UCLA School of Law

The Williams Institute

(University of California, Los Angeles)
Year  Paper badgett 

Unequal Taxes on Equal Benefits: The

Taxation of Domestic Partner Benefits
M.V. Lee Badgett
Williams Institute, UCLA School of Law

This paper is posted at the eScholarship Repository, University of California.

http://repositories.cdlib.org/uclalaw/williams/badgett 10
Copyright 2007
c by the author.
 UCLA School of Law

The Williams Institute

(University of California, Los Angeles)
Year  Paper badgett 

Unequal Taxes on Equal Benefits: The

Taxation of Domestic Partner Benefits
M.V. Lee Badgett
Williams Institute, UCLA School of Law

This paper is posted at the eScholarship Repository, University of California.

http://repositories.cdlib.org/uclalaw/williams/badgett 10
Copyright 2007
c by the author.
 Unequal Taxes on Equal Benefits: The
Taxation of Domestic Partner Benefits

Abstract
In this report, we will detail how employees with partners now pay on av-
erage $1,069 per year more in taxes than would a married employee with the
same coverage. Collectively, unmarried couples lose $178 million per year to
additional taxes. U.S. employers also pay a total of $57 million per year in
additional payroll taxes because of this unequal tax treatment. Because the
number of unmarried couples is growing, over time this unfair treatment will
affect millions of families.
Unequal Taxes on
Equal Benefits
The Taxation of Domestic
Partner Benefits

M. V. Lee Badgett
December 2007
D E C E M B E R 2 0 0 7

Acknowledgements
This study was made possible through a generous donation from Merrill Lynch.

The author thanks Deborah Ho for her assistance with the analysis. She also thanks
Winnie Stachelberg, Brad Sears, Christian Weller, and John Irons for their helpful
comments on earlier drafts, and Dr. Gary Gates for providing Census figures.

About the Author

M. V. Lee Badgett is the research director at the Williams Institute. She is also the
director of the Center for Public Policy and Administration, and Associate Professor
of economics at the University of Massachusetts Amherst. Her book, Money, Myths, and
Change: The Economic Lives of Lesbians and Gay Men (University of Chicago Press) presents
her ground-breaking work on sexual orientation discrimination and family policy. She
is currently working on a new book asking whether same-sex marriage will change
marriage or change GLB people, drawing on the U.S. and European experiences with
same-sex marriage.
 D E C E M B E R 2 0 0 7

Introduction and Summary

E
mployer-provided health insurance is the backbone of health coverage for
American families. Most people who have health insurance get it through their
own employer or a family member’s employer.1 Public policy encourages em-
ployers to provide health insurance by exempting that form of compensation from taxa-
tion. As a result, married workers who get family health insurance benefits get a double
benefit—they get health insurance coverage for their spouses and children and are not
taxed on the value of that coverage.

In sharp contrast, workers who have an unmarried domestic partner are doubly bur-
dened: Their employers typically do not provide coverage for domestic partners; and
even when partners are covered, the partner’s coverage is taxed as income to the em-
ployee. Employers who cover domestic partners are also penalized under current law,
since employer payroll tax responsibilities increase along with employees’ income and
Social Security taxes.

As a result, the taxation of domestic partner health care benefits sets up a two-tiered
tax policy that costs many American families and their employers millions of dollars
each year. This report estimates the financial impact of this extra tax on employees
and employers.

In this report, we will detail how employees with partners now pay on average
$1,069 per year more in taxes than would a married employee with the same cover-
age. Collectively, unmarried couples lose $178 million per year to additional taxes. U.S.
employers also pay a total of $57 million per year in additional payroll taxes because of
this unequal tax treatment. Because the number of unmarried couples is growing, over
time this unfair treatment will affect millions of families.

To remedy this situation, we recommend that Congress enact legislation now under
consideration that would eliminate this federal tax on equal benefits. The legislation
would exclude the value of domestic partner benefits, or DPBs, from income subject to
taxation just as the value of employee and spousal benefits is excluded.


