Skip to content
NCBNational Capability Benchmark
Navigate with โ†‘ โ†“Enter to openEsc to close

Challenge the benchmark

Choose a score and give the benchmark a specific reason to reconsider it.

When you submit, the form attaches the current score and confidence for the country and capability you chose.

Submissions are public and start as awaiting review. A maintainer can add a response and signature after review.

Documented examples

Each record describes something a country built that the indicators do not capture. The source provides the published number; the mechanism is our interpretation.

Scores describe outcomes. These records describe how a country acted, what it needed and what happened. 742 of 742 records include a mechanism. They do not affect scores.

Narrow the list

11 of 742 deliveries

Anticipation

1 delivery linked to Government foresight capacity.

United Statessince 1972dismantledbears on Government foresight capacity

Office of Technology Assessment, defunded and closed in 1995

Congress gave the Office of Technology Assessment, created under the Technology Assessment Act of 1972, the statutory function to provide early indications of the probable impacts of technology; it appropriated $21,970,000 for the office with a cap of 143 staff for fiscal year 1995, then appropriated $3,615,000 for fiscal year 1996 to carry out its orderly closure, with no more than 17 employees paid except for severance.

3,615,000 US dollars, appropriation for the orderly closure of the office of technology assessment, FY1996. 21,970,000 US dollars, appropriation for office of technology assessment salaries and expenses, the last operating year, FY1995. US Government Publishing Office, Public Law 104-53, Legislative Branch Appropriations Act, 1996 (109 Stat. 514), retrieved 2026-10-02.

The move, and what it needed

Congress built its own analytic office, governed by a bipartisan board of members, to give legislators early warning on technology choices, and later closed it through the appropriations bill rather than by repeal, cutting its funding by roughly five sixths in one year and capping staff for the wind-down.

Preconditions

  • A legislature with its own budget for in-house analytic agencies
  • Bipartisan support strong enough to protect a small agency through spending rounds, which failed here
  • A statutory mandate that survives defunding, so the function can be revived without new law

Where else it has been used: European parliaments built technology assessment offices on the model, including the German Bundestag's TAB, which cites the US office as its model; the US original closed while the copies continued.

Limits: Technology assessment is adjacent to strategic foresight, not the same thing: the office studied technologies on request from congressional committees rather than running whole-of-government scenarios. It is filed here because its statute, 2 U.S.C. 472(c), states an anticipatory function, to provide early indications of the probable beneficial and adverse impacts of the applications of technology (govinfo, US Code 2023 edition, https://www.govinfo.gov/content/pkg/USCODE-2023-title2/html/USCODE-2023-title2-chap15-sec472.htm). Both figures are appropriations, not spending or output: they show the money Congress gave, not what the office produced, and the 143 and 17 figures are statutory staff caps, not headcounts. The FY1995 figure comes from a different source, Public Law 103-283 in the Statutes at Large (108 Stat. 1423), inside a PDF on page 11 of the file (https://www.govinfo.gov/content/pkg/STATUTE-108/pdf/STATUTE-108-Pg1423.pdf). The authorising statute was not repealed and still appears in the US Code, so the office was ended by withdrawing funds rather than by abolishing its legal basis; it could be refunded without new legislation. The 1996 act also carried $2,500,000 of FY1995 funds forward to September 1996, so the FY1996 closure figure is not the whole of the wind-down money. Started is the year of the authorising Act, not the year the office first reported.

Coordination

3 deliveries linked to Public-private collaboration, University-industry collaboration.

United Statessince 2021dismantledbears on Public-private collaboration

Affordable Connectivity Program, and its lapse

The Federal Communications Commission ran the Affordable Connectivity Program from 31 December 2021, paying monthly broadband discounts through voluntarily participating internet providers to more than 23 million enrolled households, and ended it on 1 June 2024 when the funding Congress had provided ran out.

23 million households, households enrolled in the acp when new enrolments stopped, 2024-02. Federal Communications Commission, press release 'FCC brings Affordable Connectivity Program to a close' (DOC-402930A1), retrieved 2026-10-01.

