The Economics of a Targeted Economic Development Subsidy

November 21, 2019

Many government officials consider targeted economic development subsidies key to economic development. In a recent survey of 110 mayors, for example, more than 8 out of 10 said targeted incentives are a good idea. In reality, economic development subsidies only help their corporate recipients and the politicians that supply them. Other companies, local residents, and the economy at large are harmed.

In “The Economics of a Targeted Economic Development Subsidy,” Matthew D. Mitchell, Michael D. Farren, Jer­emy Horpedahl, and Olivia Gonzalez provide a comprehensive analysis of the effects of economic development sub­sidies. Their estimates are based, in part, on the broad body of peer-reviewed academic research that finds that sub­sidies have little to no effect on where companies choose to invest. This means that the expected gross benefits of such subsidies should be substantially reduced. Furthermore, the authors incorporate the economic impact of the higher taxes needed to pay for the subsidies. They find that in the case of Wisconsin’s subsidies to Foxconn, the net effect of the subsidies will likely reduce future economic activity in Wisconsin by $370 million to $19.2 billion.

Foxconn in Wisconsin: A Case Study of Subsidy Failure

In 2017, Wisconsin struck a deal with Taiwanese company Foxconn to manufacture large LCD screens within the state. Foxconn was supposed to make a $10 billion investment and create up to 13,000 jobs. In return, Wisconsin would do the following:

  • Provide up to $3.6 billion to Foxconn in tax breaks and other subsidies
  • Exempt Foxconn from certain environmental regulations
  • Provide billions more in local government, utility, and infrastructure subsidies

Just two years after the deal, Foxconn is already reneging on its commitments and is building a much smaller $2 billion to $3 billion facility that will employ far fewer workers. This should be no surprise, given a recent Wisconsin state audit finding that, on average, subsidized firms create only 34 percent of the jobs they promise.

The Public Does Not Win With Economic Development Subsidies

Despite their promises, subsidies are bad for the communities that provide them. Subsidies cause economic harm in the following ways:

  • Subsidized companies are made less efficient. By allowing firms to shift costs onto taxpayers, subsidies allow firms to have higher production costs and to be less attentive to customer desires.
  • Entrepreneurs are encouraged to seek favors. Subsidies encourage entrepreneurs to develop new ways to obtain political privilege rather than new ways to lower costs or enhance consumer welfare.
  • Nonsubsidized companies are harmed. They are saddled with the tax cost of the subsidies given to their competitors.
  • Taxpayers foot the bill. Scarce public resources, which could otherwise fund public services or tax cuts for all, are instead wasted encouraging business decisions that would likely be made anyway. These taxes, in turn, discourage other economic activity.
  • Communities are put at risk. Subsidies can encourage overspecialization within a region, making commu­nities more vulnerable to economic downturns.

Key Takeaway

Economic subsidies rarely sway where a company chooses to invest. Instead, companies prefer locations that offer productive workers, efficient business logistics, and access to region-specific resources. Subsidies turn companies’ attention away from satisfying consumers, cost taxpayers billions of dollars, and generally don’t create the economic development they claim. Subsidies may harm the long-term health of the companies that receive them. And from a broader perspective, they are almost certainly harmful for economic development.

There's Nothing Natural about the State of Government Spending in Arkansas

August 8, 2017

Despite being one of the poorest states in the nation, Arkansas has a high state government spending level, even in comparison to similar states. Arkansas has higher state spending per capita than its bordering states and other regional “competitor states,” specifically Alabama, Florida, Georgia, North Carolina, South Carolina, and Virginia. When federal transfers are subtracted from state government spending, Arkansas still has far higher spending than its competitor states.

In “There’s Nothing Natural about the State of Government Spending in Arkansas,” Jeremy Horpedahl and Jacob Bundrick undertake a comprehensive analysis of how Arkansas got to where it is today as well as the reforms nec­essary to put Arkansas on a more sustainable path. Arkansas’s spending trends are very similar to those of other states, but the magnitude of Arkansas’s spending and the institutional environment that has shaped it are unique. Given Arkansas’s history of state spending growth, it would be wise to consider some lim­its to future increases in spending.


Arkansas’s spending increased by 97 percent from 1991 to 2013, even after adjusting for inflation and population growth. The state spends $7,674 per capita, 59 percent more than its competitor states, which spend an average of $4,815 per capita. Considering just own-source revenue, Arkansas is spending almost double what its competitor states are spending on average. The state government has not taken on many new roles since 1991, other than an expanded role in healthcare, so for most categories the increase is just more spending needed to fund the same services as in the past. The primary drivers of Arkansas state spending growth have all experienced cost-per-capita increases of around 100 percent or more in just the past two decades.

  • Education transfers to local governments. Education transfers have increased dramatically and constitute a relatively large portion of the budget. A series of changes that resulted from the Lake View court case took Arkansas’s education expenditures from $1,206 per capita in 1992 to $1,807 per capita in 2007.
  • Insurance trust benefits. Public pensions, the most significant item in this category, have not only grown more expensive over the years—they have also developed a significant funding problem. Arkansas’s pen­sion system has been assessed as 72 percent funded using state discount rates and as only 51 percent funded using fair-value discount rates.
  • Public safety. Public safety spending, most of which relates to corrections, has grown by more than 100 per­cent since 1991, much greater than the average of 31.5 percent across all 50 states.
  • Welfare spending. Owing to the state’s Medicaid expansion, welfare will continue to be the largest and one of the fastest-growing components of state spending in Arkansas.


Several of the current budgetary processes and informal institutions in Arkansas can be traced to Great Depres­sion–era budget issues. In 1933, Arkansas became the first and only US state in the 20th century to default on its debt. As Arkansas suffered financially, the state issued a series of laws and amendments to its constitution. In addition to the Futrell Amendments of the 1930s and the Lake View court case and related cases since the 1990s, the following leg­islative acts stand out as institutional factors that have shaped Arkansas’s spending.

The Revenue Stabilization Act of 1945

The Revenue Stabilization Act of 1945 established a process whereby all spending bills passed in a legislative ses­sion are prioritized by the legislature. Programs are funded only if there is enough revenue to pay for them, and programs ranked lower on the priority list may be only partially funded, or not funded at all. Although the act has successfully prevented budget deficits, it has made state spending very sensitive to changes in the tax system.

Removal of the Cap on Targeted Economic Development Incentives

Targeted economic development incentives have become a thorn in the side of Arkansas’s budget. The state uses these incentives to attract and retain business in an effort to boost the state’s economy, but empirical evidence reveals that targeted incentives are largely ineffective at stimulating the economy. In 2013 alone, tax incentives and subsidies cost the state of Arkansas more than $218.56 million—nearly $74 per capita. In 2016, voters approved a measure that removed the cap on issuing new state debt for large development projects.


  • Instituting a new and improved Revenue Stabilization Law. Arkansas’s Revenue Stabilization Law has served the state well by preventing budget deficits and allowing the state to forgo a large “rainy day fund,” but it could be improved. A modified version of the law would include a feature refunding taxpayers when tax growth outpaces inflation and population growth. Additionally, Arkansans could make the law a permanent feature of the budget process by amending the state constitution.
  • Improving the efficiency of education spending. Education is the largest and one of the fastest-growing spend­ing categories in Arkansas. The state spends more per pupil on education than neighboring states, yet its test scores generally lag behind those of these same neighbors. Rather than instituting spending restrictions, Arkansas could improve the efficiency of education spending by increasing school choice. Increasing the number of open-enrollment charter schools would increase competition with traditional public schools on the basis of quality, and this competition would improve educational outcomes.
  • Ending corporate welfare. Rather than funneling tax dollars toward a select few companies, Arkansas should create a more competitive tax environment. Reducing the number of corporate tax brackets, indexing corporate tax brackets for inflation, using the same corporate tax base as the federal income tax code, and removing antiquated manufacturing taxes are all steps that the state can take to simplify its tax code.