D E C E M B E R 2 0 0 7

Unequal Taxes on Equal Benefits

Background: The Uneven Playing Field for Benefits

Families depend on employers for health insurance coverage. In 2007, 60 percent of

employers offer health benefits to employees and typically to the spouses and children
of employees as well.2 A much smaller percentage of employers, however, offer cover-
age to the unmarried domestic partners of employees. Findings from a recent survey
suggest that only 22 percent of employers cover same-sex partners of employees, and
28 percent cover different-sex partners.3

But coverage is uneven. A 2007 Kaiser Family Foundation survey shows that large em-
ployers (with 200 or more employees) are much more likely to cover same-sex partners
than small employers,4 as also seen in the fact that 53 percent of Fortune 500 firms offer
such benefits.5 Coverage of partners is also much more common in the Northeast and
West Coast states than in the South or Midwest, as shown in Figure 1 on page 3.

Some of the regional differences might reflect differences in the legal status of same-
sex couples, since the states that give rights to same-sex couples are clustered on the
West and East coasts. Massachusetts allows same-sex couples to marry, but no oth-
er state does. Vermont, California, Connecticut, New Jersey, New Hampshire, the
District of Columbia, and Oregon allow same-sex couples to enter into a registered
domestic partnership or civil union status, which provides most of the state-provided
legal rights and responsibilities of marriage. Hawaii, Washington, and Maine give
same-sex couples a lesser package of rights.

Legal recognition may have led same-sex couples to push employers in these states to
cover partners. Alternatively, increased social tolerance in these regions may account
for both formal recognition by the states and increased rates of offering DPBs by pri-
vate companies.

The recent increase in partner coverage is likely tied to various economic and social
trends.6 In the 1990’s, the temporary slowdown in the rate of increase of health care
costs plus low unemployment rates allowed employers to add partner benefits to re-
cruit and retain valued employees. Gay, lesbian, and bisexual employees pressed their
employers to offer health care benefits to domestic partners as part of a larger move-
ment for equal rights.


 D E C E M B E R 2 0 0 7

FIGURE 1: COVERAGE OF PARTNERS BY REGION

40%
38%
Percentage of Employers

34%
30%
26%

17%
16%

11%

Same-Sex Diff-Sex
Covered Covered

Northeast Midwest South West

Source: Author’s calculations from findings in Kaiser Family Foundation survey, note 2.

As companies increased their coverage of couples, as shown in Figure 2 on page 4.8

partners, evidence accumulated that the The number of different-sex unmarried
cost of coverage was very small and the couples increased to 5.2 million in 2006
possibility that domestic partners would from 3.1 million in 1990. The number of
have higher-than-average health care same-sex couples increased to 780,000 in
costs was also very low. One recent study 2006 from 145,000 in 1990.
suggests that 0.1 percent to 0.3 percent of
employees will sign up a same-sex partner Over the last sixteen years, the big-
for health care benefits, and 1.3 percent gest percentage increases in the counts
to 1.8 percent of employees would sign of same-sex couples have come in the
up a different-sex partner.7 Midwest and South, suggesting that the
relatively low rates of partner coverage
Pressure on companies to treat employ- by employers in those regions may soon
ees with partners equally when com- come under increasing scrutiny.9 More
pared with married employees is likely than 6 million unmarried couples are
to increase. Over the last few decades, now living together as partners. These
the number of families who fall into families often include children, as well.
this category has been increasing as a In Census 2000, one in four same-sex
result of changes in marriage patterns couples had children under 18 living in
for different-sex couples and an increas- the home, as did 39 percent of different-
ing willingness to be open for same-sex sex unmarried couples.10


D E C E M B E R 2 0 0 7

FIGURE 2: NUMBERS OF UNMARRIED COUPLES

6,000,000 900,000

800,000
5,000,000
700,000

4,000,000 600,000
Different-Sex Couples

Same-Sex Couples
500,000
3,000,000
400,000

2,000,000 300,000
Different-sex couples
Same-sex couples
200,000
1,000,000
100,000

0 0
1990 2000 2002 2003 2004 2005 2006

Source: Data from U.S. Census and American Community Survey provided by Dr. Gary Gates.

