Perspectives on Two Proposals for Tax Filing Simplification

POINT: Simplified filing would be a boon for all taxpayers, not just those who file the simplest returns, and would save substantially on filing costs.

By Joseph Bankman, Ralph Parsons Professor of Law and Business, Stanford Law School, Stanford, CA

The Financial and Other Costs of the Current Tax Filing System

Filing taxes is expensive. The annual cost of individual filing for taxpayers without self-employment income has been estimated at about $40 billion a year.¹   Although absolute costs of filing rise with income, filing imposes significant costs, as a percentage of income, on all taxpayers. The primary cost is the value of the taxpayer’s time.

Filing also imposes other costs not incorporated in the above estimates. Filing leaves many taxpayers with feelings of anxiety, frustration and anger. The basic problem is that filing requires expertise and skills most taxpayers don’t have, as well as detailed record-keeping. This is true even for non-itemizers, who only have to keep track of a few pieces of information (such as their W-2s) and can file a short form such as Form 1040A.

As tax lawyers, we may regard these tasks as simple. We sometimes forget, or do not realize, how difficult those tasks are for citizens who are less literate and numerate than we are. Scholars today define literacy in terms of task (rather than grade-level). Only 30% of Americans can use a bus schedule to determine how long a particular bus ride will take.² Taxpayers with these levels of literacy and numeracy don’t find filing simple.  As income falls below average, the difficulty of self-filing increases, because of the need to make additional determinations relevant for the earned income tax credit. As income rises, levels of literacy and numeracy rise, but the complexity of the return increases.

Of course, taxpayers don’t have to do their returns themselves—they can and do farm out the task. Figuring out whom to hire for that task, however, is itself anxiety-provoking. More importantly, the largest single component of filing costs is record-keeping, and that task can’t be farmed out. Faced with the difficulty of the task, many low income taxpayers who would receive refunds simply don’t file.

The remaining taxpayers who do file must bear the financial and psychological costs of doing so. Some of the complexities they face can be tackled only by simplifying the basic structure of the tax code, but that is close to impossible in the current political climate. In contrast to simplifying the tax code, simplifying filing is relatively easy. Almost all the information that the government asks for it gets from third parties. Under the data-retrieval proposal outlined above, the government would provide that data to the taxpayer as downloadable entries on an electronic tax return.

The Benefits of Simplified Filing with Data Retrieval and Pro Forma Returns

It is easy to see how this system would simplify filing for non-itemizers. Their electronic return would show wage income and (possibly) some interest income. All they would need to do to file is to add or check on dependents and hit “calculate.”

What may surprise readers is that it would simplify filing for itemizers as well. Consider, for example, someone with taxable income of $800,000. According to IRS data, the average taxpayer in this position has income from wages, taxable and tax-free interest income, dividends, capital gains and losses from stocks, and deductions for state taxes, mortgage interest and charitable contributions. All of these items except charitable contributions are already subject to third-party reporting.³

Some taxpayers who now use preparers or pay for e-filing programs, like TurboTax, would find it easy enough to file their own taxes with the new system. Others would still want to use a trusted third party to double-check the information the government provides or do the necessary calculations. In such cases, they could authorize the government to provide the information it already has directly to the e-filing program or tax preparer. With all third-party income already accurately reported by the government, the equilibrium price of professional programs or preparers is likely to fall.

For most non-itemizers, third-party reported data constitutes all the information on the return other than marital status and dependents. For such taxpayers, the government could provide a tentative, or pro forma, tax return using third-party data together with the prior return year’s information on marital status and dependents. Taxpayers would be expected to verify the accuracy of the information and make any necessary corrections. If and when the return is accurate, they could simply hit “accept.”

Pro forma tax returns work well only for those whose profile is characteristic of simple returns. For taxpayers who itemize (or who show expenses for which software can predict the option to itemize), a data-retrieval system will be more accurate. Taxpayers in the latter group would not be eligible for pro forma returns. Thus, even if a data-retrieval system includes the option of pro forma returns, those returns will make up a minority of total returns.

One of the goals of either a data-retrieval system or a data-retrieval system with pro forma returns is to reduce taxpayer anxiety or frustration. Accordingly, each system would be optional. Taxpayers who do not want to access the government’s record of third-party information for their own use or their preparer’s use would not be forced to do so. (It is difficult to imagine why any taxpayer would not want to access that information, a fact that testifies more eloquently than anything else to its being in taxpayers’ interests for the government to make it available).