The Charitable Contributions Deduction

January 14, 2016

Taxpayers who make contributions to approved charitable organizations can deduct those contributions from their income before computing their tax liability. In fiscal year 2014, the deduction lowered taxes by $47 billion, with over 93 percent of the benefits going to tax filers under the individual income tax rather than the corporate tax. Most taxpayers would benefit from removing the deduction and lowering tax rates since most taxpayers do not use the deduction regularly. The only economic justification for the deduction would be to encourage donations to organizations that provide public goods or quasi-public goods. It is unclear, however, that the charitable deduction actually encourages private sector provision of these goods.

One of the primary purposes of government is to provide certain public goods that it is not possible for the private sector to provide. Providing public goods is one of the primary roles of government as defined by almost every economist from Adam Smith through the authors of modern public finance textbooks. But just because government should assist in providing certain public goods, it does not necessarily have to provide them directly. Government could finance the private provision of public goods, or simply encourage their provision in some way.

The deduction for charitable contributions is a form of this last option: government will let private philanthropic organizations provide some public goods, and it will encourage taxpayers to fund these organizations by offering a tax benefit rather than funding the public goods out of tax revenue. To be justified economically, this tax deduction must pass several tests. Two threshold tests are: Are the organizations actually providing public goods? Does the deduction actually increase charitable donations?

Are Charitable Organizations Providing Public Goods?

For the charitable deduction to be justified economically, the government cannot simply encourage more donations to nonprofit organizations. The government must encourage donations to organizations that are providing goods undersupplied by the normal market process—or, as economists refer to them, public goods. A public good has a precise definition in economics, which is that the good in question must be nonrival in consumption (at the margin, adding more users does not diminish the quantity available) and that it must be difficult to exclude non-payers from benefitting from the good (“difficult” meaning prohibitively expensive). In common language, the term “public good” is used in a less precise way, often to mean “something nice” or “something the government supplies,” but economists have in mind a clear definition that has a very high burden. I acknowledge that governments may have other goals in mind when the tax code rewards charitable contributions, such as encouraging charity generally even if there are no positive externalities, but such other goals are hard to justify using economic criteria.

For quasi-public goods, those with some elements of both private and public goods, the role for government is less clear than for pure public goods. Education and healthcare are two clear examples of quasi-public goods. While most of the benefits of consuming these services accrue to the individual consuming them, there are also clearly some external benefits to society. (For example, society benefits from educating voters and containing contagious diseases.) And as we will see below, education and healthcare are some of the most widely supplied services by public charities.

In the case of quasi-public goods, there is a real question of how government should be involved. To facilitate access to quasi-public goods that have primarily private benefits, one possible solution is for government to encourage private provision with subsidies, with government refraining from providing the goods directly. Encouraging private provision is certainly one of the purposes of the charitable deduction, though as I will discuss below, it is questionable whether the deduction has achieved this goal.

What activities are the charitable organizations engaged in? Economists Cyril Chang and Howard Tuckman examined tax-exempt organizations in the United States. They found that the majority of them provide a mixture of private and public goods, which may call into question whether they should be subsidized at all (or whether the subsidy should be based on the outputs they produce). The table below uses IRS data compiled by the Urban Institute on 501(c)(3) public charities, the main group of organizations that the charitable deduction applies to. It shows total revenue for the sectors that these charities fall into. The data are for total revenue of these organizations, not just donations, as many of these organizations obtain a significant amount of revenue from other sources (e.g., charging for services provided).

While the list of charity categories is long, the two largest—health and education—have nearly 73 percent of the revenues among public charities. These two categories certainly produce a mix of private and public goods, but arguably most of their output is private goods. As discussed above in this paper, the charitable deduction is often defended as a way of encouraging the private provision of the quasi-public goods.

Has the deduction been effective in achieving the goal of private provision? There is no easy way to answer this question, but a cursory look at health expenditure data suggests that the charitable deduction has not been powerful enough to encourage health spending to stay in the private sector. In 1960 about 70 percent of health spending was from private sources and 20 percent was from government sources, while in 2013 about 49 percent was private and 45 percent was government (the balance in each case is from categories that are a mix of private and public sources). Even starting from a later date, such as 1970, after Medicare and Medicaid had been established, private health spending was still about 56 percent and government spending was only about 33 percent.

It appears that, at the very least, the charitable deduction has not stopped the funding of quasi-public goods in healthcare from drifting into the public sector. On the other hand, the United States does still have much more private provision of education and healthcare compared to other industrialized countries.

Most other categories in the table also provide some mix of private and public goods, but there are only a few on the list that likely provide mostly public goods: national security, science and technology research, medical research, environment, crime, social science research, civil rights, and public safety. But these categories combined only represent about 5 percent of the total public charity revenues, meaning that 95 percent of the charities, based on revenue, are not clearly providing mostly public goods. Thus, the benefits to society are not clear from this tax deduction, and many of the benefits of charitable activities flow to private groups and individuals.

Does the Deduction Increase Charitable Donations?

The first important test for justifying the charitable deduction is whether or not it increases donations to charity beyond what they would be without the deduction. Ideally, a justification would also provide a benefit-cost analysis of the increase in donations since this implies that marginal tax rates must be higher than otherwise (and thus lower economic growth), but I am not aware of any studies that provide a direct comparison.

Several studies do provide a kind of benefit-cost analysis on this issue, but the “cost” they use is forgone tax revenue to the government. Nearly all of the studies in this literature find that there is some increase in charitable giving from its current tax treatment. However, one study by William C. Randolph casts doubt on the claim that the deduction increases giving in the long run. Randolph’s paper analyzes both major tax reforms in the 1980s and follows individuals for 10 years, finding that taxpayers alter the timing of their giving in response to changes in tax policy, but not necessarily the total amount of giving.

Recent experimental research attempts to answer similar questions in a slightly different venue. For example, Dean Karlan and John List find that offering to match charitable contributions ($1:$1) does increase both the size of the gift (by 19 percent) and the probability of donating (by 22 percent). But they also find that larger matches ($2:$1 and $3:$1) do not further increase an individual’s donations beyond the $1:$1 match. The current match in the federal tax code is, of course, much less than $1:$1, being determined instead by the taxpayer’s marginal tax rate. But one possible application of this result to the charitable contributions deduction is that the government may be able to make the deduction less generous but still elicit roughly the same amount in contributions.

Distribution of Benefits Across Income Groups and Suggestions for Reform

The benefits of this tax deduction skew largely to upper-income earners, as with most tax expenditures. There are three primary reasons for this, among many others: higher-income earners are more likely to itemize deductions; higher-income earners have a larger tax burden as a share of their income; and the progressive tax schedule means higher-income earners have a stronger incentive to reduce their tax burden. For the lowest-income households, those with annual income under $20,000, only 0.3 percent claim the charitable deduction. By contrast, for households with over $100,000 in annual income, almost 59 percent claim the deduction in a given year—and for those over $200,000, it is still higher at over 72 percent. For households closer to the middle of the income distribution (between $40,000 and $75,000), around 20 percent claim the charitable deduction.

