The Great Reinvention logo THE GREAT
REINVENTION

Government Reinvention: A Leaner, Smarter, Digital Federal Government

Proposed legislation: The Government Reinvention and Digital Services Act (bill text coming soon)

Government Reinvention: A Cost-Benefit and Policy Analysis

The federal government is the largest, most complex organization on earth, and it runs on machinery much of which was designed for the typewriter era. It comprises hundreds of agencies, bureaus, commissions, and offices, many with overlapping missions, incompatible computer systems, and processes that still move paper between buildings. The Government Reinvention and Digital Services Act proposes to consolidate this sprawl into roughly six mission-based departments, to modernize the technology underneath, and to digitize the services citizens actually touch — permits, benefits, filings, and identity. The headline claim attached to such proposals is dramatic: savings on the order of $1 trillion or more per year. This page takes that claim seriously, and assesses it honestly. The short version: real, large savings are achievable; the trillion-dollar-a-year figure, as a recurring number, is almost certainly overstated.

The Problem: A Government Built in Layers

The core problem is accretion. New programs are created without retiring old ones, each arriving with its own statute, funding stream, oversight committee, and IT system. The result is duplication, fragility, and waste that the government's own auditors document every year.

Two figures frame the opportunity. According to the Government Accountability Office (GAO), the federal government made an estimated $162 billion in improper payments in fiscal year 2024 across 68 programs at 16 agencies — and cumulative improper payments since 2003 total roughly $2.8 trillion. Three-quarters of the 2024 total was concentrated in just five program areas, chiefly Medicare, Medicaid, the Earned Income Tax Credit, SNAP, and a pandemic-relief fund. Separately, GAO has estimated that the government could lose between $233 billion and $521 billion per year to fraud. These are not the same as "waste that disappears with reorganization," but they define the scale of the integrity problem that better data systems are designed to attack.

The Mechanism: Consolidation, Automation, Digitization

The Act has three pillars.

Consolidation would group agencies by mission rather than by historical accident — for example, a single department for economic affairs, one for human services and health, one for security, one for natural resources and energy, one for science and infrastructure, and a central digital-and-administrative service. Consolidation does not, by itself, save enormous sums; the real prize is the shared back office it makes possible: one payroll system instead of dozens, one identity layer, one procurement pipeline.

Automation targets the routine. A large share of federal employee time goes to rules-based processing — eligibility checks, form review, data re-entry — that modern software can handle with human oversight. Used carefully, automation can both reduce cost and reduce error, which is where it intersects the improper-payments problem.

Digitization addresses the citizen-facing experience: digital permits, online filings, and a "tell us once" principle in which a citizen who updates an address with one agency updates it everywhere.

The Money Underneath: Federal IT

The technology numbers explain both the opportunity and the difficulty. The federal government spends more than $100 billion a year on IT and cyber-related investments, and according to GAO roughly 79 percent of planned IT spending — about $83 billion in fiscal year 2025 for the major agencies — goes to operating and maintaining existing systems rather than building better ones. Some of those systems are decades old. GAO's 2025 review of the most critical legacy systems found that of ten systems flagged for urgent modernization back in 2019, only three had been modernized by early 2025, and several agencies still lacked complete modernization plans.

This is the heart of the savings case: every dollar spent keeping a mainframe alive is a dollar not spent on a system that would prevent fraud, speed service, and cut staff time. But it is also the heart of the difficulty — modernization is precisely what the government has repeatedly failed to execute on schedule and on budget.

Evaluating the Trillion-Dollar Claim

Here the analysis must be candid. Total federal spending is on the order of $7 trillion a year, but the large majority is mandatory spending — Social Security, Medicare, Medicaid, and interest on the debt — and net interest alone now exceeds $1 trillion annually. Reorganizing agencies and modernizing IT does almost nothing to those line items, which are set by benefit formulas and bond markets, not by org charts. The discretionary budget that reorganization could actually touch is far smaller.

So where could a trillion in annual savings even theoretically come from? It would require eliminating essentially all improper payments ($162 billion), a large share of fraud (a few hundred billion at the optimistic end), and substantial program and workforce reductions — and sustaining all of it every year. Capturing even half of improper payments and a meaningful slice of fraud would be a historic achievement worth tens of billions annually; eliminating them entirely is not realistic, because much improper-payment "loss" reflects documentation errors rather than recoverable cash, and fraud is an adversarial moving target.

