Technology Policy Institute Says Return-Free Filing Would Burden Employers

Simplifying the complex U.S. tax code is the most direct way to reduce both the public and private costs of complying with the federal income tax, but tax reform is extraordinarily difficult to achieve. Some analysts have argued that return-free filing systems, such as those used in other countries and in the state of California, could substantially reduce the costs for many individual taxpayers with relatively simple returns at little or no net administrative cost to the government.

There is substantial uncertainty as to how a federal return-free system would affect the costs of government and individual tax filers. California's program appears to have reduced the state's administrative costs, but the net savings are largely attributable to e-filing rather than to the return-free system itself. The vast majority of California's eligible filers have declined to use it (only 3.2 percent do), 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.

A return-free tax system would increase third-party tax compliance costs -- those of employers, financial institutions and other payers of income to individuals -- largely because reporting deadlines would have to be advanced in order to provide timely returns and tax refunds. Calculations of such costs range from $500 million to as much as $5 billion, offsetting or exceeding any potential savings for taxpayers or for government. Added costs would fall disproportionally on small businesses.

Any cost savings for the IRS and individuals would likely be modest and additional costs to employers and other payers of income could be significant. Further, return-free filing would introduce a host of issues at the policy, economic, and operational levels. These include:

• The Internal Revenue Service (IRS) would face a conflict of interest in functioning as both tax preparer and enforcer.

• Risks of error would result from stretched IRS capacities. The IRS lacks the essential electronic processing capabilities and would face challenges in carrying out new responsibilities, particularly while its mission is expanding under health reform. Compressed income reporting schedules would also increase risks of error.

• Taxpayers who are unwilling to challenge an official IRS document -- even when it may be erroneous -- would nevertheless retain responsibility and liability for errors in government-prepared returns, a problem particularly for lower-income filers and those with English as a second language.

• Taxpayers would become less cognizant of the incentives in the tax code and their personal finances.

• IRS preparation of individual returns could compromise taxpayers' privacy.

• IRS tax preparation may pose greater security risks because private providers face stronger financial incentives to strengthen security practices in an environment of rapidly advancing technology and changing threats.

• The government would enter into competition with the private sector, reducing incentives for investment in innovation and electronic tax preparation systems.

INTRODUCTION

Collecting U.S. federal income taxes is complex and imposes both public and private costs. The Internal Revenue Service (IRS) and other taxing authorities incur costs in administering the income tax and monitoring taxpayer compliance. Individual taxpayers spend time and money on a range of tasks, including keeping records, tax planning, and completing forms. The U.S. Government Accountability Office (GAO 2005, Table 2) estimated costs of the federal tax system to the government and individuals to be in a range of $67 billion to $100 billion annually, or approximately 7 to 11 percent of personal income taxes collected. These figures do not include the costs to third parties, such as employers and financial institutions, of maintaining and providing information to taxpayers and the government about taxpayer incomes and withheld taxes. Neither do they include the economic cost associated with tax-induced distortions in market decisions (the deadweight loss of taxes), which have been estimated to be as high as 30 percent of income tax revenue collected.¹

While some costs of collecting any tax are unavoidable, it is important to identify ways of reducing those costs. In the case of the income tax, the most direct approach would be to reform and simplify the tax code itself, but tax reform is extraordinarily difficult to achieve.

Recently, some analysts and government officials have argued that the costs of administering and complying with the federal income tax could be reduced by changing how individuals file and pay their income tax. Instead of the current practice, in which individuals are responsible for gathering the necessary information, calculating tax liability, and ultimately filing and paying federal taxes, they suggest creating what is technically referred to as a Tax Agency Reconciliation (TAR) system -- more popularly described as a return-free system -- in which "pre-populated" tax returns would be sent to taxpayers, who would have the option to accept and file the already-prepared return.²

Although return-free tax systems in other countries, most notably the United Kingdom, have been used for many years, the current policy interest in return-free filing in the United States seems to stem from the digital revolution, which has reduced the costs of communicating financial information between income payers (e.g. employers) and the government, and between the government and individuals. Supporters of return-free filing argue that resources invested in such a system would yield significant dividends in the form of reduced time and money burdens for individual taxpayers. They point to other developed countries that provide some form of pre-prepared tax return and to the positive feedback received from taxpayers who voluntarily participated in the ReadyReturn program, a TAR program piloted by the California Franchise Tax Board.

Another motivation for adopting a system of government-prepared returns is its potential to increase revenues and reduce the so-called "tax gap," the difference between taxes owed and taxes collected. That issue is briefly considered in the President's Economic Recovery Advisory Board (PERAB) report on options for changing the current tax system in its discussion of the Simple Return option.³ The report notes that the types of federal investment in technology and manpower that would be needed to implement the Simple Return would also be required to increase overall tax compliance. Capturing unreported income might be one reason to expect revenues to increase; another reason is that the use of low-income tax credits and incentives could decline. As noted in the PERAB report, however, under-reporting of income could increase and revenues could decline as a result of incomplete IRS information. Most policy discussion of return-free filing has focused on how it would affect compliance costs and its potential as a revenue-raising strategy has not been systematically examined.

There is reason to be skeptical about the net benefits of implementing a return-free system.⁴ Because programs such as ReadyReturn do not actually simplify taxes, they may simply shift compliance burdens from individual taxpayers to the government and third parties. While shifting these responsibilities might be expected to lower compliance costs for taxpayers filing certain types of tax returns, it is an empirical question whether implementing such a system on a larger scale would lower overall compliance costs in the economy. Moreover, tax preparation software and other assistance have sharply reduced the cost of tax preparation for filers, reducing the potential savings from return-free filing.

To address these issues, we first describe different approaches for making pre-prepared tax returns available to taxpayers and then present a simple framework for assessing how shifting compliance burdens in the manner contemplated by return-free proposals affects the total cost of income tax compliance. We summarize what is known empirically about the potential effects of such programs on compliance costs, including the effects of the ReadyReturn program, and discuss the implications of this evidence for introducing a return-free option in the United States at the federal level.

We also assess orders of magnitude of increased compliance burdens on third parties -- employers, financial institutions and other payers of income -- resulting from the need to move up required reporting dates, which would be necessary for the government to provide timely returns and tax refunds. Empirical evidence on such costs is quite sparse, although clearly relevant to assessing the overall costs of moving to a return-free system. Increased costs to third parties could be substantial and exceed any savings for individuals and government.

Having the government assume responsibility for preparing individual tax returns raises other important issues aside from the effects on time and money spent on tax compliance. These include taxpayer privacy, risks of error, system security, and the role of the tax system as an instrument of social policy. A return-free system also raises issues of taxpayer risk in accepting a government-prepared return even when it may be erroneous and exposing themselves to potential liability and penalties. Low-income tax filers, those with English as a second language, and those with lower levels of formal education, could present particular concerns with regard to this risk, as they may be more likely to accept government-prepared returns without critical review. Responses to taxpayer surveys and low participation rates in the California ReadyReturn program indicate that most taxpayers are not persuaded that the benefits they might realize outweigh the costs of participating in a return-free system.

RETURN-FREE TAX SYSTEMS IN OTHER COUNTRIES

Broadly speaking, there are two types of return-free systems. The United Kingdom has used a Pay As You Earn (PAYE) system for income tax collection for more than sixty years. Under that approach, employees' payroll deductions are designed to match exactly the tax liability for most employees, so that no end-of-year filing, payment, or refund is typically needed (APPTG 2009).

The principal alternative to exact withholding is the Tax Agency Reconciliation (TAR) system, used by a number of countries, including Denmark, Sweden, and Spain. With TAR, taxpayers can elect to have the tax authority prepare their return based on information the authority receives from employers, financial institutions, other sources of income, as well as on information from the taxpayer. (Gale and Holtzblatt 1997, p. 477). The California ReadyReturn system is an example of TAR.

Based on a survey of TAR systems in the Nordic countries, the Organisation for Economic Co-operation and Development (OECD) identified several factors as critical to the success of a reconciliation system. These factors include: accurate withholding; high integrity taxpayer identifiers allowing easy matching; comprehensive systems of third party reporting to the tax authority; a compatible tax code that provides relatively limited scope for factors such as deductions, rebates, and credits that cannot be predicted using third party reports; and automated and minimal interactions with taxpayers to avoid costly vetting of completed pre-populated returns. (OECD 2006, pp. 15-16)

The OECD emphasized the critical feature of accurate and calibrated withholdings at the source, "to ensure that aggregate withholdings over the course of a fiscal year more or less approximate to taxpayers' annual liabilities." The OECD concluded that TAR systems in effect may not be so different from PAYE in that more or less exact withholding is needed to minimize large tax refunds and allow taxpayers to file returns shortly after the end of the fiscal year.

Gale and Holtzblatt (1997) note that both exact withholding and TAR systems are more easily adapted to structurally simple tax codes than to the highly complex U.S. code. Return-free systems cannot readily handle capital gains, itemized deductions, business income, employee business expenses, or individual retirement accounts. The countries that have adopted such systems generally have far simpler tax codes than the United States.

Gale and Holtzblatt also found that TAR systems add costs because payers of income and benefits need to report information as close to the end of the year as possible, while tax authorities have to absorb, process, and match millions of information returns more quickly than they do under the current U.S. tax system.