These millions of couples face a significant A recent study found that if employers
disadvantage in getting health insurance offered equal coverage for spouses and
coverage for their families, since employers unmarried partners, the gap in insurance
typically do not provide health care ben- coverage between the two groups would
efits to domestic partners of employees. fall by at least a third.12
This practice leaves people with partners
particularly vulnerable to being uninsured. One significant inequality remains for
people with domestic partners. Even
Recent research shows that people in when employers offer equal benefits, the
unmarried couples are two to three Internal Revenue Code treats the value
times more likely to be uninsured than of the benefits as taxable income, or “im-
people who are married, mainly because puted income,” to the employee.13 The
of their exclusion from employer-pro- only exception to taxation comes when
vided plans.11 Figure 3 shows that while the partner qualifies as a dependent of
16 percent of the whole United States the employee.14
population is uninsured, only 11.5 per-
cent of married people lack insurance. In addition, employers must also pay
In contrast both different and same-sex taxes on this imputed income for their
couples have higher rates of uninsurance share of the employee’s payroll tax. In
than the general population: 20 percent contrast, benefits for an employee or for
of people with same-sex partners and an employee’s spouse are not considered
almost one-third of people with different- taxable income, regardless of the depen-
sex unmarried partners are uninsured. dence or independence of the spouse.


 D E C E M B E R 2 0 0 7

Although we do not know the precise by unmarried partners and their employ-
impact of the taxation of partner benefits ers millions of dollars each year. Employ-
on an employee’s choice of coverage for ees pay both income tax and the 7.65 per-
a partner or on an employer’s choice to cent payroll tax (for Social Security and
offer such benefits, standard economic Medicare) on the imputed income. Em-
reasoning suggests that the taxation of ployers also pay 7.65 percent on imputed
benefits increases the price of cover- income for their share of the payroll tax.
age for employees and employers. That
higher price, therefore, would reduce the A related tax disadvantage is that em-
number who would enroll a partner for ployees cannot use pre-tax dollars to pay
benefits and the number of employers for a partner’s coverage. In this section,
who offer such benefits. we calculate the average impact of un-
equal taxation on the average employee
with a partner as well as the impact on
Equal Benefits, the employer.
Unequal Taxes
To estimate the amount of taxes paid by
While we do not know the effect of the employees and employers on the value of
taxation of domestic partner health care partner benefits, we use data from several
benefits on the willingness to sign up a government data sources: the Current
partner, this unequal taxation will clearly Population Survey, the U.S. Census 2000,
put a dent in employees’ budgets. The the Medical Expenditure Panel Sur-
two-tiered tax policy costs families headed vey, and Internal Revenue Service data.

FIGURE 3: PERCENT UNINSURED

32.4%
Percent Uninsured

20.0%

15.8%

11.5%

U.S. Population Married People Same-Sex Diff-Sex

Unmarried Partners Unmarried Partners

Source: Ash and Badgett (see note x).


D E C E M B E R 2 0 0 7

(In the appendix on page 9 we outline when moving from individual to family

the three steps involved in calculating our coverage or when moving from employ-
estimate in detail.) ee-plus-one to family coverage.

