Bonnie M. Wongtrakool

On July 22, 1998, President Clinton signed the Internal Revenue Service Restructuring and Reform Act of 1998 ("RRA") into law. ¹ RRA addresses a wide range of issues, including executive branch governance and management of the IRS, personnel staffing, customer service and compliance, technology modernization, tax law simplification, and taxpayer rights, with the goal of modernizing and improving taxpayer services. ² These ideas originated from a report entitled "A Vision for a New IRS," which had been released by the National Commission on Restructuring the Internal Revenue Service (the "Commission") on June 25, 1997, one year prior to RRA’s enactment. ³ The Commission’s recommendations became the basis for H.R. 2676, RRA’s predecessor, which the House passed by a vote of 426 to 4 on November 5, 1997. ⁴ As in the House, the bill enjoyed strong bipartisan support in the Senate, which passed the currently enacted version by a vote of 96 to 2 on July 9, 1998. ⁵

One of the major objectives of RRA is to promote electronic filing. The RRA sets a long term goal for the IRS to increase the percentage of returns filed electronically from its current level of 20% to 80% by the year 2007. ⁶ The IRS, which has been offering an electronic filing option to a growing number of taxpayers since 1986, supports this goal largely because of its demonstrated potential to reduce errors, decrease costs, and expedite the tax filing process. ⁷ The initial error rate in processing tax returns is less than 1% for electronic filing, compared to approximately 20% for paper filing; refunds on electronic returns are returned in two weeks, rather than six weeks for paper returns; and service center costs to process electronic returns are five to 40% lower than costs to process paper returns. ⁸

While its past success with electronic filing is encouraging, the IRS must accomplish three major tasks as it continues to move towards a paperless tax system. First, the IRS must remove the last pieces of paper from the electronic filing process. ⁹ Previous to amendment by RRA, Internal Revenue Code Section 6061 stated that tax forms be signed as required by the Secretary of the Treasury. ¹⁰ Thus, a taxpayer who elected to file electronically still had to file a paper Form 8453, which contained her signature information. ¹¹ Under Section 2003(a) of RRA, the IRS must develop procedures for the acceptance of digital or electronic signatures, thereby eliminating by the year 2007 the need to file any paper form, and in the interim, may provide for alternative methods of signing or may waive the signature requirement. ¹² Second, RRA requires the IRS to develop procedures that allow an electronic filer to review her account electronically, without compromising taxpayer privacy, by the year 2006. ¹³ Third, the IRS must install a return-free tax system under which individuals may pay taxes without filing a return for taxable years beginning after 2007. ¹⁴

This paper will examine plans for a paperless tax system from the perspectives of privacy, implementation, and fairness. ¹⁵ Part I will discuss privacy concerns raised by RRA. ¹⁶ Part II will explore fairness issues, including possible discrimination between those with Internet access and those without, and unintended adverse market effects that may occur during the disintermediation of the tax filing industry. Part III will address the issues related to implementation, namely, the challenge of spreading Internet access and the possibility that digitizing could lead to a more complex Internal Revenue Code. Finally, Part IV will conclude that a paperless tax system, if implemented intelligently, has the potential to reduce cost and increase efficiency, without unduly compromising values of privacy, administrability, and equity.

I. Technology and Taxpayer Privacy

As a result of RRA, in a future world, opportunities to correspond with the IRS electronically will be greatly expanded. Along with this growth in opportunity comes added concern about the government’s ability to protect the privacy of individuals by maintaining information in a secure way. RRA Sections 2003 and 2005 set two mandates that in particular may impact taxpayer privacy. The latter provision requires the IRS to make taxpayer records accessible electronically. ¹⁷ The former provision requires the IRS to accept a taxpayer’s digital signature in lieu of a handwritten signature when filing a return or when requesting or remitting taxpayer information. ¹⁸

Because of the possibility of network attack and the impossibility of designing a foolproof encryption process, placing taxpayer information online, as required by RRA Section 2005, poses a legitimate risk to privacy. Computer networks cannot guarantee absolute security. There are numerous examples of system assault by both insiders, persons that work for legitimate entities, and outsiders, persons that break into the computer system operated by legitimate entities. ¹⁹ Indeed, the security of IRS operations, both its networks carrying electronic information and warehouses storing tangible information, has been criticized in the past. ²⁰

The adoption of digital signature technology may further compromise taxpayer privacy. From one viewpoint, digital signatures "represent one of the most explosive clusters of privacy-threatening technologies, motivations, and processes that has yet been invented." ;²¹ In simplified terms, a digital signature is an encrypted version of a message, to which a digest containing a "hash" resulting from encryption, or a fingerprint of the message, is attached. ²² The encryption is performed with the signer’s private key, so that the hash result is derived from and unique to both the signed message and the private key. ²³ The digital signature and identity of the owner is verified by a digital certificate, which is an electronic format of a physical certificate such as a passport. ²⁴ This digital signature and certification process creates a repository of vital information and a library of aliases for wrongdoers to discover. Because taxpayer records provide a concise summary of personal and financial information, an electronic tax system may merit heightened concern. ²⁵