There is, however, an interesting twist to the numbers for this deduction, because lower-income households also donate to charities in large numbers. Using data from the Panel Survey of Income Dynamics, economist John List shows that, first, the relationship between income and propensity to give to charity is as we would expect. Over 90 percent of higher-income households (above $100,000 in annual income) give to charity, but only 37 percent of lower-income households (under $20,000) do so. A majority of middle-income households (between $40,000 and $75,000), over 70 percent, also donate to charity. But looking only at the households that donate to charity, a reverse pattern occurs: lower-income households give a large percent of their household income (12 percent) to charity, and this declines as income increases to just 2 percent for the highest-income group (over $130,000). Thus, it is not that lower-income households do not donate to charity: many do, and on average they give a substantial portion of their income. However, very few of them benefit in terms of their tax burden, because many lower-income households have no positive tax liability.

For middle-income households the disparity is most evident: over 70 percent donate to charity, but only 20 percent can claim the tax deduction. And for those that do claim the deduction, the average tax benefits are quite small: $229 for filers in the $40,000–$50,000 group and $348 for filers in the $50,000–$75,000 group. These small dollar amounts are no doubt important to households receiving the benefit, but if the deduction were eliminated, they could easily be offset by a small reduction in marginal tax rates.

For the 80 percent of middle-income filers who do not currently claim the charitable deduction, any cut in marginal tax rates is a pure benefit. Most taxpayers would be better served by eliminating the charitable contributions deduction and using the additional revenue to lower tax rates. 

The Child Tax Credit

May 8, 2015

First introduced in 1997, the Child Tax Credit (CTC) has grown in size and eligibility to become one of the single largest tax credits available to middle-class families.1 It provides qualifying taxpayers with a tax credit of up to $1,000 per eligible child under the age of 17. The credit currently provides over $57 billion in benefits to taxpayers through both its nonrefundable and refundable parts.2 This $57 billion figure is comparable to the deduction for state and local taxes ($56.5 billion)3 and the Earned Income Tax Credit (EITC), which provides a benefit of over $69 billion to taxpayers.4 It is also close to the more well-known Supplemental Nutrition Assistance Program (SNAP, sometimes described as “food stamps”), which provided $76 billion in benefits in 2013.5 

Unlike SNAP, the CTC is not primarily targeted at low-income families. In fact, as I describe in the next section, very low-income families are either excluded from the credit completely or receive a reduced credit. Instead, the CTC is a broad-based credit available to families earning at least $3,000, and begins to phase out for incomes over $110,000. Of all taxpayers, around 22 percent receive the credit, but for most income groups, around one-quarter receive it. Figure 1 shows the percentage of taxpayers in each income category receiving the CTC. 


For taxpayers with annual incomes of $3,000 or less, no credit is available.6 Above $3,000 of income, the CTC is phased in at a rate of 15 percent. The basic credit is limited by the amount of the taxpayer’s tax liability, but there is also a refundable portion of the credit, limited to 15 percent of the taxpayer’s income above $3,000. 

Thus, for a family with one qualifying child, the $1,000 credit is fully phased in at $10,000 of income. For a family with two children, the $2,000 credit is fully phased in at $16,400 of income. The credit begins to phase out over $110,000 for married joint filers ($75,000 for head of household) at a rate of 5 percent, meaning that $50 of the credit is lost for every additional $1,000 in income. For a married couple with two children, the credit goes to zero once they reach an annual income of $150,000. 

There is no limit to the number of children a taxpayer can claim, but the credit is effectively limited because it phases out based on tax liability and income. For example, for a family (married filing jointly) with income 

of $50,000, there is essentially no tax benefit beyond seven children. For a family with $25,000 in income, the fourth child only adds a $300 credit, and there is no tax benefit beyond the fourth child. 


The CTC is clearly a benefit to a large number of working families, so removing the credit without some offsetting decrease in tax rates would likely be politically unpopular. But is the credit justified as a matter of public policy? The tax code already has a number of provisions that lower the tax burden on families with children. 

For example, the personal exemption is based on family size, thus for most families, taxable income decreases as a family has more children (and consequently the tax they owe decreases). The EITC is also very small unless a family has children, and it gets more generous with each child up to the third child. And the federal income tax code is not the only place where tax policy subsidizes families with children. A large share of state and local taxes goes to fund local public schools. The tax code has many other features that ease the financial burden of raising children. These provisions also potentially increase the number of children, as one study of a tax credit in Quebec suggests that for every Can$1,000 increase in tax credits, the rate of childbirth increases by 16.9 percent.7 

Defenders of lowering the tax burden for families with children often offer two kinds of justifications. First, families with children have greater expenses than families without children, thus they have a lower “ability to pay” taxes. However, if having children is a choice that families make, there is little economic justification for subsidizing the choice to incur more expenses, as much as there would be for incurring any other expenses.8 

Instead, subsidizing children through the tax code is only economically justified if an increase in children leads to external benefits to society. This is the second justification offered for lowering the tax burden for families. In order to justify lowering the burden for families, these external benefits to society must be large enough relative to the private benefits that families would gain from their choice of having fewer children.9 For example, children will eventually contribute to Social Security, which is a benefit that is available whether one has children or not. The contributions of these children to a program that benefits even those without children might justify tax subsidies that encourage families to have children. 

It is extremely difficult, perhaps impossible, to measure whether families are having the “right” number of children from a social perspective. But it is much easier to show that the US federal tax code has greatly expanded the subsidy to families with children since the CTC was introduced. 

Table 1 shows the percent of income paid in federal income taxes in 1997 and 2013 for three different income levels and filing statuses, and for different numbers of children. First, in both years and for all three family incomes, there is a tax subsidy for having more children. But the table also shows that the tax subsidy is much bigger in 2013 than in 1997, the last tax year before the availability of the CTC. 

The first family in the table, a married couple earning the median household income, saved about 1 percentage point of their income in the form of lower taxes per child in 1997 (more concretely, just under $400). In 2013, the savings per child jumped to over 3 percentage points, or almost $1,600. Of this amount, $1,000 is due to the CTC, which is most of the $1,200 difference in nominal terms (and virtu- ally the entire difference in real terms, as $400 in 1997 is worth almost $600 in 2013). For the other two family types shown in the table, the tax subsidy for children grew even larger in percentage point terms between 1997 and 2013, as these families also qualify for the EITC. 


The CTC provides a significant subsidy to almost all tax- paying families with children, and the US federal and local tax codes contain many other provisions that subsidize child rearing. In the aggregate, the CTC subsidy to families with children has grown to nearly $60 billion, placing it among the list of the largest “tax expenditures” as defined by Congress’s Joint Committee on Taxation. 

This is a topic that requires much more scrutiny, primarily to determine if, given other policies subsidizing child rearing, there is a social need for $1,000 more per child. One alternative is to remove the subsidy and lower tax rates by an offsetting amount. This tax cut would benefit all taxpayers, not just parents. It would also provide social benefits in terms of increased productivity and economic growth, and this is the most relevant comparison to any social benefits from additional children. 

Principles of a Privilege-Free Tax System, with Applications to the State of Nebraska

September 16, 2014

Nebraska is one of several states considering ways to make its tax code more efficient and fair. Every state’s code includes special privileges for politically favored industries, businesses, or groups of people, which typically require higher taxes on other economic activity.