The first step is to estimate the number After estimating the number of people
of people who are currently paying taxes who receive DPBs and the value of those
on DPBs. The Current Population Sur- benefits, in step three we estimate the
vey provides information on how many amount paid in additional taxes. The tax
employees are getting employer-provided rate paid by employees depends on their
benefits for their partners, and we apply tax bracket, which in turn depends on the
the percentages of those covered to counts employee’s income.
of couples from Census 2000, providing
an estimate of 210,000 people receiv- Employees whose highest tax rate is
ing partner benefits. Slightly more than a 10 percent will pay 10 percent of the val-
third of those people have children. ue of partner benefits in additional taxes,
or an extra $303 (10 percent of $3,027)
Some of these individuals covered are for an employee who has a partner but
likely to qualify as dependents, making no children. An employee whose income
their benefits nontaxable. Taking out puts him or her in the 33 percent bracket
those partners who are most likely to be would pay $1,009 in additional taxes.
dependents leaves 166,000 people receiv-
ing taxable partner benefits. We use IRS statistics on the highest rate
paid by taxpayers to estimate how many
Second, we estimate the likely average employees receiving partner benefits are
imputed income on which those individu- in each tax bracket. Finally, we calculate
als pay taxes. The IRS requires no set 7.65 percent of imputed income to get
procedure for measuring this imputed the extra payroll tax payments for em-
income, but one common method is to ployees and for employers.
measure the increased employer con-
tribution that results from signing up a After putting the three steps together, the
partner. When an employee moves from average employee who receives partner
single coverage to “employee-plus-one” benefits pays an additional $771 per year
or family coverage, the employer often out of pocket in federal taxes based on
pays part of that higher premium. the value of those benefits. Out of this
average, $523 is for the federal income
The Medical Expenditure Panel Sur- tax. The remaining $248 is the average
vey (discussed further in the appendix) paid for the FICA tax.
of employer contributions to health
care plans suggests that the increase in Employers pay the same $248 in ad-
the employer contribution from add- ditional federal payroll taxes. These tax
ing a partner with no children would be payments are in addition to any pay-
$3,027 per year for the average em- ments made for premiums, deductibles,
ployer. The increase for someone adding or co-payments. Moreover, in most states,
a partner and children would likely be employees would also pay state income
$4,068, which is the employer contribu- tax on partner benefits, perhaps adding
tion averaged across two situations— hundreds of dollars to their tax bills.


 D E C E M B E R 2 0 0 7

In contrast, a married employee and his employers pay, since there are more em-
or her employer pay no additional taxes ployees who have different-sex partners
on the value of health care benefits for than same-sex partners. Employees with
their spouses. To put this added tax in same-sex partners account for 21 percent
perspective, an employee with a partner of the total of $235 million that workers
who gets employer-provided coverage and businesses pay in additional taxes.
pays 8 percent more in taxes than he or
she would otherwise.15
Fiscal Implications for
A second related disadvantage adds to Eliminating the Equality Tax
the extra taxes paid on the employee
contribution for coverage. Since em- Legislation has been introduced that
ployees cannot pay for partner coverage would eliminate this federal tax on
with pretax dollars, they also lose out equal benefits. The 2007 “Tax Equity
on a potential $298 in savings, which is for Domestic Partner and Health Plan
calculated based on the employee contri- Beneficiaries Act” (S. 1556) and “Tax
bution in the same way as the extra tax Equity for Health Plan Beneficiaries Act”
on imputed income. Thus the total tax (H.R. 1820) would exclude the value of
disadvantage is on average $1,069, or DPBs from income subject to taxation.
11 percent of taxes paid by the typical Since the extra taxes on partner benefits
single taxpayer. paid by employers and employees consti-
tute tax revenue for the federal govern-
Table 1 adds up taxes paid by all employ- ment, this bill would have the effect of
ees and employers with partner benefits, reducing federal tax revenue.
including both out-of-pocket taxes on
imputed income and lost pre-tax savings Such a law could have state fiscal impli-
(assuming that all would otherwise pay cations as well. Some states, including
the employee share with pre-tax dol- California and Massachusetts, however, al-
lars).16 Employees pay $178 million more ready exclude the value of partner benefits
per year in taxes. Employers pay $57 mil- from state taxation for registered domestic
lion more per year. Together they pay partner or married same-sex couples.
$235 million in additional taxes.
The calculations in this report provide
Overall, employees with different-sex an estimate of the change in federal tax
partners account for 79 percent of the revenue if the bill were passed. The loss
higher taxes that either they or their in tax revenue would be equal to the
extra taxes now paid, estimated above
as $235 million per year, depending on
how many partner benefit recipients now
Table 1: Added taxes paid on
domestic partner benefits
qualify as tax dependents.

