Where the Public Money Goes: The Index of Programs, Lif...

This is the index to a running series that reads federal spending one program at a time. The method is always the same, borrowed from the way the Government Accountability Office and the Congressional Research Service read a budget: take a program most people have never heard of, find what it actually costs from a primary source, find the honest efficiency critique, find the honest public-good defense, and let the two verdicts sit next to each other instead of picking one for a headline. Almost every figure links to a government report, a statute, or an agency filing. Here is the whole set, grouped by what kind of money problem each one is.

The single idea underneath all of it is that efficiency and access are different yardsticks, and most fights about public money are really fights about which yardstick to use. A program can fail one and pass the other. Which number you lead with is usually a decision you made before you opened the spreadsheet.

Rural lifelines: the cost-per-unit versus cost-per-community problem

These are the programs that look wasteful measured per passenger or per rider and look essential measured per community, because the thing they buy is a connection that the market will not provide at any price.

Essential Air Service: subsidized flights to 183 small and remote communities, sometimes over 1,900 dollars a passenger, and the roadless Alaskan villages where it is the only way out.
The Alaska Bypass Mail subsidy: the postal freight system that keeps groceries flowing to roadless Alaska, and quietly makes the passenger flights on the same planes viable.
The Alaska Marine Highway: the state ferry that recovers only about a third of its cost from fares and is the only road to Juneau, a state capital you cannot drive to.
Amtrak's long-distance trains: the routes that lose the most money per passenger in the system and are the only public transit for hundreds of small towns.
PILT and Secure Rural Schools: two programs that pay counties for the federal land inside their borders that they are not allowed to tax.
The Homeowners Assistance Program: $555 Million to Cover a Loss the Government Caused: the Army program that makes wounded warriors, Gold Star families, and base-closure movers partly whole on a home-sale loss, and the $555 million the 2009 stimulus poured through it.

Health care: the safety net and the spread

The federal government pays for rural and safety-net medicine through mechanics that look strange until you see what they hold together.

Critical Access Hospitals: the hospitals Medicare pays 101 percent of their own reported cost, because a per-case rate would close them and leave counties without an emergency room.
The 340B Drug Pricing Program: the mandated drug discount that moves tens of billions of dollars, is captured as revenue by large nonprofit systems that also run enormous investment portfolios, and divides sharply on whether the savings reach patients.
Federally Qualified Health Centers: the community clinics that deliver primary care to about one American in eleven, on a sliding scale, funded through a grant stream that keeps hitting a cliff.
The Yellow Ribbon Reintegration Program: $181 Million to Reach the Half of the Force That Lives Nowhere Near a Base: the roughly $181 million a year reintegration and behavioral-health net for Guard and Reserve troops who live nowhere near a base.

Insurance and trust funds in the red

Programs designed to pool a risk the private market will not, which end up owing the Treasury money by design or by neglect.

The National Flood Insurance Program: the only flood insurer most Americans can buy, deep in debt to the Treasury, repeatedly rebuilding the same flooded houses.
The Black Lung Disability Trust Fund: a coal tax that has never quite covered the disabled miners it pays, borrowing from the Treasury nearly every year while the disease resurges.
The Pension Benefit Guaranty Corporation: the federal backstop behind private pensions, self-funded on one side and rescued by a taxpayer bailout on the other.

The nuclear enterprise: the most expensive corners of the budget

The nuclear-weapons and nuclear-waste complex is where public money meets the highest stakes and some of the largest failures.

The Office of Secure Transportation: the unmarked armored fleet and armed federal agents that move nuclear weapons across the country, 140 million miles without a release, for about 450 million dollars a year.
FOGBANK: the classified warhead material the government forgot how to make, and paid tens of millions and a year to relearn.
The Nuclear Waste Fund and Yucca Mountain: a 49-billion-dollar account for a repository that was never built, while taxpayers separately pay billions in damages.
The MOX Fuel Fabrication Facility: a plant to burn bomb plutonium into reactor fuel that consumed 7.6 billion dollars and was cancelled before it processed a gram.
The Uranium Processing Facility: the replacement plant at Y-12 that was designed too small to fit its own equipment, then redesigned at enormous cost.
The Sentinel ICBM cost breach: the Minuteman III replacement whose cost jumped about 81 percent to roughly 141 billion dollars, mostly on the ground silos, and was continued anyway.
The WIPP accident: how the wrong absorbent in a single waste drum shut the nation's only operating nuclear-waste repository for three years.
Nunn-Lugar: the $13 billion that pulled 7,600 Soviet warheads out of service: the rare nuclear line that made the world safer, deactivating thousands of former Soviet warheads and later destroying Syria's chemical stockpile at sea.
The Army Wants Reactors on Its Bases Again. Project Pele Is the $300 Million Test.: the Army's bet on transportable microreactors for bases, a roughly $300 million contract to build the first one, and the 1950s program that tried the same thing and failed.

Wildfire: the cost curve and the vendors behind it

Who pays to put out the fire: federal suppression costs that crossed a billion dollars in 2000 and now run several times that, and the interagency machine that spends them.
Who gets paid to fight the fire: the private air-tanker operators, the sole-source retardant contract, and the contractors who receive the suppression dollars.
Who owns the firefighting fleet: the private-equity, public-market, and debt money behind the tankers, and the military and ex-agency service that runs them, including a founder who is now a sitting United States senator.

Boondoggles and the watchdog that counts them

The programs above include several honest failures, and there are others whose mission was real but whose execution was not. Several of these are documented by the agencies' own Inspectors General or by the GAO, which is the point: the government indicts itself here, and both the failure and the mission behind it are shown.

The VA electronic health record modernization: a multibillion-dollar Oracle-Cerner rollout meant to unify military and veteran records, paused after cost growth and patient-safety failures.
The FBI case files that never went digital: Virtual Case File, scrapped after roughly 170 million dollars, and its late, over-budget successor Sentinel, the decade it took the bureau to move off paper.
NPOESS, the weather satellite that doubled in cost: a joint civilian and military program that grew from about 6.5 billion dollars toward 15 billion, breached the Nunn-McCurdy limits, and was torn up and rebuilt as two separate programs.
The DEA and 4 billion dollars in cash: a decade of civil-forfeiture cash seizures, roughly 3.2 billion of it forfeited with no criminal charge, and the Inspector General finding that the agency rarely tracked whether the seizures advanced any case.
Operation Fast and Furious: the ATF gun-walking operation that put roughly 2,000 firearms into trafficking channels, two of them found where a Border Patrol agent was killed.
The GAO High-Risk List and improper payments: the congressional watchdog behind most of these stories, the list of program areas most vulnerable to waste, and the roughly 160 to 240 billion dollars a year the government pays out improperly.
The Pentagon that has never passed an audit: the only major federal department without a clean opinion, seven straight disclaimers meaning its books cannot be verified, which is not the same thing as money proven missing.
The Chemical and Biological Defense Program: One Army, Twenty-Six Organizations, $1.4 Billion a Year: the roughly $1.4 billion a year chemical and biological defense enterprise the Army runs for the whole department, and the fragmentation GAO has flagged for thirty years.
The Two-Million-Dollar Board That Governs Half a Million Tons of Explosives: the cheapest safety win in the Pentagon, a tiny board born of a 1926 depot disaster that sets the blast-radius rules every ammunition depot lives by.
Anti-Tamper: The $50-Million Office That Guards the Secrets Inside America's Exported Weapons: the roughly $50 million a year office that guards the crown-jewel technology inside exported weapons, and the $499 million figure that is a contract ceiling, not spending.
The Defense Production Act's Title III: A Few Billion Dollars, and Almost No Scorekeeper: the Pentagon's quiet reshoring bank, about $3.6 billion committed since 2018 to rare earths, chips, and masks, with an effectiveness GAO says is essentially unmeasured.
The Defense Logistics Agency: the quiet backbone that feeds, treats, and fortifies the force: the roughly $47 billion logistics backbone that supplies the military's food, medicine, and barrier materiel, mostly works, and keeps drawing GAO excess-inventory findings.
WHINSEC: A $10 Million School, a New Name, and a Closed Roster: the former School of the Americas, now a roughly $10 million a year institute that teaches human rights and closed the student roster outsiders once used to check it.
The $108 Million Classroom: The Six Pentagon Schools No Audit Can Price: six little-known DoD schools that spend about $108 million a year building ties with foreign security elites, on a mission GAO struggled for years to prove works.
The Military Working Dog Program: The Mission Worked, the Kennels Did Not: the single bomb-dog pipeline the Air Force runs for the whole military, mostly a success, and the 2026 inspector-general report that found dogs dying in mold and heat.
CATCH: The Pentagon's Quiet Database That Waits for a Serial Offender to Repeat: a rare cheap federal database that works but almost nobody uses, letting restricted-report assault victims anonymously link repeat offenders.