The move, and what it needed

Congress funded a fixed pot, the FCC turned it into a monthly discount that private internet providers applied on their own bills, and enrolment climbed until the pot was exhausted; when Congress did not refill it, the FCC froze enrolment, cut the benefit to a partial discount and closed the program.

Preconditions

  • A national regulator able to certify eligibility against existing benefit rolls such as Lifeline and SNAP
  • Thousands of private providers willing to apply a federal discount on their own billing systems
  • A one-time appropriation with no automatic renewal, which made the program's life a matter of each budget fight

Where else it has been used: The design reused the Lifeline telephone subsidy's provider-reimbursement model at much larger scale; its end shows how a program financed by one appropriation, rather than by a standing fund, can be stopped by inaction.

Limits: The 23 million figure is the regulator's own enrolment count, taken at the February 2024 freeze, and it reports sign-ups, not households that would otherwise have gone without broadband: the FCC's claim that 68 percent of ACP households had inconsistent or no connectivity before rests on a self-reported agency survey. The program was a voucher paid through private providers whose participation was voluntary, so a hostile reviewer would call it a subsidy channel rather than joint delivery; the public-private element is that providers carried enrolment, billing and the discount. It grew out of the pandemic-era Emergency Broadband Benefit, so the 2021 start understates how long the capability was built. It ended through a funding lapse rather than a repeal, and the FCC said it could resume if Congress appropriated money, so 'dismantled' records the end of delivery, not the destruction of the administrative machinery. The release is an agency press statement, not an audited statistical series, and the PDF is the script-readable copy; the fcc.gov program pages block automated requests.

United Statessince 2020delivered and closedbears on Public-private collaboration

Operation Warp Speed

Operation Warp Speed, an HHS and DOD partnership, financed private vaccine makers' development, at-risk manufacturing and advance purchases, including a $5.97 billion contract for 300 million Pfizer/BioNTech doses. Three of the vaccines it bought held FDA emergency use authorization by 1 March 2021, the first from 11 December 2020.

3 vaccines, ows-purchased covid-19 vaccines holding fda emergency use authorization, 2021-03-01. 5.97 USD billion, ows contract value for 300 million pfizer/biontech doses, 2021-03-01. Congressional Research Service, Insight IN11560 'Operation Warp Speed Contracts for COVID-19 Vaccines and Ancillary Vaccination Materials', version 7, retrieved 2026-10-01.

The move, and what it needed

HHS and DOD pooled BARDA's countermeasure contracting with defense logistics and paid private manufacturers to build production capacity while trials were still running, carrying the financial risk of failure so that doses existed the day authorization came.

Preconditions

  • A standing countermeasure agency (BARDA) with contracting authority and existing industry relationships
  • Domestic firms with mRNA and vaccine platforms already in development
  • Emergency appropriations large enough to fund parallel bets, some of which failed (Merck/IAVI)

Where else it has been used: Advance market commitments of this kind were proposed for global vaccine supply, but COVAX lacked the purchasing budget and manufacturing ties that made the US version fast.

Limits: The metric counts authorized products, not doses delivered or people vaccinated. It does not show what share of the speed came from federal money rather than private R&D already under way: Pfizer/BioNTech took no federal development funding and joined only through a purchase contract, as the CRS notes. The contract values are awards, not outlays, and CRS cites BARDA and DOD pages without auditing them; the GAO's audit of OWS sits on gao.gov, which blocks automated requests, so it is not the source here. The CRS Insight is a congressional research summary rather than a statistical series, and it does not state the May 2020 launch date, so 'started' is the year vaccine delivery began. Ancillary-supply shortages and distribution problems are flagged in the same document. The Biden administration said it planned to restructure and rename OWS, so 'concluded' marks the end of the vehicle, not of federal vaccine purchasing.