In the UK, a bipartisan committee in Parliament (the All Party Parliamentary Tax Group) reported in 2009 that PAYE, rather than simply reducing filing burdens, encouraged taxpayers to absent themselves from compliance activities. People are effectively presented with a tax bill which they assume must be correct or decide not to challenge in order to avoid unwelcome attention from the tax authority. The parliamentary group concluded that the system places cost burdens on employers, particularly small employers, and is strained by increasingly dynamic labor markets in which government data cannot keep up with people's changing employment and life situations. The parliamentary tax group further concluded that adopting features of a TAR system such as those pioneered in Nordic region countries would not make the UK tax system more compatible with dynamic labor markets or ameliorate the imbalance between administrative costs for large and small firms because both TAR and PAYE systems require accurate withholding.

To provide accurate data for determining eligibility for socio-economic tax credits, Her Majesty's Revenue and Customs (the UK's IRS) has instituted a requirement under PAYE to file a pre-return form, setting forth personal and financial information. The form and accompanying instruction booklet resemble the IRS 1040 form and instructions.⁵ UK's PAYE system has encountered major problems in maintaining accuracy. Almost 6 million cases of incorrectly calculated taxes of the 40 million covered by PAYE were found over two years and, as a result of widespread errors in rebates, almost 1.5 million workers face demands to pay back funds.⁶ The All Party Parliamentary Tax Group has called for fundamental reform of the PAYE system (APPTG 2010).

It is a widely accepted principle of tax policy that, all else being equal, tax systems should minimize the total resources in time and money devoted to administration and compliance (Slemrod 1990). As noted in a U.S. Treasury (2003) report, substituting a TAR system for the current system would not reduce the complexity of the federal income tax per se, but rather would shift the burden of income tax compliance from individual taxpayers to the government and to third parties (e.g. employers and financial institutions).

Whether shifting compliance burdens in this manner can reduce the total costs depends on the comparative advantage of the various parties in carrying out various compliance and administrative activities. If we define C_T as the total cost of compliance, C_G as costs of the government (i.e., IRS), C_I as the cost of compliance for individual taxpayers, and C_P as the cost of compliance for third parties (e.g., employers and financial institutions), then (1) defines the net change in compliance costs resulting from a shift in responsibilities from individuals to the government and to private parties resulting from adopting a TAR system.

 

 

IRS and GAO Estimates

In its 2003 report, the Treasury Department summarized the findings of two government studies, one by the IRS and one by the GAO, which attempted to estimate some of the components of (1) for two hypothetical TAR systems. The results of these earlier studies are summarized in Table 1 below.

 

Table 1

 

IRS and GAO Analyses of TAR Systems

 

 

 

As noted in the Treasury report, the wide divergence in estimates of changes in compliance burdens for individuals and government can be attributed to several factors.

First, the IRS report is based on data from 1985 tax returns, whereas the GAO report was based on 1994 tax return data. Second, compared with the GAO report, the IRS report made more modest assumptions about participation in, and time cost savings from, participation in a TAR: the IRS report assumed that only those filing form 1040A and 1040EZ would be eligible, and that 45 percent of those eligible would choose to participate; whereas the GAO report assumed that some IRS form 1040 filers would be eligible to participate in addition to 1040A and 1040EZ filers, and that 100 percent of those eligible would participate.

The IRS study assumed form completion times of 26 minutes (1040EZ) and 60 minutes (1040A) compared with the GAO study's estimates of 3 hours (1040EZ) and 7 hours (1040A) and 12 hours for IRS 1040 filers. The GAO estimates include filers' time devoted to record-keeping and tax planning, which would not be reduced by participation in a TAR program. Recent IRS estimates indicate that form completion accounts for only about 25 percent of the taxpayers' total time burden.⁷ Thus, the GAO analysis overstated taxpayer time savings from participating in a TAR program.

Importantly, both the GAO and IRS studies predate the broad adoption of computer tax preparation software programs, which have greatly simplified the task of tax preparation for taxpayers. For this reason as well, these studies likely overstate filers' time savings associated with a TAR program.⁸

Lastly, as observed by Holtzblatt (2006), the IRS study accounted for the fact that providing pre-populated returns to taxpayers in a timely manner would require the IRS and the Social Security Administration to dramatically accelerate their normal procedures -- i.e., to process "one billion returns during the normal filing season and within a 30 to 60 day time frame." This, in turn, would require significant additional investments in staff, equipment, and facilities. Holtzblatt's description of the normal process for gathering and matching taxpayer information by the IRS indicates the modifications that would be necessary:

Adjusting the GAO estimates both for differences in assumptions and for some of the additional costs included in the Treasury report brings them closer to the Treasury estimates. For example, multiplying the GAO estimate of hours saved for individuals by 45 percent to adjust for a lower participation rate assumed by the IRS, and then again by 25 percent for form completion time saved, would lower the estimated time savings in the GAO study to 17.5 million hours.¹⁰

The Treasury report also observes that a less-than-100 percent participation increases the estimated net cost to the government of a TAR program. Presumably this is because, when a large fraction of those eligible for a TAR fail to participate, the government incurs additional costs associated with processing taxpayer information, but fails to realize much of the savings that might be associated with providing taxpayers with pre-populated tax returns. Making an adjustment for this factor, and adding in costs of the sort described by Holtzblatt for the additional staff, equipment, and facilities required by the IRS and the Social Security Administration to process the information needed to prepare returns in a timely manner would further reduce, and perhaps eliminate the net savings in administrative costs estimated in the GAO report.

The Treasury report concludes that the two studies "highlight the uncertainty in estimating the changes in administrative and compliance costs" of adopting a TAR system, with estimates differing not only "in terms of magnitude, but also as to whether a return-free system would result in savings or costs to the federal government." In particular, neither study provides the information needed to assess whether adopting a TAR program at the federal level would lower overall compliance costs of the federal income tax, or instead shift these costs from individual taxpayers to the IRS and to third parties.

Third-Party Costs

Both the IRS and the GAO studies acknowledge that TAR proposals would impose additional reporting and compliance burdens on third parties -- employers, financial institutions and other payers of income to individuals, including government entities -- in moving toward more automated systems of tax compliance. Neither study, however, estimates those costs.

An important component of those costs to employers is preparing and sending wage data to the government. Timely and accurate wage data are essential to a system of government-prepared returns, as the 2003 Treasury Report noted. Wage data currently must be sent to taxpayers by January 31^st of each year but are not due to the federal government until the end of February, or if filed electronically, the end of March. Due to filing extensions, the IRS does not receive all W-2s until April 30.¹¹ IRS-prepared returns could not be provided consistently before April 15^th without moving up those deadlines, which would impose additional cost burdens on all employers and on smaller employers in particular. Taxpayers would have to be given additional time to review the government's returns and accept or reject them. This would substantially delay refunds for the vast majority of taxpayers who receive them.¹² Moving up reporting deadlines would also present problems with accuracy and increase the risks of error, as discussed below.

As noted by Joel Slemrod, Professor of Economics at the University of Michigan, there is virtually no systematic data for the United States on the costs incurred by third parties acting on behalf of individual taxpayers.¹³ The best that can be done is to gauge some plausible orders of magnitude based on limited research.

Vaillancourt (1989) estimated in a study of Canadian taxes that compliance costs incurred by employers for personal income taxes and payroll taxes equaled 3.5 percent of taxes collected.¹⁴ Applying this percentage to the total amount of personal income taxes collected in the United States in 2009 ($915.3 billion according to the Economic Report of the President) yields an estimate of federal third-party compliance costs of approximately $32 billion. If moving to a return-free system increased employers' compliance costs by 0.5 percent, the estimated added cost, would be roughly $1.6 billion per year.

The Canadian tax system differs from the U.S. system, so it is difficult to know whether that estimate is high or low. Another approach to assessing added costs for third parties is to start with the number of employers in the United States, and apply a rough estimate of added costs per employer. According to the U.S. Census, approximately 5 million firms in the United States have paid employees.¹⁵ If earlier reporting and more detailed data collection cost each employer $100 per year on average, their increased costs of compliance, , would come to $500 million per year. If the average cost per employer was $1000, the total would be $5 billion per year. Added costs would be felt disproportionately by small businesses that lack dedicated staff to handle more demanding processing and filing requirements.

In addition to employers, financial institutions that need to report investment income and government agencies that make transfer payments to individuals, such as the Social Security Administration and state agencies that provide unemployment compensation, would also incur added costs. This would be true for all relevant third parties, regardless of the share of taxpayers who might participate in a TAR system.

These estimates suggest that third-party costs are likely to be much larger than any changes in costs for individuals, , and the government, . Compliance burdens on private third parties would necessarily be passed on to others in the form of reductions in wages paid to employees or increases in prices to customers.