One way to think about data retrieval is to compare tax filing to paying other bills like credit card charges. Suppose VISA sent each cardholder a blank piece of paper each month, asking the taxpayer to list their charges and calculate their payment due. In addition, suppose VISA assumed that any discrepancy between the taxpayer’s bottom line and its own calculations was due to cardholder error and levied a penalty on any amount not paid. Cardholders would rebel. Since VISA has all the charges in its computer system and is going to use its data in any event, it would seem senseless to all concerned for VISA not to list that information in the monthly statement so that the basis for the charge is clear. It makes paying infinitely easier, and makes it easier for customers to spot errors in VISA’s internal accounting.

Pro forma returns are the norm in many countries, including OECD nations such as Finland, Sweden, Norway and Denmark.⁴ (Other OECD countries make greater use of withholding at the source, thus eliminating the equivalent of April 15 for most citizens.) California also offers a form of data-retrieval, after a successful pilot program on pro forma returns.  The IRS offers a limited form of data retrieval as well, though most current-year data is not in the system. A recent proposal for a federal program of data-retrieval and pro forma returns was supported by scores of tax scholars, including Austin Goolsbee and Emmanual Saez.⁵

The Opposition’s Arguments Are Not Convincing.

Not everyone agrees that this kind of filing simplification is desirable. Professor Maule, the author of the related Counterpoint, has been a determined and indefatigable opponent. Intuit, the maker of Turbo Tax, has also opposed filing simplification, going as far as to make a $1 million independent campaign expenditure in support of a politician who promised to eliminate the California pilot program mentioned above.⁶ Finally, the filing simplification proposals outlined above have been opposed by conservatives such as Rush Limbaugh and Grover Norquist.

In the Counterpoint, Professor Maule levies thirteen arguments against simplified filing. A number of these arguments are effectively made in the alternative. For example, Maule at one point argues that taxpayers won’t use simplified returns, but later he argues that tens of millions of taxpayers will fail to notice problems in the substantive tax law when they use pro forma returns. In order to fit this reply in the space allotted, I confine my responses below to what I believe are the most significant and plausible concerns Maule raises.

1. Taxpayers won’t benefit from, or use, simplified returns.

Maule argues that data retrieval offers taxpayers no real benefit, since entering information from W2s, 1099s, and other third-party reports received by taxpayers is one of the “quickest and easiest” steps in the filing process. In fact, as noted above, tax filing is enormously time-consuming and expensive, even though the process for most taxpayers consists solely of keeping information returns and putting the data from the returns onto the correct line of the tax return. One way to reconcile Maule’s intuition with the literature on compliance cost is to realize that there are 160 million taxpayers who spend over 10 hours each on record keeping and filing, so reducing time by even an hour (and imputing a value to that time) produces cost savings of billions of dollars a year. More generally, what seems “quick and easy” to us as tax lawyers is befuddling to the average citizen.

2. Simplification will cause taxpayers to pay too much tax.

Maule argues that the government will make systemic errors in its favor under a simplified filing regime, and taxpayers will miss deductions. This echoes the main argument made against the proposals by Grover Norquest and Intuit.

This argument is unfounded with respect to data retrieval. Data retrieval merely gives a taxpayer access in easily usable electronic form to third-party information, such as wage data, that the government already has in its file. This is the same data that the third parties have sent directly to the taxpayer. Using the government’s copy of that data should have no effect on tax liability. Moreover, unless notified that it is mistaken, the government is going to use its own third-party data to determine tax liability anyway. Data retrieval lets the taxpayer know ahead of time if there is a mismatch, so she can work to correct the government’s file (for example, by getting an information-provider to send a corrected return to the government).

Pro forma returns include one additional piece of information not supplied by data retrieval: information on the taxpayer’s marital status and dependents, generally acquired from the taxpayer’s prior year’s return—the same place that most accountants and e-filing programs acquire it. In all cases, taxpayers would be asked to verify that information. While mistakes are always possible, there is no reason to think they would be more frequent with pro forma returns than without them, or that such errors would systematically favor the government.

There is also no reason to think that pro forma filers would lose deductions. Pro forma returns would be limited to non-itemizers. Taxpayers whose reported data indicates that it may be to their advantage to itemize would be denied a pro forma return on that ground. So, for example, a taxpayer with mortgage interest who applied for a pro forma return would be told that she was not eligible because the presence of mortgage interest suggests she might be better off itemizing. Far from causing the taxpayer to lose deductions, a pro forma return system would alert taxpayers to deductions to which they are entitled.