The debate surrounding the Access Certificates for Electronic Services ("ACES") initiative reflects these privacy concerns. The goal of the ACES project is to create a certificate authority that verifies the identities of computer users so that they might access private information or transact business with government agencies electronically, including checking their tax records. ²⁶ Critics initially resisted ACES as a "national identification card" because they were concerned about the government amassing a centralized bank of information, "escalating the collection and centralization of data about an individual’s identity and activities, potentially increasing government intervention in our daily lives." ²⁷ ACES representatives responded that the certificates would be managed by several vendors and that citizens would be able to use different certificates for different purposes, so that there would not be only one warehouse for record keeping. ²⁸ Still, the fear remains that any government involvement in the digital certification process is undesirable because it centralizes personal information, and independent contracting has been strongly urged by privacy advocates. ²⁹

In evaluating the risks posed by the RRA provisions, one must consider the risks posed by the current hybrid system of tax information management. It is important to realize that a great deal of taxpayer information is already accessible through computer networks. Although only 16% of returns filed for the tax year 1997 and received by the end of April 1997 were submitted electronically (excluding the 5% which were telefiled), 67% of these returns were prepared on a computer. ³⁰ Thus, the risk of placing taxpayer information online already exists; conceivably, a hacker could target the computer systems of an online electronic tax filing service today, just as she could target the IRS network of the future.

The reasonableness of RRA’s perceived threat to taxpayer privacy also depends on how technology is used, rather than the mere existence of its use. For example, the Social Security Administration’s ("SSA") past experience operating an Internet site provides models of both poorly designed and well designed access systems. The SSA’s initial attempt in March 1997 to provide personal earnings and benefit estimate statements ("PEBES") on the Web was curtailed after privacy experts criticized its security as inadequate. In order to access the original site, users had to supply five items of personal information: name, address, telephone number, place of birth, Social Security number and mother’s maiden name. Because many of these items were easily obtainable from public databases, it could not effectively prevent people from obtaining confidential information about others, including salary record by year, Social Security taxes paid, and estimated retirement and disability benefits. ³¹ After holding public forums and consulting with technology and policy analysts, SSA reopened the PEBES Web site in September 1997, earning the approval of privacy advocates. Its new system minimized threats to privacy by revising the content of information provided on the site and by requiring a password (previously delivered via email) in addition to the original five items for access to personal information at the site. ³²

Moreover, some privacy advocates argue that the access mandated by Section 2005 is actually necessary to protect taxpayer privacy. ³³ Taxpayers are entitled to know exactly what quantity and quality of personal information government files contain. At the same time, RRA requires the IRS to make taxpayer information available for review "only if all necessary safeguards to ensure the privacy of such account information are in place." ³⁴ Thus, RRA recognizes that taxpayers cannot be given access without providing a secure environment for that access. ³⁵ The IRS itself has launched a massive upgrade of its facilities in conjunction with the private sector, with the goal of "creating new IRS technology systems that will improve taxpayer service, add new security features and carry the agency into the new century." ³⁶ If systems are well designed and well maintained, the IRS will have advanced RRA’s promise to protect taxpayer privacy.

In addition, the IRS is not limited to any particular technology by RRA, which provides that "The Secretary shall develop procedures for the acceptance of signatures in digital or other electronic form." ³⁷ The IRS is thus exploring a range of alternatives to handwritten signatures. For example, it is now testing the use of passwords to replace paper Form 8453, which contains a taxpayer’s signature information. ³⁸ In 1999, the IRS will conduct two signature pilot tests, the "On-Line Pilot with Taxpayers" and the "Practitioner Pilot with Tax Preparers." The former program will mail an e-file Customer Number ("ECN") to any person who files her own taxes using IRS-approved PC software and whose taxable income is comprised only of wages, dividends, and interest. Similarly, the latter program provides a Personal Identification Number ("PIN") to any person with similar income components but who files taxes using certain paid preparers. ³⁹

Although using a password may seem radically different from signing on the dotted line, it should pose minimal additional risk, because there is little danger that someone hostile to a taxpayer would file a fraudulent return using her ECN or PIN. ⁴⁰ As additional reassurance, in the event that a vengeful divorcee or embezzling accountant does abuse the alternative signature, RRA places the burden on the government to prove that the signature truly represents the identified taxpayer. ⁴¹

Beyond password-based systems, the IRS is exploring the role that technologies such as digital certificates and biometric signatures will play in the authentication of returns. Although it has not determined whether it will ultimately operate the PKI or outsource it to third party providers, the IRS considers development of a public key infrastructure ("PKI"), which is a system to manage the signatures, to be an essential step in the development of the procedures required by RRA. ⁴² Designing those procedures presents a more challenging set of issues than the password systems, for example, what levels of security and authentication should be required and how to apportion liability when the authentication process is adulterated. ⁴³

Because there are no models for this system, some caution may be justified. However, since the risks of the process are known, it should be possible to design a system that maintains taxpayer privacy. ⁴⁴ This can be accomplished by upholding operating standards for the Certification Authority ("CA"), which is, in basic terms, the entity responsible for the verification of the taxpayer’s identity. ⁴⁵ The CA maintains a directory of all of its subscribers and guarantees the accuracy of these listings. ⁴⁶ The CA thereby plays a vital role in the prevention of "spoofing," the creation of a bogus public key that allows a wrongdoer to assume another’s identity." ⁴⁷ Typical problems that Certification Authorities encounter include loss of records, inaccurate or fraudulent records, and insecure encryption software. ⁴⁸ As long as the CA adheres to a suitable verification process and invests resources in the integrity and protection of its databases, it can minimize the risk of identity fraud and credit fraud. ⁴⁹

Although the preceding suggestions refer to the Certification Authority in the singular, the IRS could further protect privacy by giving taxpayers the option to choose among different Certificate Authorities and certification technologies to the extent practicable. In addition, the IRS could involve multiple entities in the storage and backup of taxpayers’ private keys. By decentralizing the certification process, these actions can decrease the risk of unauthorized access. ⁵⁰