Estimates by Return-Free Filing Proponents

Proponents of adopting a federal TAR system have estimated the benefits to taxpayers but have said little about the costs. Drawing on IRS estimates of taxpayer burden, Austan Goolsbee (2006) estimates that providing pre-prepared tax returns to 1040A and 1040EZ filers would result in annual savings in out-of-pocket and monetized time costs of at most $1.6 billion.¹⁶ This estimate assumes that all 1040A and 1040EZ filers eligible to use a government-prepared return would choose to do so and would reduce their compliance time by 80 percent.¹⁷

The evidence suggests, however, that time savings and participation rates would be much lower. As discussed above, form completion time comprises roughly 25 percent of total compliance time; tax filers' time burden includes activities such as record keeping, tax planning, and checking as well. Although return-free filing proponents often cite enthusiastic responses from a user satisfaction survey conducted under the California's ReadyReturn initial pilot program, only 20 percent of those invited chose to participate in the pilot. Moreover, in the first and second years of the actual program (tax years 2007 and 2008), participation rates were much lower -- only 1.5 percent and 3.2 percent, respectively. California projected in April 2009 that roughly 8.0 percent of eligible taxpayers would participate in 2010.¹⁸

Adjusting Goolsbee's estimate by adopting the California FTB projection of an 8.0 percent rate instead of his assumed 100 percent participation rate and for a reduction in compliance time of 25 percent (which assumes all form completion time is saved) instead of 80 percent reduces estimated taxpayer cost savings from $1.6 billion to approximately $40 million.¹⁹ In order to assess whether implementing a federal TAR system would lower total compliance costs, these estimated taxpayer savings would need to be compared to the net change in government costs plus additional costs imposed on third parties. Goolsbee discusses some of the administrative steps that would need to be taken to implement a federal TAR and references the IRS and GAO reports discussed above as well as California's ReadyReturn pilot, but does not attempt to estimate third-party costs. Goolsbee recognizes that data processing and file transfer in the current federal system would have to be accelerated to avoid delays in refunds.²⁰ He acknowledges costs to third parties of a speedup in reporting and suggests that organizations that do not file electronically or are below a certain size could be exempt and "would probably [be given] a small tax credit to offset the minor inconvenience."

In testimony before the President's Economic Recovery Advisory Board (PERAB) Joseph Bankman stated that implementing a federal TAR system could save up to $2 billion in monetized costs, and change the way "20 million+ taxpayers view government."²¹ Although Bankman presented no specific support for his estimate, it is comparable in magnitude to Goolsbee's upper bound estimate of cost saving if a federal TAR were made available to and chosen by all eligible 1040A and 1040 EZ filers.

In an earlier, detailed analysis of issues that would need to be resolved in order to implement a federal TAR, Bankman (2008, p. 777-8) points out that the state of California gets wage data earlier than the federal government, and in cleaner form, which enabled that state to implement its ReadyReturn system.

Bankman observes that making pre-populated returns available at the federal level would require changing the date by which third party information is provided. He conjectures that the costs of changing timing in this manner would be minimal -- perhaps on the order of pennies per employee. However, the discussion of third-party costs above suggests that costs would likely be much higher.

California ReadyReturn Program

In 2005 and 2006 the California Franchise Tax Board (FTB) launched a pilot study of a voluntary TAR program offered to taxpayers with the simplest of tax returns. Based on the results of the pilot, the California FTB has moved forward to broader implementation of ReadyReturn.²² The California program is often cited by proponents of a federal TAR, so it is useful to examine the California experience in more detail.

Participation

Eligibility to participate in the initial pilot program was limited to taxpayers with the simplest tax returns: those with single filing status, no dependents, wages from a single employer, and wages from all four quarters of the year. In both years of the pilot, approximately 20 percent of those who were sent a prepared ReadyReturn participated, while 80 percent declined to do so.

Based on results from the initial pilot program, the FTB made ReadyReturn available to taxpayers meeting the following qualifications: (1) single or head of household filing status, (2) income only from wages, (3) only one employer, (4) no more than five dependents, (5) no credits other than the renters' credit, and (6) filing the standard deduction.

In the first year of implementation following the pilot (tax year 2007) 732,724 California taxpayers were eligible for ReadyReturn and 11,253 (1.5 percent) used the program. In the second year (tax year 2008), 1,936,741 taxpayers qualified and 62,134 (3.2 percent) used the program.²³ Final data have not been made available for tax year 2009, but the number of participants appears to have changed only slightly.²⁴ In its April 2009 report, the FTB projected that about 2 million taxpayers would be eligible in 2009, and that 4.5 percent (90,000 taxpayers) would participate, rising to 8.0 percent (160,000) in 2010.

Thus, participation rates in California have been much lower than those assumed in the IRS and GAO studies discussed above and significantly lower than the rates assumed by Goolsbee and Bankman. One reason for low participation may be that carefully checking a government-prepared return for completeness and accuracy requires much of the same compliance work as tax preparation from scratch. Also, increasingly widespread use of electronic tax preparation programs has lowered tax compliance time and costs for filers over time, thus reducing the value of government-prepared returns. More than 90 percent of federal taxpayers used either commercial software or paid preparers in 2009, compared with less than 75 percent in 2001. Almost 30 percent of taxpayers self-prepared their returns with software assistance in 2009, up from slightly over 15 percent in 2001.²⁵

Taxpayers may also be reluctant to participate on account of other issues discussed below, including retained liability for errors, as well as privacy and security concerns.²⁶ Another reason may be that that ReadyReturn does not prepare a federal return along with the California state return, thus insufficiently alleviating taxpayers' total tax compliance burdens.

Costs/Savings of ReadyReturn

After completing the pilot study, the FTB initially projected that actual implementation would entail net fiscal costs to the state of California. Those projections are presented in the Appendix and shown in Table A-1.

These initial cost projections, however, were revised and sharply reduced in the FTB's April 2009 report to the California legislature, which estimated that the program would reduce government expenditures on net. Specifically, the report estimates that in 2009 ReadyReturn would serve 90,000 taxpayers at a budgetary saving of $77,000, and in 2010 would serve 160,000 taxpayers at a budgetary saving of $235,000.

Those estimated budgetary savings seem to stem from two related factors. First, the FTB is using web presence combined with a public information campaign to alert taxpayers to their potential eligibility for ReadyReturn. For example, taxpayers who are about to file their return can learn whether they qualify for ReadyReturn by going to a Franchise Tax Board website. Taxpayers who are eligible then have the option to review their Ready Return on-line or request that the ReadyReturn be sent to them by mail. The principal source of estimated savings, however, is the implicit assumption that each taxpayer choosing the online ReadyReturn option would otherwise have filed a paper return rather than an electronic return. A ReadyReturn filing costs the state $2.25 less than a paper return.

To the extent that taxpayers are induced by ReadyReturn to file electronically instead of on paper, it is appropriate to attribute such savings in processing costs to the ReadyReturn program. However, these savings appear to result from filing electronically, rather than specifically from the ReadyReturn program. This suggests that savings in federal tax system costs could be achieved by other means of facilitating electronic filing than government preparation of returns.

Increasing the proportion of individual tax returns filed electronically is consistent with current IRS procedures and goals and would substantially streamline tax administration and result in significant budget savings. Approximately 69 percent of taxpayers filed their returns electronically in 2009, rising to 72 percent in 2010.²⁷ The IRS goal of an 80 percent e-filing rate was originally promulgated in 1998, to be achieved by 2007.²⁸ The current goal is to achieve 80 percent e-filing by 2012.²⁹

California has extended ReadyReturn eligibility from single taxpayers to heads of household, but still limits participation to those with only wage income from one employer, no credits other than the renters' credit, and who use the standard deduction. Although the IRS would face challenges in implementing a simple TAR system patterned along the lines of California's pilot ReadyReturn, advocates of a federal return-free system support extending it to a broader group of taxpayers who file more complicated returns.

Goolsbee proposes implementing a federal TAR in three waves, representing different groups of taxpayers. The first wave would comprise taxpayers with attributes similar to those who were eligible to participate in the ReadyReturn pilot program: single filers with no dependent children who are not dependents themselves, had only wage income on their last return, have no other credits, and did not itemize in previous tax filings.³⁰ Presumably this simplest group also would be restricted to filers with one employer and no change in family status during the year.

Goolsbee's first wave also includes a second group -- married filers and those with dependents. That extension would require employees filling out W-4 forms to provide additional information about themselves and their family status to their employer, such as a spouse's Social Security number.³¹ Progressive second and further waves of eligible filers would include increasingly more complex returns with sources of income such as that reported on 1099s and from government entities making transfer payments such as the Social Security system and state unemployment benefits systems.

Collecting additional data about employees' family and economic circumstances would impose further burdens on the government and employers and would raise further privacy and security concerns, discussed below. In the case of including family information in his expanded version of "Wave 1" of implementation, Goolsbee acknowledges that some people may not want their employers to have detailed personal information about themselves or their family status. He suggests that employees could provide the required information on a postcard mailed directly to the IRS, avoiding the need for their employer to see or record it. Such separate reporting to the IRS by filers, however, would present potential data matching problems for the agency and obvious privacy and security problems, especially if the information were sent in by postcard. Data matching would also become progressively more costly and complex with risks to privacy and security increasing with additional types of eligible filers.

As has been discussed above, there is little information on what it might cost the federal government and other parties to implement a ReadyReturn-like "Wave 1" program. Whatever those uncertainties, however, the costs of a return-free system that included more complex returns can be expected to be higher.

Discussion of the Estimates

We have presented a simple framework for assessing how a TAR system would affect total compliance costs. Only fragmentary evidence exists, however, on the relative magnitudes of the major cost elements.

In equation (1), presented above, is the change in total compliance costs, and is the sum of the change in government compliance costs, , the change in compliance costs for third parties, , and change in compliance costs for individuals, .

 

 

California's ReadyReturn program appears to have reduced processing and compliance costs for both the state, , and for some taxpayers,, but the net government savings are largely attributable to savings from e-filing, rather than to the return-free system itself. Comparable federal savings might be achieved by simply encouraging electronic filing, which is current IRS policy. As already mentioned, the agency is working toward the goal of an 80 percent e-filing rate by 2012.

