Exclusion of Capital Gains for Owner-Occupied Housing

^Pr^ep^ar^{ed f}^o^r^M^e^m^b^ers^an^d^Com^mⁱ^t^t^ees^of^C^ong^r^e^ss

For more than 50 years, capital gains on sales of taxpayers’ homes have been preferentially
treated. A revision in 1997 replaced two longstanding provisions—a provision allowing capital
gains tax deferral when a new residence was purchased, and a provision allowing a one-time
exclusion of $125,000 for sellers over age 55—with a capped exclusion for each sale. While the
cap adopted in 1997 was higher than the cap for the over-age-55 sellers, it was less generous than
the uncapped rollover provision. In addition, the dollar cap was not indexed for price changes,
and, unlike the previous over-age-55 cap, was half as large for unmarried taxpayers—$500,000
for married couples and $250,000 for single taxpayers.
Two factors in recent years, the rapid rise in housing prices and interest in tax reform, suggest the ^th
capital gains exclusion, including the dollar cap, might be reconsidered. In the 109 Congress,
two bills were introduced to address this issue. H.R. 2127 would have allowed taxpayers over the
age of 50 to double the current exclusion, once in their lifetime. H.R. 2757 would have indexed
the exclusion to price changes. Other legislation (H.R. 3803 and S. 4075) was introduced to ^th
change the amount of the exclusion for surviving spouses to that of a married couple. In the 110
Congress, S. 138 was introduced to allow a surviving spouse to exclude up to $500,000 of gain
from the sale or exchange of a principal residence owned jointly with a deceased spouse if the
sale or exchange occurs within two years of the death of the spouse. That provision was also
included in H.R. 3648, the Mortgage Forgiveness Debt Relief Act of 2007, which became public
law (P.L. 110-142) on December 20, 2007.
Some criticisms have been made that there are significant tax benefits for owner-occupied
housing. Capital gains treatment is one of those benefits. Yet, there is an efficiency argument for
eliminating or excluding a large portion of the gain on homes from tax. Capital gains taxes on
homes create barriers to labor mobility in the economy. Imposing capital gains taxes on homes
also creates significant compliance costs, requiring individuals to keep records for decades and to
make fine distinctions between improvements and repairs. Capital gains taxes also tend to distort
housing choices, discouraging individuals from selling their homes because of changing family
and health circumstances. And, while the exclusion favors homeowners relative to renters, the
taxation of gains in excess of a cap creates inequities between homeowners with different job
circumstances, between those living in different parts of the country, and between those with
different health outcomes. Exclusions of gains on homes do, however, contribute to tax avoidance
schemes, especially ones that allow gains on investment properties to escape tax.
The current treatment of capital gains could be maintained. However, some consideration might
be given to changing the dollar ceiling. One option is to eliminate the ceiling. Another option is to
adjust the ceiling. This report examines the capital gains exclusion and the ceiling and will be
updated to reflect legislative developments.

Current Tax Treatment.....................................................................................................................1
Development of the Current Rules..................................................................................................3
Is Relief From the Capital Gains Tax on Residences Justified?......................................................4
Labor Mobility..........................................................................................................................5
Other Distortions.......................................................................................................................6
Equity Issues.............................................................................................................................7
Record Keeping.........................................................................................................................8
Contribution of Provision to Tax Sheltering and Avoidance.....................................................9
Converting Rental Property to Owner-Occupied Property.................................................9
“Like Kind” Property Exchanges........................................................................................9
Sharing Capital Gains.......................................................................................................10
Including Investment Property with the Home.................................................................10
The Professional “Fixer-Upper”.......................................................................................10
Cottage and Home.............................................................................................................10
House Swapping................................................................................................................11
Options for Change........................................................................................................................11
Eliminating the Ceilings...........................................................................................................11
Indexing the Dollar Cap..........................................................................................................14
The Single vs. Joint Exclusion................................................................................................16
Changing the Structure of the Exclusion.................................................................................17
Tax Sheltering of Investment Gains........................................................................................17
Conclusion ..................................................................................................................................... 18
Author Contact Information..........................................................................................................19

or more than 50 years, capital gains on sales of taxpayers’ homes have been preferentially
treated. A revision in 1997 replaced two longstanding provisions—a provision allowing
capital gains tax deferral when a new residence is purchased and a one time exclusion of F

$125,000 of capital gains for sellers over age 55—with a capped exclusion for each sale. While
the 1997 cap was higher than the previous cap for the over-age-55 sellers, it was less generous
than the uncapped rollover provision it replaced. In addition, the dollar cap was not indexed for
price changes, and, unlike the previous over-age-55 cap, was half as large for unmarried
taxpayers—$500,000 for married couples and $250,000 for single taxpayers. The exclusion is
allowed once every two years, subject to taxpayers meeting ownership and use tests.
The ceiling was presumably meant to eliminate any capital gains tax on home sales for the vast
majority of taxpayers, but the rapid rise in housing prices and the passage of time have reduced
the value of the exclusion. With no revision and a continued upward march of housing prices, an
increasing share of gains will be subject to tax.
Two bills in the 109^th Congress addressed this provision. H.R. 2127, introduced by Representative
Filner, would have allowed taxpayers over the age of 50 to exclude an amount that is double the
current cap, but it is available only once in their lifetime. H.R. 2757, introduced by
Representative Andrews, would have indexed the exclusion. Other legislation (H.R. 3803 by
Representative McCarthy and S. 4075 by Senator Schumer) was introduced to change the amount
of the exclusion for surviving spouses to that of a married couple.
In the 110^th Congress, S. 138 was introduced to allow a surviving spouse to exclude up to
$500,000 of gain from the sale or exchange of a principal residence owned jointly with a
deceased spouse if the sale or exchange occurs within two years of the death of the spouse. That
provision was also included in H.R. 3648, the Mortgage Forgiveness Debt Relief Act of 2007,
which became public law (P.L. 110-142) on December 20, 2007.
This report examines the capital gains exclusion and the ceiling. The first section describes the
current tax rules, and the second section presents the historical development of the capital gains
provisions. The third section discusses potential justifications for capital gains relief, as well as
tax avoidance problems that may arise. The final section discusses various options for change,
primarily focusing on the dollar ceiling. The modification to current law enacted by the Mortgage
Forgiveness Debt Relief Act of 2007 is also addressed in the final section.

When an individual sells a personal residence, the excess of the sales price over the original cost
plus improvements is a capital gain and is subject to tax. The individual is able to deduct any
costs of the sale (such as commissions and advertising), and may also be required to include gain
that was deferred from previous home sales.
Gain up to $250,000 for single taxpayers and $500,000 for married couples filing joint returns is
excluded if the taxpayer meets a use test (has lived in the house for at least two years out of the
last five years) and an ownership test (has owned the house, also for two years out of the last

five). The use and ownership years may be the same or different. The exclusion can be used every ¹
two years.
In order to minimize the tax, assuming the individual is subject to it, he or she must keep records
of any improvements during the entire time the home is owned and also, to comply with the law,
appropriately distinguish between expenditures that are repairs, which do not reduce gain, and
those that are improvements, which do reduce gain.
For homes that were acquired before the 1997 change in tax law, there may also be deferred gain
from previously owned residences. Under pre-1997 tax law, a taxpayer could defer the gain on a
home sale if another residence was purchased. If the new residence cost as much or more than the
old residence sold for, the tax on the entire gain was deferred. If the new residence cost less than
the sale price of the old residence, gains tax was due on the difference between the value of the
old and new residence (if less than the gain) and tax on the remainder was deferred. The
additional gain from previously owned residences makes it more likely that total gains will
exceed the cap; had the 1997 law been in place for many years, much or all of the prior gain
would have been excluded.
Capital gains on assets that are held at least a year are taxed at special, lower rates: 5% for
taxpayers in the 10% and 15% marginal income tax brackets and 15% for taxpayers in marginal
income tax brackets above 15%. These provisions are temporary. Although the 5% rate will be
reduced to 0% in 2008, the lower rates will expire at the end of 2010, absent legislative change, ²
and the rates will revert to their previous levels, generally 10% and 20%.
A special relief provision for military families and the Foreign Service allows them to expand the
five-year period for the ownership and use tests to up to 10 years while on qualified official duty.
Another provision that may influence a taxpayer’s decision about selling a residence is a long-
standing provision that allows the gain to be forgiven entirely if the taxpayer does not sell the
home and leaves it as part of his or her estate. If an individual keeps his house until death and
leaves it to heirs, no tax on gain accumulated would be due, since the heir would be able to
deduct the fair market value at time of death from sales price, when the house is sold (resulting in ³
a zero capital gain with instantaneous sale). This rule is called a step-up in basis.

