If the Internal Revenue Service (IRS) already knows so much about our taxes, why does the process to file them have to be so onerous? A group of advocates and technologists got together to make tax filing suck less, and it worked really well. That is, until President Donald Trump killed it.

Launched in 2024, Direct File—the first public online tax filing tool in U.S. history—became one of the defining domestic policy successes of the Biden Administration, helping taxpayers access our increasingly complex tax system. For two years, it cut costs for ordinary tax filers, helped close the tax benefits coverage gap for the most vulnerable, broke the tax prep industry’s stranglehold on a core government function, and proved that public institutions can deliver.

Yet, thanks to the opposition of special interests and anti-government ideologues, Direct File predictably became a short-lived political football, canceled in the first year of the Trump Administration after the 2025 filing season ended. The Administration’s only rationale was a Treasury report echoing the tax prep industry’s argument that Direct File is duplicative of Free File—the failed private-sector option that funnels tax filers to the same companies Direct File was built to bypass. That rationale rang hollow, however, given that the Administration and Department of Government Efficiency (DOGE) had systematically undermined the program from day one. A new administration should bring it back—and make it better than before.

For decades after the advent of electronic tax filing and the digital migration of most basic services, the IRS offered no public solution for online filers: Online tax filing was run exclusively by third-party private tax filing companies, putting a sometimes unreliable middleman between taxpayers and the government, and serving as a potent symbol of a federal government unable to provide basic services. The arrangement cost taxpayers tens of billions of dollars in filing fees, subjected their data to misuse by third parties, and exacerbated a tax benefits coverage gap by making the tax system harder and more daunting to access.

Direct File, in contrast, sought to solve these issues for taxpayers. First, it was unconditionally free to use. Many private or semi-public tax filing options are advertised as free, are sometimes free, are partially free (e.g., free federal filing but paid state filing), or are free with caveats (e.g., no filing fees, but your data will be monetized). Direct File was always unconditionally free.

Second, it was easy and automated. Direct File was built around the plan to use IRS data to streamline the filing process. This meant that, in time, the Direct File experience would eradicate tax filing as we know it for many taxpayers, turning a stressful chore into a quick, painless, and largely automated process.

Additionally, it was publicly owned and an official offering of the government. Taxpayers did not have to engage with third-party providers they may not trust and could have faith that the product’s tax interpretations were fully endorsed by the government and guaranteed to be accurate. Direct File could be straightforwardly promoted as the official tax filing option, breaking through the overload of dozens of private offerings, none of which could be lawfully promoted by the government as a simple, default option. Being the single official option would have helped Direct File get through to hard-to-reach populations, including new and intermittent tax filers and the people who are eligible for valuable tax credits like the earned income tax credit (EITC) but never claim them.

Finally, Direct File was able to integrate with other IRS services across the tax system. As an IRS product, Direct File had the potential to be positioned within a suite of services that collectively allowed taxpayers to access refunds and fulfill their filing obligations—all within their online IRS account.

Direct File is a policy and political no-brainer. With widespread backing in Congress and the overwhelming support of taxpayers and users—94 percent of Direct File users rated it “excellent” or “above average,” and interest in it is high (73 percent) across all tax filers—it should be a day-one priority for any future administration.

A New Administration Should Bring Direct File Back Faster, Bigger, and Better

We were part of a group of veterans of the Direct File project who spent the months following the program’s demise documenting what we learned and making the case for how to build on the product in the future, as part of the Economic Security Project’s Future of Tax Filing Fellowship. The full recommendations and lessons run the gamut of Direct File, from the tax situations it did and should cover to its identity verification system to its promotion and marketing.

In particular, there are a few things a future administration should do to reanimate Direct File.

1. Launch in year one of a new administration

Bringing Direct File to the public required a years-long runway inside the Biden Administration, from building policy consensus to laying the technological groundwork. Internal discussions began in mid-2021, but the tool didn’t launch to taxpayers until March 2024. Ultimately, this timeline damaged the project’s longevity: With only one year to prove its value under the waning Biden Administration, Direct File did not become an entrenched program Americans would fight to keep, making it easier for the Trump Administration to cancel it the following year.

A new administration should have the wherewithal to move faster. The technological groundwork is present (Direct File’s code is open source and well documented, and the pathways to hire appropriate personnel into the government are better known), and the policy consensus exists, if it can be maintained in the interim. With an aggressive timeline, Direct File could relaunch within weeks of a new administration—in time for that administration’s first filing season. At the very least, Direct File should relaunch within six months, missing the core filing season of the first tax year but allowing millions of extension-period filers to use the product in year one. There is no reason—and no excuse—for Direct File to wait longer than that to relaunch.

2. Prioritize more and better pre-population

Among Direct File’s virtues, the one that has arguably been the most animating for reform advocates is its ability to streamline the filing process by using data the IRS already has on file. The IRS has detailed information on filers from third-party information returns (W-2s, 1099s), from prior-year returns, and from IRS systems. Why make taxpayers provide information the agency already has? Pre-populating or automating this information would take the time and headache out of filing and is proven to break down barriers in particular for the millions of households that cannot track down their tax documents or answer complex tax questions.