The IRS would face much greater challenges in attempting to implement a TAR system along the lines of ReadyReturn. Importantly, the IRS lacks the electronic data processing capabilities that California has developed and the scale of IRS operations is vastly greater than that of California. While the IRS has had to continue reliance on its older master file system, implementation of ReadyReturn in California was facilitated by a computerized system that contains detailed income records, the Enterprise Customer Asset, Income and Return (ECAIR) Data Warehouse.³² The IRS business systems modernization project has experienced delays and cost overruns for many years, and Congress has been concerned that the IRS was redirecting funding for modernization activities to other priorities.³³ The agency, moreover, faces significant new responsibilities under health reform.³⁴

Although many of the survey respondents in the original pilot group were enthusiastic about ReadyReturn, the low participation rates in the actual program suggest that the vast majority of eligible filers do not perceive that they would be advantaged by using it. Participation rates were 20 percent in the ReadyReturn pilot program; 1.5 percent in the first year of implementation; and 3.2 percent in the second year. Such low participation rates call into question whether individuals generally believe they would save time and money if they used the system. Return-free federal filing is generally envisioned only for filers with the simplest federal returns, IRS 1040A and 1040EZ. Under the complex U.S. tax code, a TAR system is not likely to cover taxpayers with more complicated returns, such as IRS 1040 filers.

Analysts and return-free filing proponents agree that third parties would bear additional costs, . Our analysis shows that increases in third-party compliance costs (i.e. to employers and other payers of income) could be significant, ranging from $500 million to $5 billion each year. Third-party costs could dominate other changes in compliance burdens under a federal TAR system and increase the economy-wide costs of compliance with the individual income tax, . Employers would have to report their employees' wages to the government earlier than they do now and they would have to collect more demographic and earnings data from their employees on W-4 forms in order for tax filing under a broadly-based TAR to proceed more or less on the current schedule and to enable tax refunds to be issued promptly. In the absence of earlier reporting, tax refunds would be delayed by several months.

OTHER ISSUES FOR GOVERNMENT AND TAXPAYERS

The objective of reducing total tax administration and compliance costs is a central benchmark for evaluating proposals for return-free filing. It is, however, not the only consideration. Return-free filing would introduce a number of other concerns including competition with the private sector and effects on innovation in tax preparation technologies, conflicts of interest, increased risks of errors on returns, and taxpayer privacy and data security.

Competition with the Private Sector and Effects on Innovation

Preparing individuals' tax returns raises issues of the government's role in a market economy. It goes without saying that the government is intimately involved in the taxation process and should employ information technologies to make that process efficient. Government involvement in personal income tax preparation, however, would reduce competition in existing markets, violating a "red light" principle set forth by respected economists Joseph Stiglitz and Peter Orszag.³⁵ It could also have the unintended effect of reducing innovation in rapidly evolving software markets.³⁶

In the United States, income tax preparation has traditionally been the responsibility of individuals, many of whom are assisted by private tax preparers and firms, which offer taxpayers an expanding array of electronic products and services.³⁷ Free assistance is available to lower-income taxpayers through the Free File program, sponsored since 2002 by the IRS and 21 states, which provides both federal and state income tax preparation and electronic filing through a partnership between the IRS and private sector tax software companies. Taxpayers are eligible for Free File services if their adjusted gross income is below a certain level; for the 2009 tax year the eligibility level is $57,000. Many interactive forms are available to higher income filers, including a calculator that checks for arithmetic errors. In 2009, approximately three million taxpayers made use of the Free File program.³⁸ While the program is sponsored by government, the returns are prepared by the taxpayer and the services are provided by private companies under rules governed by the IRS.

Direct government involvement in preparing tax returns can affect incentives for private investment in tax compliance technology. In principle, it could increase incentives for investment if private preparers try to induce taxpayers to use their services rather than the government's services, or decrease investment if the pool of potential customers shrinks.

The evidence to date suggests that return-free systems reduce investment incentives. Critics of the UK's PAYE system have noted that, unlike the United States, the UK has had relatively little private investment in tax software or related innovation.³⁹ A recent example illustrates the point. In 2010, the financial software firm Intuit released an iPhone application called SnapTax that allows taxpayers to prepare tax returns on their mobile device. Whether this particular "app" will be widely used remains to be seen. However, it is indicative of the incentives that exist to spur the private sector to develop information technologies for simplifying tax compliance in the United States. Based on the UK experience, such incentives would largely be absent if the United States were to move to a full-blown exact-withholding system; they would likely be reduced if not eliminated if tax returns were prepared by the IRS on a wide scale. Put somewhat differently, the more the government becomes directly involved in the direct preparation of tax returns, the more the responsibility for innovation in the development of tax compliance technologies would fall to the public rather than to the private sector.

Conflict of Interest

The Stiglitz-Orszag-Orszag study cited above raises the question of whether a conflict of interest exists if a tax agency functions as both tax preparer and enforcer.⁴⁰ For example, IRS preparation of tax returns would make it possible for the agency to target for audit those filers who overrode its calculation and most sharply reduced their reported tax liability below the IRS's calculation.⁴¹ Many respondents to the Treasury's survey reported conflict-of-interest concerns.⁴²

The potential conflict between the desire of the IRS to maximize tax revenues and the desire of taxpayers to take advantage of the full range of deductions and credits available to them is likely to be more than merely hypothetical. Under the United Kingdom's return-free system, the tax authority has on occasion pointed out overpayments but in general appears to disproportionally allocate resources to cases of underpayments.⁴³

Relationship of Taxpayers to the Government

Another issue that policymakers should weigh in deciding whether the federal government should prepare tax returns is whether individuals would, as a result, become less engaged as taxpayers and citizens and whether they would become less cognizant of their personal financial affairs.

Many American taxpayers have little awareness of their federal income tax obligations and little knowledge of the incentives embodied in the tax code. Of approximately 155 million tax returns filed for tax year 2007, the vast majority had overpayments refunded. Because of extensive withholding of earned and other income, only a small minority of filers had taxes due at the time of filing.⁴⁴ Many people barely glance at their periodic withholding statements, W-2, or 1099 forms, and for most filers the expectation of a tax refund is their most concrete interaction with the tax system.

Some advocates of a return-free filing suggest an analogy with property tax bills sent out by local governments. When their bills arrive in the mail, property owners typically pay them with a minimum of fuss and few would say that the near-zero compliance burden makes them less engaged in property tax assessments or local government budget issues. However, there are important differences between property tax bills and prefilled income tax forms. Property owners may challenge an assessment they believe to be too high but are under no obligation to point out an assessment they believe to be too low. In the case of government-prepared income tax returns, taxpayers would retain responsibility for truthful and complete reporting of their income and legal liability for inaccuracies and omissions.

In addition to the issue of tax salience and citizen engagement with the tax system, government-prepared returns raise the concern that taxpayers using it would become less knowledgeable about their personal finances. Nina Olson, the IRS National Taxpayer Advocate, has explained the role that annual tax preparation plays in personal financial management: ". . . for many, many individuals [preparing their taxes] is the only time that they ever really sit down and look at what happened to them financially over the last year. And maybe -- I wouldn't want to lose that in a return-free system because I think that for the broader health of the country -- the financial health of the country, that is an important ritual."⁴⁵ Eric Toder, of the Tax Policy Center, agreed that that there are positive aspects associated with people preparing their returns and knowing what taxes they are paying.⁴⁶

In a related vein, Kay and King (1990) described the British PAYE system as one that few people understand. This point was recently reaffirmed by the All Parliamentary Taxation Group, which concluded that, after about 60 years, the return-free system of the United Kingdom has resulted in many taxpayers absenting themselves from compliance activities and lacking awareness of their tax obligations.⁴⁷

William Frenzel of the Brookings Institution and a member of the 2005 President's Advisory Panel on Tax Reform expressed concern that government preparation of tax returns would be inconsistent with traditional resistance to "big-brotherism" in the United States.⁴⁸ The 2003 Treasury survey, cited above, corroborates that view: about 70 percent of respondents agreed completely or agreed somewhat with the statement "It gives the government too much control of your life."⁴⁹

Risks of Error

No system is entirely free from errors, of course, and a system of tax returns prepared by the IRS would be expected to make its share, particularly during startup. The 2003 Treasury study noted, "accelerated processing of information returns might increase errors, while late filings of information by employers and other payers could result in incomplete and inaccurate tax returns."⁵⁰ Indeed, in January 2009, the IRS moved back the due date for payers' 1099-B reports from January 30 to February 15, in order to reduce the error rate in those forms.⁵¹ The Electronic Tax Administration Advisory Committee warns that accelerating reporting obligations will result in increased errors in W-2 and 1099 Forms and notes that many companies already struggle to meet the current deadlines for processing and electronic filing.⁵²

Goolsbee has argued that some mistakes, such as the wrong amount of income reported on a taxpayer's W-2 form, could be corrected more quickly in a return-free system.⁵³ Errors in a centralized system, however, can also be more severe and widespread than in a decentralized one because they can affect many people at the same time.