The use of digital signatures in tax administration, when considered discretely, is not central to any privacy concerns. Rather, it is the interaction of the electronic tax system with other online systems that could compromise taxpayer privacy. Different levels of security and disclosure are appropriate for different purposes. Taxpayer privacy will be more secure if the IRS does not sign on to a "one size-fits-all" project such as ACES. ⁵¹ Unfortunately, there will be efficiency-driven reasons to use a single certification authority. Not only would a single-CA system simplify the management of multiple government agencies’ correspondence with citizens, it would also be less costly. ⁵² However, the IRS should recognize the value of decentralization for both privacy and competitive reasons (discussed below). ⁵³

II. Fairness Issues

RRA states that "paperless filing should be the preferred and most convenient means of filing Federal tax and information returns" for American taxpayers. ⁵⁴ This goal applies to all return filers, regardless of income or form of return filed. On the one hand, RRA seems to aspire towards horizontal equity, because it seeks to benefit taxpayers across all groups. On the other hand, it does not address the principles of vertical equity and distributive justice on which current tax law is predicated. ⁵⁵

RRA also mandates the IRS to "cooperate with and encourage the private sector by encouraging competition to increase electronic filing of such returns." ⁵⁶ Partnerships with industry could not only help the IRS to reach RRA’s goals but also help achieve them more quickly. Conversely, government intervention could hinder rather than help market progress. These statements and their implications invite consideration from a fairness perspective.

At least in the short term, the benefits of the online tax system may not reach the entire population. Although there is no stereotypical user, factors such as income, education, race, and gender currently divide Internet users from non-users. According to a Business Week survey, out of the 40 million people who used the Internet in 1997, over 42% of them had household incomes over $50,000. ⁵⁷ In comparison, 78% of returns were filed in 1996 by people with adjusted gross incomes under $50,000. ⁵⁸ The Business Week survey also found that 73% of Internet users in 1997 have at least a college degree, only 6% were Hispanic or African-American, and 41% were women. ⁵⁹ Meanwhile, the Census Bureau reports that less than 25% of adults age 25 and older have completed college, 23% of the total population is black or Hispanic, and women constitute 51% of the population. ⁶⁰ These disparities between user and non-user profiles suggest that certain socioeconomic groups will benefit more directly from RRA than other groups.

In addition, RRA makes specific instructions that may create a discriminatory structure, with an impact that can be measured in time value rather than current dollars. For example, taxpayers who file electronically receive their refunds in two weeks rather than six. ⁶¹ Also, RRA Section 2002 extends the due date for electronic filers of information returns from February 28 to March 31 of the year following the calendar year to which the return relates. ⁶²

Provisions considered during drafting but not ultimately included in RRA provide other illustrations of potential discrimination. One idea that was proposed was to give electronic filers a 30-day extension as an incentive for taxpayers to file electronically. ⁶³ Another idea was to increase electronic filing was to decrease the audit exposure period for electronic filers from 36 to 24 months. Opponents argued that although such provisions might appeal to high-income, balance-due taxpayers, they would do so at the cost of middle and lower-income refund taxpayers, whose taxes might be increased to compensate for the resulting loss of tax revenue. ⁶⁴

RRA also sanctions a two-tiered tax system that discriminates between those who file electronically and those who file conventionally. RRA Section 2001(f)(2) specifically authorizes the Secretary to implement procedures to provide for the payment of appropriate incentives for electronically filed returns. ⁶⁵ Essentially, this would be equivalent to giving a tax credit to those who file electronically, at the expense of those who file tangible tax returns.

Furthermore, if and when the IRS launches a digital certification system, funding that system presents another veiled threat to the progressivity of the Internal Revenue Code. To illustrate, under the ACES model, digital certificates would be free to electronic filers, and a vendor would be paid each time a certificate was used in a transaction with the IRS Web site. ⁶⁶ Presumably, certificate users would be those with access to technology, but fees for their use of the certificates would be spread across the entire tax base, including those without access to technology. If the certificate user tends to be higher-income like the online user, this would result in a regressive effect, with the lower-income taxpayers subsidizing the higher-income electronic filers.

There are several counterarguments to these fairness-related concerns. First, policymakers to date have demonstrated awareness of these issues. For example, in 1997, the IRS issued a Request for a Proposal (RFP) to pay electronic tax return transmitters $3 for every return they filed electronically. Tax software developers and tax practitioners argued that this strategy was assured neither to increase electronic filing nor to route the subsidy to the intended recipient - the taxpayer – and the proposal was never enacted. Second, even though many taxpayers do not yet have Internet access, they still can take advantage of the option to electronically file by filing through a tax preparer. Indeed, the typical electronic filer historically has been a lower-income, refund due taxpayer attracted by the prospect of a quick return. ⁶⁷ Third, as tax software developers offer incentives to taxpayers in order to secure market share, competition in the private sector should push down the costs of electronic filing. ⁶⁸ Fourth, community service programs such as Volunteer Income Tax Assistance ("VITA") and Tax Counseling for the Elderly ("TCE") can assist in providing electronic filing to taxpayers who would not otherwise take advantage of the option. ⁶⁹

Still, the tension between encouraging electronic filing and maintaining neutrality between online and offline taxpayers remains. Incentive programs offered by software developers will not reach the taxpayers who do not have Internet access and who will have to use tax preparers in order to file electronically. Some tax preparers absorb the costs of transmitting returns electronically, compensating for these costs by providing and charging for additional services, while other tax preparers charge the taxpayer for the cost directly. ⁷⁰ If a lower-income taxpayer who seeks an expedient refund electronically files through the latter group of preparers, her refund is decreased and her tax rate effectively increased.