In addition, a pro forma return regime would no doubt be monitored closely by the tax preparation industry. One would expect the industry to offer to review the returns for accuracy and publicize any cases in which the taxpayer was disadvantaged.

One final point: Maule uses the error rate (as of 2010) in the United Kingdom’s withholding-at-the-source system to support his claim of systemic error in pro forma systems. Withholding at the source is a much different system than the data retrieval or pro forma systems recommended here. It offers many advantages, essentially eliminating tax filing altogether, but is much more prone to errors than a pro forma system.⁷

3. Simplification will cause taxpayers to pay too little tax.

Maule is correct to assert that pro forma returns and data retrieval might reduce tax compliance, by letting taxpayers know what information the government doesn’t have. Taxpayers with unreported cash income, for example, know they can misreport that income without (much) fear of detection. But the effect is likely to be small. Studies of compliance rates show that taxpayers with cash income already know the income is not subject to third-party reporting – that’s why they report so little of it. In a recent article (and other articles on filing simplification), I discuss this form of evasion and suggest easy-to-implement modifications of these proposals that are likely to actually increase taxpayer compliance.⁸

4. There are significant administrative costs to implement filing simplification.

Maule argues that filing simplification imposes significant costs on third-party reporters and the IRS. In fact, the only change required for third-party reporters is to send reports to the government contemporaneously with sending those reports to taxpayers (currently, employers and other reporters send data to the government after they send it to employees). The cost of that change is negligible.⁹ The IRS would bear costs to develop software and improve the data-retrieval system it already maintains for taxpayers with preparers. California, whose income tax is based on the federal tax, developed the software needed for its program for a few million dollars, and it needs only a handful of employees to run the program.

5. Security is a major concern with simplified filing.

Security is an important concern of any new or existing tax program. California, which has adopted a version of data retrieval and pro forma returns, adheres to industry best practices (such as the so-called “NIST standards”). Any proposal for filing simplification must be based on the government’s ability to conform to those practices and standards.

Of course, even financial institutions or third-party vendors, such as Intuit, are vulnerable to break-ins. The same will be true of the most secure government agency. It is in part for this reason that participation in filing simplification (like participation in on-line banking) should be elective.

6. Filing acts as a check on Congress.

Maule argues that filing helps citizens “become aware of what Congress has done with the tax law” and that taxpayer frustration from filing is the “necessary seed of grassroots income tax reform.” Tax filing simplification, on the other hand, creates “tax zombies” who will “never feel a thing” as the government “extracts a wee bit more” from their paycheck. He suggests that having the government’s assistance in incorporating data into a return somehow means taxpayers no longer “take ownership” of their tax obligations, though taxpayers still must review their returns and pay the tax due.

I am less confident than Maule that the majority of taxpayers dislike the compromises that comprise our substantive tax law. It is not clear that there is another tax system out there that would win in a head-to-head vote. I am also skeptical of the claim that the level of government services and the tax rate required to pay for them is inconsistent with citizen preferences. It just might be that elected representatives give their constituents what they want, or at least what they think they want.

It is a moot point, though, because preparing one’s own return tells a taxpayer nothing about the substantive tax law, the magnitude or recent direction in government expenditures and budget deficits, or the distribution of the tax burden, and does not make salient individual average or marginal tax rates. It is instead an exercise in record-keeping, literacy and numeracy, requiring that taxpayers save information returns and copy numbers on those returns to the right lines of their 1040s. Taxpayers sensibly reject this exercise and farm the task out to preparers, or they purchase software that requires them to passively answer questions before spitting out a completed return. They are left with the compliance costs mentioned at the start of this essay, but without any more substantive knowledge of the tax law than they’d get with a simplified filing system.

COUNTERPOINT:  Although theoretically appealing, simplified filing systems entail too many risks of security, resource, and compliance costs.

By James Edward Maule, Professor of Law, Villanova University Charles Widger School of Law, Villanova, PA

The data-retrieval system and the pro forma return approach, the latter officially proposed under the name Real-Time Tax System, are both appealing theoretical ideas that fail miserably when they meet practical reality. Though the ideas are inspired by the best of intentions, implementation of either one requires resources and protections currently unavailable and unlikely to be provided in the future.