¹^T^h^{ere are so}^m^e^e^x^cep^tⁱ^oⁿ^s^t^o^t^h^e^s^{e ru}^l^e^s.^If^t^a^xp^a^y^er^{s h}^a^{ve n}^o^t^lⁱ^v^edⁱⁿ^t^h^{e p}^rⁱ^m^ar^y^re^si^d^eⁿ^{ce f}^o^{r a t}^o^t^a^l^o^f^t^w^o^years
^o^u^t^o^f^t^h^{e l}^a^st^fⁱ^ve,^t^h^ey^ar^{e el}ⁱ^gⁱ^b^l^e^f^o^{r a p}^a^rtⁱ^a^l^excl^u^sⁱ^oⁿ^capⁱ^f^t^h^{e real}^est^a^t^e^w^a^{s so}^l^d^b^ecau^{se o}^f^a^ch^an^{ge i}ⁿ
^e^m^p^l^o^y^men^t^,^h^eal^t^h^,^o^r^un^f^o^reseeⁿ^ci^rcu^m^st^an^ces.^T^h^{e t}^a^xp^a^y^{er ca}ⁿ^recei^{ve a p}^o^r^tⁱ^oⁿ^o^f^t^h^{e excl}^u^sⁱ^oⁿ^cap^,^b^a^sed^on^t^h^e
^por^ti^on^of^the^tw^o-^y^e^a^r^pe^rⁱ^{od t}^h^e^y^r^e^sⁱ^de^{d in t}^h^e^h^o^m^e^{. For}^e^x^am^ple^,^a^sⁱ^ng^le^ta^x^p^ay^e^r^w^{ho liv}^e^dⁱⁿ^the^h^ous^e^f^o^r^oⁿ^e
^y^e^a^r^a^{nd qua}^lif^ie^{d f}^o^r^aⁿ^e^x^c^e^p^tio^{n w}^ould^ha^v^e^a^$1^25,⁰⁰^{0 c}^a^p^.^For^pe^op^le^liv^ingⁱⁿ^aⁿ^u^r^s^ing^h^o^m^e^{, the}^ow^ne^r^s^{hip a}^nd
^use^te^{st is low}^e^re^{d to}^one^out^of^fⁱ^v^e^y^e^a^r^{s be}^f^o^re^eⁿ^te^ri^ng^the^f^a^c^ili^ty^.^A^{nd tim}^e^speⁿ^{t in t}^h^eⁿ^u^rsing^hom^e^{still c}^o^uⁿ^ts
^t^o^w^a^rd^o^wⁿ^e^rshⁱ^p^tⁱ^m^{e an}^d^u^s^{e o}^f^t^h^{e resi}^d^eⁿ^ce.^F^o^{r exam}^p^l^e,ⁱ^f^{a t}^a^xp^a^y^{er l}ⁱ^v^edⁱⁿ^a^h^o^u^s^{e f}^o^{r a yea}^r^,^an^d^t^h^{en sp}^en^t
^the^ne^x^t^fⁱ^v^e^y^e^a^r^{s in a}^nursi^ng^ho^m^e^be^f^o^re^se^lling^the^hom^e^,^the^f^u^ll^$2^50,⁰⁰^{0 e}^x^c^l^usio^{n w}^ould^be^a^v^a^ila^ble^.
²^For^f^u^r^t^he^r^de^ta^ils^oⁿ^t^h^e^g^e^ne^r^a^{l c}^a^p^ita^{l g}^a^ins^pr^ov^is^io^ns^{, s}^e^e^C^R^S^R^e^por^t^96-⁷⁶^9,^C^apit^a^{l G}^a^ins^Ta^x^e^s^{: An}
^Overvi^ew^,^b^y^Jan^e^G^.^G^r^av^el^l^e^.^Asset^s^h^e^l^d^f^o^{r at}^l^east^fⁱ^ve^{years h}^a^{ve a p}^e^r^m^an^en^t^t^a^{x rat}^e^o^f⁸^%^an^d¹⁸^%
^re^spe^c^tiv^e^l^y^.^Note^tha^t^sta^t^e^inc^o^m^e^ta^x^ra^te^s^w^{ould still}^beⁱⁿ^pla^c^e^e^v^eⁿ^w^{ith a}^z^e^r^{o f}^e^de^ra^{l ra}^te^.
³^Unde^r^prov^isio^{ns e}ⁿ^a^c^t^e^dⁱⁿ²⁰⁰¹^{, w}^h^ic^{h p}^h^a^s^e^ou^{t the}^e^s^ta^te^ta^x^a^{nd re}^pe^a^l^it^by²⁰¹^{0, t}^h^{is ste}^p^pe^d-u^p^ba^{sis r}^u^le^w^ill
^b^e^rep^ealedⁱⁿ²⁰¹⁰^an^d^rep^l^aced^b^y^{a carr}^y^-o^v^e^{r b}^a^{sis ru}^le^{. T}^h^{is ru}^{le w}^{ill n}^o^t^im^p^o^s^{e a cap}^{ital g}^aⁱⁿ^s^tax^{at d}^eath^,^b^u^t
^w^{ill i}^m^pos^e^the^ta^x^if^the^he^{ir s}^e^ll^s^the^h^o^m^e^(a^lthoug^{h if}^aⁿ^he^ir^liv^e^s^{in the}^ho^us^e^f^o^{r the}^re^q^u^ire^d^pe^riod,^the^he^{ir w}^ill
^be^e^lig^ible^f^o^{r the}^re^g^u^la^{r c}^a^p^ita^l^g^a^{ins e}^x^c^l^{usion). How}^e^v^e^{r, $1.3}^m^{illion of}^g^a^{in c}^aⁿ^still^be^e^x^c^l^ud^e^d^pe^rm^aⁿ^eⁿ^tly^.
^T^h^e^s^e^prov^{isions a}^r^e^te^m^pora^r^y^a^{nd w}^{ill e}^x^pire^{, u}ⁿ^le^{ss e}^x^te^nde^d.

The gain realized upon the sale of a personal residence was taxed as capital gain until the passage
of the Revenue Act of 1951 (P.L. 82-183). At that time, Congress enacted a rollover provision that
allowed for the deferral of capital gains tax if the proceeds of the sale were used to buy another
residence of equal or greater value within a year before or after the sale of the old residence.
Congress stated that the rollover provision was in response to transactions that were:
ⁿ^eces^s^{itated b}^y^s^u^c^h^f^acts^a^s^aⁿⁱⁿ^creas^{e i}ⁿ^t^h^{e s}ⁱ^{ze o}^f^t^h^e^f^a^mⁱ^l^y^{or a ch}^aⁿ^g^e^{in th}^{e place o}^f
^th^{e tax}^p^a^y^er’s^e^m^plo^y^m^eⁿ^t^{. In th}^es^{e s}ⁱ^t^u^atioⁿ^s^th^{e tra}ⁿ^s^actioⁿ^partak^es^of^th^eⁿ^a^t^u^{re of}^an
ⁱⁿ^v^o^l^uⁿ^t^ar^y^coⁿ^v^e^rsi^oⁿ^.^C^a^s^e^{s of}^t^hⁱ^s^t^y^{pe are part}ⁱ^c^u^l^arl^yⁿ^u^m^e^rousⁱⁿ^peri^ods^of^rapi^d⁴
^ch^ang^e^s^u^c^h^as^m^obilizat^ion^{or recon}^v^e^rs^ion^.
At that time, the economy had grown as a result of industrialization and residential moves were
more frequent due to business transfers and other employment related changes. Congress also
recognized that capital gains from home sales were, in part, a result of general inflation. During
the congressional debate, the rollover provision was justified on the grounds that homeowners
were changing homes, not to make a profit as investors, but rather in response to employment
and/or family size changes.
The first exclusion from taxation for capital gains on the sale of a primary residence was enacted
by the Revenue Act of 1964 (P.L. 88-272). The provision was available only for the elderly (age

65 and over) and applied to residences sold after 1963. It was available on a one-time basis only,

and was at the same level for both single and married individuals. To qualify, the house had to be
occupied for five of the previous eight years. The exclusion of gain was limited to the amount
attributable to the first $20,000 of sales price. Above that level, a ratio was used to determine the
gain subject to taxation, such that the amount of the exclusion depended on the relationship of
sales price to basis, as well as the relationship between $20,000 and sales price. For example, if
the sales price were $40,000, one-half of the gain ($20,000/$40,000) could be excluded. But the
actual amount excluded would be less than $20,000, unless the house originally cost $20,000. For
example, if the basis in the house was $10,000, the gain on the $40,000 sale would be $30,000.
Of that $30,000, one half, or $15,000, would be excluded because $20,000 was half the sales
price.
The reason given for the exclusion was to reduce the burden on elderly taxpayers who would
have to tie up all of their investment in a new home to avoid paying capital gains tax. The dollar
restriction was due to a focus on the average and smaller home, thus suggesting a distributional ⁵
motive.
The amount of capital gains excludable from taxation for older taxpayers was increased three
times in response to higher housing prices. The three increases were enacted by The Tax Reform
Act of 1976 (P.L. 94-455), the Revenue Act of 1978 (P.L. 95-600), the Economic Recovery Tax
Act of 1981 (P.L. 97-34) and finally, the Taxpayer Relief Act of 1997 (P.L. 105-34). The limit for
elderly homeowners rose to $35,000 in 1976, $100,000 in 1978, and $125,000 in 1981. The 1978

⁴^{U.S. C}^ong^re^{ss, Se}^na^te^Com^m^itte^e^{on F}ⁱ^na^nc^e^,^The^Re^v^e^nue^Ac^t^of¹⁹⁵^1:^Re^por^t^{to Ac}^c^o^m^p^aⁿ^y^H^.^R.⁴⁴⁷³^{, 82}^nd
^C^ong^{., 1}^st^s^e^s^s^{., S.}^P^r^t^.⁸²^-⁷⁸¹⁽^W^a^s^hing^ton:G^P^O^,¹⁹⁵¹⁾^,^p.³⁴^.
⁵^U^.^{S. C}^ong^r^e^s^s^,^H^ous^e^W^a^y^s^a^{nd Me}^aⁿ^s^C^o^m^m^itte^e^,^R^e^p^o^r^t^t^o^A^cco^mp^an^{y H.}^R^.⁸³⁶³^{, 88}^th^Cong^.,¹^st^sess.^,^H.^P^r^t.⁸⁸^-
⁷⁴⁹⁽^W^a^s^hing^ton:G^P^O^,¹⁹⁶³⁾^.

provision also liberalized the benefit by simply allowing an exclusion rather than a proportional
share that depended on basis and lowered the age limit to 55. (There was consideration of
eliminating the age requirement altogether.) The 1978 change also reduced the holding period
requirement to three out of the previous eight years.
In each case of capital gains relief, Congress cited the rising sale prices of homes as the source of
large amounts of taxable capital gains on residences and the reason for adjusting the amount of
capital gains that could be excluded from taxation.
When the rollover provisions and one time exclusion for the elderly were replaced by the current
exclusion (in the Taxpayer Relief Act of 1997, P.L. 105-34), a major reason given was to reduce
the record keeping burden and to eliminate the need for referring to records and making
judgements about what expenditures are improvements.
Other reasons cited for changing the tax law were to limit the distortions in behavior arising from
the rollover treatment and from those elderly who had exceeded the exclusion limit or had already
used it. Because the full deferral of tax required the purchase of a new residence of equal or
greater value, the law may have encouraged taxpayers to purchase more expensive homes than
they otherwise would have. The pre-1997 rules also discouraged some elderly taxpayers from
selling their homes to avoid possible tax consequences. As a result, elderly taxpayers who had
already used their one-time exclusion and those who might have realized a gain in excess of
$125,000, may have retained their homes even though it was desirable for them to move.
It was also clear from statistical data that between rollovers, exclusions, step-up in basis (which
allowed capital gains to be avoided if the home were held until death and left to heirs), and under-
reporting, very little capital gains on owner-occupied housing were taxed. Thus, little revenue
was gained from a set of provisions that, nevertheless, caused distortions in behavior and
complicated compliance. These observations supported a simple elimination of the capital gains
tax on principal residences.
In 1997 Congress imposed what was characterized as a “relatively high” ceiling on the amount of
excluded gain, $500,000 for married couples. In a departure from the historic treatment of
lifetime exclusions, however, the exclusion was only half as large for single taxpayers as for
married couples—$250,000. The previous treatment had cut the exclusion in half for married
couples filing separately but not for single taxpayers, an important difference given that most
married individuals who do not divorce are eventually widowed. Unlike the lifetime exclusion,
however, the exclusion could be taken in each period. In contrast to many other dollar limits in
the tax code, the amount of the exclusion was not indexed, so that it, like the previous exclusion,
might need to be periodically revisited.