The rise of the Internet has already transformed the way that business is transacted in industries such as securities, retail, and banking. ⁷¹ Some refer to this trend - revision of the role of the middleman - as disintermediation. ⁷² If the IRS is truly to succeed in its quest to make the tax system paperless, the structure of the tax filing industry is due to experience similar change. The parties at stake include tax practitioners, tax preparers, and computerized tax software manufacturers. As has occurred in other industries, disintermediation is likely to affect one or more of these parties. The issue, then, is not whether disintermediation will occur, but whether the IRS’s involvement will interfere in the progress of technology.

There is little danger that government collaboration with tax practitioners and preparers will impede the progress of technology. As the users, rather than the creators, of technology, practitioners epitomize the middleman referred to in disintermediation. Should their role evolve, this would be a neutral and unsurprising development ⁷³.

Rather, the IRS should be concerned with the effects of endorsing particular products or technologies. The IRS has already begun to work with the private sector to increase electronic filing. For example, it recently announced several partnerships with private firms beginning in 1999, including two agreements that allow taxpayers who file electronically to pay federal income taxes by credit card. ⁷⁴ Ironically, government programs such as these run the danger of inadvertently stifling commercial development, by interfering with the market forces that RRA encourages it to foster. ⁷⁵ A fine line separates advocating the use of technology, on the one hand, and on the other, unfairly favoring particular technologies, industries, or certificate providers at the expense of further development. Tax preparation software manufacturers, for example, have complained that the provision of electronic forms by the IRS puts the agency in direct competition with commercial electronic tax products. ⁷⁶ Also, skeptics of the government’s plan to develop digital signature technology fear that the result will be "a single government approved identity verification system undermining both individual privacy and the healthy development of the Web." ⁷⁷

The government should remain aware of the possibility that its involvement could adversely affect the tax filing industry. Specifically, it should appraise its choices in the context of three factors: technical interrelatedness, economies of scale, and quasi-irreversibility of investment. The first refers to the need for system compatibility between hardware and software, the second to the decrease in user costs as a system gains acceptance, and the third to the lock-in effect that results from high software costs. These factors must be considered because their presence indicates the potential for entrenchment of inefficient technology. ⁷⁸

With respect to the tax software market, these factors are not alarmingly present. Software developers can design software that is compatible with IRS infrastructure at a reasonable cost, as demonstrated by the wide variety of developers involved in the IRS’s 1999 "On-line Pilot with Taxpayers." ⁷⁹ Similarly, neither economies of scale nor the lock-in effect are of great concern because consumers enjoy an unrestricted option to switch between software. In general, the government appears to be on the right track with software developers. As long as the government continues along this path, it has little risk of upsetting the competitive balance.

A similar analysis suggests more caution when dealing with digital signature technology. The second and third factors of the analysis – economies of scale and the lock-in effect – cannot be ignored. Digital certification systems are generally expensive to implement, since the Certification Authority must not only handle the verification process but must maintain a current database of private-public key pairs and corresponding subscriber identity information. ⁸⁰ Thus, efficiency and administrability considerations favor a structure with fewer rather than a greater number of certification authorities. Because the fixed costs of providing large-scale services are high and the additional cost per user is low, there may be a natural tendency towards a market with only a few large Certification Authorities.⁸¹ In addition, as more federal agencies begin to offer services online, the government may find it cheaper and less confusing to make all transactions through one Certification Authority. Despite the privacy concerns of some groups, Internet users themselves may even prefer to use a single CA not only for all correspondence with the government but also with private third parties, including electronic commerce, because of the convenience and simplicity.⁸²

Nevertheless, while the government is trying to develop a certification model, the IRS should resist these forces. It should not at this stage limit taxpayers to one digital signature standard or provider.⁸³ One could argue that allowing consumers to choose their own certificate providers faces interoperability problems, a hurdle that programmers are currently trying to surmount.⁸⁴ One could also argue that even if the IRS uses multiple contractors, the public key infrastructure requires that there still be some degree of centralization.⁸⁵ However, by doing what it can to maintain competition, which at a minimum would be using multiple contractors, the government can minimize the danger of entrenching the "wrong" technology. In the context of an online tax system, where sensitive personal data is required to be collected and verified, the need for accuracy and security is particularly important.

III. Administrability of the Conversion

RRA sets as a goal for the IRS to file at least 80% of all returns electronically by the year 2007, and to the extent practicable, to file electronically 100% of all computer-prepared returns after the year 2001. ⁸⁶ Proponents of RRA have emphasized that electronic tax filing may significantly reduce costs and errors. ⁸⁷ The initial error rate in processing tax returns is less than 1% for electronic filing, compared to approximately 20% for paper filing. ⁸⁸ And service center-related costs to process an electronically filed Form 1040 return are $1.90, compared to $3.25 for a paper Form 1040 tax return. ⁸⁹ Yet these appealing statistics may be offset by factors that complicate rather than simplify the tax system.