A more defensible framing: a serious, sustained reinvention program could plausibly save on the order of tens of billions of dollars per year in recurring administrative and integrity costs once mature, plus one-time and avoided-cost savings from retiring legacy systems — meaningful money, but an order of magnitude below the trillion-dollar headline as a steady-state annual figure. Reform should be sold on that honest basis. Overpromising is itself a risk: it invites cynicism when the trillion fails to materialize and discredits genuine, achievable gains.

Administrative and Implementation Considerations

The hardest part is not the vision but the transition. Reorganizations are notorious for costing money and degrading performance for years before any payoff. Merging incompatible IT systems is risky; the same agencies that struggle to modernize a single system would be asked to integrate dozens. Civil-service rules, congressional committee jurisdictions, and union agreements all constrain how quickly workforces can be reshaped. And automation introduces its own hazards: an algorithm that wrongly denies benefits can do real harm at scale, and several state-level automated eligibility systems have produced exactly that kind of failure. The Act would therefore require phased rollouts, mandatory human review of adverse automated decisions, independent auditing, and a dedicated, professionally staffed digital service with the authority to set standards across departments. Crucially, savings should be reinvested in modernization at first, not booked as immediate cuts, or the effort will stall the way past modernization efforts have.

International Comparisons and Precedent

The clearest precedent is Estonia. According to the Estonian government, by December 2024 essentially every state service had become digital. Citizens file taxes in a few minutes, building permits that once required five separate processes now run through one, and a backbone data-exchange system (X-Road) handled over 2.7 billion queries in 2024 under a legal "ask once" rule: an agency may request a given piece of information from a citizen only once, after which it is shared across government. Estonia is small, which makes it an imperfect template for a continental federal system — but it proves the model works and quantifies the kind of efficiency available. Domestically, the U.S. Digital Service and 18F have shown that small teams of skilled technologists can rescue failing systems, and the long arc of GAO's "High-Risk List" demonstrates that sustained attention can save billions over time.

Comparison to the Status Quo and Alternatives

The status quo is expensive in ways that do not appear on any single budget line: time lost by citizens navigating broken processes, fraud and error that better systems would catch, and the compounding cost of maintaining technology that should have been retired a decade ago. The main alternative to wholesale reinvention is incrementalism — modernize one critical system at a time, as GAO recommends. Incrementalism is lower-risk but has proven painfully slow. A second alternative is across-the-board budget cuts, which save money on paper but tend to gut capacity (including the very oversight and IT staff who prevent waste) without fixing the underlying machinery. Reinvention aims for the middle: structural change paired with disciplined, phased execution.

Risks, Trade-offs, and Counterarguments

The strongest counterargument is that reorganizations rarely deliver promised savings and frequently cost more than they save for years. The history of federal IT modernization is a graveyard of over-budget, behind-schedule projects; betting a trillion dollars on a flawless execution of dozens of them simultaneously is not prudent. A second objection is that consolidation can reduce accountability — when many agencies become a few mega-departments, it can become harder for Congress and the public to see who is responsible for what. A third is the human cost and disruption: large workforce changes affect real people and can hollow out institutional knowledge. A fourth is automation risk: poorly governed algorithms can deny benefits, entrench bias, or create new attack surfaces for fraud rather than closing them. A fifth, already discussed, is the credibility risk of the savings claim itself. None of these is a reason to keep a typewriter-era government; all of them are reasons to reform carefully, measure honestly, and resist the temptation to promise more than the structure can deliver.

Conclusion

The federal government can and should be reinvented for the digital age. The case is strongest where it is most concrete: retiring decades-old systems that consume four out of every five IT dollars, attacking the $162 billion in annual improper payments with better data, and giving citizens services that work like the rest of modern life. Those gains are real and worth pursuing on their own terms. The trillion-dollar-a-year headline, however, does not survive contact with the structure of the federal budget, most of which reorganization cannot touch. The honest promise is a government that is faster, more trustworthy, and tens of billions of dollars a year leaner once mature — achieved through patient, well-governed execution rather than a single dramatic stroke. Sold that way, reinvention is both credible and overdue.

Sources

← Back to The Great Reinvention