To the extent that tax refunds could be delayed, tax information could be misdirected, or IRS forms sent to filers could be erroneous in some systematic way, consequences would be serious in terms of lost confidence in the system. For example, although a primary motivation for boosting the use of electronic medical records in the American Recovery and Reinvestment Act of 2009 (ARRA) was to reduce medical errors, there have been reports of computer systems increasing errors, adding to doctors' workloads, and compromising patient care.⁵⁴ Other examples are the 2008 crash of the website of the United Kingdom's centralized taxation authority, which halted online filing of tax returns for tens of thousands of taxpayers, requiring the filing deadline to be extended,⁵⁵ and, as noted above, widespread errors recently discovered in PAYE tax calculations that affect millions of UK taxpayers.

Various independent audit reviews of IRS processes raise questions about the ability of the agency to carry out new responsibilities with reasonable accuracy. For example, the Treasury Inspector General for Tax Administration determined that the IRS is unable to verify eligibility for the majority of ARRA benefits at the time a tax return is processed.⁵⁶ The Inspector General in another report indicated that individual taxpayer identification numbers are being issued to alien individuals without sufficient supporting documentation.⁵⁷ The National Taxpayer Advocate, in her annual report to Congress, reported that an expanding slate of duties is stretching IRS capacities, resulting in poor customer service and undermining its ability to collect taxes owed.⁵⁸ Since that report, the IRS has assumed substantial new administrative responsibilities under health reform.⁵⁹

Taxpayer Liability and Behavioral Analysis

When faced with a choice, people will tend to make the default or easier selection.⁶⁰ This means many are likely to sign and send in a return filled in by the IRS, whether or not it is correct. Taxpayers, however, would retain responsibility for accuracy and completeness of their returns, creating problems for people who unwittingly submit erroneous returns. Low-income filers present particular concerns with regard to Earned Income Tax Credit (EITC) errors, take-up rates for refundable credits, and recovery of erroneous refunds.

If a taxpayer realizes his IRS-prepared return shows insufficient tax liability, i.e., he is aware of an error or omitted source of income, he may be likely nevertheless to take the path of least resistance and sign it, exposing himself to future liability and significant IRS penalties and legal actions. Taxpayers who submit returns that show tax liabilities that are too low could be subject to fines and legal actions, even though they accepted the government's calculations. Acknowledging this problem, Goolsbee suggests the IRS could establish a safe harbor level, perhaps $250.⁶¹ Thus, if a taxpayer failed to report certain income or a change in family status, and the resulting tax owed was below $250, the IRS would not impose a penalty and the taxpayer would only be responsible for the tax owed plus interest.

On the other hand, a taxpayer may receive a return that is erroneous because it indicates excessive tax liability. An IRS-prepared return could reduce the extent to which a taxpayer takes full advantage of ways to reduce his tax liability. This could occur for several reasons, including incorrect deductions or failure to include tax credits for which the taxpayer is eligible, including the Earned Income Tax Credit. In addition to the simple appeal of choosing the default option rather than making his own calculations and populating his own tax form, a taxpayer might feel intimidated by a personal document sent to him by the IRS even if it is erroneous or incomplete, and for that reason fail to reject it.

In principle, accepting a government-prepared return would presumably be voluntary and taxpayers could simply override it. Goolsbee reports that evidence from California shows many people did override the ReadyReturn and calculated taxes on their own. An official document prepared by the U.S. Internal Revenue Service, however, would carry more weight and potential enforcement clout than one prepared by the California tax authority and may be more likely to be accepted as a default, even if it shows an excessive amount of tax owed. Evidence from the United Kingdom's return-free PAYE system shows that taxpayers are often unwilling to challenge the tax-authority's calculations because they do not wish to call attention to their affairs.⁶²

In many cases, however, taxpayers are unlikely to know whether the government's calculated tax liability is too high or too low without keeping records and doing some work on their own. Thus, determining whether the government's estimates are correct would require taxpayers to continue calculating their own taxes, eliminating or substantially reducing the potential benefits of having the forms prefilled in the first place.

Low-income filers and those with English as a second language may be particularly intimidated by official IRS documents and thereby disadvantaged in a return-free system, as pointed out by Robert Greenstein of the Center on Budget and Policy Priorities.⁶³ Greenstein notes that EITC eligibility depends on complex determinations that can be difficult to gauge accurately with return-free filing, especially since low-income filers have more frequent family status changes than do upper-income filers. Low-income taxpayers may also be less likely to correct erroneous pre-populated tax forms or claim tax credits generally for which they are eligible. Recovering erroneous refunds would be particularly burdensome for low-income filers, especially in view of the fact that there is considerable evidence that many taxpayers plan for and count on tax refunds, and seem to like being overwithheld.⁶⁴

Privacy

Federal preparation of tax returns would also introduce risks to privacy, which would increase with broadened eligibility. Advocates of return-free filing state that taxpayers would not have to disclose any more information about themselves to their employers or the government than they provide now, nor would the IRS receive more information than it currently gets about wages, federal benefits, investment income, or family status.⁶⁵ Additional information the government could potentially garner about taxpayers, however, would come from comparing the IRS-prepared returns with those taxpayers actually send in, and such information could be and has been used by state tax agencies for purposes such as audits and return analysis.⁶⁶

Prefilled returns can be delivered by mail or online. If the IRS mailed out millions of forms, some would undoubtedly get delivered to the wrong address.⁶⁷ Electronic forms can also be misdirected or caught up in spam filters. Although under the current system mailed information such as W-2 forms and Social Security lifetime earnings histories can also be delivered to the wrong address, tax forms present greater privacy and security risk than do those documents. The government's tax forms would combine information from a number of sources and would contain multiple personal identifiers. If heads of household were included as well as single filers, as they are under ReadyReturn, information on spouses' incomes would have to be matched and combined.

In addition, to the extent the success of a return-free system depends on prompt and accurate employer withholding in order to get refunds out on time,⁶⁸ many employees would need to reveal more information than they do now about their family and financial circumstances for withholding purposes.

Security

Security and privacy are related but not identical issues. Privacy generally relates to intended uses of data whereas security issues are about unintended uses of information such as identity theft, blackmail, extortion, and embarrassment.⁶⁹

In terms of technical security for electronic returns, privately offered tax preparation software and returns prepared by government tax authorities can provide common measures such as the Secure Sockets Layer protocol for Web communications.⁷⁰ To the extent that pre-populated tax returns require the matching of data from various sources, however, such as filers' previous returns, heads of household and spouses' information, earnings from capital income, and other reports from employers, banks, financial firms, and other government sources such as Social Security benefits, and earnings and benefits information from state unemployment authorities, additional weak spots in security may emerge. Each point of access to data constitutes potential for a security breach.⁷¹

In addition to potential problems with system security, threats arise from human error. A corporate official of Symantec, a major Internet security firm, observed that "98 percent of the data loss is through mistakes of human error and process breakdown."⁷² Such threats require investments in time and money to minimize, if not eliminate, security and other breaches.

Some analysts believe that government tax-preparation systems may pose greater security risks than private tax preparation software because private providers face stronger financial incentives to invest in sound security practices and the maintenance and improvement of security systems in an environment of rapidly advancing technology and changing threats. Credit card firms, for example, must constantly update and improve their security systems.⁷³

Private providers face the risk of being put out of business or substantially harmed by security breaches and are clearly liable for resulting harms.⁷⁴ Government agencies face administrative and political sanctions, but recent experience suggests that such incentives may be not be as effective as one would like. In a data breach at the Department of Commerce on December 4, 2009, employees' names and Social Security numbers were released on the Internet; employees received notice of the breach by mail more than seven weeks later. According to the letter to employees, "a Department of Commerce employee inadvertently transmitted over the Internet a file containing the PII [Personally Identifiable Information] of Commerce employees to other employees."⁷⁵

Federal law sets a bar for security in industry, which must be implemented in a practical way. The Federal Trade Commission requires a "reasonable" level of data security through the FTC Act and the Safeguards rule;⁷⁶ the agency brings enforcement actions against firms when consumers' financial data are compromised.⁷⁷ In addition to federal data security requirements, most states have laws requiring that individuals be notified of security breaches. The California law, implemented in July 2003, was the first of its kind in the nation.⁷⁸

Private firms learn from the mistakes of others more readily than do government agencies as they are penalized in their stock price (sometimes substantially so), by the loss of business due to decreases in consumer trust, and by losses of intellectual property after a data breach. They also incur the costs of hiring forensic experts and notifying customers. Companies that experienced a data breach paid out an average of $6.6 million in 2008 to restore their brand image and help retain customers, according to a study by the Ponemon Institute, a research firm based in Tucson, Arizona.⁷⁹

At a Senate Committee on Finance hearing in 2008 on Identity Theft in Tax Administration,⁸⁰ National Taxpayer Advocate Nina Olson expressed concern that the IRS does not know how many taxpayers are impacted by identity theft and only recently began to track incidents that are reported and documented. She also testified that the agency lacks a coordinated effort to address identity theft issues.⁸¹ Concerns about security were also expressed in an IRS Oversight Board report on electronic filing: ". . . the potential for identity theft raises still more difficulties for the IRS to achieve its goals of a secure and smooth-flowing electronic tax administration environment. . . . The danger of identity theft discourages some taxpayers from considering e-file and the prospect of engaging in other electronic tax interactions."⁸²

The Government Accountability Office in March 2010 pointed out longstanding weaknesses in IRS information security systems and concluded that the agency had not consistently implemented controls intended to prevent and detect unauthorized access to its systems and information.⁸³ The Treasury's Inspector General for Tax Administration, in its fiscal 2009 audit issued March 31, 2010, found that the IRS failed to implement adequate security measures to protect sensitive data that tax professionals entered into a Web portal.⁸⁴ Auditors noted that the agency failed to establish strong password controls and to analyze audit logs to detect unauthorized access.

RETURN-FREE FILING, TAX REFORM, AND THE POLITICAL ECONOMY OF TAX INCENTIVES

A return-free system is most easily implemented for taxpayers with relatively simple tax returns; and, as the OECD and others have found, countries that have adopted various forms of return-free filing have relatively simple income tax structures, especially as compared to the United States. Moreover, countries with return-free filing also tend to make less use than does the United States of income tax incentives as a means of implementing social policies. As noted by Gale and Holtzblatt (1997) in comparing the United States with the British system, such differences have potentially important implications for "both the structural and the administrative features of tax policy."

One important example is that of the Earned Income Tax Credit (EITC), which is widely regarded as a cornerstone of American welfare policy to assist the working poor. Presently, those eligible for the Earned Income Tax Credit receive benefits through the federal income tax. In Britain, a similar EITC-like credit is administered as a spending program by the UK tax authority. Errors in administering the Family Work Credit through PAYE resulted in Her Majesty's Revenue and Customs creating the requirement that families submit documentation of their eligibility, which the government then uses as input to preparing their tax return. As noted above, the required forms and instruction booklet resemble those for the IRS 1040 and introduce a significant taxpayer compliance burden that the return-free system had been designed to obviate.

Gale and Holtzblatt argue that a voluntary TAR system would still allow the EITC to be administered through the income tax. But some tax experts who are otherwise sympathetic to the objective of reducing tax filing burdens for lower income taxpayers caution that "simplified filing would need to be implemented carefully in order to avoid large increases in EITC errors or large reductions in take-up rates for refundable credits."⁸⁵ Thus, an indirect, but significant potential cost of implementing a TAR system would be added complexity and costs of administering the EITC and similar transfer incentive programs, such as the child tax credit, which are administered through the federal income tax.

Gale and Holtzblatt also observe that implementing pre-populated returns on a broader scale would make it extremely difficult to provide other financial incentives, such as the charitable tax deduction, through the income tax. They argue, however, that such incentives could still be provided in a form compatible with pre-populated returns. For example, instead of allowing individual tax deductions for charitable contributions, contributions in the UK are directly matched by the Treasury. While such adjustments are possible in principle, they would require significant changes in the structure and administration of a number of fiscal incentives.

The widely-acknowledged consideration that return-free filing would be much easier to implement in conjunction with a simplified income tax has led some proponents to suggest that moving toward return-free filing might create political pressures to make the income tax simpler and more streamlined.⁸⁶ Constraining the role of the income tax as a tool of social policy would require legislators to substitute more transparent and controllable spending policies for tax incentives.

As a number of scholars have noted, however, the American political culture and legislative process favor implementing economic and social policy through the tax system.^87,88 Because the U.S. penchant for using the tax code for social purposes is rooted in deeply embedded culture and political structures, a broader consequence of return-free filing could be to increase the difficulty and complexity of devising and implementing economic and social policies through the tax system.

CONCLUSION

Reducing the burden of complying with the federal income tax can yield potentially significant benefits both to individual taxpayers and to the government. Simplifying the complex U.S. tax code is the most direct way to reduce the costs of collecting federal income taxes, both public and private. Some analysts have argued that return-free filing systems like those used in other countries and in the state of California could substantially reduce compliance costs for many individual taxpayers with relatively simple returns, at little or no net administrative cost to the IRS, and might in addition create political pressures for tax simplification.

There is substantial uncertainty as to how a federal return-free system would affect the costs of government and individual tax filers. The federal government would need to upgrade its electronic systems and make significant additional investments in staff, equipment, and facilities. California's program appears to have reduced administrative costs for the state, but the net savings are largely attributable to e-filing, which some ReadyReturn users may have already been using, rather than to the return-free system itself. The vast majority of eligible filers in California have declined to make use of the return-free option. The most recent data show a participation rate of only 3.2 percent, suggesting that most people believe any savings they might realize in time and money would be outweighed by the costs, including concerns about privacy and security.

Return-free filing would increase third-party tax compliance costs -- those of employers, financial institutions and other payers of income to individuals -- because reporting deadlines would have to be advanced in order to provide timely returns and tax refunds. Although systematic analyses of third party compliance costs for the U.S. income tax system are lacking, calculations of such costs based on studies from other countries yield estimates ranging from $500 million to perhaps as much as $5 billion, offsetting or exceeding any savings for individuals or for government.

Return-free filing raises other policy concerns including a conflict of interest for the IRS in functioning as both tax preparer and enforcer. Further, the government would enter into competition with the private sector, potentially discouraging further private investment and innovation in tax software and electronic taxpayer assistance, which have already substantially reduced individuals' filing costs despite increases in tax complexity.

The IRS would face challenges in attempting to implement return-free filing, in part because it lacks the essential electronic processing capabilities and timeliness of data reporting by third party sources. Risks of error would increase as a result of compressed income reporting schedules, and reviews of IRS processes raise questions about the ability of the agency to carry out new responsibilities with acceptable accuracy, particularly as its mission is expanding under health reform. Default problems could arise when filers accept a government-prepared return, even when it may be erroneous, and expose themselves to future liability and penalties. Low-income filers and those with English as a second language present particular concerns with regard to the Earned Income Tax Credit (EITC) and reimbursing erroneous refunds to the IRS.

IRS preparation of individual tax returns could make filers less engaged as taxpayers and thus less cognizant of their personal financial affairs. It could also compromise taxpayers' privacy, particularly if eligibility were broadened to include filers beyond those with the simplest returns, because employees might have to reveal more information than they do now about their family and financial circumstances for withholding purposes. Security concerns arise because government-prepared returns entail matching data from various sources and each point of access to data poses potential for a security breach. Further, government systems may pose greater risks than private tax preparation software because private firms have strong financial incentives to strengthen security in an environment of rapidly advancing technology and changing threats.

In light of this analysis, we conclude that adopting a return-free tax preparation system is not an advisable course of action for the federal government. Any cost savings for government and individuals would likely be modest at best and additional costs to employers and other payers of income could be substantial. Moreover, a return-free system would introduce numerous challenges and issues for the federal government and the taxpayer.

Implementing ReadyReturn was initially projected to entail one-time start-up costs of $241,000 and on-going annual costs of $201,000, $366,000, and $449,000 in 2007, 2008, and 2009, respectively. These added costs of implementing and administering the ReadyReturn program were projected to be partially offset by reductions in filing costs of $96,000, $223,000, and $301,000 in those three years. After netting out these cost savings against public implementation and administrative costs, the FTB projected that the net public cost of making ReadyReturn available to eligible California taxpayers would be $346,000 in 2007, $143,000 in 2008, and $148,000 in 2009.

Taxpayers who responded to surveys administered during the pilot program reported a reduction in form completion time of approximately 30 minutes compared with both non-participants and the control group. Evidence on out-of-pocket savings from participating in ReadyReturn was mixed. Those who participated in the ReadyReturn pilot reported costs of return preparation of roughly $30 compared with members of a control group (who were not invited to participate in ReadyReturn). At the same time, taxpayers who were invited to participate in ReadyReturn but who declined to participate reported out-of-pocket tax preparation costs of approximately $0.

                           Table A-1

All Party Parliamentary Taxation Group (APPTG). 2008. "The Future of Income Tax Administration in the UK." May. http://www.techpolicyinstitute.org/files/PPRMay08.pdf

_____. 2009. "Pre-Populated Returns." Briefing Paper, September. http://www.techpolicyinstitute.org/files/PPRSep09.pdf

____. 2010. "Improving PAYE: How information is collected and managed in the administration of Income Tax." March. http://www.appgtaxation.org/APPG_Tax__Report_--_Improving_PAYE.pdf

Bankman, Joseph. 2005. "Simple Filing for Average Citizens: The California ReadyReturn." Tax Notes, Vol. 107, No. 11, pp. 1431-1434. Tax Analysts.

_____. 2008. "Using Technology to Simplify Individual Tax Filing." National Tax Journal, Vol. LXI, No. 4, Part 2, pp. 773-789.

Chetty, Raj. 2009. "Is the Taxable Income Elasticity to Calculate Deadweight Loss? The Implications of Evasion and Avoidance." American Economic Journal: Economic Policy. Vol. 1, No. 2, pp. 31-52.

Franchise Tax Board (FTB), State of California. 2006a. "ReadyReturn Pilot Tax Year 2004 Study Results." April. http://www.ftb.ca.gov/ReadyReturn/TY04RRFinalReport.pdf

_____. 2006b. "ReadyReturn Pilot Preliminary Studies for Tax Year 2005." September. http://www.ftb.ca.gov/law/meetings/attachments/092006/2.pdf

_____. 2009. "Report to the Legislature." April. http://www.ftb.ca.gov/readyReturn/ReadyReturnReport2009.pdf

Feldstein, Martin. 1999. "Tax Avoidance and the Deadweight Loss of the Income Tax." Review of Economics and Statistics, Vol. 81 No. 4, pp. 674-680.

