Design and Cost Considerations in Free-file Tax Systems

Executive Summary

  • The Inflation Reduction Act provided $79 billion to the Internal Revenue Service, most of which is devoted to enforcement actions, including $15 million to study options on how to develop a “free direct efile” tax filing system.
  • A simplified or potentially “return free” tax filing system may be feasible for taxpayers with simple tax situations; however, due to the complexity of the tax code, estimates from the literature suggest that such a system risks being infeasible or unappealing for most taxpayers.
  • A new system would nevertheless require new resources to run in parallel with the existing tax-filing system.
  • On net, relatively low participation, additional burdens on third parties, and additional administrative costs could exceed the modest compliance savings realized by some taxpayers.


Tax filing season officially begins January 23, foreshadowing the annual ritual of tax reconciliation that confronts to one degree or another every American. It is also a ritual that most would likely choose to forgo. The allure of easy tax filing has obvious appeal and reducing the burden of tax compliance has always been a part – at least rhetorically – of any major tax reform proposal in the United States. Yet there is a reason why complying with the tax code remains, at least in the aggregate, a costly proposition: tax complexity. That complexity bedevils the feasibility of simplified or return-free tax system at scale. The essential challenge is that such a system risks being infeasible or unappealing for the majority of taxpayers. The new system would nevertheless require additional resources and impose additional burdens on third parties. The evidence suggests that such a system could ultimately impose greater costs than it ultimately saves.

Examining Alternative Approaches to Tax Filing 

After a lengthy congressional debate, President Biden signed the Inflation Reduction Act (IRA) into law in August 2022. Among the more significant elements of the law is a $79 billion direct appropriation to the Internal Revenue Service (IRS), most of which is devoted to enforcement actions. The law also provided the IRS with $15 million to study options on how to develop a “free direct efile” tax filing system. The provision funding the study of a potential return-free filing system specifies that the study include, among other provisions, “taxpayer opinions, expectations, and level of trust, based on surveys, for such a free direct efile system; and…the opinions of an independent third-party on the overall feasibility, approach, schedule, cost, organizational design, and Internal Revenue Service capacity.”

This is not the first time Congress has sought options to reorient the tax-filing experience. Indeed, a study conducted in 2003 by the Treasury Department at Congress’s behest notes that the IRS, the Government Accountability Office (GAO), and the Department of Treasury itself had previously examined potential eligibility for return-free tax systems. The 2003 study assessed design issues and implications of a return-free tax filing system, among other considerations, and examined similar systems in other nations. The study identified two alternative design approaches to a return-free tax system. The first approach is exact withholding, in which employers withhold the exact tax liability owed by a wage-earner. This can be performed on an ongoing basis, or through periodic adjustments such as at the end of the year. A second approach is tax agency reconciliation. This design in some ways mirrors the existing system, but for the key distinction that the tax authority prepares a taxpayer’s return and the taxpayer reviews, verifies, or contests the government-prepared return.

According to the Treasury study, most return-free systems hew to the exact-withholding approach, though Treasury observes that the underlying tax laws in the relevant nation tend to be simpler and otherwise lend themselves to such a system. The U.S. system, which features significant income-support policies, health policies, and is household-centric (as opposed to taxpayer-centered) would complicate the implementation of an exact withholding system.

Senator Elizabeth Warren, an advocate for the IRA-funded study, has introduced legislation that would, among other provisions, require the IRS to develop a return-free tax filing system for certain eligible taxpayers. This renewed effort in Congress reflects longstanding interest in simplifying the tax administration. Yet as the relevant literature, on which any IRS study would necessarily rely, indicates, simply reordering the tax-filing process does not render the process costless. Indeed, despite claims from advocates that having the IRS pre-fill tax returns will reduce compliance costs, it may simply shift them.

It is important to note that any tax administration and tax filing system is ultimately a reflection of the tax code. The United States now substantially delivers health, anti-poverty, and family benefits through the code. These elements add complexities and complications that move the code away from simple administration. Among the key conclusions of the 2003 Treasury study was the observation that, “While it is clear that a return-free tax system would shift compliance costs among affected parties, it is not clear whether such a system would reduce overall compliance burdens and administrative costs if it were unaccompanied by tax simplification.” In short, unless Congress makes a concerted effort to simplify the tax code, which to some degree would come at the expense of other policy priorities, a reorientation of tax administration would not necessarily reduce compliance burdens and may increase them.

Tax Compliance Cost Under the Current System

The appeal of reorienting tax filing is rooted in the relative pain of complying with the tax code. Were filing taxes a painless exercise, there would be little interest encouraging the government to assume a greater share of the undertaking. A number of studies have attempted to capture the cost to the taxpayer and the economy of administering the U.S. tax system. These estimates provide valuable context and, despite some differences, are noteworthy for the relative similarity in magnitude and direction. They also reflect the sheer volume of information collection required to administer the tax code. That information must be recorded, verified, communicated, and otherwise reconciled against a body of law – the internal revenue code – that changes somewhat frequently to varying degrees.