First, RRA does not address how to increase the number of electronic filings when not all taxpayers have access to the Internet. Originally, 80% had been drafted as a mandate rather than a goal. ⁹⁰ Parties such as the Treasury successfully argued that the 80% goal should be flexible rather than mandated, because meeting the goal could unduly strain fiscal resources if it proved to be unreasonable. ⁹¹ One who peruses statistics on today’s Internet user could well doubt the feasibility of this goal. ⁹² Presently, only 37% of Americans have Internet access. ⁹³ Of those with access, only 25% of them have access at home, and almost half of those without home Internet access have no plans to add it. ⁹⁴

At the same time, a variety of consultants have issued optimistic prognostications on the growth of Internet use, which may naturally be followed by a growth in electronic tax filing. ⁹⁵ Furthermore, a significant portion of returns are already being filed electronically – almost 20% of returns filed in 1998, up from 16% in 1997. ⁹⁶ In addition, increasing individual Internet access may not be wholly relevant to increasing electronic filing, since half of taxpayers do not file their own taxes but instead use tax preparers. Tax preparers, over 85% of whom used computers to prepare returns in 1997, have expressed enthusiasm for electronic filing and would be responsive to a campaign to promote a paperless system. ⁹⁷ Another five percent of taxpayers choose to Telefile. ⁹⁸ The 80% goal thus appears quite attainable, considering that an electronic filing campaign can effectively target a large percentage of taxpayers. ⁹⁹

Regardless, RRA’s 80% benchmark does at least bear on the issue of how to spread Internet use, and the corresponding question of how to fund it. ¹⁰⁰ These issues will no doubt be fueled by much debate and discussion, as they are pertinent to many groups, aside from the IRS and drafters of RRA. Although the answers to these questions are beyond the scope of this paper, one potential solution is to place terminals in public locations such as libraries and postal offices. ¹⁰¹

Second, RRA could lead to an increase in the administrability and complexity of the Internal Revenue Code. At a minimum, the IRS will be burdened with the additional work of processing electronic claims separately from paper claims during the transition from paper to electronic tax filing. ¹⁰² Beyond this, there is still a possibility that the Code itself could become more complex. Taxpayers who perceive the tax system as being complex already may be surprised to learn that the IRS uses only 40% of the data that the Code requires taxpayers to submit. This low utilization rate is largely a result of the limited capability of the IRS computers, which were designed in the 1960’s, and the difficulty of transcribing paper tax returns into electronic format. If taxpayer information is transmitted electronically, it increases the potential for increasingly opaque tax administration. ¹⁰³

Technology already exists allowing taxpayers to file taxes without leaving their desktops; for example, taxpayers who use Intuit TurboTax for MacInTax software can file electronically and then pay electronically using any NOVUS credit card. ¹⁰⁴ As software like TurboTax, which can perform calculations quickly and effortlessly, becomes more available, Congress could create more Byzantine regulations without having to worry that tax preparers would be unable to interpret them, while taxpayers would have less incentive to question calculations that are done by a machine. ¹⁰⁵

The return-free system that RRA requires the IRS to create could lead to an even more extreme scenario. ¹⁰⁶ In this situation, taxpayers may be able to elect a return-free system whereby they do not have to file taxes at all because the IRS will automatically compute them. These taxpayers, who presumably chose the return-free system for reasons of convenience, could be presented with a tax bill or refund which was the output of an incomprehensible formula. Should they later need to contest their taxes, these taxpayers would be faced with the severe inconvenience of untangling the complex machinations of the system.

There are two safeguards against an increase in tax complexity. First, RRA requires development of a return-free tax system "for appropriate individuals" only. ¹⁰⁷ Currently, this group would be limited to taxpayers who receive only wages, interest, and dividends, because the corresponding W-2, 1099-INT, and 1099-DIV forms may be remitted electronically. Even if more obscure information returns become electronically fileable, the group eligible for the return-free system will remain selective. Beginning in the year 2000, the Secretary of the Treasury must report annually to the House Ways and Means Committee on the types of taxpayers for whom a return-free option would be "appropriate" and the related costs. ¹⁰⁸ Therefore, expansion of the scope of the RRA provision beyond simple tax returns, which was not the intention of the drafters, is an unlikely possibility. ¹⁰⁹

Second, RRA takes the unprecedented step of giving the IRS a formal role in minimizing the complexity of the Code. Each year after 1998, the Commissioner will present to the House Ways and Means and Senate Finance Committees an analysis of the sources of complexity in the Code and recommendations to reduce its complexity, including repeal or modification of Code provisions. ¹¹⁰ This dialogue between Congress and IRS should provide resistance to the adoption of more complex tax law.

Moreover, electronic filing itself provides little potential for increased complexity. Taxes have become more complicated in past years because of the myriad of tax credits and accompanying phaseouts peppered throughout the Code. ¹¹¹ The availability of software that can easily and accurately make these computations, rather than the implementation of electronic filing, has facilitated the implementation of these provisions. As noted earlier, tax preparation software has already gained wide acceptance by today’s taxpayers and tax preparers; over 65% of tax returns filed are electronically prepared. ¹¹² Because the process of electronic filing is decoupled from the process of computing tax liability, shifting to a paperless tax system is unlikely to result in radical change.