Employees
If eliminating the unequal taxation of
Imputed income tax $128,432,315
partner benefits leads more employers to
Loss of pre-tax savings $49,614,095
offer those benefits or leads more employ-
Employers $57,276,839 ees to sign up for benefits, the number
TOTAL $235,323,248 of recipients would obviously increase.
Source: Author’s calculations (see appendix for details)


D E C E M B E R 2 0 0 7

The fiscal impact of the policy change, The Equality Tax Dilemma:
however, would not need to reflect such
Taxed if you do, uninsured if
changes, since those individuals are not
you don’t
now paying taxes that would be lost after
the policy change. The federal tax treatment of employer-
provided health care benefits has devel-
To put the revenue loss in perspective, we oped to encourage employers to provide
compare it to the total “cost” of the exclu- health care benefits to employees. The
sion of all employer health benefits from taxation of domestic partner benefits,
taxation. In fiscal year 2008, the total cost however, is likely to discourage the take-
of the exclusion for all employees, their up of family coverage by people with
spouses, and their children or other de- domestic partners. Furthermore, the tax
pendents is expected to be $160 billion. In gives businesses a disincentive to provide
other words, because employer-provided coverage to domestic partners.
health benefits are not taxed as income,
the federal government loses $160 billion Employees face the uncomfortable
per year in income tax payments.17 choice of paying more in taxes or risking
their family members’ health. In effect,
In contrast, expanding the exclusion to the extra $1,069 in taxes that an em-
cover DPBs would cost only $235 mil- ployee pays to cover his or her domestic
lion, or 0.1 percent of the total cost to partner runs counter to the purposes of
the federal government of the health policies that promote both health and
benefit exclusion. equality among Americans.


 D E C E M B E R 2 0 0 7

Appendix
Calculating Taxes Paid on Domestic Partner Benefits

We calculate the direct impact of taxing domestic partner benefits in three steps. First,
we estimate the number of people who are currently paying taxes on domestic partner
benefits, or DPBs. Second, we estimate the likely average imputed income on which
those individuals pay taxes and their lost tax savings. Third, we allocate people from
step one into their likely tax brackets and calculate the extra taxes paid on the imputed
income and lost pre-tax savings from step two.

Estimating the number of people paying equal benefits tax

We have estimated the percentage of people in same-sex and different-sex couples who
currently have health insurance from their partner’s employer. In the Current Population
Survey, 3.2 percent of men in same-sex couples, 4.2 percent of women in same-sex cou-
ples, and 1.7 percent of people in unmarried different-sex couples are covered by DPBs.18

We apply these percentages to the U.S. Census 2000 calculation of the number of un-
married partner couples to estimate a total number of individuals who receive DPBs.19
Combining the Current Population Survey and Census figures in this way, we estimate
that a total of 210,000 people currently get taxable partner benefits. Of that total,
79 percent are people in different-sex unmarried partner couples, and 21 percent are
people in same-sex couples.

Some of those individuals have children, which will influence their health insurance costs
and plan choices. Based on the Census data, we estimate that 77,000 of those individuals
have children under 18, while 133,000 have no children under 18 living in their homes.

Note that these estimates overstate the number of people currently paying taxes, since
partner benefits are not taxable if the partner qualifies as a taxpayer’s dependent under
Internal Revenue Code Section 152. In general, to qualify as a dependent of a taxpayer,
a domestic partner would have to live all year in the taxpayer’s household, receive more
than half of his or her support from the taxpayer, and meet the IRS citizenship/residency
test. Furthermore, the relationship with the taxpayer cannot be “in violation of local law.”

Unfortunately, we do not know how many of the 210,000 might qualify as dependents.
We can estimate the number of dependents from the proportion of couples in which
one partner participates in the labor force but the other does not—in those cases, the
nonparticipating partner is likely to rely on the other for meeting basic needs.


D E C E M B E R 2 0 0 7

We assume that those partners would be considered “dependent” under IRS rules.
In the Current Population Survey 9.2 percent of men and 10.1 percent of women in
same-sex couples are in this situation, while 23.4 percent of different-sex unmarried
couples have this labor force configuration.

After adjusting the earlier figure to remove likely dependents, we estimate that 106,000 in-
dividuals without children and 60,500 people with children are likely to be paying taxes
on their domestic partner benefits. Because this assumption might overstate the number
of legal tax dependents, we also present figures below that do not make this assumption.

Estimating imputed income

People receiving domestic partner benefits for a non-dependent partner must pay taxes
on the imputed income they are receiving from their employers in the form of health
care benefits. They are also unable to pay the employee share of partner benefits with
pre-tax dollars, which also increases their taxes compared with married couples. As
noted earlier, the value of health care benefits provided to a legal different-sex spouse is
not considered taxable income.