One estimate places the compliance costs at $200 billion on taxpayers. The U.S. tax system imposes billions in costs on taxpayers in terms of pecuniary expense and time spent. Most Americans aren’t directly exposed to the totality of the compliance burden. Rather, the roughly $40 billion in compliance costs imposed by the individual tax code is sufficiently onerous to inspire alternative approaches.

A 2019 study by the Bipartisan Policy Center provides an excellent survey of recent estimates. The Taxpayer Advocate Service (TAS) has also reviewed recent attempts at capturing the cost of the U.S. tax code, noting that experts have embraced a range of methodologies for these calculations. TAS, for example, estimated the 2015 cost of income-tax compliance at $195 billion. The Tax Foundation estimated that compliance costs amounted to $406 billion in 2016. Subsequent estimates that include additional cost considerations and alternative approaches to monetizing the hours spent complying with the tax code alter these estimates considerably. Fichtner and Feldman completed a thorough assessment of the costs that the U.S. tax code extracts from the economy through complexity and inefficiency, beyond TAS’s estimate. According to the authors, in addition to time and money expended in compliance, foregone economic growth and lobbying expenditures amount to hidden costs estimated to range from $215 billion to $987 billion.

While the tax code imposes hundreds of billions in compliance costs on the economy, taxpayers also fund the Internal Revenue Service that administers the tax code. The IRS had a budget of just under $14 billion in fiscal year 2022. Out of this budget, the agency devotes $2.7 billion to Taxpayer Services, the division tasked with processing tax returns and providing assistance and education to the public.

The taxpayers’ return on this investment is questionable. In its most recent report to Congress, the IRS internal watchdog, TAS, observed that the IRS answered only about 1 in 10 telephone calls in fiscal year 2021, down from the 24 percent from the prior year. This decline in responsiveness corresponded with new responsibilities for COVID-19-related relief programs and, somewhat related, a doubling in taxpayer phone calls, but historical performance was relatively poor. This suggests policymakers should approach the notion of adding to the agency’s mission set with caution.

IRS funding had been essentially flat in real terms over the last two decades. Over the same time, the IRS’s workforce has declined, and the number of tax-filers has grown. There was a legitimate case to be made that declining customer service was at least in part a function of resource constraints. That case has now been rendered somewhat moot by the IRA’s appropriation of $79 billion – more than five times the agency’s annual operating budget. One caveat though is that of this amount, only $3 billion has been devoted to “Taxpayer Services,” while the majority of the funding is being devoted to enforcement and other expenses. Any new tax filing system would likely prompt a request to Congress for start-up and operational funding. The legislation, sponsored by Senator Warren, authorizes “such sums” as necessary to implement the policy. 

Policy Implications of a Return-Free Tax System

The principal appeal of a return-free system is the degree to which taxpayer is spared the time, effort, and cost of filing taxes. This proposition involves significant tradeoffs for the taxpayer, however. As noted above, many of the features of the modern U.S. tax code make administration more difficult. Yet those features are the product of deliberate policy choices within a democracy. Relatedly, the more complex a taxpayer’s tax situation, the less likely the IRS will be able to accurately pre-populate tax return information. The simpler a taxpayer’s tax situation is, the simpler it is to provide return-free filing. But the more taxpayers by a return-free system, the more diverse and complex the tax situations would be. The scope of any potential return-free system is therefore elemental to assessing cost. 

Eligibility and Participation

Estimates of potential participation and eligibility for a return-free tax system vary, both by estimator, as well as by design feature. In general, the sweep of estimates is fairly similar. At the lower bound, a return-free tax system that was restricted to the simplest tax situations – such as for a single taxpayer with only one source of income – would allow for around 8 million taxpayers to benefit from a return-free tax system.[1] Scaling up eligibility necessarily expands the scope of the tax situations that the IRS must be prepared to administer and the inherent cost of the exercise. Estimates from the IRS, GAO, Treasury, and Gale and Holtzblatt (1997) found that at an upper-bound 51–63 million taxpayers could benefit from a return-free system. A more recent estimate found that a return-free tax system could be feasible for a similar proportion of taxpayers.[2] But in the absence of more material changes to America’s tax laws – a challenging proposition – eligibility is closer to the 8 million lower bound, according to a 2007 study.[3]

Expanding the program to taxpayers with multiple sources of income introduces additional complexity to the tax reconciliation process. With additional complexity comes increased risk of miscalculations, which reduces compliance and participation.[4] It is important to note that while the relative share of eligible taxpayers may be stable, individuals’ tax situations can vary substantially year to year due to circumstances the IRS would be unable to foresee. This heightens the risk of incorrect pre-population of tax forms, which can suppress participation in addition to the individual risks to taxpayers.