IV. Conclusion: Progress, Not Merely a Change in Process

A paperless tax system holds the promise of convenience, efficiency, and accuracy unparalleled by the current system. At the same time, the IRS’s goal should be not just to increase electronic tax filing but to develop an online tax system that maintains taxpayer confidentiality while upholding principles of neutrality and progressivity. These issues are pertinent whether or not the IRS is successful in getting the majority of taxpayers to file online. ¹¹³

In terms of administrability, RRA’s 80% goal does appear realistically attainable. Over half of the returns filed every year are prepared by tax professionals, and nearly all in this subset are prepared using a computer, providing a well-defined market for the IRS to channel towards electronic filing. Tax preparers, who have expressed frustration over limitations on their ability to file electronically in the pre-RRA system, are eager to cooperate with the IRS, and software developers, who see taxpayers as consumers with the lucrative potential, will provide additional assistance by creating innovative, user-friendly programs and offering consumers cost-based incentives.

Once achieved, a paperless tax system should not facilitate an increase in the complexity of tax law. Complexity results from the legislative process of drafting tax law, which is independent of the process of administering tax law. If complexity is attributable to any aspect of the digitization process, it is attributable to the use of tax preparation software, which is capable of computing tax pursuant to complicated provisions.

Nor should a paperless tax system pose added threat to taxpayer privacy. Taxpayers are already vulnerable to invasion of privacy in today’s system, in which all taxpayer data is eventually placed on the IRS’s network, even if only a fraction travels there via modem. Given secure technology and well designed procedures, tomorrow’s taxpayers should be able to submit and request information without increased anxiety. Likewise, the IRS’s acceptance of alternative signatures should not make taxpayers any more susceptible to fraudulent returns than the present handwritten system. The law will also treat an electronic signature just as it would a handwritten one, placing the burden on the government to prove its validity.

However, RRA does place fairness values in some jeopardy. Because not all taxpayers have Internet access, not all taxpayers will bear the same costs or reap the same benefits from a paperless system. Also, the government’s efforts to work with the private sector to encourage electronic filing may misintentionally stifle competition.

To avoid these problems, the system should draw on the resources of the private sector by opening all markets to competition as fully as possible, particularly when dealing with electronic signature technology. Innovations in technology and competitive promotions by the private sector can spread the advantages of electronic filing across a more diverse group of taxpayers. Open competition will also avoid privacy problems that would stem from not the electronic tax system itself but the temptation for the federal administration to centralize its digital operations.

In summary, the benefits of moving to an electronic tax system outweigh the costs. Although the conversion to the system will not be carefree, it is administrable and maintains privacy, in addition to being less costly and more accurate. The resolution of these issues raised by plans for an electronic tax system are particularly important because of the general trend for government agencies to move activities online. For example, the Department of Treasury is now piloting the use of PKI tools to interact securely with Treasury contractors over the Internet. The Lawrence Livermore National Laboratory of the Department of Energy ("DOE") is building a PKI for its which can interface with other DOE laboratories, facilities, and vendors. And the Department of Defense ("DOD") is creating a Defense Travel Management System in which DOD travelers will utilize digital signatures issued by the DOD’s PKI to make travel and reimbursement arrangements. ¹¹⁴ As it breaks ground in the world of digital operations, the IRS can and should serve as a model for other federal agencies seeking to transform government bureaucracies into cyberservice providers.

IRS RESTRUCTURING AND REFORM ACT OF 1998

TITLE II - ELECTRONIC FILING

SEC. 2001. ELECTRONIC FILING OF TAX AND INFORMATION RETURNS.

(a) IN GENERAL.--It is the policy of Congress that--

1. paperless filing should be the preferred and most convenient means of filing Federal tax and information returns;
2. it should be the goal of the Internal Revenue Service to have at least 80 percent of all such returns filed electronically by the year 2007; and
3. the Internal Revenue Service should cooperate with and encourage the private sector by encouraging competition to increase electronic filing of such returns.

(b) STRATEGIC PLAN.--

1. IN GENERAL.--Not later than 180 days after the date of the enactment of this Act, the Secretary of the Treasury or the Secretary's delegate (hereafter in this section referred to as the "Secretary') shall establish a plan to eliminate barriers, provide incentives, and use competitive market forces to increase electronic filing gradually over the next 10 years while maintaining processing times for paper returns at 40 days. To the extent practicable, such plan shall provide that all returns prepared electronically for taxable years beginning after 2001 shall be filed electronically.
2. ELECTRONIC COMMERCE ADVISORY GROUP.--To ensure that the Secretary receives input from the private sector in the development and implementation of the plan required by paragraph (1), the Secretary shall convene an electronic commerce advisory group to include representatives from the small business community and from the tax practitioner, preparer, and computerized tax processor communities and other representatives from the electronic filing industry.

"(f) PROMOTION OF ELECTRONIC FILING.--

SEC. 2002. DUE DATE FOR CERTAIN INFORMATION RETURNS.

SEC. 2003. PAPERLESS ELECTRONIC FILING.

"(a) GENERAL RULE.--Except as otherwise provided by subsection (b) and"; and

"(b) ELECTRONIC SIGNATURES.--

(b) ACKNOWLEDGMENT OF ELECTRONIC FILING.--Section 7502(c) is amended to read as follows:

"(c) REGISTERED AND CERTIFIED MAILING; ELECTRONIC FILING.--

SEC. 2004. RETURN-FREE TAX SYSTEM.

SEC. 2005. ACCESS TO ACCOUNT INFORMATION.

FOOTNOTES

6. RRA § 2001(a)(2). See Appendix, IRS Restructuring and Reform Act: Part II, “Electronic Filing,” for full text of RRA provisions cited herein. Out of a total of 113.1 million returns filed for tax year 1997 and received by April 30, 1998, 18.4 million (16.3%) were filed electronically, and 5.9 (5.3%) were telefiled, for a total of 24.4 million (21.6%). See Taxpayer Statistics, “Table 1 - Taxpayer Usage Study: Number of Returns Received by the End of April for Tax Years 1996 and 1997, with Selected Filing Characteristics” [hereinafter Taxpayer Statistics] (visited Dec. 17, 1998) <http://www.irs.ustreas.gov/tax_stats/soi/tpus.html>.