As economist Jeremy Horpedahl demonstrates, these tax favors have real costs for working taxpayers. In a new Mercatus Center at George Mason University study, he defines government-granted tax privileges and outlines key principles for reforming sales, property, and income taxes, and economic development tax incentives.

The basic principle of a privilege-free tax system is simple: No individual or business should be treated differently from any other similarly situated individual or business. After removing privileges, policymakers should next lower tax rates across the board, in a way that applies equally to all taxpayers. 

To read the paper in its entirety, click here. See below for the author's case study on Nebraska, and for general principles all states can apply.

Case study: Nebraska

If Nebraska eliminated tax privileges and used the resulting revenue to lower tax rates, the average family in the state would save over $3,200 dollars, with no reduction in government services or spending.

  • Nebraska doled out over $2 billion in economic favors through its tax code in 2012, the latest year for which full data is available.
    • These privileges equal about ten percent of Nebraska’s total 2011 revenue, and about 29 percent of its total sales, income, and property tax collections.
  • Sales tax favors accounted for $654.8 million of this figure—equal to about 27 percent of total 2011 sales tax collections.
    • Horpedahl recommends that Nebraska begin taxing services, which should not be distinguished from physical items for economic purposes.
    • Tax exemptions on specific goods including fuel, groceries and motor vehicle trade-ins should be removed (the benefits these provide to the poor could be offset in other ways).
    • Finally, taxes on intermediate business-to-business transactions should be eliminated, because these amount to double taxation after business-to-consumer transactions are taxed.
  • Property tax favors accounted for $475.6 million.
    • These privileges take three main forms: the Homestead Exemption, Property Tax Relief Credit, and the 25% tax break for agricultural and horticultural land, which is treated differently than other business real estate.
  • Income tax favors accounted for $841.5 million.
    • Horpedahl recommends removing itemized deductions that don’t apply equally to all Nebraskans, such as the Home Mortgage Interest Deduction.
    • Expanding the state’s Earned Income Tax Credit could offset the costs of other tax reforms for low-income Nebraskans.
  • Tax credits for economic development accounted for $123.3 million.
    • These incentives are so generous that Nebraska has the lowest effective tax rate for new businesses, but only the ninth-lowest for established businesses.
    • Companies earned a total of $332 million in tax credits (not all of which have been spent yet). Together the used and unused credits equal 71 percent of the corporate taxes collected by the state since 2006.

Principles of a privilege-free tax system for all states

The author estimates that nationwide, state tax privileges cost American taxpayers $432.5 billion annually—about 40 percent of the roughly $1 trillion they spend on federal tax expenditures.

 Sales Taxes

  • Intermediate business-to-business transactions should be exempt from taxation because different industries have different production processes which require varying amounts of transactions. Levying sales taxes only on final consumer purchases avoids this problem.
  • There is generally no sound economic reason for treating services differently from goods. As the economy becomes increasingly service-based, state sales tax revenues as a percent of the economy will tend to decline, requiring higher taxes elsewhere to fund a given level of government spending.
  • Exempting basic necessities like groceries from taxation, while well-intentioned, is not an effective method for assisting those in poverty. Better options include making Earned Income Tax Credits more generous.

 Property Taxes

  • The primary form of economic privilege found in the property tax is the practice of taxing property at different rates (or assessing it differently) based on how the property is used, such as differentiating between personal, business, and agricultural properties. It is somewhat strange to single out agricultural property as being distinct from business property.
  • While a privilege-free property tax is ideal, a policy of property tax exemptions for low-income taxpayers would be preferable to current system.

Income Taxes

  • Economic privilege in the income tax code arises when individuals or corporations are able to lower their tax burden by engaging in certain activities. Even if it does have some benefits for society, the costs must also be considered.
  • One major example of privilege in federal and state income tax codes is the home mortgage interest deduction. This is the largest explicit form of tax privilege offered to individuals at the federal level, yet only about one-fifth of taxpayers take this deduction. The benefit for middle-class families (those with $30,000–$75,000 in annual income) is on average between $100 and $200. Meanwhile, about three-quarters of high-income taxpayers with more than $200,000 in annual income take the deduction, and receive almost $1,800 on average.

 Economic Development Tax Incentives

  • Tax incentives intended to spur economic development through job creation and increased economic activity are often justified by politicians on efficiency grounds. However, upon close examination, this is frequently overstated.
  • Many state governments offer businesses that agree to relocate to their state the right to pay lower tax rates, or avoid one or more taxes, typically for a limited time. Most recent research suggests that, at least on average, these tax incentives do not pay off for local and state governments.
  • Not only do bureaucrats lack the information to pick which companies will succeed, they may also lack the incentive due to the lobbying of potential beneficiaries. Thus, states may find it in their interest to avoid all such development-based tax incentives in a privilege-free state tax code.

The Deduction of State and Local Taxes from Federal Income Taxes

March 6, 2014

Taxpayers who itemize their deductions are allowed to deduct state and local taxes from their federal taxable income. This deduction is limited to either income or sales taxes, but not both. Personal property taxes, such as local taxes on housing and real estate, can also be deducted. The main benefit of this deduction is to provide some tax relief to taxpayers living in states with higher taxes, since they have less disposable income. However, this benefit also points to the main cost of the deduction: it subsidizes higher taxes and spending at the local level, since taxpayers in those states will not feel the full burden of the taxes. Instead, the burden of these taxes is in some sense “exported” to taxpayers in other states, since federal tax rates must be higher than otherwise to fund the same level of federal spending. 

When discussing taxation in the United States, it is important to consider all levels of taxation rather than just focusing on one at a time, such as federal income taxes, because these tax levels affect each other. One way in which these taxes interact is found in the federal income tax code, whereby taxpayers who itemize deductions are able to deduct a variety of state and local taxes when calculating taxable income. In most years it is one of the five largest tax expenditures in the individual income tax, and it is thus one of the largest “tax expenditures” as defined by government agencies, such as the Office of Management and Budget.1 OMB estimates that in fiscal year 2012, this deduction reduced federal tax revenue by about $45 billion, and this amount will roughly double over the next five years. As a result, this deduction has had a large impact upon the overall tax system of the nation at all levels of government, which warrants further analysis of its overall desirability. 

Two Concerns For Equity

The distribution of the benefits of this tax expenditure, as well as the costs of removing it, can be thought of in two ways. First, there is the question of which taxpayers within the distribution of income benefit the most. According to estimates by the Joint Committee on Taxation for 2012, almost 95 percent of the benefits of this deduction go those earning over $75,000 per year, and over half (55 percent) of the benefits are for those earning over $200,000. The average deduction for those in the over $200,000 income group is over $5,000, while it is only about $250 for those in the $50,000 to $75,000 group (around the national median).2 These averages only include those taxpayers that claimed the deduction, which is only 27 percent of all taxpayers since most do not itemize their deductions. 

Second, there is the distribution of benefits across the 50 states, based on how high taxes are in each state. States with higher taxes will have more filers claiming this deduction, and therefore the deduction will also be a larger share of the state’s income. These high-tax states also tend to be high-income states, implying that there is a transfer from low-income to high-income states through this deduction. By looking at the total deductions in this category as a percent of adjusted gross income, we can see that there is wide variation across the states. The states with the lowest state and local tax deductions (Alaska, Wyoming, South Dakota) claim deductions amounting to less than two percent of their adjusted gross income, while the highest tax states (New York and New Jersey) claim over nine percent of their adjusted gross income. The figure below plots the correlation between the total deduction for state and local taxes and per capita adjusted gross income for 2010 for all 50 states and DC.3

Redistribution Or Offset For Higher Taxes?