There are at least thirteen reasons for rejecting the data-retrieval system, the pro forma return approach, or both. Some are more serious than others, but all focus on risks that will be faced by taxpayers and the nation if either proposal is pushed through Congress or the IRS.

Both the data-retrieval system and the pro forma return approach pose huge security risks. Almost every day we wake up to news of another security breach at the IRS. For example, not long ago the IRS announced that it was cutting off taxpayer access to the electronic filing PIN tool, after supposedly fixing flaws leading to earlier intrusions affecting at least 100,000 taxpayers, because of “additional questionable activity.”¹⁰ The IRS continues to update the number of taxpayers whose transcripts were hacked, with the most recent total number approaching at least one million.¹¹ The data-retrieval system, the pro forma return approach, and tax transcripts put taxpayer information on the internet, theoretically for access by the taxpayer, but realistically for access by every two-bit hacker with evil intentions. In many respects, the data-retrieval system is nothing more than an early release of taxpayer transcripts. Until and unless the IRS fixes its security issues, which requires funding from a Congress intent on curtailing the IRS, both the data-retrieval system and the pro forma return approach almost guarantee identity theft and financial disruption for millions of taxpayers. Failure of the Congress to renew IRS streamlined critical pay authority triggered the departure of the IRS Chief Technology Officer and “harms [its] ability to recruit the best IT and cybersecurity talent out there.”¹² Worse, some advocates of these simplified filing approaches have reacted to criticisms pointing out the risk that taxpayers will fail to add transactions to those included by the IRS on the return by suggesting that the IRS collect additional details of taxpayers’ financial accounts, cleared checks, credit card statements, organization memberships, and all sorts of similar records.¹³ Some advocates of these proposals even suggest having the taxpayer provide a credit card number to the IRS so it can automatically collect additional taxes when it decides the return or return information needs to be changed. It’s possible that some proponents of these filing approaches advocate them with the goal of creating a base for such an expansion of collection possibilities. Considering the state of IRS digital security, the entire theory underlying both proposals is a recipe for a tsunami of identity theft and financial disruption. Surely those hackers living in the dark edges of the internet are rooting for one of these proposals to be adopted. I agree with Bankman that “[a]ny proposal for filing simplification must be based on the government’s ability to conform to [industry best] practices and standards.” The problem is that the IRS is not ready for this proposal and, in light of its future prospects, will not be ready for this proposal for a long time, if ever.

Both the data-retrieval system and the pro forma return approach will likely erode the value of self-compliance, a core principle of the federal income tax system.¹⁴ By taking ownership of their obligation to pay taxes through filing returns, citizens participate in government. When they focus on the details of the return, they become aware of what Congress has done with the tax law. Those “feelings of anxiety, frustration, and anger” are the necessary seeds of grass-roots-nourished income tax reform. Democracy demands transparency in taxation.

Both the data-retrieval system and the pro forma return approach hide the tax law (the former to a lesser extent) and could thus reduce the number of voices demanding tax reform. The pro forma return approach, in particular, will entice tens of millions of taxpayers into thinking that the federal income tax no longer is a problem in need of a solution. Taxpayers who do tax returns receive an education, and an educated electorate is far more valuable to democracy than is a pacified electorate.

Both the data-retrieval system and the pro forma return approach pose huge conflict of interest problems, with the latter presenting the greatest challenge. The pro forma return approach puts the IRS in the dual roles of tax return preparer and tax collector. Does one branch of the IRS audit the other? If the IRS makes an error, either in generating a pro forma return or in making available a partially filled-out return—and errors will happen—does it take the same “it’s not our fault if our advice is bad” position it offers to taxpayers calling in for assistance? Is there not a conflict between the IRS goal of maximizing tax collections and its preparer role requiring fiduciary responsibility to taxpayers?¹⁵

Contrary to the claims advanced by their advocates, the pro forma return approach (and, to a lesser extent, the data-retrieval system) will decrease tax return accuracy. A system in the United Kingdom similar to the pro forma return approach generates an error rate of 15 percent in favor of the government.¹⁶ In the less frequent instances when an error is made in the taxpayer’s favor in that system, the taxpayer is liable for the mistake. By nurturing taxpayer dependency on partial or full government-prepared returns, both the data-retrieval system and the pro forma return approach make it less likely that taxpayers who experience extraordinary transactions will bother to make changes on account of them. Worse, when the government-prepared return or partly completed return has an error in favor of the taxpayer, the odds of the taxpayer pointing out that error are very low. Advocates claim that the error rate in California’s Ready Return system is lower than that of returns not prepared in the system,¹⁷ but that rate reflects the test taker grading its own exam.