Economists have often been critical of preferential treatment of certain types of activities, because
that preferential treatment distorts behavior and causes a misallocation of capital. Tax preferences
also narrow the tax base and require higher marginal tax rates for a given revenue target; these
higher tax rates in turn magnify other distortions. Tax preferences also can be inequitable,
favoring those who engage in tax preferred activities. (Tax preferences are also sometimes

criticized because they favor higher income individuals; the appropriateness of such criticisms
depends on one’s view of how taxes should be distributed and whether such preferences are offset
by a more graduated rate structure).
It is also true that the favorable treatment for owner-occupied housing may divert investment
from business investment. Absent a market failure, this misallocation reduces the efficiency of the
economy. Favorable capital gains treatment for housing is not the only tax benefit it receives, or
even the most important in economic terms. The implicit income from housing is not subject to
tax, yet costs such as mortgage interest and property taxes are allowed as deductions, at least for
those who itemize. In general, the effective tax rate on the return to owner-occupied housing is
around zero. And although housing is often claimed to provide external benefits, such benefits
have not generally been measured; nor is there reason to believe that they justify such significant
subsidies.
It is not clear whether the prospect of future capital gains relief plays an important role in
inducing additional investment in housing. Unlike mortgage interest deductions, future capital
gains relief provides no immediate cash flow benefit, and may be heavily discounted due to the
delay and uncertainty of the benefit.
In addition, there are some good reasons to provide some relief for capital gains on owner-
occupied houses and to restrict other owner-occupied housing tax benefits if a reduction in the
preferential treatment of owner-occupied housing is desired. Perhaps the most important of these
justifications for relief is to reduce the barriers to labor mobility, contributing to economic
efficiency. Other reasons include reduction in other inefficiencies that distort housing costs; more
equitable treatment among homeowners in different circumstances; and reduction of compliance
burdens. There is some empirical evidence that significant distortions are induced by the gains tax ⁶
once an individual has a home and wishes to move.
On the other hand, the exclusion can contribute to compliance problems, by allowing a potential
for tax sheltering. These tax sheltering problems are discussed below.
The possible forms of capital gains revisions are closely tied to these rationales and issues.
Therefore, following the general discussion of the rationales, we also consider the implications of
the particular forms of these potential changes.
One of the important reasons for having some type of relief is to minimize the barriers to labor
mobility. In order to have an efficient market economy that can respond to changes in tastes and
technology, it is important to have as few barriers to labor mobility as possible. This consideration
was reflected in the rationale for the rollover provision enacted in 1951. Americans’ taste and
preference for owning their own homes inevitably creates barriers to a willingness to relocate,
barriers that cannot be avoided. But imposing capital gains tax at sale adds to that barrier.

⁶^L^e^ona^r^d^{E. B}^u^r^m^aⁿ^{, Sa}^lly^W^a^llac^e^,^a^{nd D}^a^vⁱ^{d W}^e^ine^r^{, “}^H^ow^C^a^pita^{l G}^a^ins^T^a^x^e^s^Dⁱ^s^t^or^{t H}^o^m^e^ow^ne^r^s^’^D^e^cⁱ^sⁱ^ons^,”
^P^r^o^ceedⁱⁿ^g^s^o^f^t^h^{e 89}^th^Aⁿ^nu^al^C^oⁿ^f^eren^{ce on}^T^a^xa^tⁱ^oⁿ^o^f^t^h^{e Na}^tⁱ^oⁿ^a^l^T^a^{x A}^sso^ci^a^tⁱ^oⁿ^{, Na}^tio^na^{l T}^a^x^A^ssocⁱ^a^tion:
^W^a^s^h^ing^t^{on, D}^C^,¹⁹⁹⁷^{, p}^p^.³⁸^2-³^90.

The rollover provision, as it existed in prior law, provided some relief but still left some barriers
to mobility in place. One problem arose because there were regional differences in housing
prices, which still exist. If the individual was moving to an area that generally has lower prices
(for example, from California to Arizona), it might be sensible to buy a house that was similar in
quality but cost less because of lower overall prices in the area (which might also have included a
lower salary). This shift would result in a capital gain, and the individual then might have been
discouraged from making the move, or induced to purchase a larger house than otherwise
desirable. There may also have been circumstances where it might have been more desirable to
rent rather than to purchase a new home when relocating (for example, when the family is
moving because of economic hardship or the new location is expected to be the place of
employment for only a few years). The rollover provision would not have reduced the capital
gains barrier in that case. Thus, the rollover provision was imperfect in its elimination of labor
mobility barriers.
The capped exclusion eliminates all barriers, as long as the cap is not exceeded, and reduces the
cost of labor mobility. Unless the cap is increased explicitly or indexed to housing prices,
however, an increasing share of individuals will be affected by the ceiling over time and barriers
to labor mobility will grow.
Aside from the labor mobility problem, capital gains taxes on owner-occupied housing can cause
other distortions. Capital gains taxes on assets in general cause a lock-in effect (i.e., discourage
changes in portfolio allocations by replacing old assets with new assets). One can argue, however,
that the lock-in effect for homes in particular imposes greater costs. Financial assets are more
likely to be close substitutes, so the lock-in effect is probably not very costly. [However, it is also
possible to swap real estate investments without paying a capital gains tax.] Some might argue
that people should be encouraged to hold on to investments in the stock market for a long period
of time, in order to average out the ups and downs of the market, and the lock-in effect may
actually be beneficial in some cases.
Different types of housing, however, may be less substitutable for each other; the difference
between the house to which an owner is “locked-in” and the home he desires may be more
important than for various alternative financial assets. And with no relief provisions, capital gains
taxes discourage moving for any purpose, whether to a larger house (for example, to
accommodate a larger family) or a smaller one (when children have grown and left the home or to
simplify maintenance during retirement). As noted above, a rollover treatment can cause people
to buy too much housing, or continue to own when renting might be optimal. The once-in-a-
lifetime exclusion which aided the elderly was aimed at older individuals who might wish to sell
their houses to move into smaller and more easily maintained houses, to move to a rental status,
or to “cash out” the value of the house for other purposes. If the exclusion cap does not come into
play, these distortions do not exist, but, as noted above, if the cap does not rise with housing
prices it will become increasingly binding.
The current provision permitting a capped exclusion every two years actually creates, for those
affected, the opposite distortion. It encourages higher income individuals to move more
frequently. For instance, an individual who has a capital gain at the limit can move, take
advantage of the gains exclusion, and then, within, two years take advantage of it again, while the
individual who sells only once, but has an equal gain would have to pay tax. Suppose, for
example, that one taxpayer (a single individual) realizes a capital gain of $200,000 on the sale of

a home, purchases another home, and then sells that second home two years later, earning an
additional $200,000 in capital gains. The taxpayer would be able to exclude $400,000. If a similar
taxpayer experiences a single gain of $400,000 in the same time period, he or she may exclude
only $250,000.
The caps on both the prior one-time exclusion and the current exclusion were enacted to impose
gains taxes on higher income individuals with large capital gains, and therefore the caps are
presumed to have a vertical equity objective because they limit the benefit for high income
taxpayers.
The cap, however, produces some horizontal inequities. First, the limited exclusion combined
with the step-up in basis at death causes elderly taxpayers who had to move from their homes due
to ill health to pay taxes not assessed on their healthier counterparts who remain in their homes
until their death and leave the houses to their heirs with no capital gains tax.
The cap itself also produces some inequities among individuals who sell their homes and who are
affected by the cap. These inequities are of three types: between those who move frequently and
those who do not; between people living in different regions of the country; and between married
couples where both spouses survive until the point they wish to sell and those where only one
spouse survives.
In the first case, people who buy and sell frequently (and are thus less likely to accrue a large gain
in a particular sale) are less likely to be affected by the tax. For example, a married couple who
sells every five years and gets a $150,000 gain on each sale would not pay a gains tax. If a similar
couple buys a house and sells after 20 years with the same accumulated gain (four times $150,000
or $600,000), they will pay a tax.
Taxpayers living in states and locations where the cost of living, including housing prices, tends
to be high, are more likely to be affected by the cap even in cases where their real incomes and
standard of living are the same as those who are not affected. Thus, taxpayers in California and
Massachusetts are more likely to be affected than taxpayers in Mississippi and Oklahoma.
Finally, the tax is more likely to fall on elderly taxpayers who have lost a spouse than married
couples who remain alive at the time they wish to sell their house. Although the tax laws allow
the gain for the spouse who is deceased to be excluded (half the gain at that time), further capital ⁷
gain exclusions are limited by the lower ceiling that applies to singles. For example, suppose the
gain on a house is $400,000. In the case of a married couple who sells, the entire gain will be
excluded. In the case of a surviving spouse, the exclusion will be $250,000 plus half of any gain
that accrued during the deceased spouse’s life; if that gain is less than $150,000 some tax will be
due. Since most women marry husbands who are older than themselves, and since women live

⁷^I^f^a^s^u^r^v^iv^ing^s^pous^e^a^nd^the^d^eced^en^t^o^wⁿ^e^d^t^h^{e ho}^m^e^j^oⁱⁿ^t^l^y^,^t^h^{e b}^a^si^{s i}ⁿ^t^h^e^h^o^m^{e ch}^an^{ges a}^f^t^e^{r t}^h^{e d}^eat^h^o^f^t^h^e
^d^eced^en^{t. T}^h^{e n}^e^w^b^a^{sis f}^o^{r th}^{e h}^a^lfⁱⁿ^{terest th}^{at th}^e^d^eced^en^{t o}^wⁿ^e^d^w^{ill b}^e^oⁿ^e^-^h^alf^o^f^th^{e f}^a^ir^m^a^r^k^{et v}^a^lu^{e o}ⁿ^th^e
^da^te^of^de^a^t^{h (}^o^r^a^lte^rⁿ^a^t^e^v^a^lua^tⁱ^on^da^te⁾^.^A^s^aⁿ^e^x^am^ple^,^s^uppos^e^a^c^oup^le^joiⁿ^tly^ow^ne^{d a}^hom^e^tha^t^ha^{d a}ⁿ^a^d^jus^t^e^d
^ba^sⁱ^s^of^$10^0,⁰⁰^{0 o}ⁿ^t^h^e^da^te^of^one^s^p^o^u^s^e^’^s^de^a^t^{h. T}^h^e^f^a^ir^m^a^r^k^{et v}^a^lue^on^tha^t^da^te^w^a^s^$200,⁰⁰⁰^.^T^h^e^ne^w^ba^sⁱ^sⁱⁿ
^the^h^o^m^e^is^$150^,0^{00 (}^$⁵⁰^,⁰⁰⁰^f^o^r^one^-^h^a^l^f^of^the^a^d^j^u^s^t^e^d^ba^sⁱ^s^pl^us^$¹⁰⁰^,0⁰⁰^f^o^r^one^-^h^a^l^f^of^the^f^a^ir^m^a^r^k^e^t^v^a^lue⁾^.