Many of the relevant studies assess eligibility or feasibility – essentially testing who could participate in such a system, not necessarily who would participate. Participation is essential to the value proposition of a return-free system. The IRS commissioned a telephone survey to gauge attitudes among potentially eligible participants in a hypothetical return-free tax system. The survey revealed that roughly a third of potential users were interested, a nearly equal share of potential users was uninterested in such a system, while the remainder of the population was neutral or didn’t know. Thus, for a given return-free system design, the total eligible population is very much a theoretical bound. Just applying the survey results to the eligibility estimates could essentially halve the participation rate relative to the share of taxpayers eligible to participate. Evidence from California’s return system should caution policymakers against the attractiveness of such a system. According to a 2010 study, only 3.2 percent of California’s taxpayers participated in the state’s return-free system. The study observed that “the vast majority of California’s eligible filers have declined to use it…, suggesting that most people believe the savings they would realize in time and out-of-pocket spending would be outweighed by the costs, including risks to privacy and security.”[5]


The Treasury aptly summarized the extant literature on the cost of a potential return-free tax system: “Estimates of the benefits and costs of a return-free tax system diverge widely, revealing some of the uncertainty surrounding how such a system would be implemented as well as how taxpayers would respond.” The Treasury cited a past IRS estimated that a tax agency reconciliation system would cost the federal government, on net, more than $284 million (2001 dollars), and would reduce annual taxpayer burdens by 8.8 million hours.[6] For context, the current tax code imposes a paperwork burden estimated at 6.5 billion hours. The GAO (1996) estimated that the same system would reduce the federal government’s costs by $60 million (2001 dollars), while reducing taxpayer burdens by 155 million hours a year. These studies diverge due to different estimates and assumptions about participation and administrative costs. Critically, the GAO study assumed all eligible participants would participate, an assumption somewhat at odds with the relevant research.

A new tax-filing system would involve both start-up and operational costs. With respect to start-up costs, there is recent experience to suggest some caution and humility with respect to how quickly, ably, and affordably the federal government can stand up a new tax filing system. The Affordable Care Act of 2010 established the Federally Facilitated Marketplace to serve as an online marketplace for the purchase of federally regulated health insurance products. Despite years to prepare, and over $9.6 billion in initial costs and over $21 billion in lifetime costs, the online portal faced a number of technical challenges that severely impaired the service. 

The upshot of eligibility and participation estimates suggest that most taxpayers would still file their taxes consistent with recent practices, such as self-filing, or with the assistance of a tax preparer or software. Were the federal government to establish a free-file system, it would need to run that system in parallel with the existing system – which, narrowly defined, costs taxpayers $2.7 billion per year. While there are perhaps some institutional efficiencies that could be realized, adding a new mission to the IRS would require additional resources, while likely imposing new burdens on third parties such as employers and payroll firms, offsetting or exceeding any savings that the federal government or individuals would get from such a system. A 2010 study found that additional third-party compliance costs of a federal return-free tax preparation system could range from $500 million to $5 billion a year, which would swamp the potential compliance savings for taxpayers.[7] For example, assuming the 8.8-million-hour reduction in compliance burden estimated by the IRS, taxpayers would save about $320 million based on the Bureau of Labor Statistics’ estimated average hourly wage for compliance officers ($36.45).[8] It would take nearly nine times the savings to make up for half the upper-bound costs estimated by Cordes and Holen (2010).

Proponents of a return-free tax filing system note that such a system would reduce the “tax gap” – the difference between the amount of “true tax” and the amount of tax actually paid – and could mitigate some of the new costs of the system. It is important to note, however, that the underreporting tax gap for employee wages only amounts to about 1 percent, whereas other forms of income, such as partnership and pass-through income, face higher rates of underreporting. Yet it is alternative income sources that generally render taxpayers ineligible for return-free systems.


A simplified tax filing experience holds tremendous appeal and is regularly invoked as a potential benefit of tax reform. But simplified tax-filing is generally confounded by a complex tax code. Congress directed the Treasury to study a direct efiling tax system, as well as assess costs and other design considerations. This not the first attempt at such a system. Past research reveals that challenges scale with the scope of participation, but the scope of participation necessarily relates to the value proposition of the system. While a simplified tax-filing experience may be efficient and feasible for a relatively modest share of taxpayers, deploying such a system at scale may ultimately fail a cost/benefit test.


[1] See Tax Administration: Alternative Filing Systems. Oct. 1996, U.S. General Accounting Office (GAO); and On the Possibility of a No-Return Tax System 1997, William Gale and Janet Holtzblatt

[2] Automatic Tax Filing: Simulating a Pre-Populated Form 1040. April 2022, Lucas Goodman, Katherine Lim, Bruce Sacerdote, and Andrew Whitten

[3] Implications of Return-Free Tax Systems for the Structure of the Individual Income Tax Sept. 2007, Janet Holtzblatt

[4] See: Less cheating? The effects of prefilled forms on compliance behavior July 2018, Martin Fochmannb , Nadja Müllerc , & Michael Overeschd; and How choice architecture can promote and undermine tax compliance: Testing the effects of prepopulated tax returns and accuracy confirmation Aug. 2020, Wilco W. van Dijk, Sjoerd Goslinga, Bart W. Terwel, & Eric van Dijk

[5] Should the Government Prepare Individual Income Tax Returns? Sept. 2010, Joseph Cordes and Arlene Holen

[6] See: United States Department of the Treasury, Internal Revenue Service, Current Feasibility of a Return-Free System, Washington, D.C. October 1987; and GAO (1996)

[7] Should the Government Prepare Individual Income Tax Returns? Sept. 2010, Joseph Cordes and Arlene Holen

[8] See: