IRS Report to Congress

May 16, 2023

Publication 5788 (5 -2023) Catalog Number 93968P Department of the Treasury www.irs.gov
Table of Contents
Executive Summary
Introduction 4
Report Structure 4
Background 5

Section 1: TAXPAYER OPINIONS 6

Taxpayer Interest 7
Taxpayers who would be interested 9
Trust
Reasons taxpayers would not adopt
Experience expectations
Factors that May Affect Interest in Direct File

Section 2: COSTS, POTENTIAL BENEFITS AND OPERATIONAL CHALLENGES 16

Costs
Additional

Operational Challenges
Conclusion

APPENDIX A: RESEARCH METHODOLOGY AND QUESTIONS 21

2022 Taxpayer Experience Survey
MITRE Survey
Simple Tax Return Scenario #1: Same Functionality and Cost 5
Simple Tax Return Scenario #2: No State Return
Simple Tax Return Scenario #3: Pre-Population
Simple Tax Return Scenario #4: No State Return and Pre-Population
Complex Tax Return Scenario #1: Same Functionality
Complex Tax Return Scenario #2: No State Return
Complex Tax Return Scenario #3: Pre-Population
Complex Tax Return Scenario #4: No State Return and Pre-Population
MITRE Survey Demographics
Qualitative User Research

APPENDIX B: OPINIONS OF AN INDEPENDENT THIRD PARTY 36

Opinions Of An Independent Third Party

IRS Report to Congress 2

Executive Summary

potential Direct File option was conducted in the context of the IRS’s IRA Strategic Operating Plan (SOP), which
details the vision for the overall transformation of taxpayer services. Filing taxes can be time-consuming and

the experience of the IRS and other federal government agencies providing new services to taxpayers and other
users, as well as the experience of other governments around the world that offer free government preparation

annual Taxpayer Experience Survey (TES). To better understand taxpayer opinions, expectations, and level of
trust in a potential Direct File option, the 2022 TES surveyed taxpayers on these additional topics. The IRS also

IRS supplemented data from those taxpayer surveys with user research and usability testing that was conducted
using a functioning internal prototype to better understand taxpayer perspectives.

A majority of taxpayers across both surveys and user research report interest in using an IRS-provided tool to

Some taxpayers report concern about the motives of the IRS in providing a Direct File tool and potential

Taxpayer preferences regarding Direct File tend to be based on the assumption that a Direct File option
would be about the same as or easier to use than other tax preparation software. Taxpayers shown a
functioning internal prototype of Direct File, developed for research purposes, reported that the software
exceeded their expectations in terms of ease of use and simplicity.
The study also investigated the costs that the IRS would incur to develop and operate a Direct File tool. The cost
estimates are subject to considerable uncertainty, and substantially depend on assumptions regarding both the
number of taxpayers who would choose to use a Direct File option and the complexity of their respective tax
situations. The assumptions underlying cost estimates in this report are not intended to foreshadow an agency
position on the scope of a potential Direct File option. Under varying scenarios relating to the scope and usage
of Direct File, the study estimates that annual costs of Direct File may range from $64 million (assuming 5 million
users and a narrow scope of covered tax situations) to $249 million (assuming 25 million users and a broad scope
of covered tax situations). Customer support accounts for more than half of the cost in all scenarios, and 84% of
the cost in the 25 million users broad scope scenario. These estimates do not include potential cost savings due

Finally, the study highlights several operational challenges that the IRS would need to address if a decision were
made to implement Direct File. These challenges include fostering technical product development expertise within
the agency, developing customer service capabilities to support taxpayers using the product, and coordinating
with states and other stakeholders to support state tax administration.
With the resources provided by the IRA, the IRS is already delivering improved customer service to taxpayers
and now has the capacity to expand service options for taxpayers. This study provides important data on
taxpayer preferences, cost, and feasibility of a Direct File option, so as to inform future Treasury Department
decisions about a Direct File option.
IRS Report to Congress 3
Introduction

169, enacted August 16, 2022), which requires the Internal Revenue Service (IRS) to deliver a report to Congress on

– For necessary expenses of the

Internal Revenue Service to deliver to Congress, within nine months following the date of the enactment of
this Act, a report on (I) the cost (including options for differential coverage based on taxpayer adjusted gross

costs to build and administer each release, with a focus on multi-lingual and mobile-friendly features and
safeguards for taxpayer data; (II) taxpayer opinions, expectations, and level of trust, based on surveys, for

approach, schedule, cost, organizational design, and Internal Revenue Service capacity to deliver such a

these amounts shall be in addition to amounts otherwise available for such purposes.
As outlined in the IRS Strategic Operating Plan (SOP), the Congressional directive for the IRS to explore and

through a free IRS-provided tool, was conducted in the context of broader IRS transformation efforts under the
IRA to improve service to help taxpayers meet their tax obligations and receive the tax incentives for which they
are eligible.2 The IRS created a Direct File Task Force, a cross-functional group of IRS employees, supported by
the United States Digital Service (USDS), to conduct this study.
Report Structure

to the 2022 Taxpayer Experience Survey (TES).3

commissioned and designed by MITRE Corporation without the involvement of the IRS or the Treasury
Department (the MITRE Survey),4 which asked taxpayers about expectations and level of interest and trust in

usability testing about a potential Direct File option.

The IRS, with the help of USDS, built a functioning internal prototype5
tool. This prototype provided the Task Force with a deeper understanding about the type of support that taxpayers
want and expect from the IRS. The prototype, while limited in its capabilities, allowed taxpayers participating in

an abstract description of a hypothetical product. This introduction to the look and feel of a potential tool – albeit

with the IRS might impact those expectations. The prototype uses an interview-based experience to ask users

generate or submit an actual return.6 Taxpayers interacted directly with the prototype in controlled usability testing
sessions, providing the IRS with insight into whether the experience of such an IRS-provided tool would meet
taxpayer expectations. That analysis is also included in this section.
––––––––––––––––––––––––
1

3
The Taxpayer Experience Survey is an IRS-sponsored annual phone and online-based survey, and engages taxpayers regarding their

(FFRDC) sponsored in part by both the U.S. Department of the Treasury (Treasury) and the IRS. MITRE conducted this survey
independently in its role as FFRDC administrator. Treasury and IRS did not commission this survey and were not involved in its
development, design, or implementation.
5
In this context, a prototype is a working model of a computer software product with limited functionality, which is built to test a concept and process.
6
An interview-based experience allows taxpayers to complete a return without having to reference external instructions.
IRS Report to Congress 4
deploying Direct File. Cost estimates explore the effects of supporting different levels of return complexity as
required by the IRA, including the resources necessary to keep pace with tax law changes over time. The
estimates in this section are informed by the IRS’s experience in building a functioning internal Direct File
prototype for usability testing.
As required by the IRA, Appendix B of this report provides the opinions of an independent third-party. With the
exception of the “Overview” section to provide context for the appendix, the IRS did not contribute to or edit the
contents of Appendix B prior to the publication of this report.
Background

approximately 95% of the revenue to support the operations of the U.S. government.7 The amount of time the

activities makes up approximately 71% of all Federal paperwork burden annually.8 On average, including all
associated forms and schedules for non-business income, individual taxpayers spend approximately 8 hours and
$140 preparing their taxes each year.9

returns. Tax returns can be prepared and submitted in numerous ways, including via paper through the mail,
electronically through IRS’s Free File or Free Fillable Forms, through a free tax return preparation site, such as

commercial software, both paid and free, and/or with support from professional tax preparers.

meet their obligations and receive the tax incentives for which they are eligible.”10 Each objective includes a
number of initiatives, which are strategic bodies of work that will help drive transformation within the IRS. As part

IRA.11
outcomes of the Direct File study.