The nuclear pieces above, the MOX plant, the Uranium Processing Facility, and the WIPP accident, are boondoggles too, filed by subject rather than by failure.

Spies, satellites, and the black budget

The most secret corners of the budget are also the hardest to oversee. These are programs funded through classified or covert channels, where the dollar figures are estimates rather than audited numbers, and where the failure and the mission are both real. Every figure here is sourced to what the record actually shows.

Future Imagery Architecture: the reconnaissance-satellite program that spent an estimated 10 billion dollars on an optical satellite that was never built, cancelled in 2005 with only its radar half surviving.
The spy agency headquarters Congress did not approve: the NRO's secret 300-million-dollar building and the roughly 1 to 3.8 billion dollars in unspent funds the oversight committees say they were never clearly told about.
EnhancedView, the commercial imagery commitment: the roughly 7.3 billion dollars the map agency committed to buying commercial satellite imagery, and the question of whether it bought more capacity than it used.
The NSA data centers and the electrical failures: the roughly 1.5-billion-dollar Utah Data Center whose arc-fault surges destroyed equipment and delayed its opening, and how the black budget builds through Military Construction while its mission money stays classified.
Trailblazer and the cheaper one it passed over: the billion-dollar NSA data program that failed, the in-house alternative that was rejected, and the whistleblower prosecution that collapsed.
How the CIA paid for its secret prisons: the more than 300 million dollars in non-personnel costs, the cash to host governments, and the 81 million to the contract psychologists, all funded through covert-action channels the public budget never showed.
In-Q-Tel, the intelligence community's venture arm: how roughly 100 million dollars a year of public money flows through a private nonprofit into startups, from the mapping company that became Google Earth to Palantir, and the family of government venture arms cut from the same cloth.
The $13 Million Prisoner: The Army's Job as Owner of the DoD Detainee Program: how the Army became the single owner of DoD detainee policy after Abu Ghraib, and the roughly $13 million a year per prisoner that detention at Guantanamo costs.

Systems that failed and programs that were cancelled

The most common federal boondoggle is not exotic. It is a large modernization or a big-ticket acquisition, a contractor, years of cost growth, and a program that delivers little or nothing before it is cancelled or quietly written off.

The census technology that failed, then recovered: the 2010 handheld computers that were scrapped for paper at a cost of billions, and the 2020 count that, unusually, came in under its own estimate.
A billion dollars for nothing: the Air Force logistics system cancelled in 2012 after roughly 1.1 billion dollars with, in the service's own words, no significant military capability.
Deepwater and the cutters that buckled: the Coast Guard patrol boats lengthened until their hulls deformed, removed from service at a loss of about 100 million dollars.
The new Marine One that cost more than Air Force One: the VH-71 presidential helicopter cancelled in 2009 after billions, with no aircraft delivered.
SBInet, the virtual border fence: nearly a billion dollars for about 53 miles of Arizona sensor towers that underperformed, cancelled in 2011.
SPOT, the airport behavior-detection program: the roughly 900 million dollars TSA spent watching travelers for deception, on a method GAO said the science did not support.
US-VISIT and the exit that was never finished: the biometric border-exit system Congress mandated for decades and billions, delivered at entry but never completed as designed.

Weapons and systems that never delivered

The largest single-program failures tend to be defense acquisitions: enormously ambitious systems, a contractor, years of cost growth, and a cancellation before anything reaches the field. Each figure below keeps the amount spent separate from the amount merely projected.