longer than men, a significant number of widows are likely to live in the house after the spouse
has died.
Record keeping required to deal with the capital gains tax on residences is complex. To comply
with tax regulations, taxpayers have had to keep detailed records of the financial expenditures
associated with their home ownership. The taxpayer needs to record the original cost of the
residence, any costs added at the time of purchase, and any capital improvements. In the latter
case, taxpayers also have had to differentiate between those expenditures that affected the basis of ⁸
the property and those that were merely for maintenance or repairs. In many instances these
records have to be kept for decades. Congress has addressed this issue, stating in the reasons for
the 1997 increase in the exclusion that
^calcu^lati^ng^{capital g}^aⁱⁿ^f^r^o^m^t^h^{e s}^a^{le o}^f^{a prin}^{cipal res}ⁱ^den^c^e^w^a^s^a^m^oⁿ^g^t^h^e^m^o^s^t^co^m^p^le^x
^tas^k^s^f^{aced b}^y^{a t}^y^{pical tax}^p^a^y^er....^[A^]s^{a res}^u^{lt of}^th^{e rollover pr}^ov^is^ions^an^{d th}^{e $125,0}⁰⁰
^oⁿ^e-^ti^m^e^e^x^cl^u^s^ioⁿ^uⁿ^d^e^r^p^r^io^r^la^w^,^d^e^tailed^r^eco^r^d^{s o}^f^tr^an^sactioⁿ^s^aⁿ^d^ex^p^eⁿ^d^itu^r^e^{s o}ⁿ⁹
^h^o^m^eⁱ^m^prov^e^m^en^t^s^h^a^{d to be k}^e^{pt, in}^m^o^s^t^cas^es^,^f^o^r^m^a^ny^decades^.
The 1997 tax law simplified income tax administration and record keeping by providing a
“relatively high threshold, few taxpayers will have to refer to records in determining income tax ¹⁰
consequences of transactions related to their house.”
The capital gains exclusion, however, was not indexed for inflation, or for housing price changes.
Since 1997, the value of homes has increased as much as 70%, depending on the price index used
(see the discussion of indexing the cap below). If this appreciation were used to adjust the 1997
figure, the exclusion would rise from $500,000 to as much as $850,000 for couples (and from ¹¹
$250,000 to $425,000 for single taxpayers). In addition, gain on houses increased proportionally
more. For example, if the basis (the original cost) of the house in 1997 were half the market value
purchase price, a 70% increase in value would mean a 140% increase in the gain. This
appreciation means that many more taxpayers would be subject to the ceiling. Without an
indexing procedure, some of the potential record keeping benefits from the 1997 revision have
been lost. It would be unwise to abandon record keeping given that the exclusion is covering
fewer and fewer sales over time and there is no commitment from the government to index the
provision, so that the simplification from less record-keeping is likely to be diminished.

⁸^Ca^lc^ula^ting^c^a^p^ita^{l g}^a^{ins re}^q^u^ire^s^a^m^e^a^s^ure^of^ba^{sis. A}^ta^x^p^a^y^e^r’s^ba^{sis in re}^a^l^e^s^ta^te^{is c}^o^{st (}^o^{r f}^a^{ir m}^a^rk^e^t^va^lue^if
^a^c^quir^e^d^by^inhe^r^ita^nc^e⁾^{. T}^h^e^c^o^s^t^of^pr^o^p^e^r^ty^is^the^am^ount^pa^{id f}^o^r^{it in}^c^a^s^{h, de}^bt^oblig^a^tio^ns^,^othe^r^pr^o^p^e^r^ty^{, or}
^se^rv^ic^e^s^{, w}^h^ic^{h c}^aⁿ^inc^l^u^d^e^the^p^u^rc^ha^se^pric^e^aⁿ^d^c^e^r^ta^{in se}^ttle^m^e^{nt or}^c^l^osiⁿ^g^c^o^{sts. W}^h^eⁿ^c^a^l^c^u^la^tⁱ^ng^the^g^aⁱⁿ^{or l}^o^ss
^oⁿ^t^h^{e sal}^e^o^f^{a resi}^d^eⁿ^ce,^t^h^e^b^a^si^{s i}^s^ad^j^u^st^{ed f}^o^{r ch}^an^{ges m}^a^d^e^siⁿ^{ce t}^h^{e acq}^uⁱ^si^tⁱ^on^o^f^t^h^e^p^r^op^ert^y^.^In^crea^{ses t}^o^b^a^si^s
ⁱⁿ^cl^ud^{e t}^h^{e co}^st^o^f^capⁱ^t^a^lⁱ^m^p^r^o^v^e^m^en^t^s^,^su^ch^{as ai}^{r co}^ndⁱ^tⁱ^oⁿⁱⁿ^{g or a n}^e^w^ro^o^f^;^sp^eci^al^assess^men^t^{s f}^o^{r l}^o^cal
^im^pr^ov^em^eⁿ^ts^{; a}^{nd a}ⁿ^y^othe^r^a^d^d^itions^tha^t^ha^v^e^a^us^e^f^{ul lif}^e^of^m^o^r^e^thaⁿ^oⁿ^e^y^e^a^r^{. Ex}^am^ple^s^of^dec^r^e^a^s^es^{to ba}^sⁱ^s
^inc^l^u^d^e^aⁿ^y^c^a^p^ita^{l g}^a^{in tha}^t^w^a^s^pos^tp^one^{d f}^r^om^the^s^a^le^of^a^pr^e^v^ious^hom^e^be^f^o^r^e^May^{7, 19}^97;^d^e^duc^tible^c^a^s^ua^lty
^l^o^{sses an}^d^/^o^rⁱⁿ^s^u^ran^{ce p}^a^ym^en^t^s^recei^ved^f^o^{r casu}^a^l^t^y^l^o^sses;^p^a^ymen^t^s^recei^ved^f^o^{r gran}^tⁱⁿ^{g an}^ease^m^en^t^o^r^ri^gh^t^-^o^f^-
^w^a^y^;^de^pr^e^cⁱ^a^{tion a}^l^l^o^w^e^{d if}^the^hom^e^w^a^s^us^e^d^f^o^r^busⁱⁿ^e^s^s^or^r^e^nt^a^l^pur^pos^e^s^{; a}^fⁱ^r^s^t-^tim^e^hom^e^buy^e^r^c^r^e^d^{it (}^a^llow^e^d
^{to c}^e^r^ta^{in f}ⁱ^r^s^t-^tim^e^buy^e^r^s^of^a^hom^e^{in the}^Dⁱ^s^t^rⁱ^c^t^of^C^o^l^u^m^b^ia⁾^;^a^{nd e}ⁿ^e^r^g^y^c^ons^e^r^v^a^{tion s}^ubs^idy^e^x^c^l^ude^{d f}^r^om
^gro^{ss i}ⁿ^co^m^e^b^ecau^{se i}^t^{was recei}^ved^t^o^bu^y^o^rⁱⁿ^st^al^l^an^y^en^erg^y^c^oⁿ^s^ervatⁱ^oⁿ^m^easu^r^e.
⁹^{U.S. C}^ong^re^{ss, Join}^{t Com}^m^itte^e^{on T}^a^x^a^tion,^G^e^ne^r^a^l^Ex^pl^an^atio^{n of}^Tax^Le^gis^l^atio^{n E}^nac^te^dⁱⁿ¹⁹⁹⁷^{, 1}⁰⁵^th^Co^ng^.,
¹^st^s^e^s^s^{., (}^W^a^s^hing^to^n:G^P^O^,¹⁹⁹⁷⁾^,^p^p^.^54-^55.
¹⁰^Ibid^.
¹¹^Iⁿ^te^r^m^s^of²⁰⁰⁵^pr^ic^e^s^.

The presence of a special exclusion contributes to the possibility of using the tax benefit to avoid
capital gains taxes in unintended ways. This section discusses several ways this might occur.
Capital gains avoidance can occur by converting rental property to owner-occupied property.
After this conversion, the property can be sold and the capital gains excluded up to the allowable
amount, as long as the property has been owned and used as a principal residence for at least two
years during the five-year period ending on the date of the sale of the residence. As an example,
consider a married couple who have a primary residence and a rental property and both properties
have substantially appreciated in value. The couple can sell the primary residence and claim a
capital gain exclusion of up to $500,000 on that residence. The couple can then move into the
rental property and use that as the primary residence. After two years the taxpayers could then sell
the property and realize a gain of which up to $500,000 can be excluded under the law. It makes
no difference that most of the appreciation on the second property was realized when it was a
rental unit. Current tax law, however, does require that any depreciation on the rental property be ¹²
recaptured and taxed.
An example of the depreciation recapture can be seen in the following example. A married couple
sells their primary residence which had an adjusted basis (purchase price plus capital
improvements) of $100,000 for $200,000. In prior years, the property had been a rental property
and the couple had claimed $50,000 in depreciation deductions on the home. The taxable gain for
the sale would be $100,000, which is the sales price minus the adjusted basis. Of that gain,
$50,000 is tax-free and the $50,000 taken as depreciation deductions in the past would be subject
to a 15% capital gains tax.
Taxpayers can avoid paying tax on the gain from the sale of their real estate property by
participating in a “like kind” property exchange. Under section 1031 of the Internal Revenue
Code, like-property exchanges offer business owners or investors a way to trade their property for
something of similar value and type without reporting a profit and, thereby, defer paying taxes on
the gain. In the case of residential property, this exchange can be combined with the exclusion of
capital gains to allow taxpayers to avoid capital gains tax on some of the deferred gain. Some
gain on the original investment property would be taxed as recaptured depreciation, and some
may be taxed if the total remaining gain exceeds the cap. For example, a taxpayer who owns a
rental property can participate in a “like kind” exchange for another residential property which
then becomes the taxpayer’s primary residence. The taxpayer must meet the ownership and use
tests for a minimum of five years before the taxpayer can then sell the property and exclude
capital gains up to the allowable amount. Essentially, the taxpayer will have sold real estate and
avoided capital gains taxation on some, and perhaps most, of the capital gains earned on both
properties.

¹²^T^h^{e recap}^t^u^{re ru}^l^e^,^en^act^ed^b^y^t^h^{e T}^a^xp^a^y^{er Rel}ⁱ^e^f^A^c^t^o^f¹⁹⁹⁷^,^app^lⁱ^e^{s t}^o^d^e^p^r^eci^atⁱ^oⁿ^cl^ai^m^e^d^af^t^e^r^M^a^y⁶^,¹⁹⁹⁷^.
^Aⁿ^y^d^e^p^r^eci^atⁱ^oⁿ^t^a^ken^b^e^f^o^{re t}^h^a^t^d^a^t^eⁱ^s^“f^o^r^gi^ven^”^an^dⁿ^o^t^recap^t^u^red^.