––––––––––––––––––––––––
7
IRS. “Final Monthly Treasury Statement for Fiscal Year 2022 through September 30, 2022, and other Periods.” Page 38.

8
IRS. Research, Applied Analytics & Statistics. Tax Compliance Burden Report 2018. Page 1.
9
IRS. Tax Year 2022 Instructions for forms 1040 (and 1040-SR). Page 107. . These

10

11

IRS Report to Congress 5

Section

Taxpayer Opinions, Expectations, and Level of Trust

The IRA charged the IRS with surveying taxpayers on their “opinions, expectations, and level of trust” with
an IRS-run Direct File system. The IRS’s 2022 Taxpayer Experience Survey (TES) included additional survey

2021 Federal tax return in 2022. Participants were randomly selected using NORC’s AmeriSpeak panel12 to
obtain a representative sample of taxpayers covering approximately 97% of U.S. households, with an additional

by phone, in English or Spanish.13

The TES provided taxpayers with a description of a hypothetical free IRS-provided online tool used to prepare
(See
for a full list of related questions). The survey responses
provide information regarding taxpayers’ opinions, expectations, and level of trust with a Direct File option, as well

The TES results were supplemented by results from the MITRE survey. The MITRE survey was entirely online,
and 2,000 total respondents opted in to complete the survey. While opt-in surveys may not be representative of

another for more information about the MITRE survey).14

the Task Force consisting of two components: unstructured interviews and usability testing. The unstructured

a hypothetical Direct File tool.15 The usability testing consisted of interviews with 14 taxpayers using a functioning
internal prototype of a potential IRS-run, free Direct File tool. This additional research (referred to collectively
as “user research” below) provided qualitative insight into taxpayer opinions, expectations, and level of trust in a
Direct File tool.

––––––––––––––––––––––––
12
NORC at the University of Chicago. “AmeriSpeak: NORC’s Breakthrough Panel-Based Research Platform.”

13

14
In conjoint analysis, participants choose between a set of options with different attributes. For more information on methodology, see
Appendix A.
15
Interviews spanned taxpayers with a range of demographics including geography (with all Census regions represented), race, gender,
income, and educational attainment. for more information.

IRS Report to Congress 6

Taxpayer Interest
The 2022 TES asked taxpayers whether they would be interested in using a free IRS-provided online tool to
16

interested” or “somewhat interested.”17 Among taxpayers who currently self-prepare their taxes using commercial
software, 68% of taxpayers reported that they would be either “very likely” or “somewhat likely” to switch to a free
IRS-provided online tool.

Source: 2022 Taxpayer Experience Survey

among all taxpayers among current users of self-preparation software

• Very interested 28% • Very likely 24%

• Somewhat interested 45% • Somewhat Likely 44%
• Not very interested 17% • Somewhat unlikely 21%
• Not interested 11% • Very unlikely 11%

The MITRE survey asked respondents to choose between three hypothetical options for tax preparation and

in which taxpayers would manually enter their information into the IRS-provided product, and a free IRS-provided

that it receives from your employer.” When faced with these three choices – all of which assume the product is
free to the taxpayer – a majority (52%) of respondents stated that they would prefer to use one of the two IRS-
provided options (split 15% manual entry option, 37% “return-free” option), and 48% stated they would continue to

Taxpayers’ survey responses were based on high-level descriptions of an IRS-provided option or options.

Direct File” later in this report). When considering a real IRS-provided tool, taxpayer interest could shift based on
design, functionality, and other product choices.
The survey data from TES and MITRE, along with the Task Force conducted user research, reveal a range

taxpayers who were interested in using an IRS-provided option typically reported one of the following three
reasons as to why they would be interested.

––––––––––––––––––––––––
16
The prompt reads, “Based on the description, how interested would you be in using this online tool to prepare your taxes if it becomes
available?”
17

differences in individual vs. combined rounding.

18
Percentages provided in all charts in this report may add up to greater than 100% due to rounding.

IRS Report to Congress 7

Reason 1: IRS’s role

proper role for the IRS. On the MITRE survey, 46% of taxpayers who expressed interest in an IRS-provided tool stated

Others disagree. Among taxpayers who chose a free commercial option19 in

the MITRE survey, 11% cited “I don’t think it’s the IRS’s role to prepare taxes”

“I don’t want the IRS to prepare my return” as their most important reason
(see “Reasons taxpayers would not adopt” later in this report).
Some taxpayers hadn’t previously considered the possibility of IRS providing a
Direct File option. As one interviewed taxpayer said in a user research session,
“I don’t know why I never really thought about them doing it, so I didn’t even

consider using it. It just didn’t occur to me before.”

Some interviewed taxpayers also felt that the IRS should make taxes easier. One user research participant said,
“At the end of the day, I am a big believer in paying taxes but the enormous stress I’m incurring with my family,

sorry! Okay, why don’t you provide software for us to do it? […] So here, I’m going to send it to you on this sheet
of paper instead, so it didn’t cost me any money.”
Reason 2: Cost

most. In one of scenarios covered in the MITRE survey, respondents were given a choice between a commercial
product costing $80 and a free IRS option. In that scenario, 70% of taxpayers chose the IRS option. Even when

described their frustration with upselling tactics, as well as cost and complexity, during their “I can just do it
interview. “I guess really it comes down to the cost. The software is not cheap. I think the on paper and
minimum you’re going to pay is $70, $80 to $100 bucks, and then I have to look further into,
you know, there’s different versions if you’re going to use this form or this form, and you
need this. […] I can just do it on paper, and it costs me $0.”
Reason 3: Security and Privacy

provided tool, and was selected as the most important reason by 18% of taxpayers who chose an IRS tool from
a set of multiple-choice options (See Figure 7 below).
Taxpayers expect to entrust government with sensitive information. As one
interviewed taxpayer said in a user research session, “Usually when it’s a
government website, I don’t feel uncomfortable. I don’t put my personal info
much out there, but government is okay.” (See “Trust” later in this report.)
Some taxpayers are aware that some commercial entities sell or repurpose their
data. Describing their expectations of an IRS tool in a user research interview,
one taxpayer said, “I think it’d be very protected. I don’t think the IRS is going
to sell [my personal data] off, they’re not that interested in making money. So, I
mean, not in that respect, not like a third party, where they would be.”
19
In the MITRE survey, respondents were offered scenarios in which their current commercial software were made free to use.

IRS Report to Congress 8

In user research sessions, taxpayers also noted that the IRS is already receiving their data. One interviewed

other means] they send it to the IRS… I’m going straight to the source.”
Taxpayers who would be interested
It is important to understand which taxpayers would be interested in an IRS-provided option. There are different
motivations for taxpayers to choose one option over another. However, a few high-level themes emerge.
Theme 1: Self-preparation
Taxpayers who self-prepare taxes20 are more likely to be interested in a self-preparation tool from the IRS. The
2022 TES data shows that 83% of taxpayers who self-prepared their return would be very or somewhat interested
in an IRS-provided tool, as compared to 57% of those who use a paid preparer.