United Statessince 2012still operatingbears on University-industry collaboration

Manufacturing USA innovation institutes

In fiscal 2023 the 16 reporting Manufacturing USA institutes, public-private consortia sponsored by the Departments of Commerce, Defense and Energy, managed 929 applied research and development projects and held 2,921 memberships, 655 of them universities, community colleges and other academic members alongside 1,315 small and 493 large manufacturers.

929 projects, ongoing applied research and development projects across the institutes, FY 2023. 655 memberships, academic members (universities, community colleges, etc.) of the institutes, FY 2023. National Institute of Standards and Technology, Advanced Manufacturing National Program Office, Manufacturing USA Report to Congress 2025 (covering FY 2023), retrieved 2026-10-02.

The move, and what it needed

Federal agencies fund a standing consortium per manufacturing technology for a fixed term; firms, universities and colleges pay membership and cost share to sit on a shared applied research and workforce agenda that no single member would fund alone.

Preconditions

  • Multi-year federal base funding with an expectation of non-federal cost share
  • Research universities and manufacturers in the same technology area willing to share pre-competitive IP
  • A national program office that sets common metrics across institutes run by different agencies

Where else it has been used: The model borrowed explicitly from Germany's Fraunhofer institutes, but institutes are member consortia with term-limited federal money rather than permanent contract-research labs, which leaves each one to find a sustainable funding model after its federal term.

Limits: The figures are the institutes' own reports aggregated by the program office, in Table 2 on PDF page 13 (printed page 11) of the report, and they cover fiscal 2023 although the report was posted in March 2026, so they are about three years old. Sixteen of the 17 institutes report; EPIXC joined mid-year and is excluded. The counts are memberships, not distinct organisations: a university that belongs to three institutes counts three times. The academic count mixes research universities with community colleges, which join mainly for workforce training, and the report does not say how many of the 929 projects actually pair a university with a firm; some are workforce or firm-only projects. Federal base funding was 160.4 million dollars in the year against 379.5 million of cost share and other funding, all as reported by the institutes, with no counterfactual for what members would have done anyway. The start year is the first institute, America Makes, established August 2012 per the report's Appendix B; the program's statute is the Revitalize American Manufacturing and Innovation Act of 2014.

Learning

1 delivery linked to Adult learning participation.

United Statessince 1966operating below its peakbears on Adult learning participation

State-administered adult education under the Adult Education and Family Literacy Act, and its erosion

Federally funded, state-administered adult basic, adult secondary and English language programmes served 2,629,643 participants in fiscal year 2000, and by fiscal year 2021 served 770,179, after a fall through every year the table shows up to fiscal year 2020 (627,728) and a partial rebound in 2021.

770,179 participants, participants in state-administered adult basic, adult secondary and english language acquisition programmes, FY2021. 2,629,643 participants, participants in the same programmes in fiscal year 2000, the highest year the table shows, FY2000. U.S. Department of Education, National Center for Education Statistics, Digest of Education Statistics 2022, table 507.20 (from the OCTAE National Reporting System), retrieved 2026-10-02.

The move, and what it needed

The federal government funds states by formula to run adult basic education, high school equivalency preparation and English classes through local providers, and requires every state to report participants and outcomes into one national reporting system.

Preconditions

  • A federal formula grant with a state match that keeps every state in the system
  • A national reporting system that defines a participant the same way across states
  • Local providers, from school districts to community colleges, willing to run low-margin classes

Where else it has been used: Formula-funded adult basic education is common across OECD countries; the American series shows how a stable statute can still deliver a shrinking service when appropriations flatten and accountability rules raise the bar for who counts.

Limits: The table starts in fiscal year 2000, so the second metric is the highest year it shows, not a proven all-time peak. FY2021 is a pandemic year and a partial rebound from FY2020's 627,728; FY2019, before the pandemic, already stood at 948,934, so most of the fall predates it. The fall is computed from two published table cells (about 71 percent), and part of it may reflect tighter participant definitions and federal performance reporting rather than lost provision; the table carries no note on definitional changes. The count covers only programmes reporting to the federal National Reporting System, not community college, employer or private adult learning, so it is not a national adult participation rate. Funding is shared between federal and state governments and the record does not attribute the decline to a single decision.