The IRS does not provide explicit guidance on how to value partner benefits.20 One
common method of valuing the imputed income measures the increased employer con-
tribution when an employee signs up a partner. Employer plans typically have two or
more tiers of coverage. Plans often include a tier for a single employee, a different tier
for an employee and one other person (“employee-plus-one” or “plus-one”), and a third
tier for a family.

Premiums rise across tiers as more people are covered. The employer contribution to
the premium also usually rises as an employee with a partner moves from his or her
initial base tier to the plus-one or family tier. That increase in employer contribution
constitutes the imputed income that employees must then report to the IRS as taxable
income under this method of calculation.

The imputed income of an employee will depend on the tier that the employee starts in
and the tier that the employee ends up in after signing up a partner and, in some cases,
the partner’s children. The employee might have children of his or her own, in which
case the employee starts in the employee-plus-one or family tier.

Limitations in available data mean that we do not know how families are distributed
across the nine different possible combinations of family configurations (each partner
with zero, one, or two or more children). We know the proportion of unmarried part-
ner households with a child, though, and it seems reasonable to assume that half of
those children are the legal children of the householder who are already covered by the
employee’s plan, and half are the children of the partner and are not initially covered
by the employee’s plan. As a result, we assume that:

10
 D E C E M B E R 2 0 0 7

ß Imputed income for childless couples is the difference between individual and plus-
one coverage

ß Imputed income for half of couples with children is the difference between individu-
al and family coverage

ß Imputed income for the other half of couples with children is the difference between
plus-one and family coverage.

Two sources of information on typical health care plan premiums provide similar esti-
mates of the difference in cost between individual and family plans. The 2005 Medical
Expenditure Panel Survey Insurance Component found that the average premium for
individual coverage was $3,991, while the average family premium was $10,728. The
difference between the two plans was $6,737, and employers paid $4,875 of that differ-
ence, on average.21 Employees paid the rest of the difference.

A 2006 survey of employers by the Kaiser Family Foundation found that the average
premium for a single employee was $4,242, increasing to $11,480 for a family, for a dif-
ference of $7,238. Employers paid $4,892 of that increase, a figure comparable to the
Medical Expenditure Panel Survey figure.

The MEPS study also looked at employee-plus-one options, finding that the typical pre-
mium increase was $3,680 when moving up from the single tier, with employers paying
$2,644. Because the MEPS study offers all of the imputed income figures that we need,
we focus on those values, presented in Table 2. We use past rates of health care cost
increases (7 percent per year from 2004 to 2005) to inflate those 2005 values into 2007
dollars. The part of the premium paid by the employee is used to calculate the potential
savings from paying with pre-tax dollars.

Table 2: Imputed income for Recipients of Domestic Partner Benefits

Type of Family Movement Across Tiers Imputed Income

Childless couples From individual to plus-one $3,027
Half of couples with children From individual to family $5,581
Half of couples with children From plus-one to family $2,554
Source: Calculations from findings of the Medical Expenditure Panel Survey.

Estimating the tax rate on the value of partner benefits

The tax rate paid by employees depends on their tax brackets, which in turn depend
on their incomes. Employees whose highest tax rate is 10 percent will pay 10 percent
of the value of partner benefits in additional taxes, whereas those with higher incomes
will have higher tax rates. For example, using the imputed income values in Table 2, an

11
D E C E M B E R 2 0 0 7

employee who has a partner but no children would pay additional taxes of 10 percent
of the imputed income of $3,027, or $303. An employee whose income puts him or her
in the 33% bracket, however, would pay $1,009 in additional taxes.

To calculate the total taxes paid and the average taxes paid by a DPB recipient, we need
to put the employees in the appropriate tax brackets and filing statuses. First, we assume
that DPB recipients without children use the single filing status, and DPB recipients
with children file as heads of household. Next, we use IRS figures on how many taxpay-
ers paid a particular tax rate in 2005, the most recent year available, to figure out how
many DPB recipients are in each tax bracket.

