into less expensive sectors. Prices at the bot-
tom of the value spectrum would increase, as households already in
this market would be joined by those moving downward from the more
expensive sectors.
Perhaps the most important point to emphasize is the fact that the
potential adverse impacts under any scenario could be avoided by an
appropriate phase-out procedure. We describe such a procedure in the
next section.

Income Tax Concessions for Owner-Occupied Housing
541
Phasing out the income tax deductions
Proposals to abolish or limit the mortgage interest and real estate tax
deductions are always met with strong opposition on the grounds that
existing homeowners would suffer the huge capital losses just described.
Phasing in the tax reform over a period of 15 to 20 years should, how-
ever, make this transition almost painless. All that is necessary is fix-
ing a date after which no more deductions are allowed. If the cutoff
date were, say, January 1, 2021, for all buyers who purchase after De-
cember 31, 2000, then buyers in 2001 would get 20 years of deductions,
buyers in 2002 would get 19 years of deductions, and so on. Households
buying before January 1, 2001, would be allowed to deduct mortgage
interest and real estate taxes until their original mortgage loans were
paid off. If the deductions were phased out slowly in this fashion, there
would be virtually no adverse effect on demand for, or capital values
of, expensive homes in the existing stock. On the supply side, as home
builders reduced the amount of expensive construction during the
phase-out period, home sellers in this price bracket would face less
competition from the new construction market. On the demand side,
the gradual reduction in demand for expensive homes caused by phas-
ing out the deductions would be offset by a corresponding gradual in-
crease in demand due to income and population growth.
At the lower end of the market, home builders would gradually increase
construction of less expensive housing in response to the shift in de-
mand. If, however, it seemed necessary at some point to stimulate de-
mand in this sector of the market, various subsidy methods could be
used to do so.
Conclusions
In this article, we have reviewed the four main tax concessions for
owner-occupied housing: nontaxation of net imputed income, exemp-
tion of most capital gains, deductibility of mortgage interest, and de-
ductibility of real estate taxes. We conclude that the first two conces-
sions should remain and that the last two should be abolished. As a
package, these four recommendations are precisely the opposite of
what good tax policy would seem to demand: that imputed rental in-
come and real capital gains should be taxed and that mortgage inter-
est and property taxes should be allowable deductions. But both con-
ceptual and administrative problems argue against taxing imputed
income and capital gains, and if the IRS leaves these two concessions
untouched, the justification for retaining the mortgage interest and
property tax deductions largely evaporates. A major source of resis-
tance to eliminating these deductions is the belief that doing so would
bring about significant declines in the market value of most existing

542
Steven C. Bourassa and William G. Grigsby
owner-occupied homes. Phasing out the deductions over a period of 15
to 20 years, however, would prevent capital losses from occurring.
Some analysts have proposed that the savings from eliminating the two
deductions be reallocated to promote homeownership programs for low-
and moderate-income households. For example, in a recent review of
U.S. housing policy, Dreier (1997) recommended replacing the mortgage
interest deduction with a $50 billion progressive homeownership tax
credit. Even more recently, Green and Vandell (1999) recommended a
revenue-neutral replacement of the mortgage interest and property
tax deductions with a fixed tax credit available to all homeowners.28
They estimated that the homeownership rate would increase by three
to five percentage points and even more in lower-income categories.
Because the tax credits would be unaffected by the standard deduction
(and, in Dreier’s proposal, they would be progressive), this approach
would be much better targeted than the mortgage interest deduction.
While we agree with this conclusion, the question of whether redirect-
ing the mortgage interest deduction in this fashion would be the most
appropriate way of providing homeownership assistance for the groups
Congress wishes to help goes well beyond the scope of this article, par-
ticularly since our proposed phase-out period, if adopted, would leave
the present structure of deductions relatively unchanged for a consid-
erable length of time.
Authors
Steven C. Bourassa is Professor and Chair of the Department of Urban and Public
Affairs at the University of Louisville. William G. Grigsby is Professor Emeritus of
City and Regional Planning at the University of Pennsylvania. This article has bene-
fited from comments provided by Patric Hendershott, Richard Voith, and two anony-
mous referees.
References
Aaron, Henry. 1970. Income Taxes and Housing. American Economic Review
60(5):789–806.
Australian Bureau of Statistics. 1999. 1995–96 Survey of Income and Housing Costs.
World Wide Web page <http://www.statistics.gov.au/websitedbs/…> (last modified
February 25).
Baer, William C. 1975. On the Making of Perfect and Beautiful Social Programs. Pub-
lic Interest (39):80–98.
Bourassa, Steven C., Alastair W. Greig, and Patrick N. Troy. 1994. The Limits of
Housing Policy: Home Ownership in Australia. Housing Studies 10(1):83–104.
28 Follain, Ling, and McGill (1993) also recommend a tax credit.