Prior to October 2004 when the American Jobs Creation Act of 2004 (P.L. 108-357) was enacted,
there was no minimum holding period for properties acquired through “like kind” exchanges. The
exclusion of gain on the sale of a principal residence applied after two years, when the ownership
and use tests for the provision would have been met. The 2004 law requires the taxpayer to hold
the exchanged property for a full five years, as opposed to two, before the residence can qualify
as a principal residence. This change reduced, but did not eliminate, the attractiveness of ¹³
combining “like-kind” exchanges with the principal residence exclusion.
If taxpayers expect huge gains from owning, then selling a house—more than can be excluded
from tax under the new rule—they could divide ownership of the house in order to maximize the
amount of capital gain that could be excluded. If, for example, a married couple owns their
residence together with an adult son and he meets the ownership and use tests for one-third of the
property, the son may sell his share for a $250,000 gain without incurring a tax. His parents could
simultaneously sell their share for $500,000 without tax, sheltering as much as $750,000 in
capital gains. Note that this avoidance technique arises not from the exclusion, but from the
presence of a cap. This approach to tax avoidance involves some constraints and risks: the child
must live in the residence, and the property could be subject to attachment for the child’s debts.
This avoidance technique might be termed the “land with a small house” strategy. A taxpayer can
purchase a house with a significant amount of land as an investment, and then use the exclusion
for the residence to also exclude gain on the land. You can also sell vacant land adjacent to your
home separately from the home itself, as long as the home is also sold either two years before or
two years after the sale of the vacant land. There are some restrictions on this exclusion for land,
one being that the land must not only be adjacent to the home but used as part of the home (which
would rule out farm land, timber land, and other uses but not simple speculation).
An individual can buy a house that needs substantial renovation as a principal residence, fix it up,
live in it for two years, and then sell the home. This gain reflects untaxed labor income of the
individual, which is now excluded from tax. In fact, this approach can be used by professional
builders who would normally be paid for their services.
An individual who has both a regular home and a vacation home can take measures to shift the
vacation home to principal residence status. Such an individual may effectively continue to live in
the original home in part, but after the required holding period can sell the vacation home, avoid
capital gains, and move back to the regular home as a permanent residence. Which home is
determined to be the principal residence is based on a facts and circumstances assessment,

¹³^Se^e^C^R^S^R^e^por^{t R}^S²²¹¹^3,^T^h^e^S^a^le^of^{a Pr}ⁱⁿ^cⁱ^p^a^{l Re}^sⁱ^de^nc^e^Ac^qu^ir^e^d^Thr^o^u^gh^{a Lik}^e^-^Kⁱ^{nd Ex}^c^h^an^ge^,^b^y^Gr^e^g^g^A.
^Esen^w^eⁱⁿ^{, f}^o^r^m^o^{re in}^f^o^rmatioⁿ^.

including the length of time the taxpayer lives in each home, the location of employment and the
principal residences of family members, mailing addresses (on bills and correspondence, tax
returns, drivers’ licences, and car and voter registrations), the locations of banks used, and the
location of recreational associations and churches where the taxpayer has a membership. Thus, it
is not easy to establish the vacation home as the principal residence, though it may be feasible in
some cases and, of course, the IRS cannot audit every case of this type.
In this avoidance technique, wealthy individuals sell their homes back and forth periodically to
qualify frequently for the capital gains tax exclusion. If they mutually agree to this arrangement,
the transactions costs could be minimal (a lawyer to search the title and record the transaction).
They may not even live in the exchanged homes. Such an arrangement is illegal (a sham
transaction) but may be difficult to detect. This avoidance technique arises from the existence of
the cap.

While there are any number of potential ways to deal with capital gains taxes on owner-occupied
housing, including retaining the current rules, allowing no exclusions, or returning to the pre-
1997 rules, the two areas where changes might be considered are the ceilings on the exclusion
and in rules relating to investment property and tax sheltering.
Options for addressing the ceilings could include eliminating the ceilings altogether, indexing the
ceilings either with respect to general inflation or housing prices, changing the relative ceilings
between single and joint returns, or changing the basic structure of the ceilings.
One policy option could be to remove the ceilings altogether. For some taxpayers, the exclusion
with ceilings enacted in 1997 increases their tax liability, because they might otherwise have used
the rollover provision and then held the asset until death.
One advantage of eliminating the ceilings would be the elimination of any remaining distortions
(such as incentives not to sell a house even if it would be desirable) and record-keeping
requirements. This efficiency gain would reflect not only benefits to high income individuals who
actually pay the tax, but also the much larger group who, because of uncertainty, need to keep
records. As the section below on indexing indicates, by one index, houses have, on average, ¹⁴
increased in value by 70% in the seven years since the existing ceilings were imposed. Suppose
in 1997 a married couple had a home worth $300,000 with a $200,000 gain. A 70% increase
would have caused the new home value to rise to $510,000, for a $410,000 gain. The gain would
have doubled in only seven years, and could exceed the limit of $500,000 in a few years.

¹⁴^Offⁱ^c^e^o^f^Fe^de^ra^{l Housing}^Eⁿ^te^r^p^rise^Ov^e^r^sig^h^{t (OFHEO),}^U^.^{S. H}^ous^e^Pr^ic^e^s^C^oⁿ^tin^ue^to^Ris^e^Ra^pⁱ^dly^{, J}^u^ne¹^,
²⁰⁰^{5, w}^e^bsite^htt^p^://w^ww^.of^h^e^o^.gov^/m^e^d^ia^/^pdf^/1q⁰⁵^hpi^.p^df^{, v}ⁱ^site^{d July}^{1, 20}⁰⁵^.

Other circumstances could cause such a taxpayer to be over the limit. The percentage gain figure
cited above is averaged across the country. In several cities across the country, including some in
California, the average value of houses doubled in only five years. If the couple had a house in
those cities that appreciated at the local average rate, they would have exceeded the $500,000
ceiling. Another circumstance in which the ceiling would be exceeded is if a spouse dies before
most of the additional appreciation occurs. In that case, the surviving spouse could be over the
exclusion limit because the exclusion would be limited to about $350,000 ($100,000 for half the
appreciation that had already occurred and the $250,000 limit for single individuals). Finally,
while some individuals sell houses more frequently, others live in them for a very long time.
Appreciation that occurs more than 20 or 30 years, or even more, will be much greater than that
over a seven-year period. While appreciation has been higher than normal in recent years, a
normal rate of appreciation (that would probably be similar to the nominal growth in per capita
income over the long run) would lead to huge gains. At a 5% rate, the gain would be almost
$700,000 after 20 years. If inflation rises, or the holding period is longer, the gain will be larger.
These examples illustrate that individuals who are living in houses that may currently have
accrued gains or have values well below the limit on the capital gains exclusion would need to
keep records given the uncertainty about how long the house will be owned, what the
appreciation rate will be, whether and when Congress might act to change the ceiling, and
whether a spouse might die.
Eliminating the ceilings would also eliminate the inequities that arise among homeowners. These
inequities tend to arise because of differences in housing prices across states and localities,
differences that lead to more or less frequent sales of houses, and differences among elderly
homeowners that arise from different health outcomes that require the sale of a house.
Eliminating the ceilings has some disadvantages. It would involve a revenue loss. In addition,
some people might see its expected favoritism of high income individuals to be a disadvantage.
Any reduction in tax progressivity, however, would be minor. The revenue loss from eliminating
the ceilings is relatively minor in comparison to the revenue loss for the exclusion in general and ¹⁵
the taxes collected on other assets of high income individuals. For example, in 1999, reported ¹⁶
taxable capital gains on the sales of residences were in the range of $3.7 billion to $4.9 billion.
Assuming a 20% tax rate, the tax on this amount was less than $1 billion and much of this amount
may be due not to the cap but to failing to qualify in other ways. For that same year, the reported ¹⁷
revenue loss from untaxed capital gains was estimated at $5.8 billion. Thus, the presence of the
cap limited the amount of revenue loss from the exclusion of capital gains by less than—perhaps
much less than—17%.
Total gains on all capital assets were $719.7 billion in 1999. Thus, taxable gains on residences
accounted for around ½ of 1% of all gains. These capital gains are highly concentrated in high
income classes, so the cap on the capital gains exclusion for residences most likely played a

¹⁵^U^.^{S. D}^e^pa^r^t^m^e^{nt of}^T^r^e^a^s^ur^y^,^Iⁿ^te^rⁿ^a^l^R^e^v^e^nue^Se^r^v^ic^e^,^Statistic^{s of}^Inc^o^me^{, Sh}^or^{t T}^e^r^m^a^{nd L}^ong^T^e^r^m^C^a^pita^l
^G^a^ins^b^y^A^s^s^e^t^Ty^pe^{, Ta}^x^Y^e^a^r¹⁹⁹⁹^{, w}^e^bs^ite^http^://w^w^w^.irs^.us^t^rea^s^.g^ov^/^p^u^b^/ⁱ^rs^-s^oi/⁹⁹ⁱⁿ^01a^b.x^l^s^,^vⁱ^s^ite^{d J}^u^ly^{1, 2}⁰⁰⁵^.
¹⁶^Iⁿ^c^ons^is^te^nc^ie^sⁱⁿ^how^the^e^x^c^l^us^{ion w}^a^s^a^c^t^ua^lly^r^e^por^te^{d on r}^e^t^u^rⁿ^s^r^e^s^u^ltsⁱⁿ^s^o^m^e^unc^e^r^ta^inty^a^{bout t}^h^e^a^c^t^ua^l
^siz^e^of^the^g^a^{in, bu}^{t it s}^h^o^u^ld^f^a^{ll be}^tw^e^eⁿ^the^s^e^t^w^{o v}^a^lue^s^.
¹⁷^Joint^Com^m^itte^e^{on T}^a^x^a^tion,^Es^tim^ate^s^of^Fe^de^r^a^l^Tax^Ex^pe^ndit^u^r^e^s^for^Fis^c^{al Ye}^ar^s¹⁹^99-²⁰⁰³^{, J}^o^int^Com^m^itte^e
^P^rⁱⁿ^t,^W^a^s^h^ing^t^on,^D^C^{, U}^.^S^.^G^o^v^e^rⁿ^m^e^{nt P}^rⁱⁿ^ting^of^fⁱ^c^e^,^D^e^c^.^14,¹⁹⁹^8,^p.^18.