During the 1985 tax debate, NYU law professors Brookes Billman and Noel Cunningham offered an economic justification for the deduction: state and local taxes reduce an individual’s income and ability to pay federal taxes, which should then be considered by the federal tax code.4 After all, if federal taxes should be based on an individual’s ability to pay, then local taxes could reasonably be considered as reducing one’s ability to pay federal taxes.

Economist Bruce Bartlett took a contrary position to Billman and Cunningham, arguing that this deduction is a subsidy to high-tax states from low-tax states, and high-tax states tend to have higher per capita incomes. He also found that, in general, the deduction is associated with higher state and local taxes because the federal government is paying a portion of these taxes, with most estimates suggesting state and local taxes are about 13 to 14 percent higher.5 In this case, more services may be provided publicly, even if it is more efficient to provide them privately. This deduction also influences the types of taxes that state and local governments use, biasing them toward choosing taxes that are deductible rather than those that are most efficient.6 More recent estimates confirm that state and local spending “could fall in the absence of deductibility,” indicating that the deduction does indeed increase government spending.7

Furthermore, the ability-to-pay reasoning of Billman and Cunningham overlooks where state and local taxes go, since taxes are not only collected, but spent. Unless one views all government spending as complete waste, it is a reasonable assumption that the local taxes are providing some services to those that pay them (though likely not equal to the full value of taxes paid in many cases). Thus, local taxes don’t reduce an individual’s willingness to pay by the full amount of the tax, or possibly even at all. The local taxes an individual pays are returned to the individual as local government services, minus the costs and wastes associated with government provision of services. 

For example, in cities where garbage collection is provided by the local government, the property or sales taxes that fund this service are deductible. For cities where garbage collection is provided privately, the fees paid to the garbage company are not tax deductible. But from an economic perspective, these two situations are identical and should be treated the same in the tax code. The current tax code gives preference to city-provided garbage collection, which potentially violates both equity and efficiency. The tax code also biases municipalities towards providing services, such as garbage collection, even though it may be more efficient for these services to be provided privately.

Tax Rates And Macroeconomics Effects

Eliminating the deduction for state and local taxes will likely lead to increased macroeconomic activity by removing the economic distortions listed above. It will also potentially lead to more federal tax revenue, though this depends on how individuals react to what would be, in effect, a tax increase. A better policy would be to simultaneously decrease tax rates at the same time that this deduction is eliminated, generating additional economic activity without increasing the amount of revenue concentrated at the federal government.

Estimating the economic and tax-revenue effects is a difficult matter. Two recent studies by the Tax Foundation attempt to examine the effects of removing the deduction for state and local income, sales, and property taxes. These estimates should not be taken as definitive point estimates, but as indicative of the direction of the effects. For the state and local income and sales tax deduction, they estimate that eliminating the deduction combined with an across-the-board cut in individual income tax rates by 5.8 percent (i.e., the 10 percent rate would drop to 9.42 percent) would result in an increase in total employment by around 300,000 jobs.8 They found similar effects for the property tax deduction, though this would also require a decrease in business property taxes because “capital (even owner-occupied housing) is quite sensitive to taxes.” 9


Removing the federal tax deduction for state and local taxes would make taxes more equitable throughout the nation, as both high-tax and low-tax states are treated equally by the federal government. It may also provide an efficiency boost for states and localities, as they abandon some services that could be better provided by private companies. The removal of this deduction would also allow federal marginal tax rates to be cut across the board, providing a secondary boost to the economy while still remaining revenue-neutral at the federal level.


  1. For a discussion of the largest individual and corporate tax expenditures, see Jeremy Horpedahl and Brandon Pizzola, “A Trillion Little Subsidies: The Economic Impact of Tax Expenditures in the Federal Income Tax Code” (Mercatus Research, Mercatus Center at George Mason University, Arlington, VA, October 25, 2012). 
  2. See Table 3 of Joint Committee on Taxation, Estimates of Federal Tax Expenditures for Fiscal Years 2012–2017 (Washington, DC, February 1, 2013), JCS-1-13. The JCT estimates do not provide any further details on the about $200,000 group, but due to AMT and phase-outs of deductions, it is likely that the benefits are not flowing to taxpayers with extremely high incomes. 
  3. Deduction totals include property taxes and either income or sales taxes. Tax data from Internal Revenue Service, SOI Tax Stats—Historical Table 2, updated April 23, 2013, /SOI-Tax-Stats---Historic-Table-2. 
  4. Brookes D. Billman Jr. and Noel B. Cunningham, “Nonbusiness State and Local Taxes: The Case for Deductibility,” Tax Notes 28 (September 2, 1985): 1105–1120. 
  5. Bruce Bartlett, “The Case for Eliminating Deductibility of State and Local Taxes,” Tax Notes 28 (September 2, 1985): 1121–25. 
  6. Martin Feldstein and Gilbert Metcalf, “The Effect of Federal Tax Deductibility on State and Local Taxes and Spending,” Journal of Political Economy (1987): 710–36. 
  7. Gilbert Metcalf, “Assessing the Federal Deduction for State and Local Tax Payments,” National Tax Journal 64 (June 2011): 565–590. 
  8. Michael Schuyler and Stephen J. Entin, “Case Study #4: The Deduction of State and Local Income Taxes or General Sales Taxes,” Tax Foundation Fiscal Fact, no. 382 (August 2013). 
  9. Stephen J. Entin and Michael Schuyler, “Case Study #2: Property Tax Deduction for Owner-Occupied Housing,” Tax Foundation Fiscal Fact, no. 380 (July 2013).

The Tax Exclusion for Retirement and Pension Plans

September 17, 2013

The US federal tax code contains a number of provisions designed to encourage individuals to save for retirement. These provisions allow individuals to avoid or defer taxes if they choose to set aside a portion of their income for future consumption. When all of these provisions are combined, they are the second largest “tax expenditure” category as defined by the Joint Committee on Taxation. The exclusion of retirement savings from taxation causes some economic distortions, which we will discuss in this paper. However, unlike some other tax expenditures, there is a strong economic rationale for not taxing savings. Higher rates of investment lead to higher rates of economic growth, and it may be sound policy for the tax code to encourage this behavior, even after considering the economic costs. Excluding retirement income from taxation may also make the tax system more efficient, even though most other tax expenditures reduce efficiency. 

When an employer chooses to compensate employees with contributions to a retirement or pension plan, rather than with wages, that compensation is not taxed in the current year. Instead, the income will be taxed in the future when employees choose to withdraw it, presumably when they are in a lower tax bracket. Similarly, investment income in tax-protected plans, such as dividends and capital gains, is not taxed until it is withdrawn. 

Traditional employer-sponsored, defined-benefit pensions were the first major plans of this type to be excluded from taxable income, but over the years many other similar kinds of retirement savings have also achieved tax exclusion. Important additions were Keogh plans for the self-employed, contributions to Individual Retirement Arrangements or Accounts (IRAs) beyond employer-sponsored plans, and defined-contribution plans set up by the employer, such as 401(k)s. While these programs have important technical differences, the basic economic function is the same: contributions are made with pre-tax income and grow tax free, and the tax is paid in the future when withdrawals are made. The more recent Roth IRA operates differently from the rest, as it is funded with post-tax dollars and only the gains are tax free, but the intended economic effect of encouraging retirement savings is the same.