Both the data-retrieval system and the pro forma return approach are inefficient.  Their proponents claim that they save people time.¹⁸ Advocates claim that taxpayers will save enormous amounts of time by having the IRS enter onto “prefilled” returns information it receives on Forms W-2 and 1099. Yet the entering of information from those forms is one of the easiest and quickest steps in the tax return preparation process. Neither the data-retrieval system nor the pro forma return approach offer much in terms of saving time because, in reality, it takes much more time to check thoroughly what the government has inserted on a return than to do it on one’s own, with a preparer’s assistance, or by using tax preparation software. Only the taxpayer can serve as a screen against fake and erroneous Forms 1099 sent to the IRS, and both these filing approaches cut off the remedial techniques suggested by practitioners.¹⁹ A taxpayer is much more likely to spot a fake Form 1099 when it arrives from a third party than when it is buried in a pre-filled return. Banking on cursory reviews to make the data-retrieval system and the pro forma return approach appealing does not take the federal income tax system in the direction of efficiency.

Both the data-retrieval system and the pro forma return approach will cost businesses money and increase compliance burdens for them. Advocates of these approaches explain that Forms W-2 and 1099 would need to be filed much earlier.²⁰ Alternatively, taxpayer refunds will be delayed. The Electronic Tax Administration Advisory Committee reported that accelerating the information return filing deadlines is a significant workload increase, especially for small businesses.²¹ It has been estimated that the cost of accelerating these due dates is between $500 million and $5 billion, and that does not take partnership Schedules K-1 into account.²²

Both the data-retrieval system and the pro forma return approach present huge challenges to IRS operational capacity. What happens when a taxpayer finds an error on the IRS-prepared return, whether preliminary or pro forma? Is it fixed in time to permit the taxpayer to continue preparing the return? Does the taxpayer file a return inconsistent with what the IRS prepared? Does doing so increase the risk of audit? What happens when the issuer of the Form W-2 or Form 1099 submits a corrected copy? Is another “prefilled” return issued to the taxpayer by the IRS? Will the process of IRS amendment of “prefilled” returns be as inefficient as the current information return matching program? What happens if the data-retrieval or the pro forma return web site goes down on April 14 because of overloads or denial of service attacks? What happens if the system is outsourced and the contractor makes errors? Who pays the cost of fixing the mistakes? What happens to taxpayer protections when the outsourced pro forma return process is coupled with outsourced tax collection functions?

Neither the data-retrieval system nor the pro forma return approach will reduce the tax gap, contrary to the claims advanced by their advocates. Both the data-retrieval system and the pro forma return approach rely on Forms W-2 and 1099, but the tax gap exists because of transactions that people fail to report on those forms. To the extent the tax gap reflects deliberate cheating, neither of these approaches will shut off the opportunities. To the extent the tax gap reflects unwitting noncompliance, neither of these approaches solves the problem because very few of the mistakes made by taxpayers are within the scope of the transactions reported on information returns. To the extent the IRS makes an error in the taxpayer’s favor, unscrupulous taxpayers are likely to let it slide, thus increasing the tax gap.

Neither the data-retrieval system nor the pro forma return approach will shrink the tax return preparation business, contrary to the arguments and goals of their advocates. Instead, that business will be transformed. A comparison with the Pennsylvania local real property tax system, which is not unlike that found in other states, predicts what will happen. Similar to the dream of the data-retrieval system and the pro forma return approach lobbies, local taxing authorities prepare the real property tax return and send an invoice to the taxpayer. For many years, almost everyone blindly paid the bill. Eventually, as a few taxpayers began to question the locality’s computations (especially valuations), a new industry was born. It advertises and contacts taxpayers directly, offering to dispute a property tax bill. It is not difficult to imagine that the commercial tax return preparers so despised by the pro forma return approach advocates will go into the “dispute the IRS invoice” business. So much for the claim that the data-retrieval system and the pro forma return approach reduce tax compliance costs for taxpayers. The only thing that reduces tax compliance costs is tax simplification, and it’s a shame that the energies of the data-retrieval system and the pro forma return approach lobbies aren’t aimed in that direction. Yes, simplification is a difficult challenge, but it is essential. For it to happen, complexity supporters ought not be given another excuse for maintaining the sad state of the federal income tax system by having the opportunity to claim that return filing has become much less of a problem for taxpayers.