We assume, for instance, that if 25 percent of all taxpayers filing as single are in the
10 percent tax bracket, then 25 percent of DPB recipients without children are in the
10 percent tax bracket. These marginal tax rates associated with each bracket have not
changed since 2005. The income figures defining the tax brackets have changed, but
those changes are not likely to have dramatically changed the proportion of people in
each tax bracket since both incomes and tax brackets are subject to inflation.

We use IRS statistics that show the proportion of taxpayers who pay each possible rate
as their highest rate.22 We adjust those statistics to take out people whose highest rate
was a capital gains tax rate. Unfortunately, when a taxpayer pays a capital gains tax rate
that is higher than the regular income tax rate, the IRS only publicly reports the rate on
capital gains. When the highest rate is paid on capital gains, it is possible that the tax-
payer would pay a lower rate on earnings, so our estimate would be too high.

To focus as closely as possible on the rates paid by those who have wage and salary in-
come (which is how the value of DPBs is taxed), we first remove the 5 percent or so of
taxpayers who are in categories that indicate that those taxpayers would not have earn-
ings income or that the forms were filed on behalf of dependents who have income;
and then we remove those whose highest rate was a capital gains rate.

Once we allocate the DPB recipients across the tax brackets, it is a simple matter to cal-
culate the income tax on the imputed income (the tax rate times the value from Table
2) and then to multiply the tax on benefits by the number of people in the bracket to
get total taxes paid. To calculate the additional FICA tax, we assume that 6 percent of
taxpayers are over the maximum taxable earnings of $97,500 in 2007, so they do not
pay the 6.2 percent Social Security tax above this threshold but will still pay the 1.45
percent Medicare tax.

For 94 percent of taxpayers, we simply multiply total taxable income by 15.3 percent,
which combines the employer and employee contributions.23 We perform similar cal-
culations on the employee portion to estimate the lost savings available if paying the
employee share with pre-tax dollars.

12
 D E C E M B E R 2 0 0 7

Table 3: Total Tax Payments by Partner Benefit Recipients

and Employers

DBB recipients minus

All DPB recipients
dependent partners

Taxes on imputed income

Income tax $109,663,902 $87,105,631
Payroll tax $104,113,696 $82,653,366
Employee $52,056,848 $41,326,683
Employer $52,056,848 $41,326,683
SUBTOTAL $213,777,598 $169,758,998
Loss of savings from paying with pre-tax dollars
Income tax $42,377,183 $33,663,939
Payroll tax $40,178,203 $31,900,311
Employee $20,089,101 $15,950,156
Employer $20,089,101 $15,950,156
SUBTOTAL $82,555,386 $65,564,250
TOTAL $296,332,984 $235,323,248
Source: Author’s calculations.

Table 3 presents the totals. Column (A) shows taxes paid for all DPB recipients. Column
(B) gives the totals after taking out presumed dependents, which constitutes our best
estimate. Accordingly, we estimate that the total extra taxes paid on domestic partner
benefits are between $235 million and $296 million per year, with $235 million as the
preferred estimate, since it takes into account the highest estimate of the number of
dependents covered.