Experimentation

1 delivery linked to Firms through regulatory sandboxes.

United Statessince 2019dismantledbears on Firms through regulatory sandboxes

CFPB Compliance Assistance Sandbox, rescinded in 2022 and withdrawn again in 2025

The Consumer Financial Protection Bureau opened a national Compliance Assistance Sandbox in September 2019, issued two sandbox approval orders in December 2020, terminated one of them in June 2022 and rescinded the policy with effect from 30 September 2022; a reissued version of January 2025 was withdrawn in May 2025.

2 approval orders, compliance assistance sandbox approval orders listed as granted, 2020-12. Consumer Financial Protection Bureau, Innovation: granted applications, retrieved 2026-10-02.

The move, and what it needed

A federal consumer-finance regulator offered firms a statutory safe harbour and a written approval that a described product complied with named laws, in exchange for sharing data during the test. A later leadership judged that the policies did not advance innovation and lacked transparency, let the information-collection authority lapse and rescinded the policy.

Preconditions

  • Statutory safe-harbour provisions the agency could invoke for good-faith reliance on its approval
  • Leadership continuity across administrations, which this sandbox did not have
  • Guardrails stopping a firm from marketing an approval as an endorsement

Where else it has been used: Sandboxes that rest on an agency policy statement rather than a statute can be switched off by the next director; the CFPB's was rescinded, reissued and withdrawn within three years.

Limits: The count of two is ours, made by counting the approval orders the CFPB lists as granted (Payactiv and Synchrony Bank); the page also lists one template (Build Commonwealth) and separate No-Action Letters and Trial Disclosure waivers, which are not counted here. The rescission and the 2025 withdrawal are read from Federal Register notices on govinfo.gov (87 FR 58439, 90 FR 1974 and 90 FR 20086), not from the cited page. The CFPB page is blocked to some automated clients. Two approvals over three years is evidence that the federal consumer-finance sandbox barely operated before it was ended, not that it failed on its merits: the Bureau's stated reasons were that the policies did not advance consumer-beneficial innovation and failed to meet standards for transparency and stakeholder participation. State sandboxes (Arizona, Utah and others) are subnational and are not counted.

Adaptability

2 deliveries linked to Disaster preparedness and recovery, Institutional responsiveness.

United Statessince 1968operating below its peakbears on Disaster preparedness and recovery

National Flood Insurance Program coverage, and its erosion

OpenFEMA's records of National Flood Insurance Program contracts in force fell from 4,667,875 on 31 December 2009 to 3,589,275 on 31 December 2025, a fall of about 23 percent in the federal flood insurance base over 16 years.

3,589,275 contracts, nfip policy contracts in force (openfema records active on the date), 2025-12-31. 4,667,875 contracts, nfip policy contracts in force at the first complete year-end in the dataset, 2009-12-31. FEMA, OpenFEMA NFIP Redacted Policies dataset (v3), retrieved 2026-10-02.

The move, and what it needed

Congress created a federal insurer of last resort for flood, tied to community floodplain management and to a purchase requirement for federally backed mortgages in high-risk zones, so that recovery after floods would be paid for by premiums rather than only by disaster aid.

Preconditions

  • Flood maps that identify who must buy and that stay current
  • Enforcement of the mandatory purchase requirement by lenders, which is where take-up leaks
  • Premiums that households will keep paying once the mortgage requirement no longer binds them

Where else it has been used: Public flood pools elsewhere (Flood Re in the United Kingdom, CCRIF for Caribbean governments, DASK for Turkish earthquakes) face the same tension: pricing risk honestly drives voluntary buyers out, and keeping them in requires subsidy or compulsion.