minor role in affecting the tax liability of higher income individuals, compared to the capital
gains taxes in general.
The capital gains tax rates were cut substantially in 1997, and by almost 50% for the highest
income class, the same year the capped exclusion for gains on residences was enacted. The effect
of that cut dwarfed the revenue effect from imposing the cap. Capital gains on residences were
also a small part of the total income of high income individuals. For example, the adjusted gross
income of individuals with more than $1,000,000 in that same year was $653 billion; the income
of those with more than $500,000 was $888 billion, and the income of those with more than
$200,000 was $1,421 billion. Thus, with any of these classes, taxable capital gains on residences
accounts for less than 1% of income and not all of the gain is due to the cap.
Moreover, the data collected on tax returns filed in 1995 and 1996 (before the 1997 change)
indicate that the benefits did not solely or even largely accrue to high income individuals if ¹⁸
income is measured without including the gain itself. These data show large shifts in the
distribution between 1995 and 1996, but in both years from 20% to 25% of the tax that would
have been collected under the new law accrued to individuals with incomes below $20,000. In
1996, very little of the tax (less than 20%) would have been paid by those with incomes over
$100,000. It is possible that the distribution of the tax that one might normally think would accrue
to high income individuals may reflect sales due to divorce, job loss, and ill health. Indeed,
wealthy individuals may be more likely to have the resources to keep their houses through these
types of changes.
As these data suggest, compared to other capital gains provisions, the cap on residential capital
gains has relatively small revenue effects and plays a relatively small role in the distribution of
the tax burden. So while revenue and distributional issues may be of concern, other changes could
easily be made that would accomplish those same goals. For example, the reduction in the tax rate
on dividends enacted in 2003 cost 20 to 25 times as much ($20 billion) as the revenue gained ¹⁹
from maintaining the dollar cap on the exclusion.
Note that the appreciation in housing values should make revenue from the cap somewhat larger
than reported in 1999 relative to income and other capital gains (although not necessarily relative
to the benefit of the exclusion in general).
Another possible problem to consider, however, when weighing eliminating the cap, is the
possibility of increased tax avoidance through like-kind exchanges, through converting
investment property to residences and other techniques as described above. These activities are,
however, likely to be undertaken by a minority, perhaps a small minority, of taxpayers. One of the
challenges in designing tax rules is determining how to restrict general provisions of the tax law
in order to target abuses by a small number of individuals. In addition, in most cases, there are
alternative measures that can be undertaken to address such problems, as discussed below.

¹⁸^Se^e^G^e^r^a^{ld A}^u^te^{n a}^{nd A}^ndr^e^w^R^e^s^c^hov^s^k^y^,^“T^he^N^e^w^Exc^l^us^{ion f}^o^r^C^a^pita^{l G}^a^ins^oⁿ^P^rⁱⁿ^cⁱ^pa^{l R}^e^sⁱ^de^nc^e^s^,”
^W^o^r^k^ing^P^a^pe^r^,^O^c^t^obe^r¹⁹^98.^A^pr^e^v^ious^v^e^r^s^ion^of^this^pa^pe^r^w^a^s^pu^blis^he^dⁱⁿ^the^P^r^o^ceedⁱⁿ^g^s^o^f^t^h^e⁹⁰^th^Aⁿⁿ^ual
^Coⁿ^f^eren^{ce on}^T^a^xa^tⁱ^oⁿ^o^f^t^h^{e Nat}ⁱ^o^na^l^T^a^{x A}^sso^ci^a^tⁱ^oⁿ^{, N}^a^ti^ona^l^T^a^x^A^s^s^o^cⁱ^a^{tion: W}^a^s^h^ing^t^on^{, D}^C^{, 19}⁹⁸^{, p}^p^.²²^3-
²³⁰^.
¹⁹^S^{ee Jo}ⁱⁿ^t^Co^mmⁱ^t^t^{ee o}ⁿ^T^a^xatⁱ^oⁿ^,^“E^stⁱ^m^at^ed^Bu^d^g^et^E^f^f^ect^s^f^o^{r t}^h^{e Co}ⁿ^f^eren^ce^A^g^re^e^m^en^t^oⁿ^H.^R.²^,^{” JCX}^-⁵⁵^-⁰³^,
^Ma^y^{22, 20}^03,^ht^tp:/^/w^w^w^.hous^e^.^g^o^v^/jc^t/pubs^03.^htm^l^.

Rather than eliminating the cap, another approach would be to adjust the ceiling to reflect recent
price changes and index it for future price changes. A commitment to indexing, rather than
intermittently changing the cap as has occurred with prior exclusions, would provide individuals
more assurances that they might not need to keep records.
There are several indexes that might be considered. The Office of Federal Housing Enterprise
Oversight (OFHEO) data cited above show that house prices appreciated by 70% from the last ²⁰
quarter of 1997 through the first quarter of 2005. Applying this index to the $500,000
($250,000) ceilings would imply a ceiling of $850,000 ($425,000) for the first quarter of 2005,
and this amount would presumably be higher for the full year. This price index is based on repeat
sales or repeat financing, and thus reflects price changes for the same property over time.
The National Association of Realtors (NAR) also provides an index for existing single family ²¹
houses, which is simply the price of the average or median existing house sold. Their measures
indicate that the median house has appreciated by 60% and the average value over all houses (the
mean) has increased by 65%. Those measures would indicate an increase in the ceiling from
$500,000 to $800,000 in the case of the median and $827,000 in case of the mean (with the single
exclusion half as large).
The Census provides indexes of new single family homes, using three measures: change in
median price, change in average price, and change in a constant quality house. These measures
indicate somewhat smaller increases of 52%, 61%, and 42% respectively. The last price index is
more related to the general price indices for the economy which, in turn, measure changes in
prices of produced goods controlling for quality. These indexes would imply ceiling increases
from $500,000 to $758,000, $804,000, and $708,000 respectively.
Houses could also be indexed to general price inflation, which over the seven-year period
suggests a 13% increase, and which would bring the $500,000 cap to $567,000.
Using the 70% measure, it appears that without indexing, a substantial increase in the share of
taxpayers subject to the tax, or at least required to measure and report gain, is likely. According to

2000 Census data, 14.1% of houses were valued at amounts over $250,000 and 2.9% were valued ²²

at over $500,000. If we use the OFHEO price index to correct the distribution back to 1997 and
forward to the current time, the estimates suggest that in 1997, 9.4% of houses were valued at
over $250,000 and 1.6% were valued at over $500,000. By the first quarter of 2005, 27.4% were
estimated to be valued at over $250,000 and 6.3% were estimated to be valued at over $500,000.
These measures are consistent with Treasury estimates that in 1995, less than 3% of taxpayers
were potentially subject to the tax.
We can also use data on gains reported on tax returns in 1998 and 1999 to estimate the increase in
the share and absolute number of homeowners paying the capital gains tax. Using this approach,

²⁰^O^f^fⁱ^c^e^o^f^Fe^de^r^a^{l H}^ous^ing^Eⁿ^te^r^p^rⁱ^s^e^O^v^e^r^sⁱ^g^h^{t (}^O^FH^EO⁾^,^U^.^{S. H}^ous^e^P^r^ic^e^s^C^o^ntin^ue^t^o^Rⁱ^s^e^R^a^pi^dly^,^J^une^1,
²⁰⁰^{5, w}^e^bsite^htt^p^://w^ww^.of^h^e^o^.gov^/m^e^d^ia^/^pdf^/1q⁰⁵^hpi^.p^df^{, v}ⁱ^site^{d July}^{1, 20}⁰⁵^.
²¹^Se^e^U^.^{S. D}^e^pa^r^t^m^e^{nt of}^H^ous^ing^a^{nd U}^r^ba^{n D}^e^v^e^lopm^eⁿ^t,^U^.^S.^H^ousⁱⁿ^g^an^{d M}^a^r^k^e^t^C^o^nditⁱ^oⁿ^s^{, M}^a^y²⁰⁰⁵^.
²²^U.^S^.^Cen^s^u^s^Bu^reau^,^V^a^l^u^{e f}^o^{r S}^p^eci^fⁱ^ed^O^wⁿ^e^r-Occu^pⁱ^ed^Ho^u^sⁱ^{ng Un}ⁱ^t^s^,^h^t^t^p^:^/^/^f^act^fⁱⁿ^d^e^r.^cen^su^s.^go^v/^h^o^m^e/^sa^ff^/
^m^a^in.htm^l?^_laⁿ^g⁼^eⁿ^.

the number of homeowners paying the tax was less than 90,000 in 1997 (had the new rules been
in effect for the entire year). While some portion of these taxpayers (perhaps a significant portion)
paid tax because they did not qualify because of other rules, for those who paid the tax because
they exceeded the exclusion, their numbers could have increased by more than five fold by ²³

2005.

Eventually, the growth in the number of individuals subject to the tax may slow, in part because
housing prices are not likely to continue to grow at such a rapid pace and in part because gain will
no longer include accumulated gain on previous residences. This latter effect will cause fewer
people with houses valued above the ceiling to have gains in excess of the exclusion, but will not
reduce the numbers required to file a return and, thus, the need to keep records.
Which index would be appropriate depends on the objective of the cap. The recent change in
housing prices has far outstripped the change in the overall price level by any measure and a good ²⁴
part of that difference could reflect a housing bubble. Over a period of time, however, indexing
with respect to price levels would lag the value of housing price increases, which would tend to
grow with income. During a seven-year period after the 1997 enactment, the value of GDP grew
about 41% and, were the ceiling indexed to output growth, it would have risen to $706,000. If the
objective of the cap is to maintain a fixed inclusion to exclusion ratio, a house price index would
be appropriate, or perhaps an index to income (the latter would maintain the relationship over a
long period of time). If the objective is to fix the cap in real terms, an index to a general price rise
would be appropriate. If a general price increase is chosen, however, more and more gains will