Direct File had begun to pre-populate data in earnest in its first two filing seasons, but there is much more to be done. Pre-population should expand to a variety of data categories beyond just information returns (i.e., W-2s, 1099s), including data from the IRS taxpayer account, from other current-year returns, from prior-year returns, and from the Social Security Administration (SSA). New pre-population functionality should be prioritized based on the number of taxpayers impacted, the complexity averted for those taxpayers, and the effort to implement. The success of pre-population will depend on improved timeliness and integrity of information return reporting. This means resolving issues by which W-2 data is modified by the SSA en route to the IRS, moving up deadlines for information return reporting (which will require amending 26 USC 6071), and increasing enforcement to ensure on-time reporting.

Taken together, these measures represent the fastest and best route toward something approaching a fully automated tax filing experience, or what some advocates have called return-free filing, for many taxpayers.

Consider a married couple with three children who claim the EITC, the child tax credit, and the premium tax credit and who earn income from three W-2 jobs, some interest or dividends on various savings, and unemployment insurance when one of the parents loses their job. This is not what most observers would call a simple tax return. But Direct File should be able to automate nearly this entire return within a few years of its reanimation, making tax filing for this family a 10-to-20-minute task of simply confirming that no major life changes have occurred in the last year.

3. Provide for more and better state filing solutions

Taxpayers want and expect to be able to file their state returns at the same time they file their federal returns. Direct File supported state filing through a two-part process: After completing the federal return, taxpayers could export their Direct File data to a state filing product, where they could finish the process of preparing and filing their state return. The data export function spared taxpayers the burden of providing all their data twice, but it still did require taxpayers to navigate two separate products and required states to build and maintain their own filing products, whose user experiences may or may not have matched Direct File’s.

The process worked well for a large majority of taxpayers in the 12 states that participated in the first year and the 25 states that did in the second year. However, it proved challenging for a meaningful minority, as well as creating costs for states and failing to offer elegant solutions for taxpayers with multiple state returns.

Moving forward, we believe Direct File can provide an even better user experience for state filing at lower cost to states, enabling it to scale faster and better across the country.

Direct File can offer states the ability to configure their functionality directly in the Direct File product, rather than maintaining standalone integrated products. We believe this should be the default state support path for a reanimated Direct File. For states that choose to maintain independent standalone options, Direct File should invest in improvements to the existing collaboration model: supporting a common public-interest software platform states can adopt; refining the interaction between the products; and setting usability standards for state tools to maintain a high-quality user experience. Additionally, Congress should consider appropriating funds for states to cover the costs of their Direct File participation. Such costs are minimal relative to the federal budget, but can be prohibitive for cash-strapped states—especially when they are wary of investing in a project that was already offered and retracted once.

Just as important as any of these recommendations is that a future Direct File team must take state filing seriously as a core priority and should not sacrifice ease of state filing in exchange for faster scaling or easier development. Tax filing is only as easy and free as its hardest and least free component, and if taxpayers are left to fend for themselves with their state returns, Direct File would have little impact at all. State filing must be guaranteed free and easy for Direct File to have the transformative impact it’s capable of.

4. Commit to a new way of building

In its first two years, Direct File began to develop back-end tools and processes to make translating the tax code into accurate, human-friendly content systematized and more tractable. This included the “fact graph,” a structured map of the tax code’s underlying logic that translated dense statutory language into the conditional rules that govern what a taxpayer owes or is owed. This work was in its early stages but had the power to be one of the most transformative outputs of the Direct File project. At its most ambitious, Direct File wasn’t just building a tax tool; it was quietly pioneering a new approach to civic technology that could allow government to translate any law, from tax credits to safety net benefits, directly into services for the people entitled to them. Here’s the set of tools we envisioned:

Fact Markup Language (FactML): A standardized format for representing statute in precise, machine-readable code. Think of this as a common language that lets any application define eligibility in terms of a series of interrelated facts, which different applications can all reuse.

Factual: The engine that runs those rules. Given what is already known about a specific person—e.g., their income, dependents, filing status—Factual applies the logic encoded in FactML to determine what they qualify for, what they owe, and what the system still needs to ask them.

Formative: The taxpayer-facing front end. Rather than presenting a static list of questions, Formative displays only the questions that are relevant to a given person’s situation—dynamically, in real time, based on what Factual has already determined. The result is an interview-style filing experience that users of tax filing software are familiar with, but which can be generated partially automatically.

Formative Studio: The collaboration layer—a graphical user interface that lets lawyers, UX designers, and engineers understand the system and work on it simultaneously. A tax attorney can verify that the software accurately reflects the statute without reading a line of code; a designer can see how a legal rule will appear to a taxpayer; a product manager or tax expert can test whether the logic is doing what it is supposed to. It is the tool that makes the whole system legible to everyone who needs to build and maintain it.

We believe a reanimated Direct File should continue to pursue this type of tooling to make the product more reliable and faster to deliver, as well as to make possible the tighter state integrations defined above and even integrations across government benefit domains. In fact, civic technologists can be iterating on these tools now, in other contexts, in advance of an administration reanimating Direct File.