13
D E C E M B E R 2 0 0 7

14
 D E C E M B E R 2 0 0 7

Endnotes
1 Carmen DeNavas-Walt, Bernadette D. Proctor, and Jessica Smith, “Income, Poverty, and Health Insurance Coverage in the
United States: 2006,” U.S. Department of Commerce, Economics and Statistics Administration, U.S. Census Bureau, August
2007, Table C-1, p. 58, available at: http://www.census.gov/prod/2007pubs/p60-233.pdf, last accessed Nov. 13, 2007.
2 The Kaiser Family Foundation and Health Research and Education Trust, “Employer Health Benefits: 2007 Annual Survey”
(2007), available at: http://www.kff.org/insurance/7672/upload/EHBS-2007-Full-Report-PDF.pdf, last accessed Nov. 14, 2007.
3 Calculated from figures in 2007 Kaiser Family Foundation survey, especially Exhibit 2.7, p. 37.
4 The Kaiser Family Foundation and Health Research and Education Trust, “Employer Health Benefits.”
5 Human Rights Campaign, “The State of the Workplace for Gay, Lesbian, Bisexual and Transgender Americans, 2006-2007,”
available at: www.hrc.org, last accessed 7/25/07.
6 M. V. Lee Badgett, Money, Myths, and Change: The Economic Lives of Lesbians and Gay Men, (Chicago: University of Chi-
cago Press, 2001).
7 Michael A. Ash and M. V. Lee Badgett, “Separate and Unequal: The Effect of Unequal Access to Employment-Based Health
Insurance on Same-Sex and Different-Sex Couples,” Contemporary Economic Policy, Vol. 24, No. 4, October 2006, pp. 582-
599.
8 Gary J. Gates, “Geographic Trends Among Same-Sex Couples in the U.S. Census and the American Community Survey,”
The Williams Institute, UCLA School of Law, November 2007. Dr. Gates also provided unpublished counts of different-sex
couples in Figure 2.
9 Ibid.
10 Tavia Simmons and Martin O’Connell, “Married Couple and Unmarried-Partner Households,” Census 2000 Special Reports,
CENSR-5, February 2003, p. 9.
11 Ash and Badgett, “Separate and Unequal.” See also Julia E. Heck, Randall L. Sell, and Sherri S. Gorin, “Health Care Access
Among Individuals Involved in Same-Sex Relationships,” American Journal of Public Health, Vol. 96, No. 6, 2006, pp. 1111-
1118.
12 Ash and Badgett, “Separate and Unequal,” p. 596.
13 See, for example, IRS Private Letter Ruling 200108010, Nov. 17, 2000.
14 See IRC Section 152. In general, to qualify as a dependent of a taxpayer, a domestic partner would have to live all year in the
taxpayer’s household, receive more than half of his or her support from the taxpayer, and meet the IRS citizenship/residency
test. Furthermore, the relationship with the taxpayer cannot be “in violation of local law.”
15 Based on an average tax payment of $9,489 calculated for 2005 using Tax Policy Center data on Modified Taxable Income
for people filing as single, available at: www.taxpolicycenter.org/taxfacts/displayafact.cfm?Docid=381, last accessed Nov. 14,
2007. This value was inflated to 2007 dollars.
16 These estimates are adjusted for the fact that some partners would be considered dependents, as discussed in the appendix.
17 Office of Management and Budget, “Analytical Perspectives: Budget of the United States Government, Fiscal Year 2008,”
(Washington: U.S. Government Printing Office, 2007), p. 289.
18 Ash and Badgett, “Separate and Unequal,” p. 587.
19 Simmons and O’Connell, p. 2, Table 1.
20 IRS Private Letter Ruling 200108010, Nov. 17, 2000.
21 James M. Branscome and Beth Levin Crimmel, “Employer-Sponsored Single, Employee-Plus-One, and Family Health Insur-
ance Coverage: Selection and Cost, 2005,” Agency for Healthcare Research and Quality, MEPS Statistical Brief #175, July
2007.
22 The 2005 data are from the Internal Revenue Service. “SOI Tax Stats,” available at: http://www.irs.gov/pub/irs-soi/05in34tr.
xls, last accessed 11/13/07. For explanations, see Kyle Mudry and Justin Bryan, “Individual Income Tax Rates and Shares,
2004”, IRS, 2007.
23 Since 1983, approximately 94% of individuals have had earnings below the Social Security earnings threshold. Social Secu-
rity Administration, “Annual Statistical Supplement to the Social Security Bulletin, 2006,” SSA Publication No. 13-11700,
June 2007, Table 4.B4.

15
 Report made possible with a grant from

IMPORTANT NOTE: The author is not an employee of Merrill Lynch and is not affiliated
with Merrill Lynch, and neither the Williams Institute nor The Center for American Progress
are affiliated with Merrill Lynch. The opinions and conclusions expressed in this study are
not necessarily those of Merrill Lynch or its personnel. Merrill Lynch does not provide tax,
accounting or legal advice, and, to the extent that such advice is contained herein, you are
advised to consult with your personal tax, accounting or legal professional(s) with respect
thereto. Merrill Lynch does not provide tax, accounting or legal advice. Any tax statements
contained herein were not intended or written to be used, and cannot be used for the purpose
of avoiding U.S., federal, state or local tax penalties. Please consult your advisor as to any
tax, accounting or legal statements made herein.