Income Tax Concessions for Owner-Occupied Housing
543
Bourassa, Steven C., Donald R. Haurin, R. Jean Haurin, and Patric H. Hendershott.
1994. Independent Living and Home Ownership: An Analysis of Australian Youth.
Australian Economic Review (107):29–44.
Bourassa, Steven C., and Patric H. Hendershott. 1992. Changes in the Relative
Incentives to Invest in Housing: Australia, Sweden, and the United States. Journal
of Housing Economics 2(1):60–83.
Bourassa, Steven C., and Patric H. Hendershott. 1994. On the Equity Effects of
Taxing Imputed Rents: Evidence from Australia. Housing Policy Debate 5(1):73–95.
Brimer, Roger, Mark Lasky, and David Wyss. 1995. The Real Estate Market Impacts
of a Flat Tax. Boston: Data Resources International.
Buchanan, Neil H. 1996. A Critique of Competing Plans for Radical Tax Restructuring.
Working Paper No. 173. Annandale-on-Hudson, NY: Bard College, Jerome Levy
Economic Institute.
Capozza, Dennis R., Richard K. Green, and Patric H. Hendershott. 1996. Taxes, Mort-
gage Borrowing, and Residential Land Prices. In Economic Effects of Fundamental
Tax Reform, ed. Henry J. Aaron and William G. Gale, 171–208. Washington, DC:
Brookings Institution.
Capozza, Dennis R., Richard K. Green, and Patric H. Hendershott. 1997. Taxes and
House Prices. Working Paper Series No. 97–16. Ohio State University, Charles A.
Dice Center for Research in Financial Economics.
Cecchetti, Stephen G., and Peter Rupert. 1996. Mortgage Interest Deductibility and
Housing Prices. Federal Reserve Bank of Cleveland Economic Commentary, February,
pp. 1–4.
Crowe, David A. 1993. Comment on James R. Follain, David C. Ling, and Gary A.
McGill’s “The Preferential Income Tax Treatment of Owner-Occupied Housing: Who
Really Benefits?” Housing Policy Debate 4(1):25–32.
Downey, Kristin. 1989. Mortgage-Interest Tax Deduction in Danger? Philadelphia
Inquirer, January 31, pp. 1C, 6C.
Dreier, Peter. 1997. The New Politics of Housing: How to Rebuild the Constituency for
a Progressive Federal Housing Policy. Journal of the American Planning Association
63(1):5–27.
Follain, James R., and Robert M. Dunsky. 1997. The Demand for Mortgage Debt and
the Income Tax. Journal of Housing Research 8(2):155–99.
Follain, James R., and David C. Ling. 1991. The Federal Tax Subsidy to Housing and
the Reduced Value of the Mortgage Interest Deduction. National Tax Journal 44:147–68.
Follain, James R., David C. Ling, and Gary A. McGill. 1993. The Preferential Income
Tax Treatment of Owner-Occupied Housing: Who Really Benefits? Housing Policy
Debate 4(1):1–24.
Follain, James R., and Lisa Sturman Melamed. 1998. The False Messiah of Tax
Policy: What Elimination of the Home Mortgage Interest Deduction Promises and a
Careful Look at What It Delivers. Journal of Housing Research 9(2):179–99.
Goode, Richard. 1960. Imputed Rent of Owner-Occupied Dwellings under the Income
Tax. Journal of Finance 15:504–30.