Gale, William G. 2009. "Remove the Return." Toward Tax Reform: Recommendations for President Obama's Task Force. Tax Analysts.

Gale, William G. and Benjamin Harris. 2007. "Return-Free Filing." The Tax Policy Briefing Book: A Citizen's Guide for the 2008 Election and Beyond. Tax Policy Center, December.

Gale, William G. and Janet Holtzblatt. 1997. "On the Possibility of a No-Return Tax System." National Tax Journal, Vol. L, No. 3, pp. 476-484.

Goolsbee, Austan. 2002. "The Turbo Tax Revolution" Evaluating the Ability of Technology to Solve the Tax Complexity Dilemma." Draft, January. http://gsbwww.uchicago.edu/fac/austan.goolsbee/research/turbotax.pdf

______. 2006. "The 'Simple Return': Reducing America's Tax Burden Through Return-Free Filing." The Hamilton Project, Brookings Institution, Discussion Paper 2006-04, July. http://www.brookings.edu/~/media/Files/rc/papers/2006/07useconomics_goolsbee/200607goolsbee.pdf

Harper, James W. and Thomas M. Lenard. 2002. "Online Tax Preparation: Beyond the Bounds of E-Government." Progress on Point 9.13, Progress and Freedom Foundation, April. http://www.pff.org/issues-pubs/pops/pop9.13onlinetax.pdf

Holtzblatt, Janet. 2006. "Implications of Return-Free Tax Systems for the Structure of the Individual Income Tax." Andrew Young School of Policy Studies. http://aysps.gsu.edu/isp/files/ispwp0630.pdf

IRS Oversight Board (IRS). 2010. Electronic Filing 2009, Annual Report to Congress, January. http://www.treas.gov/irsob/reports/2010/IRSOB_Report_2009%20EFile.pdf

Kay, John A. and Mervyn A. King. 1990. The British Tax System, 5th edition. Oxford: Oxford University Press.

Organisation for Economic Co-Operation and Development (OECD). 2006. "Using Third Party Information Reports to Assist Taxpayers Meet their Return Filing Obligations -- Country Experiences With the Use of Pre-populated Personal Tax Returns," Prepared by Forum on Tax Administration Taxpayer Services Sub-group, March. http://www.oecd.org/dataoecd/42/14/36280368.pdf

Slemrod, Joel. 1990. "Optimal Taxation and Optimal Tax Systems." Journal of Economic Perspectives, Vol. 4, No. 1, pp. 157-178.

Stiglitz, Joseph E., Peter R. Orgszag, and Jonathan M. Orszag. 2000. The Role of Government in a Digital Age. October. http://archive.epinet.org/real_media/010111/materials/stiglitz.pdf

U.S. General Accounting Office (GAO). 1996. Tax Administration: Alternative Filing Systems. GAO/GGD-97-6, October. http://www.gao.gov/archive/1997/gg97006.pdf

U.S. Government Accountability Office (GAO). 2005. Tax Policy: Summary Estimates of the Costs of the Federal Tax System. GAO-05-878, August. http://www.gao.gov/new.items/d05878.pdf

U.S. Department of the Treasury. 2003. Report to the Congress on Return-Free Tax Systems: Tax Simplification is a Prerequisite. December. http://www.treas.gov/offices/tax-policy/library/noreturn.pdf

U.S. Department of the Treasury. Internal Revenue Service. 1987. Current Feasibility of a Return-Free System. October.

Vaillancourt, François. 1989. The Administrative and Compliance Costs of the Personal Income Tax and Payroll Tax System in Canada,1986. Canadian Tax Paper 86. Canadian Tax Foundation, as referenced in Jason Clemens, ed. 2008. The Impact and Cost of Taxation in Canada: The Case for Flat Tax Reform, Chapter 3, p. 61.

¹ Feldstein (1999). Chetty (2009), however, shows that when a distinction is made between responses to the income tax that represent income shifting, vs. changes in labor supply and/or saving, the estimated deadweight loss is apt to be considerably lower than that estimated by Feldstein.

² S. 3018, for example, introduced February 23, 2010, by Senators Wyden, Gregg, and Bond, would require the IRS to provide taxpayers a simplified Easyfile pre-prepared return after January 1, 2011.

³ The President's Economic Recovery Advisory Board, "The Report on Tax Reform Options: Simplification, Compliance, and Corporate Taxation," August 2010. http://www.whitehouse.gov/sites/default/files/microsites/PERAB_Tax_Reform_Report.pdf

⁴ For discussions pro and con, see, for example, William Gale and Benjamin Harris, "Return-Free Filing: What are the drawbacks"" The Tax Policy Briefing Book, Ways to Improve the Tax System: Return-Free Filing, December 14, 2007, p. III-5-5; and Jeffrey Eisenach, Robert Litan, and Kevin Caves, "The Benefits and Costs of Implementing 'Return Free Filing' in the United States," Computer and Communications Industry Association, March 2010.

⁵ See HM Revenue & Customs, Getting your tax credits claim form right. http://www.hmrc.gov.uk/forms/tc600-notes.pdf

⁶ Andrew Porter, "Tax Debacle," UK Daily Telegraph, September 4, 2010.

⁷ 1040 EZ Instructions 2009, Estimated Average Taxpayer Burden for Individuals by Activity, p. 36.

⁸ The Treasury study surveyed potentially eligible taxpayers in 2000 to assess their interest in voluntarily participating in a federal TAR program and found most respondents were not interested. Thirteen percent of respondents said they would "definitely" be interested and 26 percent said they would "probably" be interested. Respondents were also asked how much they would be willing to pay for the option to have the IRS provide them with a pre-populated return, in which: "...you would still receive a refund or owe taxes at the end of the year, but you would receive a form from the IRS showing how much taxes they calculated for you. You would then send the form back to the IRS showing whether or not you agreed with their computations." Just over half of respondents indicated they were willing to pay $0; 12 percent were willing to pay $10; 15 percent $25; 6 percent $50; 2 percent $100; and 10 percent did not know. (U.S. Treasury 2003, p. 29).

If one interprets the response as the implicit value assigned by respondents to the option to receive a pre-populated return, the average was just under $10 ($12.50 in 2009 dollars).

⁹ Holtzblatt (2006) footnote 26.

¹⁰ Note that participation rates in California's ReadyReturn program, discussed below, are much lower.

¹¹ American Institute of Certified Public Accountants, "Tax Reform Alternatives for the 21^st Century," October 2009, pp. 34-35. http://www.aicpa.org/InterestAreas/Tax/Resources/TaxLegislationPolicy/TaxReformStudies/DownloadableDouments/Tax%20Reform%20Alternatives%202009.pdf

¹² Robert Weinberger, Senior Fellow at the Aspen Institute, estimates that refunds could be delayed by five to seven months. Presentation at Tax Policy Center Forum, "Tax Complexity: Can Technology Make us Free?" April 8, 2010.

¹³ "Statement for the Record submitted to the House Committee on Ways and Means," June 15, 2004, footnote 6.

¹⁴ Vaillancourt also concluded that these compliance costs are higher for smaller firms, decreasing with firm size.

¹⁵ Survey of Business Owners Table P, Statistics for All U.S. Firms With Paid Employees by Employment Size of Firm and Gender: 2002.

¹⁶ Goolsbee (2006), Table 4, p. 16. Sum of nationwide totals of compliance burden reduction for 1040A and 1040EZ filers in Wave 1.

¹⁷ Goolsbee estimates that a federal TAR program when fully implemented could serve up to 40 percent of taxpayers with relatively simple tax situations and reduce their tax compliance burden by $44 billion over 10 years (p. 7).

¹⁸ FTB (2009), p. 4.

¹⁹ A rough adjustment is made as follows. First the $1.6 billion estimated is multiplied by 8.0 percent to account for the lower participation rate. This lowers the estimated individual compliance cost savings to approximately $130 million. This figure is then multiplied by .25/.80 to account for the lower estimated savings associated with savings in form completion time only. Applying these factors results in an adjustment of the original estimate from $1.6 billion to approximately $40 million ($1.6 billion * .08 * (.25/.80), which is scaled back to account for lower participation rates and compliance cost savings.

²⁰ Goolsbee (2006), p. 15.

²¹ Joseph Bankman, "Using Technology to Simplify Filing." Presentation to President's Economic Recovery Advisory Board, Tax Reform Subcommittee, Dec. 4, 2009, (slide 15). http://www.whitehouse.gov/photos-and-video/video/perab-tax-reform-subcommittee-meeting

²² ReadyReturn prepares only state income tax returns.

²³ FTB (2009) p. 4 and e-mail communication from the ReadyReturn program

²⁴ See John Howard, "Debate over online state tax filing heats up in Capitol," Capitol Weekly, July 29, 2010, and Dennis J. Ventry Jr., "Intuit's Nine Lies Kill State E-Filing Programs and Keep 'Free' File Alive," State Tax Notes, August 30, 2010.

²⁵ John Guyton, Chief of Taxpayer Analysis & Modeling, IRS Office of Research, presentation at Tax Policy Center Forum, "Tax Complexity: Can Technology Make us Free?" April 8, 2010. Guyton described research showing large productivity gains in taxpayer assistance in recent years, comparable to overall changes in productivity in the economy. The preliminary paper concludes "[I]t appears that technology is mitigating the compliance cost impact of an increasingly complex tax system -- at least for now." George Contos, John Guyton, Patrick Langetieg, and Melissa Vigil, "Individual Taxpayer Compliance Burden: The Role of Assisted Methods in Taxpayer Response to Increasing Complexity," presentation at IRS Research Conference, June 30, 2010.