Neither the data-retrieval system nor the pro forma return approach, contrary to the claims made by their advocates, will save taxpayers money. Advocates argue that taxpayers will pay less in preparer fees, but taxpayers will end up paying preparers to check what the IRS has prepared, a task that will take at least as long, if not longer. Those taxpayers who decide to abandon their preparers by putting their trust in the IRS will pay in the form of higher tax liabilities generated by IRS errors not caught by the taxpayers. Only the most trusting and naive of taxpayers would do anything less than prepare their own returns, or pay an independent third party to do so, even if the IRS puts a “prefilled” return on the taxpayer’s table.

Both the data-retrieval system and the pro forma return approach pose an economic threat. Both, especially the pro forma return approach, are designed to put some tax return preparers out of work. What is the impact of that disruption on the economy? Would those job losses be made up by the IRS hiring programmers to design the data-retrieval system or the pro forma return approach? Is there any guarantee that the extensive programming work required to implement either proposal would not be outsourced overseas? Where will the IRS obtain the funds to turn itself into a tax preparation service or to generate partially completed returns? As improbable as are the chances of tax simplification happening in the next year, the chances of the IRS obtaining additional resources to support its activities are even lower.  A bill passed by the House of Representatives in early July of this year would make additional cuts in the agency’s budget (reducing its funding level to less than it was eight years ago) and would also prohibit the IRS from using any appropriations “to provide to any person a proposed final return or statement for use by such person to satisfy a filing or reporting requirement.”²³ Given the choice between paying preparers and paying additional taxes to fund a government tax return preparation department in the IRS, taxpayers surely prefer the former, a conclusion reinforced by the next problem with the proposals.

Both the data-retrieval system and the pro forma return approach require taxpayer acceptance, and experience demonstrates that this is unlikely to happen, especially with the latter. Advocates of this system claim that California’s Ready Return program proves the approach’s worthiness. Fewer than 90,000 California taxpayers used the program in 2015.²⁴ That’s a rather awful “vote with your feet” response considering that almost 14 million personal income tax returns were filed in California in 2015.²⁵ But it’s not surprising. One poll discovered that 71 percent of Americans would not trust the IRS to prepare their returns.²⁶ It’s difficult to imagine support for partial preparation under the data-retrieval system would be much higher. After all, only 19 percent of Americans trust the federal government to do what is right.²⁷

Neither the data-retrieval system nor the pro forma return approach is necessary. The IRS FreeFile initiative provides free tax return preparation for low-income taxpayers. One advocate for pro forma returns claims that this free service is useless because it does not use IRS software.²⁸ Why is that a problem? In reality, it is a blessing that IRS software is not used for FreeFile, because IRS software is plagued with problems.²⁹ Extending its use beyond its current boundaries is senseless and unnecessary.

There is a reason both the data-retrieval system and the pro forma return approach find their strongest support in the corridors of academia, in the sound-bite promises of some politicians, and in the dreams of those who think that a tax return unseen is a tax obligation escaped. The current Administration included “pre-filled” tax forms in its 2008 campaign promises, but discovered, once taking office, that reality presented overwhelming obstacles.

The “here, let us take care of it for you, don’t worry, be happy” message of the pro forma return approach campaigns drowns out the “and you’ll never feel a thing as we extract a wee bit more, remove a deduction, eliminate an exclusion, terminate a credit” part of the script. Both the data-retrieval system and the pro forma return approach create tax zombies. Why do that? Both proposals open the door to “just let us tap into your bank account and all of your other information and we’ll do everyone’s return.” The data-retrieval system is simply the first step on the way to the pro forma return approach.

When tax compliance gets so complicated that the IRS must prepare, in whole or in part, taxpayer returns—even assuming the IRS is better at doing so than are taxpayers and paid tax return preparers—the tax law surely has become not only too complicated but also inefficient and unjust. IRS preparation of tax returns in an unnecessarily complicated system opens the door to government control beyond reasonableness. Both the data-retrieval system and the pro forma return approach are first steps in the wrong direction, a band-aid on a hemorrhage, and a threat to the transparency required by democracy. ■