Source: 2022 Taxpayer Experience Survey

• Somewhat
Very interested 37%
• Not very interested
interested 47%
• Not interested 5% 12%
•
• Somewhat
Very interested 18%
• Not very interested
interested 39%
• Not interested 19% 24%
•
of their tax situation. For example, one interviewed taxpayer stated in a user research session, “I’m self-

individual was currently uncomfortable doing their own taxes, they had previously self-prepared when their
situation was more straightforward. This same taxpayer continues to routinely help others prepare their taxes.
They said, “I’m lucky [I] can afford to pay this accountant, but I run into so many people who are very simple in
terms of taxes, and yet they still have to pay […].”
Taxpayer preferences are also affected by their perceptions of preparer credibility. In one user research session,

unscrupulous preparer. They were open to an IRS-provided option, saying, “I actually think that’s really cool. I
would imagine doing it directly with IRS that they’d have guidance on how to do it correctly, keep you from doing
something you’re not supposed to, getting it from a [untrustworthy] source.” This taxpayer’s view centered on
credibility and wanting to get their taxes right.

––––––––––––––––––––––––
20
“Self-preparer” describes taxpayers who prepare their return themselves, without the services of a tax preparer, regardless of

IRS Report to Congress 9

Theme 2: Age Considerations
Taxpayers of all ages demonstrated interest in an IRS-provided option, 72% of all taxpayers were “very
interested” or “somewhat interested.” However, interest was somewhat diminished among older taxpayers, with
61% of seniors “very interested” or “somewhat interested.”

Source: 2022 Taxpayer Experience Survey

• Very interested 28%

• Not very interested
Somewhat interested 45%
• Not interested 11% 17%
•
• Very interested 24%
• Somewhat interested 37%
• Not very interested 19%
• Not interested 20%
Theme 3: Limited English Proficiency

be “very interested” or “somewhat interested” in an IRS-provided tool, 72% for the overall population versus 81%

experience would be “somewhat easier” or “much easier” (see “Experience expectations” later in this report).

preparation options, or those options may not provide appropriate language access, meriting further study.

Source: 2022 Taxpayer Experience Survey

compared to other tax

in an IRS-provided tool preparation software

• Very interested 28% • Much easier 30%

• Not very interested
Somewhat interested 45% • easier 31%
Somewhat
• Not interested 11% 17%
• • About the same 29%
• Somewhat more
• Very interested 42%
• Somewhat interested 39% • 4%
• Not very interested 13%
• Not interested 6%

IRS Report to Congress 10

Trust
As outlined in the IRS’s SOP, fostering public trust is central to the agency’s mission and protecting taxpayer
data is foundational to taxpayer service. As the IRS continues to improve customer service and invest in
cybersecurity protection, it is important to examine the role of trust in Direct File. The TES asked taxpayers about
21

in the IRS to protect their information. For taxpayers responding to the TES, 72% of taxpayers said they would
“somewhat” or “very much” trust an IRS tool to protect their information. See Figure 5 below.

Source: 2022 Taxpayer Experience Survey

• Very much 18%

• Somewhat 54%
• Not very much 20%
• Not at all 8%

However, there are more dimensions to trust than just data security. In user research sessions, interviewed
taxpayers shared several common perspectives about the IRS potentially providing this service.
Trust Perspective 1: IRS is authoritative and trustworthy
Taxpayers view the IRS as the authoritative source of information about taxes and
expect that the information presented by an IRS-provided tool would be similarly

to tell you everything you need to know because it’s the IRS. The information should be
up to date. I would probably expect it to be more accurate doing it directly with IRS than
though some other company. […] Just because it’s the source. They probably more
want you to do it correctly than some other company.”
In user research sessions, interviewed taxpayers often trusted information from the IRS as much as or more
than they trust information from a third party. One said, “It’s directly from the IRS. I would trust this tool more
than a third-party tool.” For other taxpayers, their trust was more balanced: “Yes [I would trust it]. Because it’s

Trust Perspective 2: Concerns about the IRS’ role as a tax administrator

The IRS’s role as a tax administrator leads some taxpayers to be wary of any interaction with the agency. One
interviewed taxpayer said of a potential IRS-provided tool, “Maybe the IRS part is intimidating because you’re
going to the source. It’s the IRS, you don’t want to be on their radar. […] I don’t want to be on their radar. I
feel safer going through a third party. It’s funny I feel that way. My taxes are so simple. Just the whole IRS is
intimidating.”
This concern may translate to unease about taxes generally. Another interviewed taxpayer said, “I’m not really
that comfortable with [taxes]. I also don’t necessarily like it. And you have to be careful if you make a mistake, it
could be audited. You could get in trouble. […] So, I still have that mindset of the evil IRS, where you get a letter
from them or something happens, you all of a sudden lose your house and go to jail. It’s obviously not that
––––––––––––––––––––––––
21
Question wording: “Based on the description, how much would you trust this online tool to keep your information safe?”

IRS Report to Congress 11

drastic but I still have those negative connotations about that branch of the
government.” This taxpayer had received a letter about an issue that was
“more or less a mistake” more than a decade ago and described the letter
as “kind of scary.” They concluded, “I don’t mind paying taxes. I’ll gladly pay
them, but yeah, I don’t think the comfort level ever is going to go up.”
Some taxpayers believe the IRS’s motivations for providing such a tool might not match their own. As one
interviewed taxpayer described, “The only concern I would have — coming from a very biased opinion that I don’t
trust anyone ever, especially the government — that they [wouldn’t] give you the biggest return or smallest taxes
to pay. Filing directly to the IRS, they want to make the most money. But that’s illegal so they wouldn’t.” Even

belief could still make them less likely to adopt an IRS-provided tool.
Trust Perspective 3: Privacy and Compliance Concerns

return. Some interviewed taxpayers expressed some worry that this “draft” information, or other data, could
be used against them. As one interviewed taxpayer explained, “I would say I kind of don’t trust it to an extent
because I know I often like to play around with things or check things. So, I think they might be tracking things
that I’m doing […] [If you got audited] would they have the ability through whatever to pull up your session to say,
‘It says that you really struggled with this?’ […] [With current self-preparation options] all they’re going to see is the
end result. They’re not going to know how much time I spend on it […] I just think there’s a lot of variables there
that I think they could track.”
Reasons taxpayers would not adopt
TES data shows that 32% of taxpayers who self-prepare their taxes would be “very unlikely” or “somewhat
unlikely” to switch to an IRS-provided tool . These taxpayers were also asked the question
“Why are you unlikely to switch from your current tax preparation software to the described online tool?” and
could select multiple responses. As seen in Figure 6 below, over one third of taxpayers who said they were

as a driver of interest.

Source: 2022 Taxpayer Experience Survey

method/software I used previously = 44%

previous method I used than the described online

tool = 30%
I don’t like the described online tool would be as
good as tax software I pay for = 17
Other = 9%

IRS Report to Congress 12

The MITRE survey asked a similar question but provided taxpayers with different options. Taxpayers who
preferred their existing, free tax preparation software over other options were asked to select up to three reasons
for their choice.