Limits: Both numbers are computed, not published headlines: they are the record counts the OpenFEMA API returns for policies whose effective date is on or before, and termination date after, each date, which is the method FEMA's own NFIP data FAQ gives. The 2009 figure comes from the same URL with the date changed to 2009-12-31. The count is of contracts, not policies: one condominium contract can cover many units, and FEMA's own 'NFIP by the Numbers' report for 31 August 2026 shows 3,388,273 contracts but 4,460,103 policies. The API count runs about 2 percent above FEMA's contracts-in-force report (3,589,275 against 3,512,555 for December 2025), so the level carries a small method error, though both ends use the same method. The dataset starts with policies effective in 2009, so the true earlier peak cannot be shown from it. The fall is not all lost protection: private flood insurance has grown, some properties left mapped flood zones, and Risk Rating 2.0 (2021) raised premiums by design to price risk. What the record shows is that the public programme's reach shrank, not that total insured flood exposure fell by the same amount.

United Statessince 2021dismantledbears on Institutional responsiveness

Expanded Child Tax Credit of 2021, and its lapse

The American Rescue Plan, passed in March 2021, expanded the Child Tax Credit and made it payable monthly, and on 15 July 2021 the IRS paid roughly 15 billion dollars to families of nearly 60 million children; Congress did not extend the expansion beyond 2021.

60,000,000 children (nearly), eligible children in families paid in the first monthly payment, 2021-07-15. 15,000,000,000 US dollars (roughly), paid in the first monthly payment, 2021-07-15. US Department of the Treasury, press release of 15 July 2021, retrieved 2026-10-02.

The move, and what it needed

Congress turned an annual tax credit into a monthly payment and the IRS paid it automatically from prior-year tax returns, reaching most families within four months without a new application system.

Preconditions

  • A tax authority holding recent returns and bank details for most families
  • A legislative majority willing to pass a large temporary expansion
  • A political settlement to make it permanent, which was missing

Where else it has been used: Child allowances paid monthly are permanent in Canada and much of Europe; the US case shows a temporary expansion can be delivered quickly and lost as quickly when its legal basis is time-limited.

Limits: The source is the Treasury announcing its own first payment, rounded ('nearly', 'roughly'), not an audited count. The four-month interval from the March 2021 law to the July payment is derived from the release's 'passed in March' and its date. The expansion was legislated for 2021 only and lapsed when Congress declined to extend it, so 'dismantled' records a decision not to continue rather than a repeal. The loss is documented by the Census Bureau, not by this source: in Poverty in the United States: 2022 (https://www.census.gov/library/publications/2023/demo/p60-280.html) the Supplemental Poverty Measure child poverty rate more than doubled, from 5.2 percent in 2021 to 12.4 percent in 2022, and other programmes also expired in that year, so not all of the rise is the credit. Families the IRS could not find through prior tax filings were missed by the automatic payments.

Building

2 deliveries linked to Large project delivery.

United Statessince 2000operating below its peakbears on Large project delivery

Measles elimination, and its erosion

The United States eliminated endemic measles transmission by 2000 through school-entry vaccination requirements, and falling coverage brought 2,777 confirmed cases by August 2026, the most since 1991, against 285 in all of 2024.

2,777 cases, confirmed measles cases in 2026, to 20 august, 2026-08. 285 cases, confirmed measles cases in the last pre-outbreak year, 2024. US Centers for Disease Control and Prevention, measles cases and outbreaks, retrieved 2026-08-27.

The move, and what it needed

The erosion is the mechanism recorded here: vaccination requirements are enforced at school entry by states, so a delivery that took federal purchasing and decades of practice is being undone state by state through widening exemption rules and falling trust, with no single decision anywhere.

Preconditions

  • Enforcement located at fifty state legislatures rather than one national rule
  • A generation of parents with no memory of the disease
  • An information environment that carries vaccine doubt faster than health agencies answer it

Where else it has been used: The same coverage decline is visible across several high-income countries; the American case is the cleanest because elimination was formally certified in 2000, which fixes the baseline the erosion is measured against.