²⁶ The Treasury (2003) survey cited above reported that about 70 percent of respondents agreed completely or agreed somewhat with the statement that they would worry about how you resolve a problem in a return-free system; roughly half agreed that the IRS calculates taxes to their benefit and not the taxpayer's and that you can't trust the IRS to calculate your taxes this way, p. 28.

²⁷ Electronic Tax Administration Advisory Committee, Annual Report to Congress, June 2010, p. v. The Electronic Tax Administration Advisory Committee recommends tax reform and simplification given the impact on electronic tax administration, p. 9.

²⁸ IRS (2010), p.4.

²⁹ IRS, Advancing E-file Study, Phase 1 Report, The MITRE Corporation, 2008, p. 24.

³⁰ Goolsbee (2006), p. 7.

³¹ Ibid,, p. 9.

³² California Franchise Tax Board, 2008. Workload Growth: Feasibility Study Report, p. 3.

³³ See for example Internal Revenue Service Modernization, Staff Paper Prepared for the President's Commission to Study Capital Budgeting, June 19, 1998, http://clinton3.nara.gov/pcscb/rmo_irs.html and ExpectMore.gov, GAO High Risk Issues, U.S. Office of Management and Budget, http://www.whitehouse.gov/omb/expectmore/issue_summary/issues.html

³⁴ Increasing the administrative responsibilities of the IRS may be particularly difficult as the agency is tasked under health reform with implementing eligibility determination, documentation, and verification processes for premium and cost sharing credits, and with handling a greater reporting volume of 1099 Forms. The Congressional Budget Office estimates added IRS costs will probably total $5 to $10 billion over 10 years. Congressional Budget Office, Letter to Hon. Jerry Lewis, May 11, 2010. This compares with an annual IRS budget of roughly $12 billion.

³⁵ Stiglitz, Orszag, and Orszag (2000). The study concluded that only under extraordinary circumstances should government enter markets where private firms are active and that government generally should not take actions that reduce competition. Joseph Stiglitz is the winner of a Nobel Prize in economics. Peter Orszag is a former Director of the White House Office of Management and Budget and a former Director of the Congressional Budget Office.

³⁶ See footnote 25 supra, John Guyton presentation at Tax Policy Center Forum, April 8, 2010, citing large productivity gains in taxpayer assistance including tax software,

³⁷ See Harper and Lenard (2002) for a sampling of such offerings.

³⁸ Answers to Questions for Michael Mundaca, Nominee for Assistant Secretary of the Treasury for Tax Policy, from the United States Senate Committee on Finance, November 4, 2009, Answer to Question 6 from Senator Ensign. http://taxprof.typepad.com/files/mundaca.pdf

³⁹ Ian Liddell-Grainger, MP, Presentation at "The Boundaries of Government in a Digital Age: Should the Government Prepare Personal Tax Returns?" Technology Policy Institute, October 15, 2009.

⁴⁰ Stiglitz, Orszag, and Orszag (2000), p. 110.

⁴¹ Harper and Lenard (2002), p.7.

⁴² See footnote 26, supra.

⁴³ APPTG (2008), p. 9.

⁴⁴ See IRS, Table 2 http://www.irs.gov/pub/irs-soi/07in53us.xls.

⁴⁵ "Can Tax Time Be Less Burdensome?" Tax Policy Center forum, April 11, 2005. http://www.urban.org/publications/900807.html

⁴⁶ Ibid.

⁴⁷ APPTG (2008), p. 5.

⁴⁸ William Frenzel, Presentation at "The Boundaries of Government in a Digital Age: Should the Government Prepare Personal Tax Returns?" Technology Policy Institute, October 15, 2009.

⁴⁹ U.S. Treasury (2003), p. 28.

⁵⁰ Ibid., p. 39.

⁵¹ IRS Notice 2009-11, January 30, 2009.

⁵² Electronic Tax Administration Advisory Committee, op cit., pp. 31-33. The report observes that processing and filing burdens are greater for small business.

⁵³ Goolsbee (2006), p. 18.

⁵⁴ Alexi Mostrous, "Electronic medical records not seen as cure-all," Washington Post, October 25, 2009; Jacob Goldstein, "Can Technology Cure Health Care?" Wall Street Journal, April 13, 2010.

⁵⁵ APPTG (2008), p. 10.

⁵⁶ "Evaluation of the Internal Revenue Service's Capability to Ensure Proper Use of Recovery Act Funds," Treasury Inspector General for Tax Administration, November 27, 2009, Highlights.

⁵⁷ "Individual Taxpayer Identification Numbers Are Being Issued Without Sufficient Supporting Documentation," Treasury Inspector General for Tax Administration, December 8, 2009.

⁵⁸ Martin Vaughan, "Watchdog Warns IRS Is Overburdened," Wall Street Journal, January 6, 2010.

⁵⁹ In her Mid-Year Report to Congress, the National Taxpayer Advocate expressed concern about the adequacy of taxpayer service as the IRS implements health care reform. Press Release IR-2010-83, July 7, 2010.

⁶⁰ See Richard H. Thaler and Cass R. Sunstein, Nudge: Improving Decisions About Health, Wealth, and Happiness, Yale University Press, 2008.

⁶¹ Goolsbee (2006), p. 19.

⁶² APPTG (2008), p. 8.

⁶³ Video recording available at http://www.whitehouse.gov/photos-and-video/video/perab-tax-reform-subcommittee-meeting

⁶⁴ Gale and Holtzblatt (1997), among others, make this point.

⁶⁵ See for example Goolsbee (2006), p. 19.

⁶⁶ Various privacy, security, and fairness concerns with regard to California's ReadyReturn program are examined in William J. Kambas, "Reform and Modernization of the Tax Compliance Process," Tax Notes, September 19, 2005, pp. 1447-1451.

⁶⁷ The IRS reported on November 5, 2009 (IR-2009-101) that it was seeking to return $123.5 million in undeliverable refunds to approximately 108,000 taxpayers. http://www.irs.gov/newsroom/article/0,,id=215117,00.html

⁶⁸ See OECD (2006), pp. 15-16. http://www.oecd.org/dataoecd/42/14/36280368.pdf

⁶⁹ Thomas M. Lenard and Paul H. Rubin, "In Defense of Data: Information and the Costs of Privacy," Policy & Internet Journal, Vol. 2: Iss. 1, Article 7 (2010), p. 169. http://www.techpolicyinstitute.org/files/lenard%20rubin%20comments%20ntia%20docket.pdf

⁷⁰ The California Franchise Tax Board presents the following assurances of security on its website:

⁷² "Improving Online Security," Industry Roundtable, Scientific American, September 2008, cited in Lenard and Rubin, (2009), p. 39.

⁷³ Lenard and Rubin (2009), p. 35.

⁷⁴ Harper and Lenard (2002), p. 8.

⁷⁵ Joe Davidson, "Commerce Dept. Slow to Notify Employees of Security Breach," Washington Post, January 27, 2010.

⁷⁶ Federal Trade Commission: Agenda, Securing Personal Data in the Global Economy, March 16-17, 2009. http://www2.ftc.gov/bcp/workshops/personaldataglobal/agenda.pdf

⁷⁷http://www.ftc.gov/privacy/privacyinitiatives/promises_enf.html

⁷⁸ Privacy Rights Clearinghouse, Chronology of Data Breaches, November 19, 2009. http://www.privacyrights.org/ar/ChronDataBreaches.htm

⁷⁹ Fourth Annual US Cost of Data Breach Study, Ponemon Institute, January 2009. http://www.ponemon.org/local/upload/fckjail/generalcontent/18/file/2008-2009%20US%20Cost%20of%20Data%20Breach%20Report%20Final.pdf

⁸⁰http://www.irs.gov/pub/irs-utl/ntatestimonyfinanceidtheft041008.pdf

⁸¹ Ibid., pp.3 and 8.

⁸² IRS (2010) p. 32.

⁸³ Government Accountability Office, "IRS Needs to Continue to Address Significant Weaknesses," March 19, 2010, GAO-10-355.

⁸⁴ "Additional Security Is Needed for Access to the Registered User Portal." http://www.treas.gov/tigta/auditreports/2010reports/201020027fr.pdf

⁸⁵ Robert Greenstein, 2009. "Fiscal Responsibility and Fairness." Presented to the Tax Reform Subcommittee of the President's Economic Recovery Advisory Board, Dec. 4, 2009, p. 30. For a general discussion of the implications of various return-free filing systems for those receiving the EITC, see also Holtzblatt (2006).

⁸⁶ See e.g. Gale (2009).

⁸⁷ See Christopher Howard, 2007. The Welfare State Nobody Knows. Princeton: Princeton University Press; and Howard, 1997. The Hidden Welfare State. Princeton: Princeton University Press. The important, and somewhat unique, role of tax incentives in the delivery of U.S. social policy has also been acknowledged by the OECD. See Organisation for Economic Cooperation and Development, Tax Expenditures in OECD Countries, January 27, 2010.

⁸⁸ The National Taxpayer Advocate suggests that the IRS mission statement be revised to explicitly acknowledge the agency's dual role as part tax collector and part benefits administrator. Press Release IR-2010-83, July 7, 2010.