20. See, e.g., General Accounting Office (GAO), Letter Report, IRS Systems Security: Tax Processing Operations and Data Still at Risk Due to Serious Weaknesses, GAO/AIMD No. 97-49 (1997), (visited Nov. 10, 1998) <http://www.epic.org/gao_irs.html>. Among other observations, GAO noted that the IRS continued to have serious weaknesses in the areas of physical security, logical security, data communications management, risk analysis, quality assurance, internal audit and security, security awareness, and contingency planning. It discovered that the five of the IRS’s ten facilities examined by the GAO could not account for approximately 6,400 missing units of magnetic storage media, such as tapes and cartridges, which could contain taxpayer data and that printouts containing taxpayer data were left unprotected and unattended in open areas of two facilities where they could be compromised. In particular, the report focused on electronic browsing of taxpayer data by IRS employees and the IRS’s lack of a coherent monitoring and disciplinary program.

21. Graham Greenleaf and Roger Clarke, “Privacy Implications of Digital Signatures,” Abstract (March 10, 1997) <http://www.anu.edu/au/people/Roger.Clarke/DV/DigSig.html>.

22. See Mtrust, “Frequently Asked Questions” (visited Nov. 9, 1998) <http://www.mtrust.com.my/faq>.

23. See Abanet, “Digital Signature Guidelines Press Release” (visited Nov. 9, 1998) <http://www.abanet.org/scitech/ec/isc/dsg-tutorial.html>.

27. See Daniel Weitzner, “Response to letter from General Services Administration (GSA) addressing CDT’s concerns on the Access Certificates for Electronic Services (ACES) Initiative”(dated June 29, 1998) <http://www.cdt.org/digsig/gsaletterrep.html>. See also Center for Democracy and Technology and U.S. Public Interest Group, “CDT’s concerns about the privacy and consumer issues raised by pending legislation on digital signatures and electronic authentication” (dated May 1, 1998) <http://www.cdt.org/digsig/bennett.html> (letter to Senator Robert Bennett regarding S. 1594).

28. See G. Martin Wagner, “Letter from General Services Administration (GSA) Addressing CDT’s Concerns on the Access Certificates for Electronic Services (ACES) Initiative” (dated June 4, 1998) <http://www.cdt.org/digsig/gsaletter.html>.

31. See Robert Pear, “Social Security Closes Online Site, Citing Risks to Privacy,” N.Y. Times, April 10, 1997 <http://www.nytimes.com/library/cyber/week/041097social.html>.

32. “Social Security Administration Reopens Web Site” (dated Sept. 7, 1997) <http://www.cdt.org/privacy/pers_info/ssa.html>.

33. See, e.g., Organization for Economic Cooperation and Development, “Individual Participation Principle,” OECD Guidelines Governing the Protection of Privacy and Transborder Flows of Personal Data at Part II § 13, (visited Dec. 18, 1998) <http://www.cdt.org/privacy/guide/basic/oecdguidelines.html>. This principle states that an individual should have the right:

1. "to obtain from a data controller, or otherwise, confirmation of whether or not the data controller has data relating to him;
2. to have communicated to him, data relating to him within a reasonable time; at a charge, if any, that is not excessive; in a reasonable manner; and in a form that is readily intelligible to him;
3. to be given reasons if a request made under subparagraphs(a) and (b) is denied, and to be able to challenge such denial; and
4. to challenge data relating to him and, if the challenge is successful to have the data erased, rectified, completed or amended."

36. “IRS to revamp computers” (Dec. 9, 1998) <http://cnnfn.com/digitaljam/wires/9812/09/irs_wg/index.htm>. On Dec. 9, 1998, the IRS announced that it had awarded a 15-year outsourcing contract to a private consortium led by Computer Sciences Corp, a contract which some valued at $8 billion. The consortium members include IBM, KPMG Peat Marwick, Lucent Technologies, Northrop Grumman, SAIC, and Unisys Corp. Id.

38. See “Electronic Services, 1999 Signature Pilot Information for Taxpayers” (visited Dec. 16, 1998) <http://www.irs.ustreas.gov/prod/elec_svcs/sig_txpy.html>.

58. According to the IRS, 94.1 million of the 120.4 million filers in 1996 had adjusted gross income ranging from $0 up to $50,000. See Taxpayer Statistics, “Number of Individual Income Tax Returns, Income, Exemptions and Deductions, Tax, and Average Tax, by Size of Adjusted Gross Income, Tax Years 1994-96” (visited Nov. 17, 1998) <http://www.irs.ustreas.gov/tax_stats/soi/ind_agi/html>.

60. According to the Census Bureau, as of March 1997, 23.9% of the population at least 25 years of age had completed a bachelor’s degree. See Jennifer Day and Andrea Curry, “Educational Attainment in the United States: March 1997”, U.S. Census Bureau Current Population Report P20-505 (1998). The Bureau estimates that as of September 1998, the total population was 270,498, of which there were 137,992 females. 11.3% of the population was classified as Hispanic and 12.1% as black, not Hispanic. See Laura K. Yax, Population Division of the U.S. Census Bureau, “National Estimates: Annual Population Estimates by Sex, Race and Hispanic Origin, Selected Years from 1990 to 1998” (last modified August 20, 1998) <http://www.census.gov/population/www/estimates/nation3.html>.