Source: MITRE Corporation

I want to stay with my brand of I would prefer to give my

software
the IRS instead of a third party
I don’t want the IRS to prepare I believe my data is more secure
my return with the IRS

I don’t think it’s the IRS’s role to I think the IRS is better suited
prepare taxes
company

I believe my software offers I believe I’d be less likely to be

better customer service than audited
the IRS could

Since there’s no cost for any I think the IRS could provide
of the options, I’d rather use better customer service than a
commercial software commercial software company

•First •Second •Third •First •Second •Third

These results from both surveys are broadly consistent with past research.22 As shown in Figures 6 and 7 above,

for them before. Some taxpayers also demonstrate brand loyalty for tax preparation software. One hypothesis

Experience expectations
The TES asked taxpayers about their expectations for the experience of an IRS-provided tool.23 55% of taxpayers
expected the tool to be “about the same” as other tax preparation software. 31% expected it to be either

––––––––––––––––––––––––
22
See Patterson et al., , MITRE Corporation,

to use this method,” 33% selected “This method allows me to retain information from my previous year’s return,” and 30% selected

23

IRS Report to Congress 13

Source: 2022 Taxpayer Experience Survey

compared to other tax preparation software

• Much easier 13%

• Somewhat
easier 18%
• Somewhat
About the same 55%
• more

User research sessions revealed that taxpayer expectations are informed by their experiences with other
government websites. Some interviewed taxpayers described negative experiences, like the taxpayer who relies
on a non-IRS government website that is “Always on weekends, it’s down. The browser experience wasn’t great.”
Similarly, when asked, respondents noticed and appreciated when government sites were able to provide good
customer experiences.

In qualitative user research, taxpayers described how their expectations are colored by recent experiences
with government websites:

“[SSA.gov was] surprisingly easy. I was able to do it “I get VA health care and it was pretty easy [to sign
without interacting with anyone. Answering a couple up] from what I remember. … I kind of expected it
of questions and [a replacement Social Security
card] got mailed to me in the mail. So it was actually I know the government and I know from my
very simple, very easy to do.” experience, things are not that straightforward
or easy. From my past in the military and stuff, I
just know all the red tape and stuff that is usually
“[COVIDTests.gov] was really straight-forward, involved with doing things that are supposed to be
so easy, accessible. … Since it was such a good simple. But I had a pretty good experience with the
experience, I sent it to my whole family.” VA since I signed up.”

The factor that taxpayers are most likely to worry about when considering their experience is the accessibility of
the language used. One interviewed taxpayer said, “Jargon can be overwhelming sometimes. My mind doesn’t
absorb all that wording.” There is skepticism among some taxpayers that IRS would be able to describe taxes
using plain language. Another interviewed taxpayer said, “I’m very skeptical of a government website, where
things could be simple, because they have to cater to all scenarios and potentially, any lawsuits and things like
that. The reason why things are often very complicated is because of bureaucracy and edge cases, right? […]
So I’m very skeptical that government can have a good experience.” This reaction underscores the importance of
the objectives outlined in the IRS SOP, including updating notices and transcripts using clear, simple, and plain
language to ensure that taxpayers can seamlessly interact with the IRS in the ways that work best for them.

IRS Report to Congress 14

After trying the prototype of an IRS-provided tool, interviewed taxpayers
reported that the software exceeded their expectations in terms of ease of use
and simplicity. One said, “It was a lot easier than I expected it to be. I guess

exceeds my expectations of what that would look like. […] I think this tool is
simple, it’s easy to use, and it’s not unnecessarily complicated. And for that I
appreciate it, and for that it exceeds my expectations.”
Factors that May Affect Interest in Direct File
Taxpayers have preconceived notions of what an IRS-provided tool would look like and do. The functioning internal
prototype was useful in better understanding taxpayer expectations in the context of user research interviews. User

Factor 1: State and Local Tax Preparation Support

Taxpayers who currently use commercial tax preparation software may expect to be able to use a single software

for more details). The TES responses might have changed if taxpayers were

return. The MITRE survey presented a hypothetical scenario in which taxpayers’ current commercial software

state returns, 60% of taxpayers chose their current commercial software, compared with 41% who chose an IRS-
provided option. As one commenter on the MITRE survey wrote, “State taxes preparation are going to cost me
money, so the IRS being free no longer offers any value.” This data suggests that taxpayer interest in a potential

Factor 2: Tax Information Pre-population

Some interviewed taxpayers know that tax forms such as the W-2 and 1099s
are sent to both them and IRS. Some expressed a preference for this data
to be used to streamline the experience of an IRS-provided tool. After using
the functioning internal Direct File prototype without this capability in a user
research interview, one interviewed taxpayer said, “IRS already knows your
tax information. So why wouldn’t I be able to login, put in, say, my [SSN],

corrections, you know?” When asked about their preference, a larger share of respondents in the MITRE survey

you using W-2 and 1099 information that it receives from your employer,” as compared with an IRS-provided
option requiring manual input of tax information. These results suggest that taxpayer interest in a potential Direct
File tool is likely to be greater if it includes the capability to pre-populate returns with tax information.
Factor 3: Preferential Treatment for Direct File Users

would receive preferential treatment, such as reduced processing times or lower likelihood of being selected for

options. The IRS does not, and would not, give preferential treatment to taxpayers who use any particular tax

of this expectation, and how to provide additional clarity for taxpayers on this topic, consistent with the IRS SOP’s

Interest in a potential Direct File tool may be overstated by taxpayers who incorrectly assumed that Direct File
users would receive preferential treatment.

IRS Report to Congress 15

Section COSTS, POTENTIAL BENEFITS, AND
OPERATIONAL CHALLENGES
Costs
The IRA instructs the IRS to estimate “the cost (including options for differential coverage based on taxpayer

including costs to build and administer each release […]”

the initial and ongoing costs of not only the technology, but the end-to-end service. This includes the cost of

keep pace with changes to tax law, as well as the cost of providing customer support.
These estimated costs of developing and running Direct File vary widely depending on the number of taxpayers
who choose to use such a system, primarily as a result of the cost of providing customer support. The cost
estimates discussed below assume that if a decision is made to build and maintain Direct File, the IRS would
leverage iterative product development practices, starting with a limited initial tax scope and adding additional tax
situations over time. In this iterative model, supporting more tax situations has a relatively limited impact on the
technology cost.
As with any such estimates, the cost estimates presented in this report are subject to uncertainty. The demand for

the technology and product development could require more or fewer resources than estimated here.
Developing a cost estimate for a potential Direct File option, as the IRA directs IRS to do in this report, requires
making assumptions about the population of taxpayers served and the complexity of their tax situations. The
assumptions are not intended to foreshadow an agency position about the scope of a potential Direct File option.
AGI and Return Complexity
The IRA requires the examination of the cost of “options for differential coverage based on taxpayer adjusted
gross income and return complexity.” Imposing an Adjusted Gross Income (AGI) limit would indirectly affect
cost due to exclusion of some portion of taxpayers, reducing potential number of calls and call center agents
required. However, it would not directly affect the cost or timeline for technology delivery as it does not change
the necessary functionality.
Instead of imposing an AGI limit, IRS could determine a scope of covered tax situations. For example, the
Volunteer Income Tax Assistance (VITA) program aims to assist taxpayers who generally make $60,000 or less,
24
VITA covers the tax situations that are most common among
taxpayers in these groups (e.g., wage income, the Earned Income Tax Credit) and marks other, less relevant
provisions as out of scope (e.g., certain deductions associated with business income).
Taxpayers using Direct File are unlikely to think about eligibility in terms of supported forms and schedules.25 As

use an IRS-provided tool and provide support for tax provisions relevant to their tax situations. Over time, the tool

The IRS estimated costs using two potential scopes of covered tax situations. The lower end of the range
assumes Direct File eventually matches the scope of the VITA program. Taxpayers of any income would be able
to use the tool if they did not require support for provisions that are out of scope. The higher end of the range
assumes Direct File would be able to eventually support a more expansive set of tax situations, including income
types, deductions, and credits that are not supported by VITA.