These exemptions result in a loss of revenue for the federal government. The Joint Committee on Taxation estimates that in FY 2013, about $117.2 billion in income tax revenue was not collected from the “net exclusion of pension contributions and earnings,” and the Congressional Budget Office has a slightly higher estimate of $137 billion once the forgone payroll tax revenue is included.1 While much of that tax revenue is simply deferred, rather than avoided, the lost payroll tax revenue (i.e., Social Security and Medicare taxes) which employers would have paid on wages is completely forgone. These estimates place the exclusion of retirement savings as the second largest tax expenditure, behind only the tax exclusion of employer-provided health insurance.2

Do Tax Incentive Increase Savings? 

A primary question on the tax exclusion of retirement savings is whether they encourage individuals to save more than they otherwise would. The primary economic benefits associated with this tax exclusion can only be achieved if savings increase on net. The macroeconomic benefit of an increase in net savings is greater long-run economic growth from more capital accumulation. The potential benefit to individuals is the long-run increase in savings if, for behavioral reasons, they will save too little from their own perspective.3 While the same result might be achieved by mandating more saving for retirement, perhaps by increasing Social Security taxation and benefits, the tax exemption may be a more attractive policy because it does not involve direct taking and giving but merely encourages citizens to provide for their future retirement.

For there to be an increase in net saving, savers actually have to decrease their current consumption and standard of living in lieu of future consumption and standard of living. Because of this condition, the incentives provided by the tax deferral may not encourage genuine savings. Instead, it may merely encourage deposits and contributions into the account in ways that don’t require reducing one’s present standard of living. Some individuals probably would have saved for retirement even without the tax incentive; thus, looking at the aggregate amount deposited in these accounts is misleading. As a result, we need to investigate how much savings increased as a specific result of the tax benefit. The Journal of Economic Perspectives devoted a symposium to this question with contributions from the leading scholars in this debate. While the empirical evidence is mixed, there does seem to be strong evidence that there is some net increase in savings from tax incentives, even if the magnitude is debated.

Other Tax Exclusion Concerns

While it is important to know whether a tax exclusion has a positive effect on savings, that alone is not enough to justify a tax policy. The costs of the economic distortions introduced by the policy must also be considered. One cost may be that individuals put their savings in forms that are different from what they would choose independent of these incentives. Individuals with savings in the form of 401(k) accounts have much less freedom to choose their investments than those with savings in traditional brokerage accounts, and those with traditional defined benefit pensions have essentially no choice in how their assets are invested. This could lead to serious principle-agent problems between employers and employees, with employers or their chosen brokerages not making the best investment decisions from the perspective of the employees. 

Another concern is that most of the benefits of the tax treatment of retirement savings accrue to those with the highest incomes. Toder, Harris, and Lim of the Tax Policy Center estimate that about 80 percent of the benefits for tax incentives for retirement savings go to the top income quintile.5 Those in the top income quintile almost always benefit the most from tax expenditures, largely due to the fact that they pay the most taxes; however, the 80 percent benefit for this category of tax expenditures is higher than other major categories, such as the mortgage interest deduction (about 68 percent goes to the top quintile) and healthcare-related tax expenditures (about 42 percent goes to the top quintile).

Can a Loophole Make the Tax System More Efficient? 

While tax expenditures or “loopholes” are generally regarded as making the tax system less efficient overall, the exemption for retirement savings may be an exception. In fact, this exemption may make the tax system more efficient both by making the current system function more like a consumption tax and by partially correcting the double taxation of capital within the current tax code. 

Since individuals can choose when they wish to realize the tax by delaying consumption, this may very well be a desirable feature of tax incentives for retirement savings. A tax on consumption is generally more economically efficient than one on income since it does not discourage production.6 The Congressional Budget Office even admits that it may not make sense to count this category as a tax expenditure: “because a consumption tax would exclude all savings and investment income from taxation, the exclusion of net pension contributions and earnings would be considered part of the normal tax system and not a tax expenditure.” 7

However, this desirable feature is only obtained by adding yet another layer of complexity to the income tax. If a consumption tax is what is desired, then proponents should make their goal changing the income tax into a consumption tax rather than creating unnecessary complexity within the system that we have. Nevertheless, we may consider this layer of additional complexity a second-best solution in a world where most federal revenue is still derived from taxes on income.

A second efficiency benefit of this tax exemption is that it serves as a partial correction for the current double taxation of capital income in the United States. With the highest corporate tax rate in the developed world, the United States must be particularly attentive to the impact of its tax system’s effect on capital formation.9 Currently, capital income is taxed when corporations earn income, and the same income is taxed again when it is paid to individuals in the form of dividends or capital gains. The exclusion of taxes on retirement income, specifically dividends and capital gains, means that the double taxation is eliminated for some capital income. In the absence of a corporate income tax, this tax exemption may make less economic sense, but given current corporate taxes in the United States, it has a sound economic logic.


In the end, the federal tax code allows for a deferral of income tax on contributions to a retirement or pension plan, which provides an incentive for people to save for their retirement. People choose to save more because they probably will be in a lower-income tax bracket during their retirement and also because they will be able to accrue the benefits of invested funds that would have otherwise been taxed away. Despite the complexity that this deduction adds to the tax code and some economic costs, this deferral brings the income-tax function closer to a more economically efficient consumption tax. It also mitigates the problem of the double taxation of capital for the tax-deferred contributions. 

The Tax Exemption of Employer-Provided Health Insurance

September 10, 2013

The Federal government does not tax health insurance when employers provide it to their employees as part of a compensation package. This tax expenditure is the largest “loophole” in the federal tax code, resulting in nearly $300 billion in forgone revenue in Fiscal Year 2012, according to the Office of Management and Budget. Even worse for the economy, the tax exemption for employer-provided health insurance creates significant distortions in the labor and health insurance markets. 

A new study from the Mercatus Center at George Mason University looks at the major unintended consequences of the tax exemption of employer-provided health insurance and the problems created by its market distortions. The study concludes that taxing health insurance and simultaneously lowering marginal tax rates would better serve most Americans and the overall economy.

Key Findings 

Many of the United States’ current health-care-related problems—from lack of choice and competition to rising costs—stem in part from the tax exemption for employer-provided health insurance. 

Eliminating the tax exemption cannot be a stand-alone change; the resulting tax increase on all working Americans should be offset with an equivalent reduction in tax rates.

  • This would eliminate many distortions to the health insurance market without a net increase in taxes.
  • This combination of reforms would also yield higher economic growth and a more flexible labor market, which in turn would increase tax revenue in the long run.

Eliminating the tax exemption would allow employers to increase employees’ pay by the same amount (about $12,000 annually, on average) that they currently spend on health insurance. 

  • Workers could use this extra income to shop for health insurance that best meets their individual needs.  
  • It would eliminate “job lock,” where employees feel “locked in” to their current job because they will lose their health insurance if they quit to search for a new job, by allowing workers to keep their health insurance if they change jobs.
  • It would also likely leave workers with more money, even after buying insurance.


Among the greatest economic distortions created by the tax exemption for employee-provided health care: restricting workers’ freedom to leave one job to find another better suited to their capabilities; limiting workers’ ability to choose the health insurance that best suits their needs; and limiting competition and driving up costs in the health insurance market. 