The Army intelligence system soldiers did not want: DCGS-A, the fusion system units in Afghanistan found slow, and the court fight that forced the Army to consider commercial Palantir, a company cut from the same In-Q-Tel cloth.
The air-traffic-control computers mostly thrown away: the FAA's 1990s modernization that wrote off about 1.5 billion dollars of the 2.6 billion it spent before it was restructured.
DIMHRS, the payroll system that never paid anyone: about a billion dollars for a joint military personnel-and-pay system that was cancelled without ever running department-wide.
BioWatch, the bioterror alarm that could not be trusted: well over a billion dollars on an air-sampler network that GAO and scientists said might not reliably detect an attack.
The Airborne Laser, a 747 with a laser cannon: roughly 5 billion dollars for a demonstrator that hit test missiles in 2010 and was retired to the boneyard anyway.
The A-12 stealth jet cancelled in 1991: about 2 billion dollars spent, no aircraft ever flown, and 23 years of litigation that reached the Supreme Court.
Future Combat Systems: the Army's system of systems that spent roughly 18 billion dollars before its manned-vehicle core was cancelled in 2009.
JLENS, the surveillance blimp that broke loose: the 2.7-billion-dollar air-defense aerostat, one of which snapped its tether in 2015 and drifted across Pennsylvania downing power lines.
Land Warrior, the wearable soldier computer: cancelled in 2007 for weight and complexity, then revived and sent to Iraq anyway.
Aerial Common Sensor: the spy plane cancelled in 2006 because the sensor package was too heavy for the jet the Army had chosen.
Comanche, the stealth scout helicopter: roughly 6.9 billion dollars spent for two prototypes before its 2004 cancellation, with the money redirected to drones.
NLOS-LS, the missile in a box: the container-launched precision missile cancelled in 2010 after its rockets kept missing in tests.
The Expeditionary Fighting Vehicle: the Marine amphibious vehicle cancelled in 2011 after about 3.3 billion dollars, defeated by its own high-water-speed requirement.
MEADS, the air defense America would not field: the tri-national system the US decided not to buy in 2011, then paid about 804 million dollars more to finish.
The C-27J flown to the boneyard: nearly new cargo planes retired almost immediately, then handed to the Coast Guard, special operations, and the Forest Service.
Global Hawk Block 30: the surveillance drone the Air Force tried to cancel as costlier than a 1950s U-2, kept flying by Congress.
The Heat Ray That Never Fired: Inside the DoD Non-Lethal Weapons Program: the Marine-run joint program that spent hundreds of millions and fielded little, and the heat-ray that was built, shipped to a war zone, and shipped home unused.
The Jammers That Silenced the Trigger: CREW, JIEDDO, and the Single Manager Who Never Quite Ran the Show: the counter-IED jammers that saved lives on Iraq and Afghanistan roads, funded through JIEDDO's more than $18 billion, and the single manager that could not stop the duplication.

Military networks and command systems

Wiring the force together, and wiring it to allies, is its own budget line and its own recurring pattern: sprawl, cost growth, and consolidation efforts that themselves run over. A few of these are genuine successes that carry only a cost or lock-in caveat. Most are the familiar story.

Coalition networks: CENTRIXS and the Mission Partner Environment: the tangle of more than 40 stovepiped allied-sharing networks, and the quarter-century of programs, including CPN-X, spent trying to merge them.
WIN-T, the battlefield network: the Army's on-the-move network backbone, curtailed in 2017 after operational and cyber-vulnerability findings.
JTRS, the software radio: the radio family whose Ground Mobile Radio component was cancelled in 2011 after about 6 billion dollars.
GPS OCX, the ground system that arrived late: the GPS control segment that breached its cost limits and left the new satellites flying on interim control.
NMCI, the Navy's outsourced network: one of the largest IT outsourcing deals ever, roughly 10 billion dollars, and the questions it raised about a service not owning its own network.
JEDI, the cloud contract cancelled without use: the 10-billion-dollar ceiling litigated to a standstill and cancelled in 2021, its cost measured in delay rather than cash.
NECC, the command system that replaced little: the net-centric command capability cancelled around 2010 after hundreds of millions.
CPOF, the software soldiers loved: a rare DARPA-to-battlefield success, and the proprietary lock-in that made it expensive to keep and to replace.
Blue Force Tracking: the friendly-force tracking network credited with cutting fratricide, alongside its cost and version-fragmentation caveats.
TSAT, the internet in the sky: the laser-linked satellite constellation cancelled in 2009 after billions in development and no launch.
MUOS, the satellites that beat their own radios: a roughly 7-billion-dollar satcom constellation that reached orbit years before the terminals that could use it.
The Pentagon's Fingerprint Machine: DoD Biometrics and the Database That Knows Who You Are: the Army-run identity database that helped deny 92,000 people off US bases, its roughly $3.5 billion planned bill, and the single point of failure GAO warned about.
DC3: The Pentagon's Volunteer Red Team, and the $300 Million Estimate Behind It: the Pentagon bug bounty that works, 50,000 vulnerabilities found by volunteer hackers at low cost, set against its own inflated savings math.
The Army's Counter-Drone Office: A Billion-Dollar Answer That Got Reorganized: the Army office made executive agent for counter-drone in 2019, herding more than 40 competing systems as spending climbed past a billion dollars.
Unified Platform and JCC2: The Software Spine of U.S. Cyber Command: Cyber Command's shared weapons platform and joint command layer, whose cost estimate grew more than fivefold and which testers still faulted.