––––––––––––––––––––––––
24
IRS. “Free Tax Return Preparation for Qualifying Taxpayers.”

25
Information obtained through user research interviews.
IRS Report to Congress 16
Technology Costs
Technology and product costs include expenses related to development and maintenance, hosting, licenses,
and shared services. To estimate technology and product costs in the Direct File context, the IRS also needs to

all IRS systems, Direct File would have to keep pace with those changes. Just as IRS makes annual changes to
its internal systems in accordance with changes in the tax code, a taxpayer-facing Direct File tool would need to
undergo frequent updates. For the tool to be sustainable, software engineers cannot be the only ones who can
update the tool and test the underlying tax logic, the ordering of questions, and the questions themselves. This

costs and increase budget predictability.

ongoing operating costs is ill-suited to the challenge of this project. Instead, the technology budget was estimated
based on a continuous development approach, requiring a moderate, steady investment over time to develop
and maintain the product. A continuous development approach ensures the product meets the experience
expectations of taxpayers year after year, while also keeping costs predictable. However, for continuous
development to work, it is imperative that the program have consistent funding at adequate levels. Without
adequate funding for each release, the system would be at risk of rapidly becoming obsolete.

Cost Product quality Cost Product quality

Consistent with this approach, the technology and product budget estimates the ongoing annual cost of a blended

developing the functioning internal prototype, which provided insight into the level of effort required to implement
Direct File at scale.
Customer Support
Customer support would be critical to the success of any Direct File option, and is also a major cost driver.

thus estimated costs vary widely depending on usage. For purposes of this cost estimate, IRS assumes that a
potential Direct File option would have in-product support (i.e., help text and other documentation), but some
taxpayers will want to talk to a customer service agent.
Customer support would provide technical support and answer simple tax questions (e.g., “am I eligible to use
Direct File?”). It would also help taxpayers resolve issues that may have led to a rejected return (e.g., a return

would maintain an ongoing feedback loop between the product team and the customer support team to maintain
a holistic experience for taxpayers and to ensure that they are getting the level of support needed through the
communication channel of their choice, including through the phone.
In estimating the cost of Direct File, the IRS considered any potential impacts a potential Direct File option could

existing IRS call center agents, the Task Force has estimated a vendor-run managed service, inclusive of call center
agents (both English and Spanish speakers), customer relationship management (CRM) software, training, and
telephony, and interactive voice response (IVR), with a handoff of only more complex calls to existing IRS agents.

IRS Report to Congress 17

To estimate customer support costs, a ratio of 10,000 users per customer support agent was used.26 The vendor

more complex tax situations are supported, more highly trained call center agents would be required to provide
assistance (See Cost Estimates, below).
Cost Estimates
As noted above, the cost estimates presented here are subject to uncertainty due to the nature of providing a new
taxpayer-facing service. In addition, there is substantial uncertainty about how many taxpayers would choose
to use an IRS-provided option. To account for this uncertainty, a range of assumptions regarding the number of
users were considered in formulating cost estimates, ranging between 5 and 25 million.
A cross-functional team of government employees and contractors would be responsible for development and
maintenance. As described above, the costs of developing and maintaining Direct File were estimated for two
alternative scenarios regarding the complexity of the supported tax situations. Estimates were developed based
on the experiences of IRS and other government agencies that have implemented new public-facing digital
services. These estimates are shown below in Table 1.27

(Narrow Scope) of Total Cost (Broader Scope) of Total Cost

Technology & Product $23.7m 36.9% $33.8m 43.4%

Customer Support $40.6m 63.1% $44.1m 56.6%

Technology & Product $25.5m 24.6% $35.5m 29.4%

Customer Support $78.1m 75.4% $85.1m 70.6%

Technology & Product $30.7m 13.9% $40.8m 16.4%

Customer Support $190.6m 86.1% $208.1m 83.6%

––––––––––––––––––––––––
26
While customer support for a potential Direct File option might decrease demand to existing call centers, it would also create entirely

of current call frequency.

27
below.

IRS Report to Congress 18

The estimated total cost per return under the narrow scope and broader scope scenarios is presented in Figure
10. As usage increases, the estimated cost per return decreases.

Number of returns (in millions)

Limited scope Expanded scope

Additional Benefits and Challenges

above relating to taxpayer opinions and expectations, as well as costs.

Potential Benefits

Filing a tax return is one of the most common interactions most Americans have with the Federal Government,
and it can be complex, expensive, and time-consuming. During Fiscal Year 2022, individual taxpayers submitted
nearly 160.6 million tax returns to the IRS.28 An individual taxpayer is estimated to spend 8 hours and $140 out-

income.29
expensive process.

whether that option is currently commercially available or a potential IRS Direct File option. Despite the IRS only
receiving 8% of returns on paper, the cost to process those returns is nearly 70% of total processing costs.30

process.31 In addition, a potential Direct File option using an interview-based interface could reduce complexity and

––––––––––––––––––––––––
28
IRS. Data Book 2022. Page 2.
29
IRS. Tax Year 2022 Instructions for forms 1040 (and 1040-SR). Page 107.

30
IRS. Taxpayer Advocate Service. 2022 Annual Report to Congress. Page 105.

31
IRS. Taxpayer Advocate Service. “Most Serious Problems.”

IRS Report to Congress 19

As described in the SOP, one of the IRS’s goals is to make it easier for taxpayers to meet their tax obligations and
to meet them where they are by giving them choices in how they interact with the IRS. Direct File would be an

to help all taxpayers to receive the credits and deductions for which they are eligible. According to a report by the
Treasury Inspector General for Tax Administration (TIGTA), roughly 5 million potentially eligible taxpayers fail to
claim EITC each year.32 Direct File could reduce barriers for taxpayers to access the tax credits and deductions
they may be eligible for, improving the customer experience.
Operational Challenges
Developing Direct File would require the IRS to develop new skills and processes to operate, maintain, and
improve complex technological products. While the IRS has experience developing successful new products such

driven software, updating the tool to keep pace with tax law changes, and ensuring taxpayer information is kept
private and secure.
Customer support would introduce additional challenges as well. For Direct File to be viable in the long term, it
would require continued funding to meet these service goals while maintaining resources for other valuable IRS
services and programs.

one place. Any solution would require collaboration with state and local tax administrators to enable integration

potential Direct File option.