²³^T^h^{ese est}ⁱ^m^a^t^e^{s are ro}^u^g^h^,^an^{d req}^uⁱ^{re i}ⁿ^t^e^rpo^l^atⁱ^oⁿ^b^e^t^w^een^dⁱ^ff^er^en^t^ho^u^sⁱⁿ^{g val}^u^{e cat}^ego^rⁱ^e^s.^In^t^e^rp^o^l^atⁱ^oⁿ
^b^e^t^w^een^t^h^{e t}^w^o^ad^j^acen^t^cat^ego^rⁱ^e^{s i}^s^u^s^ed^f^o^{r m}^e^a^s^u^rⁱⁿ^{g t}^h^{e sh}ⁱ^f^tⁱⁿ^t^h^{e ho}^u^sⁱⁿ^{g val}^u^e^dⁱ^st^ri^bu^tⁱ^oⁿ^.^T^h^{ese est}ⁱ^mat^e^s
^indic^a^t^e^t^h^a^t^the^s^h^a^r^e^of^housⁱⁿ^g^ov^e^r^$25^0,⁰⁰^{0 w}^a^s^10.0%ⁱⁿ¹⁹⁹⁸^aⁿ^d¹¹^.9%ⁱⁿ¹⁹^99;^the^s^h^a^r^e^o^v^e^r^$500^,0^{00 w}^a^s
^2.0%^aⁿ^{d 2}^.^4%^r^e^s^p^e^c^tiv^e^l^y^.^T^h^e^r^e^w^e^r^e^111,⁰⁰^{0 r}^e^turⁿ^sⁱⁿ¹⁹⁹⁸^aⁿ^{d 1}^42,⁰⁰^{0 i}ⁿ¹⁹^{99 w}^{ith c}^a^pⁱ^t^a^l^g^a^ins^oⁿ^r^e^sⁱ^de^nc^e^s^,
^a^c^c^ounting^f^o^r²^.^2%^aⁿ^{d 2}^.^7%^r^e^s^p^e^c^tiv^e^l^y^of^hous^ing^s^a^le^s^{, b}^u^{t t}^h^e^s^e^r^e^turⁿ^s^m^a^y^be^ov^e^r^s^t^a^t^e^d^b^ecau^{se o}^f^rep^o^r^tⁱⁿ^g
^inc^ons^is^te^nc^ie^s^,^a^{nd m}^a^y^be^a^s^sm^a^{ll a}^s^84,0⁰⁰ⁱⁿ¹⁹⁹⁸^aⁿ^{d 1}⁰⁶^,⁰^{00 i}ⁿ¹⁹^99.^Som^e^of^the^s^e^r^e^turⁿ^s^m^a^y^a^l^s^o^r^e^por^t
^capⁱ^t^a^l^gaiⁿ^{s t}^a^{x b}^ecau^{se t}^h^ey^dⁱ^dⁿ^o^t^qu^alⁱ^f^y^unde^r^t^h^e^tw^{o y}^e^a^r^r^e^quir^e^m^e^{nt, b}^ecau^{se t}^h^{e resi}^d^eⁿ^ce^w^a^{s n}^o^t^t^h^e
^p^rⁱ^m^ar^y^re^si^d^eⁿ^{ce o}^r^b^ecau^{se o}^f^dep^r^eci^atⁱ^oⁿ^recap^t^u^re,^o^r^t^h^ey^m^a^y^h^a^{ve rep}^o^r^t^e^d^{a gai}ⁿ^even^t^h^o^u^g^h^t^h^ey^{were n}^o^t
^req^uⁱ^red^t^o^.^T^h^{e p}^e^rcen^t^a^{ge ch}^an^{ges i}ⁿ^t^h^{e s}^h^{ares are u}^s^ed^t^o^p^r^oj^ect^t^h^e^p^e^rcen^t^a^{ge o}^f^sal^e^{s b}^{ack t}^o¹⁹⁹⁷^an^{d f}^o^rw^ard^st
^{to t}^h^e¹^q^u^a^r^te^{r of}²⁰⁰⁵^{. T}^h^e^s^e^p^e^rc^eⁿ^ta^g^e^s^a^r^e^theⁿ^m^u^ltiplie^d^by^theⁿ^u^m^b^e^r^of^sa^le^s,^w^h^ic^{h rose}^f^r^om^4.38^{2 m}^illion
^{in 1}⁹⁹⁷^t^o^6.8⁹⁰^m^{illion i}ⁿ^Ma^rc^h^of²⁰^05.^(U.S^{. De}^pa^rtm^e^{nt of}^Ho^using^aⁿ^d^Urba^{n De}^v^e^lopm^eⁿ^t,^U^.^{S. H}^ousⁱ^ng^st
^M^a^r^k^e^t^C^onditⁱ^oⁿ^s^,¹^Q^uar^te^r²⁰⁰⁵^{, Ma}^y²⁰⁰^{5. T}^h^e^s^e^num^be^r^s^s^u^g^g^e^s^t^{, de}^pe^ndi^ng^on^w^h^e^t^he^r^the^$2^50,⁰⁰^{0 s}^h^a^r^e
^ch^an^{ge o}^r^t^h^e^$⁵⁰⁰^,⁰⁰⁰^s^h^{are ch}^an^{ge i}^s^u^s^ed^and^w^h^et^h^e^{r t}^h^e^b^o^t^t^o^m^o^r^t^o^p^o^f^t^h^{e ran}^g^{e i}^s^u^s^ed^,^t^h^at^,^at^{a m}^a^xi^m^u^m^,
^60,⁰⁰^{0 t}^o⁹⁰^,⁰⁰⁰^re^turⁿ^{s w}^ould^h^a^v^e^ha^{d ta}^x^{in 1}⁹⁹⁷^ha^{d the}^provⁱ^s^ion^be^eⁿⁱⁿ^e^f^fe^c^t^the^f^u^{ll of}^tline^y^e^a^r^a^{nd ha}^{d a}^ll^of^st
^t^h^{e ret}^u^rⁿ^{s ref}^l^ect^ed^t^a^{x d}^u^e^b^ecau^{se o}^f^exce^edⁱⁿ^{g t}^h^{e excl}^u^sⁱ^o^n.^F^o^{r t}^h^e¹^q^u^a^r^te^r^of²⁰⁰⁵^{, t}^h^eⁿ^u^m^be^r^w^{ould be}
³²⁶^,⁰⁰⁰^t^o⁴⁹⁸^,⁰⁰⁰^.^T^h^{ese n}^u^m^b^e^{rs are t}^o^o^l^a^rge,^p^e^rh^a^p^{s m}^u^ch^t^{oo l}^a^rge,^h^o^w^ever,^b^ecau^{se so}^m^e^o^f^t^h^{e t}^a^xp^ayers
^r^e^por^ti^ng^g^a^{in m}^a^y^hav^e^be^eⁿ^doi^ng^s^o^f^o^r^ot^he^r^r^e^a^s^ons^.^Som^e^pr^e^lim^ina^r^y^T^r^e^a^s^ur^y^da^ta^s^ugg^e^s^ts^t^h^a^t^oⁿ^ly^a^sm^a^ll
^sh^{are o}^f^al^l^ret^u^rn^{s repo}^rtⁱⁿ^{g gai}ⁿ^{s o}^r^l^o^{sses on}^resi^d^eⁿ^ces,^l^e^{ss t}^h^an¹⁰^%,^p^aⁱ^d^t^a^{xes i}ⁿ²⁰⁰¹^b^ecau^{se t}^h^ey^e^x^ce^ed^ed^t^h^e
^d^o^l^l^a^{r l}ⁱ^mⁱ^t^,^w^hⁱ^c^h^w^o^u^l^d^p^l^{ace l}^e^{ss t}^h^an¹^%^o^f^t^o^t^a^l^ho^m^e^sal^e^{s a}^ffect^ed^b^y^t^h^{e t}^a^{x d}^u^e^t^o^exceedⁱⁿ^g^t^h^{e cap}^.^T^h^at
^e^s^ti^m^a^te^is^ba^s^e^{d on a}^v^e^r^y^sm^a^{ll s}^a^m^p^le^{. I}ⁿ¹⁹⁹⁹^,²⁹^0,⁰⁰^{0 r}^e^turⁿ^s^r^e^por^te^{d g}^aⁱⁿ^or^l^o^s^s^oⁿ^r^e^sⁱ^de^nc^e^s^{, a}^{nd t}^h^e
^e^s^ti^m^a^te^s^note^d^a^bov^e^oⁿ^w^h^ic^{h t}^h^e^e^x^tr^a^pola^tio^ns^w^e^r^e^ba^s^e^{d r}^a^ng^e^f^r^o^m^37%^to^50%^of^the^t^o^ta^l,^m^u^c^h^la^r^g^e^r^thaⁿ
^the^s^h^a^r^e^s^u^g^g^e^s^t^e^d^{in t}^h^e^T^r^e^a^s^ur^y^da^ta^{. T}^hus^the^c^u^r^r^e^nt^num^be^r^s^of^ta^x^p^ay^e^r^s^af^f^e^c^t^e^d^m^a^y^be^w^e^ll^un^de^r¹^00,⁰⁰^{0 if}
^the^s^e^pre^lim^ina^r^y^e^s^ti^m^a^te^{s a}^r^e^c^o^rre^c^t^{. If}^pric^e^s^c^ontin^ue^to^rⁱ^se^{, h}^o^w^e^v^e^r^{, the}²⁰⁰^{5 y}^e^a^r^w^{ill be}^la^rg^e^r^{. T}^h^{is pro}^j^e^c^tion
^al^so^h^a^{s an}^asp^ect^t^h^at^un^d^e^rst^a^t^e^{s t}^h^{e resu}^l^t^s^b^ecau^{se cap}ⁱ^t^a^l^gaiⁿ^{s ri}^{se f}^a^st^{er t}^h^an^ho^u^sⁱⁿ^{g p}^rⁱ^{ces i}ⁿ^t^h^{e sh}^o^r^t^run^{as t}^h^e
^ba^{sis is g}^e^ne^ra^ll^y^fⁱ^x^e^{d. More}^pre^c^ise^de^ta^{il on}^the^a^b^s^o^lute^num^be^{r of}^ta^x^p^a^y^e^r^{s a}^ffe^c^t^e^d^w^{ill re}^quire^a^m^o^re^de^ta^ile^d
^aⁿ^a^l^y^s^is^tha^t^c^aⁿ^oⁿ^ly^be^unde^r^t^a^k^eⁿ^w^{ith c}^onf^ideⁿ^tia^l^ta^x^da^ta^.
²⁴^A^hous^ing^bu^bble^is^a^rⁱ^s^eⁱⁿ^ho^u^s^ing^pr^ic^e^s^tha^t^c^aⁿ^not^be^e^x^plaⁱⁿ^e^{d by}^f^unda^m^e^nta^l^s^.^Fun^d^a^m^eⁿ^ta^ls^r^e^f^e^r^{to the}
^e^x^plaⁿ^a^{tions f}^o^{r h}^ousiⁿ^g^pric^eⁱⁿ^c^r^e^a^s^e^s^w^h^ic^{h inc}^l^ude^f^a^llingⁱⁿ^te^re^{st ra}^te^{s, inf}^l^a^{tion, a}ⁿ^{d risi}^ng^inc^o^m^e^{s. For m}^o^re
^inf^o^r^m^a^{tion a}^bout^h^ousⁱⁿ^g^bu^bb^le^s^,^s^e^e^C^R^S^R^e^por^{t R}^L³¹⁹^18,^U^.^S.^H^ousⁱ^{ng Pr}^ic^e^s^:^I^s^T^h^e^r^e^{a B}^u^b^b^l^e^?^{, by}^Ma^rc
^L^a^bonte^.