Reforming Access to the Tax System Will Require More Than Direct File

Direct File was the linchpin of a broader effort to build a more accessible, human-centered tax system—but it wasn’t the be-all and end-all of that effort. A new administration will need to implement a variety of additional reforms and projects in service of building a twenty-first-century system for taxpayers.

Build a real-time tax system

Even in our modern digital world, the tax system functions on an annual cadence: Taxpayers pay money into the system throughout the year via withholding and estimated payments and do a once-yearly reconciliation via their annual tax return. In the best case, by tax time, they have paid in roughly the right amount. In the worst case, they have either paid in far too much, giving the government an interest-free loan for no reason, or they have paid in too little, suddenly incurring an anxiety-inducing debt to the government. Even W-2 workers, who have the benefit of withholding through their employer’s payroll system, struggle to calculate how much they should pay throughout the year to ensure they come out even. Those in the gig economy have virtually no support and often don’t even know they are supposed to pay estimated tax payments—let alone how much—or that they should track expenses throughout the year.

To better support these workers, it’s necessary to create a withholding process for gig economy workers, to spare them the hassle of administratively onerous estimated tax payments. We need better real-time reporting of income and withholding data from employers and financial institutions throughout the tax year, for both W-2 workers and gig workers. Finally, to help taxpayers track and update their tax obligations throughout the tax year, let’s create a new IRS-provided product that can detect whether their withholding or estimated payments are likely on track to match their ultimate tax obligation.

Develop a new approach to bring non-filers into the system

For years, reformers in and out of government have worked to close the persistent gap in access to tax benefits, including the alleged one in five EITC-eligible households who do not claim the credit, leaving potentially billions of dollars on the table. Direct File helps to close this gap—but it will not close it entirely. First, better data on the non-filer gap needs to be gathered. For a variety of technical reasons, we believe careful research would show that the EITC access gap (and the EITC error rate, for that matter) is much smaller than commonly stated. Additionally, we need a strategic division of the calendar year into two parts—the regular filing season from January to April, and a “non-filer season” from May to November. In this second portion of the year, special measures could be undertaken to reach non-filers, such as preparing “second-best” returns for non-filers. This should also include the creation of a dedicated non-filer team in the IRS.

Reform the dependent laws

The thicket of laws and de facto administrative practices around claiming dependents on tax returns is complex, internally inconsistent, and in conflict with most taxpayers’ intuitions about how family ought to be defined. This complexity and inconsistency cause real problems. Many seemingly appropriate child claims are determined unlawful by the letter of the law and deemed overpayments, subjecting claimants to audits and punitive consequences, and building political pressure to restrict access to tax benefits. The adjudication system makes conflicts hard to surface and resolve, allowing inappropriate family members to claim children to the detriment of appropriate claimants. Some children cannot be claimed by anyone in their family at all thanks to vagaries of the rules. This all adds up to a massive administrative burden for taxpayers and tax agencies, and perceptions of complexity that make the tax system still harder for taxpayers to feel comfortable with. This complexity has direct consequences for Direct File, too: Dependent-related tax situations were among the most difficult to build into the product, limiting the filers it could serve and slowing its path to broader coverage.

Rewriting the dependent laws to be consistent across various tax benefits and to better align with taxpayer intuitions, and rewriting the administrative procedures to align with the laws, would resolve all of these issues.

Build a tiered system of tax filing assistance

Direct File is a solution for taxpayers who are comfortable preparing their own returns, with some access to limited additional assistance via live chat. But there still may be populations with especially complex tax situations or particular barriers to navigating the system who will need more hands-on help filing a return. The IRS provides a small amount of higher-touch assistance via Volunteer Income Tax Assistance (VITA) and Tax Counseling for the Elderly sites—but this is limited to about 2 percent of taxpayers.

To better serve those taxpayers who need extra help, the IRS should explore the creation of additional service models, more intensive than Direct File but less intensive than VITA. Additionally, mediating most access to VITA (and any additional service levels) via an expanded Direct File service would allow taxpayers to get started with their tax filing online, but would dynamically determine and provide the level of assistance they need. Finally, modernizing the rules of the VITA program would better incentivize VITA sites to focus efforts on high-need taxpayers who most need the high-touch service VITA can provide.

All told, these changes would be a heavy lift to implement, requiring significant new reporting not currently part of the tax system. They may well not be changes that can reasonably occur immediately. But they are the changes that are necessary to actually build a more modern tax system.

Direct File’s significance runs deeper than administrative efficiency or even taxpayer savings. The ability to fulfill a basic civic obligation—paying your taxes—without paying a private intermediary for the privilege is a statement about what democratic government is for. In most peer democracies, free and direct government filing is simply the default; that the United States outsourced this function to an industry with a financial interest in making it harder is an anomaly, and a telling one. The choice to restore Direct File is a test of whether the government exists to serve the public or to create markets for those who do.

In 2024 and 2025, Direct File proved that the IRS can build something that works—something taxpayers trust, use, and want back. The hard work of demonstrating that is done. What remains for a future administration is not to start over, but to finish what was started: Restore Direct File, expand it, and use what it built as the foundation for a government that can deliver on its promises.