CONCLUSION
Taxpayer survey data, user research, and the prototyping work conducted by the Direct File Task Force have
demonstrated that there is taxpayer interest in an IRS-run free Direct File option. The IRS’s ability to deliver such
a system would require ongoing funding at adequate levels, and other operational considerations and challenges
remain. With these considerations in mind, Direct File should be considered among future options for agency
technological transformation and customer experience improvement.

––––––––––––––––––––––––
32
Treasury Inspector General for Tax Administration. “The Internal Revenue Service Should Consider Modifying the Form 1040 to
Increase Earned Income Tax Credit Participation by Eligible Tax Filers.” April 2, 2018.

IRS Report to Congress 20

APPENDIX A:
RESEARCH METHODOLOGY AND QUESTIONS

Selection criteria Random selection Opt-in Recruitment based on

demographic and income criteria
Sampling Weighted, representative sample Unweighted, non-representative Unweighted, non-representative
sample sample
Number of participants N = 4,219 N = 2,000 (across two surveys) N = 14
Timespan August 22, 2022 – October 10, 2022 December 7 – 12, 2022 December 16, 2022 –ongoing
Language English + Spanish English only English only
Modality Online + phone Online only Videoconference
Question and response options One survey, multiple-choice Two surveys, choice-based Unstructured, one-on-one
responses for each question conjoint analysis of four interviews and usability tests with
scenarios + reasons for each a functioning internal prototype
choice in each scenario

2022 Taxpayer Experience Survey

Methodology
The Taxpayer Experience Survey (TES) is the premier survey of the individual Form 1040 population. The

preferences. The survey aims to:

Measure taxpayer awareness and use of IRS products and services,
Solicit taxpayer feedback on various aspects of the taxpayer experience, and
Obtain satisfaction ratings.

were randomly selected using NORC’s AmeriSpeak panel to obtain a representative sample of taxpayers
covering over 97% of U.S. households.33 Only one taxpayer per household was interviewed.
Respondents could take the survey online or by phone, in English or Spanish. Spanish-speaking, limited English

In all, 4,219 individual taxpayers completed TES, including 3,885 from the AmeriSpeak/NORC sample and 334

U.S. population.

––––––––––––––––––––––––
33

IRS Report to Congress 21

Questions
The TES questions related to the possibility of an IRS-provided online tool were as follows:

(N = 4,193)
RESPONSE OPTIONS:
Very interested (28%)
Somewhat interested (45%)
Not very interested (17%)
Not interested at all (11%)

(N = 4,199)
RESPONSE OPTIONS:
Very much (18%)
Somewhat (54%)
Not very much (20%)
Not at all (8%)

(N = 4,170)
RESPONSE OPTIONS:
Much easier than other tax preparation software (13%)
Somewhat easier than other tax preparation software (18%)
About the same (55%)

IRS Report to Congress 22

(N = 4,157 – 4,188)

75% 17% 5% 3%
85% 9% 4% 2%
47% 38% 11% 3%

37% 25% 25% 13%

66% 26% 5% 3%
76% 19% 3% 2%

(N = 1,733)
RESPONSE OPTIONS:
Very likely (24%)
Somewhat likely (44%)
Somewhat unlikely (21%)
Very unlikely (11%)

(N = 538)
RESPONSE OPTIONS:

I do not think described online tool would be as good as tax software I pay for (17%)

Another reason (please specify) (9%)

MITRE Survey

Methodology

(These surveys are collectively referred to as “the MITRE survey”).34 Conjoint analysis is often used in market
research to identify which attributes of a product or service are most valued to consumers. Respondents are
presented with a set of products that have attributes that vary between them. For example, one product might

By varying the attributes of a set of hypothetical products, the researcher is able to draw conclusions about how
respondents may value those individual attributes.

––––––––––––––––––––––––
34

Center (FFRDC) sponsored in part by both Treasury and the IRS. MITRE conducted this survey independently in its role as FFRDC
administrator. Treasury and IRS did not commission this survey, and were not involved in its development, design, or implementation.

IRS Report to Congress 23

Unlike traditional conjoint analysis, the MITRE surveys only varied two attributes, and only of the IRS options:
whether the IRS options included pre-population of information return data, and whether the IRS options included

other attributes described, such as to examine the impact of the option being provided by IRS instead of a

This IdV attribute was applied to all IRS options (but not commercial options) regardless of whether the described
product would require it under National Institute of Standards and Technology (NIST) standards.
One of the MITRE surveys included multiple potential IRS services in each scenario, including both an IRS Direct
File option and a hypothetical return-free tax system. By including this return-free option in all scenarios, varying

this survey about how taxpayers would approach a more limited set of options (i.e., if given the choice between a
commercial software product and an IRS-provided tool).

choice. These reasons were pre-set and vary by the option selected.
The inclusion of a return-free tax system also affects the results from the reason ranking exercise. Taxpayers who
may have selected either an IRS Direct File option or a commercial software product may have been diverted to
the return-free tax system. Had they not been diverted, they may have preferred different reasons for their choice

Taxpayers opted in to participate in the surveys, which were administered online. Taxpayers indicated that they
were age 18 or older, a U.S. taxpayer, and had used tax preparation software or an online website to prepare
their 2021 federal tax return. Each survey was completed by 1,000 respondents.

included the return-free tax system, in addition to the commercial software company and IRS Direct File options.

The second survey posited a “Complex Tax Return,” with wage and interest income, requiring “additional forms

option cost $80, while the IRS Direct File option was free.
Questions
Each of the two surveys (Simple and Complex tax returns) included four scenarios, which were provided to all

the attributes used to describe each option is provided. Then, in the following tables, each available reason is

“Top 3” indicates the total percentage of taxpayers who ranked that reason among their top three choices.

IRS Report to Congress 24

 Federal tax preparation and
software package that was down- return for you using the W-2 and
loaded to your computer or used downloaded to your computer or 1099 information that it receives
through a company’s online website used through the IRS website from your employer
Manually enter all income Manually enter all income
information information amount owed will be sent to you
directly. Includes ability to dispute
your refund or taxes owed.
Uses secure website and log in Requires IRS account using Requires IRS account using
secure advanced identity secure advanced identity
authentication techniques such as authentication techniques such as
facial recognition or a video chat facial recognition or a video chat
with a trusted agent with a trusted agent

included included included

No Cost (Free to use) No Cost (Free to use) No Cost (Free to use)

Since there’s no cost for any of the options, I’d rather use commercial software 12% 9% 11%
I don’t want the IRS to prepare my return 13% 14% 11%
I don’t think it’s the IRS’s role to prepare taxes 11% 12% 14%
I trust a commercial tax package more than I would trust an IRS tax package 9% 9% 9%
I want to stay with my brand of software 17% 12% 9%
I don’t think the functionality of an IRS program would be as good as 6% 6% 9%
commercial software
I have audit defense and I don’t think the IRS could offer audit defense against 3% 7% 6%
itself
I believe my software offers better customer service than the IRS could 9% 11% 14%
I think my data is more secure with my software 8% 10% 8%
13% 10% 9%

46% 24% 15%

third party
I believe my data is more secure with the IRS 18% 28% 21%
11% 24% 28%
I believe I’d be less likely to be audited 15% 15% 21%
I think the IRS could provide better customer service than a commercial 10% 10% 15%
software company