Afghanistan: what we built, and what we left behind

Two decades of building an Afghan force left a documented trail of spending, some of it sound, much of it flagged by the Special Inspector General for Afghanistan Reconstruction, and a great deal of it inherited by the Taliban in a matter of days in August 2021.

DASNet, the Afghan Mission Network, and the biometrics left behind: the coalition network that actually worked, and the identity and pay databases that switched hands when Kabul fell (with the correction that there was no real network called "DASNet").
The 7 billion dollars in equipment left behind: the roughly 7.1 billion in US-funded gear that remained in 2021, and why the viral 83 billion dollar figure is the 20-year force cost, not weapons seized.
The Afghan Air Force planes sold for scrap: 20 transport aircraft that cost about 486 million dollars and ended as roughly 32,000 dollars of scrap metal.
The 43 million dollar gas station: a single filling station that cost about a hundred times a comparable one, and that the Pentagon could not explain.
The 28 million dollar forest camouflage: a proprietary woodland pattern chosen on preference for an army in a country that is about 2 percent forested.
The DoD Forensic Enterprise: The Army's Crime Lab That Went to War: the roughly billion-dollar battlefield forensics enterprise that put names to bomb makers in Iraq and Afghanistan and still runs the military's only full-service crime lab.
Money as a Weapon System: The $7.8 Billion Commanders Handed Out in Iraq and Afghanistan: money as a weapon system, the more than $7.8 billion in battlefield cash commanders handed out, which auditors could never prove actually worked.

Big-ticket acquisitions: cost, delay, and the fine print

Some of the largest programs are not cancelled. They are delivered years late, over budget, or with a catch in the fine print about who actually bore the overrun.

The Ford carrier, the most expensive warship ever: about 13.3 billion dollars, delivered in 2017 with weapons elevators that were not certified until years later.
The Zumwalt destroyer whose guns have no shells: a stealth ship whose guided shells rose to nearly a million dollars each, so the Navy never bought the ammunition.
The Littoral Combat Ship: the modular warships whose mission packages underdelivered and that the Navy began retiring barely a decade old.
The KC-46 tanker Boeing paid for: a fixed-price tanker whose more than 7 billion dollars in overruns fell on the contractor, not the taxpayer, with a camera system that kept scraping the planes it refueled.
IVAS, the goggles that made soldiers sick: an augmented-reality program with a ceiling up to 22 billion dollars whose early hardware caused nausea in testing.

Big science and space: the megaprojects

Not every megaproject is a weapon. Two of the largest were a Moon program and a physics machine, and both were cancelled with money already spent.

Constellation, the cancelled return to the Moon: NASA's roughly 9-billion-dollar Moon program, cancelled in 2010, its capsule and heavy-lift work carried into today's Artemis rockets.
The Superconducting Super Collider: the Texas particle accelerator abandoned in 1993 after about 2 billion dollars and 14 miles of dug tunnel, ceding the energy frontier to CERN.
The Space Test Program: the Pentagon's thirty-million-dollar ride to orbit: the Pentagon's cheapest space program, about $30 million a year flying experimental payloads to orbit as rideshare since the 1960s, and the ESPA ring it gave the launch industry.
Operation Deep Freeze: The Military Airline That Keeps Antarctic Science Alive: the Air Force airlift and Coast Guard icebreaker chain that has carried all US Antarctic science for seventy years, now riding on one aging icebreaker.

Contaminated ground: the cleanups that outlive their deadlines

Some of the largest and slowest public-money commitments are not programs at all. They are messes the government made decades ago and is still paying to clean up, sometimes with the cleanup itself going wrong.

Hanford and the vitrification plant: the most contaminated place in America, the leaking tanks beside the Columbia River, and the glass-making plant meant to lock the waste away that has run for decades and tens of billions of dollars without finishing the job.
St. Louis radioactive waste: Manhattan Project waste that contaminated Coldwater Creek and sits in the West Lake Landfill, with an underground fire smouldering nearby.
Hunters Point Naval Shipyard: the San Francisco shipyard whose radiological cleanup contractor was caught faking soil samples, two supervisors imprisoned, and a redevelopment left in doubt.
The Chemical Weapons Buried Under American Yards, and the Bill to Dig Them Up: the Army's open-ended cleanup of buried WWI-era chemical weapons across about 249 sites in 40 states, with a projected life-cycle cost of $2.5 billion to $17 billion.
The buried shells the Pentagon still cannot price: a tiny Fort Belvoir coordination office sitting on top of a munitions-cleanup bill GAO put at $16 billion to $165 billion.