68. For example, beginning in January 1999, Block Financial and Intuit, two tax software developers, will offer free electronic filing for users of their Web-based products. These programs will be limited to taxpayers who file a 1040EZ return in the case of Block’s products and to taxpayers who have adjusted gross incomes of less than $20,000 in the case of Intuit’s products. These two target populations are comprised of lower-income taxpayers who represent approximately half of all taxpayers. Phone Interview with Stephen Holden, supra note 8./p>

69. VITA services low to moderate income, disabled, non-English speaking, and elderly taxpayers, and TCE services taxpayers age 60 and over. Both provide free electronic filing at selected libraries, shopping malls, and local community centers. See “Electronic Services: Using IRS e-file at VITA/TCE Sites” (visited Dec. 19, 1998) <http://www.irs.ustreas.gov/elec_svcs/vitatce.html>.

71. The rise of companies such as Wit Capital, which provides online investors with access to initial public offerings that would otherwise be distributed through underwriters, and eBay, an online auction service that matches buyers with sellers, are just a few examples. See (visited Dec. 20, 1998) <http://www.ebay.com> and <http://www.witcapital.com> (home pages for eBay and Wit Capital).

72. See, e.g., “Encyclopedia of the New Economy” (visited Nov. 23, 1998) <http://www.hotwired.com/special/ene>; Eric Lundquist, “Quit Dissin’ the Web’s Middlemen,” July 14, 1997 <http://www8.zdnet.com/pcweek/opinion/0714/14last.html>.

74. The IRS has contracted with U.S. Audiotex, NOVUS Services, Inc., and Intuit, Inc. to provide credit card processing services for balance due Form 1040 taxpayers to use MasterCard, American Express, or NOVUS/Discover credit cards to pay their taxes in 1999. Both electronic and paper filers will have the option to pay taxes by credit card through a toll free number, 1-888-2PAY-TAX, but the IRS is focusing on electronic filers and telefilers. See IRS Press Release, “IRS Partners with Industry to Expand Electronic Transactions,” IR-98-55 (1998); “Electronic Payment Options for Taxpayers” (visited Dec. 16, 1998) <http://www.irs.ustreas.gov/prod/elec_svcs/cred-car.html>.

The option to pay taxes by credit card raises additional fairness issues. It could be a valuable tool to attract balance-due taxpayers to electronic filing, who unlike refund-due taxpayers are not enticed by the promise of expedited return processing. By using a credit card, these taxpayers would be able to postpone their tax liability interest-free for one credit card billing cycle. E.g., Phone Interview with Charles Lacijan, supra note 40. However, these balance-due taxpayers tend to be higher-bracket taxpayers who have income outside of wages and interest, such as capital gains. To make up for the lost float on that interest, taxes on lower-bracket taxpayers could increase, effectively creating a subsidy for higher-bracket taxpayers. Cf. Lubick, supra text accompanying note 54.

Moreover, the credit card option has been criticized by some as encouraging consumers to deepen debt, effectively increasing the tax rate through the high interest rates charged by the credit company. See, e.g., Bill Atkinson, “IRS credit-card plan called ‘insane,’” The Baltimore Sun, Aug. 30, 1998, at 1D.

78. These are the three elements that led to the standardization of the “QWERTY” keyboard, rather than a more efficient layout such as Dvorak, according to economist Paul A. David. See Paul A. David, “Clio and the Economics of QWERTY,” AEA Papers and Proceedings, Vol. 75, No. 2, at 332 (May 1985)./p>

79. These manufacturers (with on-line software in parentheses) include Atilla M. Taluy (Atilla M. Taluy), RCS Rhodes Computer Services (Taxslayer), Intuit Inc. (TurboTax, MacInTax, WebTurboTax), UTS (SecureTax.com), Block Financial Corp. (Kiplinger TaxCut), Tax Systems, Inc.(File Safe), Tax Systems, Inc. (Prep 1040 Personal), George Albert & Co. (GAC Little ELF, Home Accounts Tax Modules), UDS ElectroTax, LLC (UDS ElectroTax Individual), Computer Language Research (One Tax.com), Nelco (Tax Wizard), AM Software (Personal 1040 - AM Tax), TaxLink, Inc. (AutoTax), 2nd Story Software (Tax ACT), ETAX Corporation (ETAX). See “1999 Signature Pilot Information for Taxpayers” (visited Dec. 16, 1998) <http://www.irs.ustreas.gov/elec_svcs/sig-txpy.htm>.

84. See, e.g., “MCG: Internet Open Group on Certificate and Security” (visited Dec. 20, 1998) <http://www.mcg.org.br> (home page for the Meta-Certificate Group, a non-profit group composed of participants from 26 countries and focused on using applied cryptography to solve Internet certification and security problems).

93. According to a Fall 1998 CyberStats report by Mediamark Research Inc., 72 million (37%) of Americans have access to the Internet. “Number of Internet Users, Today” (visited Nov. 22, 1998) <http://www.emarketer.com/estats/welcome.html>.

95. One report projects the number of Internet users to reach 85 million by the year 2002. See “1998 eOverview Report” (visited Nov. 22, 1998) <http://www.emarketer.com/estats/welcome.html>.

114. See “Federal Agency Projects Using Public Key Technology” (visited Dec. 19, 1998) <http://gits-sec.treas.gov/fpkiaccesstrust.html>.