The IRS already has the information, it makes sense for them to do it if it’s a 48% 21% 10%
simple return
It would be easier for me for the IRS to do my taxes 19% 27% 22%
I believe my data is more secure with the IRS 12% 14% 20%
7% 16% 21%
I believe I’d be less likely to be audited 7% 12% 18%
I think the IRS could provide better customer service than a commercial 6% 9% 9%
software company

IRS Report to Congress 25

Simple Tax Return Scenario #2: No State Return

Unchanged Unchanged Unchanged

Unchanged Unchanged Unchanged
Unchanged Unchanged Unchanged
Unchanged Does not include state tax return Does not include state tax return

Unchanged Unchanged Unchanged

34% 11% 10%

Since there’s no cost for any of the options, I’d rather use commercial software 9% 14% 14%
I believe my software offers better customer service than the IRS could 6% 13% 13%
I want to stay with my brand of software 11% 11% 9%
I don’t want the IRS to prepare my return 10% 9% 11%
I don’t think it’s the IRS’s role to prepare taxes 8% 12% 8%
11% 9% 7%
techniques
I don’t think the functionality of an IRS program would be as good as 4% 9% 10%
commercial software
I think my data is more secure with my software 5% 6% 8%
I have audit defense and I don’t think the IRS could offer audit defense against itself 2% 5% 9%

35% 22% 15%

third party
I believe my data is more secure with the IRS 13% 28% 18%
16% 14% 25%
I believe I’d be less likely to be audited 12% 15% 18%
I think the IRS could provide better customer service than a commercial 9% 11% 17%
software company
16% 10% 8%

The IRS already has the information, it makes sense for them to do it if it’s a 37% 19% 10%
simple return
It would be easier for me for the IRS to do my taxes 15% 24% 19%
I want the IRS to do my taxes for me 9% 15% 16%
I believe my data is more secure with the IRS 7% 10% 20%
10% 9% 10%
I believe I’d be less likely to be audited 6% 9% 12%
12% 5% 5%
I think the IRS could provide better customer service than a commercial 4% 9% 8%
software company

IRS Report to Congress 26

Simple Tax Return Scenario #3: Pre-Population

Unchanged Unchanged Unchanged

Unchanged Automatically inputs W-2 and Unchanged
1099 information
Unchanged Unchanged Unchanged
Unchanged Unchanged Unchanged
Unchanged Unchanged Unchanged

I don’t want the IRS to prepare my return 16% 15% 13%

I don’t mind inputting document information 11% 11% 14%
I don’t think it’s the IRS’s role to prepare taxes 12% 13% 11%
12% 12% 11%
techniques
I believe my software offers better customer service than the IRS could 9% 12% 10%
I want to stay with my brand of software 13% 7% 9%
Commercial software is already free for simple returns 9% 7% 10%
I don’t think the functionality of an IRS program would be as good as 8% 8% 9%
commercial software
I think my data is more secure with my software 6% 8% 7%
I have audit defense and I don’t think the IRS could offer audit defense against itself 4% 7% 7%

22% 32% 19%

party
I like the idea of not having to gather up/wait for/input my W2s and 1099s 46% 15% 9%
I believe my data is more secure with the IRS 13% 17% 19%
7% 14% 18%
I believe I’d be less likely to be audited 5% 15% 17%
I think the IRS could provide better customer service than a commercial 5% 6% 18%
software company

I think the IRS could provide better customer service than a commercial software 9% 9% 9%
company
I believe I’d be less likely to be audited 8% 18% 16%
I believe my data is more secure with the IRS 9% 14% 21%
8% 16% 28%
I want the IRS to do my taxes for me 15% 24% 16%
The IRS already has the information, it makes sense for them to do it if it’s a 52% 20% 11%
simple return

IRS Report to Congress 27

Simple Tax Return Scenario #4: No State Return and Pre -Population

Unchanged Unchanged Unchanged

Unchanged Automatically inputs W-2 and Unchanged
1099 information
Unchanged Unchanged Unchanged
Unchanged Does not include state tax return Does not include state tax return

Unchanged Unchanged Unchanged

29% 10% 8%
I don’t mind inputting document information 7% 16% 14%
I don’t think it’s the IRS’s role to prepare taxes 10% 13% 8%
I don’t want the IRS to prepare my return 12% 10% 9%
Commercial software is already free for simple returns 6% 11% 12%
I want to stay with my brand of software 10% 8% 8%
10% 7% 10%
techniques
I believe my software offers better customer service than the IRS could 6% 8% 11%
I don’t think the functionality of an IRS program would be as good as 6% 8% 6%
commercial software
I have audit defense and I don’t think the IRS could offer audit defense against itself 2% 6% 8%
I think my data is more secure with my software 4% 4% 7%

I like the idea of not having to gather up/wait for/input my W2s and 1099s 27% 18% 23%
25% 21% 14%
third party
13% 14% 19%
I believe my data is more secure with the IRS 14% 17% 12%
10% 13% 8%
I think the IRS could provide better customer service than a commercial 4% 9% 15%
software company
I believe I'd be less likely to be audited 8% 9% 9%

The IRS already has the information, it makes sense for them to do it if it’s a 43% 21% 14%
simple return
11% 19% 23%
I want the IRS to do my taxes for me 12% 17% 14%
I believe I’d be less likely to be audited 6% 13% 16%
I believe my data is more secure with the IRS 6% 13% 16%
I think the IRS could provide better customer service than a commercial 11% 9% 10%
software company
11% 7% 8%

IRS Report to Congress 28

Complex Tax Return Scenario #1: Same Functionality

package that was downloaded to your computer was downloaded to your computer or used through the
or used through a company’s online website. IRS website.
Manually enter all income information Manually enter all income information
Uses secure website and log in Requires IRS account using secure advanced identity
authentication techniques such as facial recognition or a
video chat with a trusted agent

Cost: $80.00 No Cost (Free to use)

I don’t want the IRS to prepare my return 18% 15% 13%

I don’t think it’s the IRS’s role to prepare taxes 12% 19% 10%
I want to stay with my brand of software 17% 10% 10%
I don’t think the functionality of an IRS program would be as good as 6% 12% 17%
commercial software
15% 8% 11%
techniques
I believe my software offers better customer service than the IRS could 7% 10% 18%
I expect to pay for my software, cost is not an issue. 15% 7% 5%
I have audit defense and I don’t think the IRS could offer audit defense against 5% 9% 9%
itself
I think my data is more secure with my software 6% 9% 7%

60% 14% 8%
14% 33% 22%
third party
I believe my data is more secure with the IRS 9% 17% 23%
9% 16% 22%
I believe I’d be less likely to be audited 5% 11% 14%
I think the IRS could provide better customer service than a commercial 3% 9% 11%
software company

IRS Report to Congress 29

Complex Tax Return Scenario #2: No State Return

Unchanged Unchanged
Unchanged Unchanged
Unchanged Unchanged
Unchanged
Unchanged Unchanged