Toxic exposure and the delayed reckoning

A recurring pattern in the ledger: the government exposes its own people to a hazard in the course of their service, denies the connection for decades, and then pays a very large bill late, through a presumption or no-fault system that trades impossible individual proof for an administrable rule.

Military burn pits and the PACT Act: the open-air waste fires of the post-9/11 wars, the years of denied claims, and the 2022 law that replaced impossible proof with presumption at a decade cost scored near 277 billion dollars.
Agent Orange: the Vietnam herbicide, the decades-long fight, and the presumptive-disability system every later toxic-exposure law is built on.
Camp Lejeune: the contaminated Marine base water, the two-statute remedy, and a claims wave that is still under one percent resolved.
RECA and the atomic veterans: compensation for the downwinders and uranium workers the government irradiated with its own weapons tests, revived and expanded in 2025 after it briefly lapsed.
The Pentagon's blast-injury brain trust, and the research money that nearly vanished: the Army office that coordinates blast-injury research for the signature wound of Iraq and Afghanistan, and the funding that fell from $175 million to zero.

Government research: the bets, the wins, and the myths

The honest counterweight to the rest of this ledger. The same government that produced the failures above also runs some of the highest-return research bets in modern history, on a model that expects most projects to fail so that a few can change the world.

ARPANET to the internet: the modest, sustained research bet that seeded a digital economy the government now measures in the trillions.
The Grand Challenge that launched the self-driving car: a prize nobody won in 2004 that jump-started an entire industry.
The mRNA bet: the early, high-risk public funding that helped seed the platform behind the COVID vaccines, credit shared across many.
CALO to Siri: the DARPA artificial-intelligence research program that became the voice in your phone.
Total Information Awareness: the post-9/11 mass-surveillance program Congress defunded in 2003.
The terrorism futures market: the prediction-market idea cancelled within about a day.
HTV-2, the Mach 20 glider that failed twice: two failed flights that still advanced hypersonics.
The X-51 Waverider: the AFRL scramjet that failed, failed, and then set a record.
The portfolio: hits and misses: how DARPA and AFRL are built for a few huge wins and many failures.
One Manager for Every Bomb: The Navy's EOD Technology and Training Program: the one Pentagon consolidation that quietly worked, a roughly $44 million a year Navy office that is the single source of render-safe tools for every service's bomb techs.
The lab behind the MRE: cheap research, an expensive stockpile: Natick's food lab, a cheap and competent R&D program that gave the military the retort pouch and the MRE, and a war-reserve stock its own officials called too big.
The Military's Language Schoolhouse: Monterey, Lackland, and a Skill the Market Won't Supply: the military's accredited language schoolhouse, a genuine success that trains more than 2,500 students a year, with proficiency that fades in the field.
The sterile fly gambit: how the screwworm was eradicated, and why it is coming back: the USDA sterile-fly program that eradicated a flesh-eating parasite and returned roughly a billion dollars a year, now facing a live re-invasion up through Mexico.

The other side of the ledger: the wage, and who indexes it

Public money is not only what the government spends. It is also what it collects and who it protects against inflation.

COLA versus the merit raise: the government indexes its own obligations to the inflation number it prints, while the private wage is left to fall behind.
The frozen paycheck on Capitol Hill: the one salary that could index itself and votes every year not to.
When the bills rise faster than the raise: the practical playbook for the insurance, energy, and health costs that outrun a paycheck.
Why a global health plan costs half a US one: the four things a cheap international premium hides.

The master hub

This index sits inside a larger one. The working ledgers reads the same method across foundations, dynasties, family offices, and the private structures that hold wealth, not just the public programs that spend it. Start there for the whole map, or here for the government's own ledger, program by program.

This series is informational and journalistic. Every load-bearing figure is documented in the fact-check section at the foot of each linked piece, with links to the original report, statute, or filing. Contested claims are labeled as contested, and where a program can be honestly described as both wasteful and worth it, both readings are shown.

Where the Public Money Goes: The Index of Programs, Lifelines, and Boondoggles