Labor Market Distortions

Job Lock. Americans fear how losing or changing jobs will affect their health care and are thus less likely to leave one job to search for another that better suits their skills. Negative effects include: The ultimate result is lower returns for investors, including retirees.

  • Employers have more leverage over employees; voluntary employee turnover is reduced by 25 percent in businesses providing health insurance, one study found.
  • Small businesses are at a competitive disadvantage in recruiting labor, as they typically must pay more for insurance benefits, and therefore are less likely to provide health insurance than large firms. 
  • Businesses and workers are less efficient, as resources are unable to flow to their most productive outlets.

Higher Health Care Demand, Higher Costs

 The tax exemption for employer-provided health insurance has increased the number of employers that provide health insurance; it has also increased the amount of health coverage purchased.

  • This increase in demand for health insurance and health care has contributed to the rise in health care prices in recent decades and further increases the job lock effect.

Less Choice, Less Competition

American workers have little direct involvement in decisions about purchasing their health insurance. Instead, large employers, representing large segments of the population, purchase health insurance for employees.

  • This reduces health insurance choices for all Americans and introduces equity concerns about who makes important economic decisions. 
  • It also makes the health insurance market less competitive.


Reform the tax code to eliminate the tax exemption for employer-provided health insurance. Offset the significant tax increase that results from closing this loophole with an equivalent decrease in marginal rates.

  • To make the elimination of the exemption tax neutral, marginal tax rates would need to be lowered significantly. 
  • For a hypothetical family making $75,000 per year (roughly the median household income for a family of four) and currently in the 10- and 15-percent brackets (see table in full study), marginal rates would need to be lowered to about 7 percent and 12 percent, respectively, to keep the level of tax revenue the same.
  • Similarly, the payroll tax rate would be applied to a larger base of income and, therefore, should also be lowered from 7.7 percent to 6.7 percent to raise the same revenue.

Interactions with the Affordable Care Act

This study’s proposed reforms should be beneficial regardless of the future status and outcome of the Affordable Care Act (ACA).

  • If the ACA remains law: the stated goals of the ACA include increasing transparency and controlling costs in the health care market, both of which are furthered by the reforms proposed in this study. 
  • If the ACA is repealed: the study’s proposals will be beneficial for correcting labor- and health care-market distortions. 

    The Home Mortgage Interest Deduction

    January 8, 2013

    The home mortgage interest deduction is the largest explicit tax deduction for households in the federal income tax code. Politicians have been reluctant to even consider removing this deduction, believing it to be one that provides significant benefits to middle-class taxpayers and encourages homeownership. These benefits are greatly over- stated: most taxpayers do not benefit from this deduction at all or receive a very small benefit. The only taxpayers who do receive a large benefit are those in the upper income brackets. Taxpayers and the entire economy would be better served by removing the mortgage interest deduction and lowering marginal tax rates to offset the change.

    The mortgage interest deduction is a tax deduction by which the federal government allows taxpayers who own their homes to lower their taxable income by the amount of interest paid on loans for a principal residence or a second home. As such, the mortgage interest deduction, like other deductions in the tax code, allows taxpayers to subtract the costs of certain items of consumption from their taxable income—thereby providing an implicit subsidy (i.e., a lower tax bill). The purpose of such a deduction is usually to encourage activities that result in more future taxable income or to create incentives for taxpayers to participate in certain behaviors that are considered desirable. In the case of the mortgage interest deduction, one of the policy goals is to increase levels of homeownership in the United States. The existence of deductions in the tax code necessitates a trade-off between different government services, providing incentives toward certain behaviors, higher taxes, and perhaps even larger deficits.


    In the aggregate, the home mortgage interest deduction substantially lowers the amount of taxes due from US households and thus lowers government revenue. For fiscal year 2011, the Office of Management and Budget estimates that the total amount of tax revenue lost was around $72 billion. The only “tax expenditures” (as the Office of Management and Budget calls them) that were larger were due to the non- taxation of employer contributions to health insurance and retirement plans.[1] But as figure 1 shows, for the typical tax- paying household, the tax benefits of the mortgage interest deduction were quite small. With median household income in the United States around $50,000, a typical family claiming the mortgage interest deduction received somewhere between $100 and $200 in tax savings for the entire year.

    This difference between the aggregate and household savings can be explained by two facts. First, only a fraction of taxpayers claim the deduction. In tax year 2010, the Joint Commit- tee on Taxation estimates that about 21.7 percent of taxpayer returns took the mortgage interest deduction. Second, of the small number of returns claiming the deduction, a disproportionate amount of the benefits went to high-income house- holds. Over 40 percent of the $72 billion went to households with adjusted gross incomes over $200,000, and another 35 percent went to households with incomes between $100,000 and $200,000. That leaves just about 25 percent of the $72 billion going to households with incomes under $100,000. Figure 2 shows the number of households that claimed the mortgage interest deduction in 2010.

    The bottom line is that most households would not be significantly harmed by removing the mortgage interest deduction. The 78 percent of households not claiming the deduction wouldn’t be harmed, and most stand to benefit if tax rates are lowered. For the typical household receiving $100 or $200 in tax savings per year (i.e., those in the figure 1 income categories between $30,000 and $75,000), the losses could easily be offset by slightly lowering marginal tax rates. High- income households would be harmed by removing the mort- gage interest deduction, but this represents a way to raise tax revenue without raising tax rates and thereby discouraging productive activity. To the extent that lower marginal tax rates encourage more economic activity, households in all income groups would be better off.


    Many defenders of the home mortgage interest deduction claim, as their central argument, that it makes housing more affordable for middle-class families. However, because of the nature of the tax system, this deduction does not necessarily provide incentives to everyone—nor does every debtor who pays interest on a mortgage pay lower taxes. The reason is that not all tax-filers use the itemized deductions that allow them to deduct their interest payments, and those who deduct are concentrated in the higher tax brackets while those who take the standard deduction are concentrated in the lower brackets. Moreover, it is important to recognize that the mortgage interest deduction is not precisely a deduction on home ownership. Homeowners are not allowed to take it against the part of the home they own (the down payment and prior repayment of principal). Instead, they can only take it against the part of the home they borrow. For these reasons, the mortgage interest deduction is an inefficient tool for increasing homeownership, since its primary effect is to encourage Americans who would have already been able to afford a house to take on even more debt.

    Empirical evidence supports the claim that the mortgage interest deduction has little effect on homeownership rates in the United States. Between 1960 and 1997, homeownership rates stayed within a narrow range of 62 to 66 percent, despite the fact that the implicit tax subsidy fluctuated dramatically.[2] During the recent housing bubble, the homeownership rate rose to 69 percent, but it has since returned to the historical range.[3] This rise appears to have been unrelated to the mort- gage interest deduction, though it was almost certainly related to other housing policies that encouraged the bubble. More sophisticated analysis suggests that the homeownership rate would be modestly lower without the deduction, by around 0.4 percent.[4]


    If the main proposed benefit of the mortgage interest deduction, increased homeownership, is small or nonexistent, the case for the deduction is significantly weakened. But the case is weakened further when we consider the costs of the mortgage interest deduction. Beyond the obvious cost of less government revenue from some taxpayers (and thus higher taxes on others to maintain a given level of government spending), there are several economic distortions that result from this deduction.