544
Steven C. Bourassa and William G. Grigsby
Green, Richard K., and Kerry D. Vandell. 1999. Giving Households Credit: How
Changes in the U.S. Tax Code Could Promote Homeownership. Regional Science and
Urban Economics 29:419–44.
Grigsby, William, and Morton Baratz. 1986. Residential Investment: Too Much or Too
Little? Research Report Series No. 7. Philadelphia: University of Pennsylvania,
Department of City and Regional Planning.
Hendershott, Patric H. 1987. Tax Reform and the Slope of the Playing Field. In Taxes
and Capital Formation, ed. Martin Feldstein, 51–61. Chicago: University of Chicago
Press.
Hendershott, Patric H., and Sheng-Cheng Hu. 1980. Government-Induced Biases in
the Allocation of the Stock of Fixed Capital in the United States. In Capital, Efficiency,
and Growth, ed. George M. von Furstenberg, 323–60. Cambridge, MA: Ballinger.
Hendershott, Patric H., and Sheng-Cheng Hu. 1981. Inflation and Extraordinary
Returns on Owner-Occupied Housing: Some Implications for Capital Allocation and
Productivity Growth. Journal of Macroeconomics 3(2):177–203.
Joint Committee on Taxation, U.S. Congress. 1998a. Estimates of Federal Tax Expen-
ditures for Fiscal Years 1999–2003. Washington, DC: U.S. Government Printing Office.
Joint Committee on Taxation, U.S. Congress. 1998b. Present Law and Analysis Relat-
ing to Individual Effective Marginal Tax Rates. Washington, DC: U.S. Government
Printing Office.
Joint Committee on Taxation, U.S. Congress. 1998c. Present Law and Issues Relating
to the Individual Alternative Minimum Tax (“AMT”). Washington, DC: U.S. Government
Printing Office.
Laidler, David. 1969. Income Tax Incentives for Owner-Occupied Housing. In Taxa-
tion of Income from Capital, ed. Arnold C. Harberger and Martin J. Bailey, 50–76.
Washington, DC: Brookings Institution.
Ling, David C., and Gary A. McGill. 1992. Measuring the Size and Distributional
Effects of Homeowner Tax Preferences. Journal of Housing Research 3(2):273–303.
Litzenberger, Robert H., and Howard B. Sosin. 1978. Taxation and the Incidence of
Homeownership across Income Groups. Journal of Finance 33(3):947–61.
Ministry of Housing, Physical Planning, and the Environment, The Netherlands. 1992.
Statistics on Housing in the European Community 1992. The Hague.
Netzer, Dick. 1966. Economics of the Property Tax. Washington, DC: Brookings
Institution.
Office of Management and Budget. 1997. Analytical Perspectives: Budget of the
United States, Fiscal Year 1998. Washington, DC.
Ott, David J., and Attiat F. Ott. 1973. The Effect of Non-Neutral Taxation on the Use
of Capital by Sector. Journal of Political Economy 81(4):972–81.
Pierce, Bethane Jo. 1989. Homeowner Preferences: The Equity and Revenue Effects
of Proposed Changes in the Status Quo. Journal of the American Taxation Association
10(2):54–67.

Income Tax Concessions for Owner-Occupied Housing
545
Reece, Barry F. 1975. The Income Tax Incentive to Owner-Occupied Housing in
Australia. Economic Record 51:218–31.
Robinson, Ray. 1988. Urban Housing Finance. Paris: Organisation for Economic
Cooperation and Development.
Rosen, Harvey S. 1979. Housing Decisions and the U.S. Income Tax: An Econometric
Analysis. Journal of Public Economics 11:1–23.
Rosen, Harvey S., and Kenneth T. Rosen. 1980. Federal Taxes and Homeownership:
Evidence from Time Series. Journal of Political Economy 88(1):59–75.
Rosen, Kenneth T. 1989. The Mortgage Interest Tax Deduction and Homeownership.
Working Paper 89–159. University of California, Center for Real Estate and Urban
Economics.
Scholten, Ulrich. 1999. The Rate of Interest and the Share of Owner-Occupied
Housing: Why the Swiss Don’t Like Owner-Occupancy and the Irish Do. Working
paper. Princeton University, Woodrow Wilson School.
Simons, Henry C. 1938. Personal Income Taxation. Chicago: University of Chicago
Press.
Sinai, Todd. 1997. Are Tax Reforms Capitalized into House Prices? Working paper.
University of Pennsylvania, Department of Public Policy and Management.
Statistics Canada. 1999. Dwelling Characteristics of Households. World Wide Web
page <http://www.statcan.ca/english/Pgdb/People/Families/famil09.htm> (accessed
November 16).
Tretzger, Joseph W. 1998. Why Our Federal Income Tax Should Promote Home
Ownership. Illinois Real Estate Letter, Summer, pp. 10–13.
U.S. Bureau of the Census. 1998. Housing Vacancies and Homeownership: Fourth
Quarter 1997. World Wide Web page <http://www.census.gov/hhes/www/housing/hvs/
q497prss. html> (last modified January 21).
U.S. Department of Commerce and U.S. Department of Housing and Urban Develop-
ment. 1991. American Housing Survey for the United States in 1989. Current Hous-
ing Reports H150/89. Washington, DC: U.S. Government Printing Office.
Vickrey, William. 1993. Today’s Task for Economists. American Economic Review
83(1):1–10.
Voith, Richard. 1999. Does the Federal Tax Treatment of Housing Affect the Pattern
of Metropolitan Development? Federal Reserve Bank of Philadelphia Business Review,
March–April, pp. 3–16.
Werczberger, Elia. 1997. Home Ownership and Rent Control in Switzerland. Housing
Studies 12(3):337–53.
White, Melvin, and Anne White. 1965. Horizontal Inequality in the Federal Income
Tax Treatment of Homeowners and Tenants. National Tax Journal 18(3):225–39.
White, Michelle J., and Lawrence J. White. 1977. The Tax Subsidy to Owner-Occupied
Housing: Who Benefits? Journal of Public Economics 3:111–26.

546
Steven C. Bourassa and William G. Grigsby
Woodward, Susan E., and John C. Weicher. 1989. Goring the Wrong Ox: A Defense of
the Mortgage Interest Deduction. National Tax Journal 42(3):301–13.