fall above the cap and in the very long run, virtually all gain will be above the cap and thus be
subject to tax.
Another issue is whether the single exclusion should remain at a level that is half the joint
exclusion. In making the cap half as large for singles (or twice as large for married couples), the
provision departed from historical practices for the over-age-55 exclusion. The 1997 change
doubled the existing $125,000 exclusion for singles, although that exclusion had not kept pace
with house price changes because it had not been changed since 1981. If the 1981 value had been
adjusted based on the Census bureau’s average new house index, it would be $413,000 ²⁵
currently. Thus, single individuals who might have been eligible for the old age exclusion have
lost ground compared to some historical periods, while married couples, at least currently, have
gained. For an equivalent revenue cost, this approach favors married individuals relative to single
individuals, including widows and widowers. (Both types of taxpayers could have been made
worse off because there was no cap on rollovers).
Allowing the exclusion to be half as large for single taxpayers may have reflected, to some extent,
the tax planning problems faced by divorcing couples. If each taxpayer has the same ceiling, then
it is more advantageous to sell a house with a large gain after the divorce, when each individual
could have a full exclusion. This problem may have been less important for older individuals in
the past when divorce was less likely, but with the exclusion substituting for rollover treatment,
many more divorcing couples would be facing this problem. Higher income individuals who
divorced and optimized their timing may be worse off under the post-1997 changes because they
were not eligible for an uncapped rollover or larger exclusion. However, the new rules are
beneficial for moderate income divorcing couples who wish to trade down. The change in relative
exclusions could have addressed the problem of unmarried couples who own houses together, an
issue that arose as part of the discussion of the “marriage penalty” in the income tax rate
structure. This latter phenomenon is probably not very numerically important since, according to ²⁶
Census data unmarried couples were only 4.2% of households. Moreover, because they tend to
be younger, they are less likely to be homeowners.
Many of the single individuals selling homes, outside of divorcing couples, are likely to be
widows or widowers and the remainder are largely people who have never been married or who
have been divorced for some time. By reducing the exclusion ceiling for them, not only is the
benefit reduced relative to historical values, but complexity is introduced because more
individuals will be subject to filing requirements and paying taxes. As noted above, more than a
quarter of houses are estimated to have values over $250,000.
Although it eliminated the complexity for divorcing couples, the halving of the exclusion
especially magnified compliance problems for surviving spouses. Although surviving spouses can
receive the benefit of the step-up in basis for the half of the house allocated to the decedent, the

²⁵^Dⁱ^f^f^e^r^eⁿ^{t v}^a^lue^s^w^{ould be}^f^o^uⁿ^d^if^inde^xⁱ^ng^w^h^e^r^e^intr^od^uc^e^d^f^r^om^ea^r^lie^r^le^v^e^ls^{. For}^e^x^am^ple^,^us^ing^the^C^e^ns^us
^B^u^r^e^a^u^’^s^a^v^e^r^a^g^e^ne^w^hous^e^inde^x^c^ite^{d a}^bov^e^,ⁱⁿ^de^xⁱ^ng^f^r^om^the¹⁹⁶^{4 le}^v^e^{l w}^ould^ha^v^e^im^plie^{d a}^c^u^r^r^e^{nt e}^x^c^l^us^ion
^tha^t^w^a^s^s^lig^htl^y^la^r^g^e^r^thaⁿ^the^$²⁵⁰^,0⁰⁰^one^now^a^llow^e^{d: $2}⁷⁴^,5^{00. T}^h^e^v^a^lue^ha^{d de}^te^rⁱ^or^a^t^e^d^by¹⁹^76,^s^o^t^h^a^t^t^h^e
^e^x^c^l^us^{ion in}^de^x^e^{d f}^r^om^tha^t^poin^t^w^{ould be}^$2⁰⁰^,0^00.^Iⁿ^de^xⁱ^ng^f^r^om^{1978, w}^h^eⁿ^a^m^u^c^h^la^r^g^e^r^e^x^c^l^us^ion^w^a^s^eⁿ^a^c^t^e^d
^w^{ould pr}^od^uc^e^a^v^a^lue^of^$43^9,⁰^00.
²⁶^Ce^{nsus Bure}^a^u^,^“^‘Sta^y^-^a^t^-Ho^m^e^’^P^a^re^{nts T}^op^{5 Mi}^llio^n,^Ce^nsus^Bure^a^u^Re^por^ts,”^P^r^e^{ss Re}^le^a^s^e^,^Nov^.²⁰^{, 2}⁰⁰⁴^.

lower ceiling not only increases the frequency with which basis must be calculated (any time the
sales price is $250,000 or more) but also requires the measurement of the basis step-up. (Note:
Surviving spouses can get the full step up of the entire gain in community property states, such as
California). These individuals may be more likely to have houses that fall into the taxable range
because they were married, perhaps for a long time. And the lower limit in general adds to the
risk that even a couple with a modest house will have to keep complex records because of the
possibility that one of them will die before the house is sold.
One potential change would be to allow surviving spouses to opt for the $500,000 exclusion as a
substitute for the step-up in basis if they are selling a house they lived in with their spouse, a
move that would simplify compliance for those whose housing values fall between $250,000 and ^th
$500,000. H.R. 3803, introduced by Representative McCarthy in the 109 Congress, proposed
this allowance for certain surviving spouses.
In the 110^th Congress a similar proposal was introduced by Senator Schumer (S. 138) and
included in H.R. 3648, the Mortgage Forgiveness Debt Relief Act of 2007, which became public
law (P.L. 110-142) on December 20, 2007. Specifically, the new law allows a surviving spouse to
exclude from gross income up to $500,000 of the gain from the sale or exchange of a principal
residence owned jointly with a deceased spouse if the sale or exchange occurs within two years of
the death of the spouse and other ownership and use requirements have been met.
Another option is to change the structure of the exclusion in general. For example, there could be
a much larger lifetime exclusion, with each sale of a home using up part of the exclusion. A
lifetime exclusion would eliminate the incentive to turn over houses frequently and would
eliminate the penalty for holding on to one’s home for a long period of time. It would shift
benefits (even for the same revenue cost) from very wealthy people who sell houses frequently to
people who are less wealthy but have lived in their houses for a long time. It would, however, add
to administrative costs and to complexity for some people who would need to keep track of the
amount of the exclusion consumed.
A different approach would be that embodied in H.R. 2127 of the 109^th Congress, which would
have allowed a one-time doubling of the exclusion for those over the age of 50.
A final issue is whether additional measures should be taken to prevent the use of the exclusion
from sheltering gains earned from investment property. There is little or no evidence of how
serious this issue is. According to revenue estimates, the restriction on like-kind exchanges
enacted in 2005 (requiring a five-year test) will have a negligible revenue effect ($200 million ²⁷
over 10 years). The revenue loss from the converting of investment property into a residence is
likely larger.

²⁷^{U.S. C}^ong^re^{ss, Join}^{t Com}^m^itte^e^{on T}^a^x^a^tion,^E^s^tⁱ^m^a^t^ed^B^u^d^g^e^t^E^f^f^ect^{s o}^f^t^h^{e C}^oⁿ^f^eren^{ce A}^g^reemen^t^f^o^{r H.}^R⁴⁵²⁰^,
^The^“Am^e^rⁱ^c^aⁿ^J^obs^C^r^e^a^tio^{n Ac}^{t of}²⁰⁰⁴^”,^J^C^X-⁶⁹^-⁰^{4, (}^W^a^s^hing^ton:^G^P^O^,²⁰⁰⁴⁾^.

Note that these tax sheltering problems are not unique to current law; they existed under prior law
as well.
To deal with the investment and “like kind” exchanges, the provision requiring longer residence
periods for “like kind” exchanges could be expanded to property used as an investment. Another,
more direct approach, which might be used to raise revenue to finance cap expansions, would be
requiring the gain attributable to investment property to be separated out and taxed, as recaptured
depreciation is now taxed.
The conversion of investment property is probably the most important tax avoidance scheme.
Techniques such as buying a house with extensive land, establishing a vacation home as a
residence, or explicit house swapping can probably only be addressed through tax auditing. These
shelters would probably be made less attractive with longer holding periods, but longer holding
periods have other consequences (such as interfering with mobility). The effect of longer holding
periods would not be as onerous, however, because houses held for a short period of time are
likely to have little appreciation, especially after deducting real estate commissions and other
selling costs. And partial exclusion could be allowed for cases where sales were due to
employment changes, and other factors. In addition, record keeping for a few years is not the
burden that would be the case if the house was owned for decades. House swapping would be
reduced or eliminated if the ceilings were increased or eliminated. It is very difficult to deal with
the professional “fixer-upper” problem, although, as in the case of other tax avoidance schemes,
longer holding period requirements would discourage such methods.

Capital gains on sales of taxpayers’ homes have been preferentially treated in the tax code for
decades. Current law allows an exclusion from income taxation of up $500,000 in capital gains
for couples ($250,000 for singles). This preferential treatment is justified for several reasons.
Capital gains taxes on homes create barriers to labor mobility in the economy. Imposing capital
gains taxes on homes also creates significant compliance costs, requiring individuals to keep
records for decades and to make fine distinctions between improvements and repairs. Capital
gains taxes also tend to distort housing choices, discouraging individuals from selling their homes
because of changing family and health circumstances. The taxation of gains in excess of a cap
creates inequities between homeowners with different job circumstances, between those living in
different parts of the country, and between those with different health outcomes. Exclusions of
gains on homes do, however, contribute to tax avoidance schemes, especially ones that allow
gains on investment properties to escape tax.
The exclusion from capital gains tax for owner-occupied housing currently exempts most
homeowners from the tax. Although the capital gains exclusion adds to the magnitude of
homeowner preferences that may distort investment, there are reasons to exempt capital gains on
home sales from tax. The rise in housing prices juxtaposed with the fixed dollar cap for the
exclusion is, however, estimated to increase more than fivefold the share of homeowners subject
to gains tax. The possibility of capital gains tax in the future, arising from a cap that does not
keep pace with housing prices, substantially reduces the number of taxpayers who could be freed
from detailed record keeping. The problems associated with the gains tax could be eliminated or
mitigated with the elimination of the cap or by indexing it to housing prices. The ceiling on the
excluded gain, presumably adopted for revenue and distributional reasons, is, however, of small
consequence compared to other provisions (such as the general taxes and tax preferences on

capital gains and dividends). A complication of increasing or eliminating the ceiling, however, is
the increased opportunity for tax sheltering activities, which may suggest additional restrictions
aimed at these activities.
^J^aⁿ^e^{G. Grav}^elle ^Pa^m^e^l^a^J.^Ja^c^k^s^oⁿ
^Senⁱ^o^r^Sp^{ecialist i}ⁿ^Ecoⁿ^o^m^ic^P^o^licy^Section^R^e^s^earch^Maⁿ^a^g^e^r
^j^g^rav^e^l^l^e^@^c^rs^.l^oc.^g^ov^{, 7-}⁷⁸²⁹^pj^ack^s^oⁿ^@^crs^.^l^o^c.g^o^v^,^7-³⁹⁶⁷