    One distortion is that more of society’s resources, human and physical, are devoted to high-income residential housing construction than would otherwise be the case.[5] This cost was seen on a grand scale in the housing boom and bust of the past decade, but it occurs on a small scale all the time. Recent empirical research suggests that the mortgage interest deduction increases the size of homes purchased but not the overall rate of homeownership.[6] This fact is consistent with the above research on homeownership rates but also demonstrates that more capital is being allocated to residential housing than the market would allocate. When the capital structure of the economy is altered, resources are not allocated efficiently and economic growth is hindered.

    A second economic cost is the alteration of the distribution of income in the country. Because this deduction favors those with high incomes, high marginal tax rates, and many other itemized deductions, the distribution of income is skewed in favor of the wealthy when compared with a simpler tax code. This outcome may further the popular notion that the entire system is rigged in favor of the wealthy. In addition, the misallocation of capital to the construction of high-income houses is to the benefit of the wealthy and skews the structure of the economy in their favor.

    The final major cost of the mortgage interest deduction is the cost of lobbying and rent-seeking associated with the deduction. Because the large benefits of this deduction are concentrated on a small group of taxpayers, a variety of lobbying and pressure groups are willing to expend resources to preserve this deduction in Congress and in the court of public opinion. While this spending is beneficial to the recipients of the benefit, from a social perspective it is pure economic waste and leads to lower growth for the economy as a whole.

    Where we cannot be sure of the subsidy having an economic cost is in the general level of housing prices. Contrary to many claims that removing the mortgage interest deduction would reduce the price of houses, economic research suggests that it is likely to have little effect.[7] As discussed above, the rate of homeownership has remained relatively constant for almost four decades, despite fluctuations in the implicit subsidy. Furthermore, the deduction primarily affects high-income earners, so we should expect that the subsidy primarily inflates the prices of high-income housing rather than the general level of housing prices. Most within and under the median income do not benefit from the deduction (figure 2), and the benefits are on average small, even for those taking the deduction (figure 1).


    There are three main routes toward ending the home mort- gage interest deduction. The first is simply ending the deduction and using the increased tax revenue as more revenue for the federal government. The second is ending the deduction and decreasing the general level of income taxation by an amount that corresponds to the increased tax revenue. The third is similarly stopping the deduction and replacing it with a tax credit that would be enjoyed by taxpayers upon the purchase of their first home.

    A major benefit of the first two possible routes is that both of them would end the federal government’s subsidization of homeowners. A benefit of the first proposal is that it could help stem the tide of red ink seeping from the federal government, and do so without threatening programs that better advance the general welfare. However, this money might amount to a net tax increase, which could harm economic growth in the short and long run. Therefore, our favored approach, on efficiency and equity grounds, is to eliminate the deduction and simultaneously lower marginal rates so the typical homeowner is no worse off.


    1. For an overview of the major tax expenditures in the current federal tax code, see Jeremy Horpedahl and Brandon Pizzola, “A Trillion Little Subsidies: The Economic Impact of Tax Expenditures in the Federal Income Tax Code” (Mercatus Working Paper, Arlington, VA: Mercatus Center at George Mason University, October 2012).

    2. Edward L. Glaeser and Jesse M. Shapiro, “The Benefits of the Home Mortgage Interest Deduction,” Tax Policy and the Economy 17 (2003): 37–82.

    3. US Census Bureau, “Housing Vacancies and Homeownership,”

    4. Harvey S. Rosen, Kenneth T. Rosen, and Douglas Holtz-Eakin, “Housing Tenure, Uncertainty, and Taxation,” Review of Economics and Statistics 66, no. 3 (1984): 405–416.

    5. Lori L. Taylor, “Does the United States Still Overinvest in Housing?” Economic Review, Federal Reserve Bank of Dallas, Second Quarter 1998: 10–18.

    6. Andrew Hanson, “Size of Home, Homeownership, and the Mort- gage Interest Deduction,” Journal of Housing Economics 21, no. 3 (2012): 195–210.

    7. Donald Bruce and Douglas Holtz-Eakin, “Fundamental Tax Reform and Residential Housing,” Journal of Housing Economics 8, no. 4 (1999): 249–271.

    A Trillion Little Subsidies: The Economic Impact of Tax Expenditures in the Federal Income Tax Code

    October 25, 2012


    The loopholes known as “tax expenditures” reduce individual and corporate tax obligations by more than $1 trillion each year. But while these tax deductions are hugely popular and fiercely protected, are they really a good deal for most Americans?

    A new Mercatus Center study looks at the ten largest tax expenditures for individuals and corporations and weighs the economic impact of each. The study also reviews the intended-vs-actual beneficiaries and outcomes of particular tax expenditures and considers the economic and political implications of eliminating all expenditures in a single swipe.

    To read “A Trillion Little Subsidies: The Economic Impact of Tax Expenditures in the Federal Income Tax Code” and to learn more about its authors, please click here.



    • According to the Office of Management and Budget, FY2011 federal tax expenditures were more than $1 trillion—with 80 percent, or $891 billion, going to individuals, and 20 percent, or $181 billion, to corporations.
    • To put this in perspective: FY2011 tax expenditures were nearly equal to all federal income tax collected in that year, or to the entire FY2011 discretionary budget; they were also greater than the annual federal spending on Medicare or Social Security.
    • Although applied through the tax code, their effect is similar to spending provisions, hence the name tax expenditures: They encourage certain kinds of government-supported behavior by subsidizing it. Because they are part of the tax code, however—reducing revenue from what it otherwise would be, rather than overtly increasing spending—they mask the true size and scope of government.




    • Tax expenditures hinder economic growth by distorting individuals’ and corporations’ behavior toward qualifying for tax loopholes rather than making the best economic decisions.
    • By distorting behavior, tax expenditures distort the entire economic system by altering spending on goods and services; distorting capital allocation; changing the distribution of income; and encouraging lobbying and rent-seeking to maintain and expand these provisions.
    • For example, corporations must divert significant resources away from pro-growth activities to the non-productive—but critical to compete with other U.S. corporations— activity of fighting for more and bigger tax breaks.
    • Of the largest tax expenditures studied, the authors found the stated legislative intent was seldom realized.
    • Intended economic benefits seldom materialized.
    • Intended beneficiaries were seldom the greatest beneficiaries; most tax-expenditure benefits accrue disproportionately to higher-income earners, and encourage “gaming” the system by those in a position to take advantage, often resulting in cronyism and the capture of the tax code for private gain.
    • For example, while the encouragement of home ownership has become the common justification of the home mortgage interest deduction, it does not effectively achieve this goal. 
    • Of the 33 percent of taxpayers who itemize deductions, only 20 percent claim the home mortgage deduction. Of those, two thirds make more than $100,000 a year.
    • Consequently, individuals and families on the margin who could be motivated to become homeowners by incentives—that is, lower-income individuals and families—are unlikely to use this deduction.
    • With respect to tax expenditures, the authors conclude that economically optimal tax reform must include two key pieces: eliminate all tax expenditures and lower marginal rates across the board.
    • Eliminating expenditures without simultaneously lowering tax rates amounts to a tax increase on the economy as a whole. This would result in slower economic growth and, thus, lower future tax collections.
    • While a one-shot elimination of tax expenditures seems less unlikely, the authors find it more probable than a piecemeal elimination. This is because the former would give all taxpayers an immediately recognizable benefit (such as an increase in individuals’ paychecks or a significant reduction in lobbying and accounting expenses for businesses) to offset later loophole losses.


    Read the paper as a PDF