38% 12% 8%
I believe my software offers better customer service than the IRS could 7% 15% 16%
I don’t think the functionality of an IRS program would be as good as 6% 11% 15%
commercial software
12% 10% 9%
techniques
I don’t want the IRS to prepare my return 8% 9% 12%
I want to stay with my brand of software 7% 13% 9%
I don’t think it’s the IRS’s role to prepare taxes 7% 9% 11%
I think my data is more secure with my software 5% 9% 7%
I have audit defense and I don’t think the IRS could offer audit defense against 4% 8% 7%
itself
I expect to pay for my software, cost is not an issue. 6% 4% 6%

50% 15% 8%
11% 28% 23%
third party
I believe my data is more secure with the IRS 9% 15% 22%
9% 18% 19%
I believe I’d be less likely to be audited 5% 11% 12%
I think the IRS could provide better customer service than a commercial 5% 7% 11%
software company
11% 7% 4%

IRS Report to Congress 30

Complex Tax Return Scenario #3: Pre-Population

Unchanged Unchanged
Unchanged Automatically inputs W-2 and 1099 information
Unchanged Unchanged
Unchanged Unchanged
Unchanged Unchanged

I don’t want the IRS to prepare my return 16% 13% 14%

19% 8% 8%
techniques
I don’t think it’s the IRS’s role to prepare taxes 9% 13% 13%
I don’t mind inputting my income information 9% 13% 12%
I believe my software offers better customer service than the IRS could 12% 9% 13%
I don’t think the functionality of an IRS program would be as good as 4% 15% 11%
commercial software
I think my data is more secure with my software 6% 12% 10%
I want to stay with my brand of software 9% 9% 8%
I expect to pay for my software, cost is not an issue. 11% 3% 5%
I have audit defense and I don’t think the IRS could offer audit defense against 4% 5% 8%
itself

46% 18% 11%

I like the idea of not having to gather up and wait for my W2s and 1099s, or 25% 37% 12%
having to input them
8% 17% 29%
third party
I believe my data is more secure with the IRS 8% 12% 16%
8% 10% 18%
I believe I’d be less likely to be audited 6% 8% 14%

IRS Report to Congress 31

Complex Tax Return Scenario #4: No State Return and Pre-Population

Unchanged Unchanged
Unchanged Automatically inputs W-2 and 1099 information
Unchanged Unchanged
Unchanged
Unchanged Unchanged

35% 11% 6%
I don’t mind inputting my income information 10% 15% 13%
14% 8% 10%
techniques
I don’t want the IRS to prepare my return 10% 8% 13%
I believe my software offers better customer service than the IRS could 4% 12% 13%
I don’t think it’s the IRS’s role to prepare taxes 7% 11% 9%
I don’t think the functionality of an IRS program would be as good as 4% 9% 13%
commercial software
I want to stay with my brand of software 8% 9% 8%
I think my data is more secure with my software 5% 9% 8%
I have audit defense and I don’t think the IRS could offer audit defense against 3% 7% 7%
itself

46% 14% 12%

I like the idea of not having to gather up and wait for my W2s and 1099s, or 13% 34% 15%
having to input them
9% 14% 20%
third party
8% 9% 17%
I believe my data is more secure with the IRS 6% 8% 17%
10% 7% 5%
I believe I’d be less likely to be audited 4% 7% 8%
I think the IRS could provide better customer service than a commercial 3% 6% 6%
software company

IRS Report to Congress 32

MITRE Survey Demographics
Unlike the TES, which uses a representative sample, respondents opted in to the MITRE Survey rather than

the diversity of taxpayers:

Survey 1 (Simple Tax Return)
Gender
Female (51%)
Male (47%)
Age
18–29 (19%)
30–44 (32%)
45–64 (35%)
65+ (15%)
Reported household income
<$35k (32%)
$35k–<$50k (13%)
$50k–<$75k (16%)
$75k+ (29%)
Filing status
Single (47%)
Married Filing Joint / Separate (44%)
Head of Household (7%)
Widow(er) with children (1%)
Census region
Northeast (16%)
South (37%)
Midwest (22%)
West (25%)
State income tax requirement
State income tax (76%)

Education
HS Grad or less (27%)
Some College (29%)
College Degree (28%)
Postgraduate Degree (16%)
Ethnicity
White (65%)
Hispanic (15%)
Black (9%)
Asian, Native American, Middle Eastern, two or more races, and others (10%)

IRS Report to Congress 33

Survey 2 (Complex Tax Return)
Gender
Female (50%)
Male (48%)
Age
18–29 (21%)
30–44 (29%)
45–64 (24%)
65+ (16%)
Reported household income
<$35k (31%)
$35k–<$50k (15%)
$50k–<$75k (16%)
$75k+ (28%)
Filing status
Single (48%)
Married Filing Joint / Separate (42%)
Head of Household (8%)
Widow(er) with children (<1%)
Census region
Northeast (15%)
South (35%)
Midwest (22%)
West (28%)
State income tax requirement
State income tax (78%)

Education
HS Grad or less (29%)
Some College (29%)
College Degree (28%)
Postgraduate Degree (14%)
Ethnicity
White (67%)
Hispanic (15%)
Black (8%)
Asian, Native American, Middle Eastern, two or more races, and others (11%)

IRS Report to Congress 34

Qualitative User Research

Methodology
Direct File Task Force user researchers conducted two rounds of small, focused studies to gain insight into

tool in particular. This was done using industry standard human-centered design methods for technical product
development and user research.
A researcher engages a taxpayer in conversation. The researcher
has a list of topics they intend to cover, but questions are not fully scripted, and the researcher follows up on ideas
introduced by the taxpayer to explore their perspective.
Participants interact with a functioning internal prototype. A researcher provides tasks and
prompts, and observes them as they attempt to use the tool, pausing to discuss their expectations and reactions
along the way.
These techniques have also been successfully implemented across government at agencies such as the
Department of Veterans Affairs35 and the Department of Homeland Security.36
Each round of studies worked with a small number of participants. In total, 14 taxpayers participated. Keeping
each round small is the industry standard37 because these studies conducted in series throughout the process of
designing and developing products, allowing for a deeper understanding of potential user issues and helping to
iterate on design details.
The primary limitation of these methods is that, because of the smaller number of participants, each round can’t
uncover every diverse perspective that exists for the wide variety of taxpayers. However, this qualitative research
complements quantitative methods used above by illuminating potential explanations for survey results and
allowing taxpayers to evaluate a functioning internal prototype from the IRS in a controlled environment, rather
than an abstract description.
Participants were recruited by a contracted user research recruiting company. The Direct File Task Force
provided screener criteria to select individuals across a range of demographic backgrounds (age, gender,

session lasted less than an hour and participants were compensated for their time.38 Sessions were conducted
remotely via videoconference in order to include participants from across the United States (all Census regions
were represented).

––––––––––––––––––––––––
35

36
Department of Homeland Security. “Burden Reduction at DHS.” January 9, 2023.
37
Nielsen Norman Group. “How Many Test Users in a Usability Study?”
38
Contractor selection, participant selection, and participant compensation were all compliant with the requirements of the Paperwork
Reduction Act (PRA) and other applicable laws and rules.
IRS Report to Congress 35
APPENDIX B:
OPINIONS OF AN INDEPENDENT THIRD PARTY

Overview

The agency selected New America and Professor Ariel Jurow Kleiman to work collaboratively as the independent
third-party to provide their opinions as required by the IRA. Those opinions are provided below.

IRS Report to Congress 36