Limits: A case count is the lagging end of the erosion; the leading end is kindergarten vaccination coverage, which CDC reports separately as falling from about 95 percent before the pandemic to under 93 percent, below the outbreak-prevention threshold. The 2026 number is a year in progress and will rise. Elimination status is a WHO-defined property of transmission chains, not of case counts, and had not been formally revoked at retrieval; what the record documents is a maintained public-health achievement being lost while the infrastructure that delivered it still exists.

United Statessince 1956delivered and closedbears on Large project delivery

Interstate Highway System

Under the Federal-Aid Highway Act of 1956, the federal government and the states built the Interstate System. In nominal dollars, the estimated federal cost to complete it rose from $34 billion in the 1958 Interstate Cost Estimate to $114 billion in the final estimate of 1991.

114 USD billion, estimated federal cost to complete the interstate system, final (1991) interstate cost estimate, nominal, 1991. 34 USD billion, estimated federal cost to complete the interstate system, first (1958) interstate cost estimate, nominal, 1958. US Federal Highway Administration, Highway History: How many Interstate programs were there?, retrieved 2026-10-01.

The move, and what it needed

Congress set a 90 percent federal match paid from a dedicated Highway Trust Fund fed by fuel taxes, and the Bureau of Public Roads (later FHWA) apportioned it to state highway departments by repeated cost-to-complete estimates, which let the states build to one national standard on a pay-as-you-go basis.

Preconditions

  • A dedicated user-tax revenue stream that kept construction off the general budget
  • State highway departments with existing engineering capacity to design and let contracts
  • A legislated national network map and design standard fixed before money flowed

Where else it has been used: The trust-fund-plus-match model was copied in transit and airport programs; it travels less well where fuel-tax revenue is small or where no subnational agencies can deliver.

Limits: Both figures are in dollars not adjusted for inflation, so the 235 percent rise we compute from them, (114-34)/34, is not a real cost overrun. FHWA itself attributes the growth to inflation, mileage added by Congress in 1968 and later, 'more safety and more attention to the environment' in later programs and 'changes in the definition of completion', as well as to the Interstate Program 'evolving'. The page gives no real-terms figure, no schedule baseline and no per-mile comparison, so it cannot separate execution failure from scope change. The 1958 ICE was a cost-to-complete estimate used to apportion funds, not a contract budget. The source gives no schedule baseline; the final estimate came 35 years after the Act. The source says later Interstate Programs have not ended, so 'concluded' refers to the end of Interstate Construction apportionments under ISTEA. The figures are federal share only.

Shared Purpose

1 delivery linked to Political polarisation.

United Statessince 1977operating below its peakbears on Political polarisation

Federal funding gaps and the 2019 shutdown

The Congressional Research Service records 20 federal funding-gap shutdowns with at least one full day between FY1977 and FY2019, including the 35-day partial shutdown from 21 December 2018 to 25 January 2019, the longest in the period it surveys.

35 days, longest partial federal shutdown, 2019-01-25. 20 shutdowns, funding-gap shutdowns with at least one full day, 1977-2019. Congressional Research Service, Past Government Shutdowns: Key Resources, retrieved 2026-08-31.

The move, and what it needed

The US appropriations process allows a lapse in budget authority to stop or furlough parts of the federal government, turning unresolved political conflict into an interruption of administration. Continuing resolutions and contingency plans soften the shock but do not remove the incentive or the coordination cost of brinkmanship.

Preconditions

  • Separated institutions in which the executive and legislature can block or condition annual appropriations
  • A legal shutdown rule that makes funding lapses administratively real rather than automatically extending the prior budget
  • Political parties willing to use public-service disruption as leverage, plus contingency systems that limit but cannot eliminate the damage

Where else it has been used: Funding-gap reversals travel as a warning: constitutional checks need a credible continuity mechanism, or polarised bargaining can turn a disagreement into a measurable loss of state capacity.

Limits: Shutdown duration and frequency are institutional consequences that can reflect constitutional rules, divided government and bargaining strategy, not a direct measure of public attitudes or latent polarisation. The CRS series ends in FY2019 for this historical resource, and a shutdown's operational cost varies with which agencies lose budget authority.