Skip to content

US Foreign Aid: The Whole Picture

Document 25: Dollars Out, Value Back, and Every Ledger in Between

Purpose [ANALYTICAL FRAMEWORK — framing, not a factual claim]: "We give away billions while Americans struggle" and "it all comes back to us anyway" are both compressions of the most misunderstood line item in the federal budget. This document decompresses the entire system: how much, to whom, in what forms, under what conditions, what returns economically, what returns strategically, how the US compares to every other donor including its rivals, and what the accountability and effectiveness records actually show — in both directions. It completes this corpus's institutional series (Documents 22–24: the UN, universities, political money) with the largest ledger of all. A reader should leave able to evaluate any foreign-aid claim — about any country — in about a minute.

The Short Version

Most Americans believe foreign aid is about a quarter of the federal budget. It has never been close: at its recent peak it was about one penny per dollar, and in 2025 it was cut by more than half — the largest aid cut by any country in recorded history. The money comes in different kinds: weapons credits that must be spent at American factories, food grown on American farms, medicine programs that have saved millions of lives, and emergency help after disasters. Much of the money never leaves America at all — but that is also why each dollar buys less overseas. Some aid buys things money can't easily measure: Egypt's aid has kept peace with Israel for 47 years, and Israel shares intelligence and technology America cannot buy anywhere else. Other countries do it differently: China lends money instead of giving it, and expects it back. And sometimes aid fails completely — America's own inspectors documented tens of billions wasted in Afghanistan. The truth is not "we're being robbed" or "it's all fine." The truth is a balance sheet — and this page shows you the whole thing.

Every fact in the full page below carries a tag showing its evidence.

PART 1: THE SIZE — AND THE PERCEPTION GAP

  • Americans on average estimate foreign aid at ~26% of the federal budget; only ~11% correctly say ~1% or less. Told the true figure, the share saying the US spends "too much" falls by nearly half [VERIFIED via KFF polling]
  • The actual figures: $71.9B disbursed in FY2023 — 1.2% of federal spending (obligations that year ran ~$99.9B — see the trap below); ~$73.5B in FY2024; $47.3B in FY2025 — the lowest inflation-adjusted level since 2004; roughly $7B through three-quarters of FY2026 [VERIFIED via foreignassistance.gov data as analyzed by Pew and CRS]
  • What happened in 2025: a January executive order froze assistance; ~83% of USAID's programs (~5,200 contracts) were cancelled; USAID ceased operations July 1, 2025 and was absorbed into the State Department; Congress rescinded ~$7.9B more; an August "pocket rescission" cancelled a further $4.9B — a maneuver the Government Accountability Office found illegal under the Impoundment Control Act, which the administration disputes [VERIFIED for the events; the legality PARTIALLY DISPUTED — in litigation]
  • The trap that explains most contradictory headlines [ANALYTICAL FRAMEWORK]: appropriated (what Congress made available), obligated (contracts signed), and disbursed (cash actually paid) are three different numbers for the same year — FY2023 is honestly describable as $71.9B, $85.1B, or $99.9B. Pentagon-account weapons transfers sit outside all three (Part 15). Every figure on this page carries its stage-of-spending qualifier

PART 2: THE FORMS

FY2023 disbursement shares [VERIFIED via foreignassistance.gov/Pew]:

  • Humanitarian/disaster relief: ~22% — refugees, famine response, Food for Peace
  • Macroeconomic support: ~22% — dominated that year by Ukraine budget support
  • Global health: ~15%+ — PEPFAR alone ~$6.5B/yr (Part 16 for its evidence)
  • Military aid: ~11% on the civilian ledger — Foreign Military Financing grants (~$6B/yr: Israel $3.3B, Egypt $1.3B, Jordan ~$425M+) — plus a parallel Pentagon ledger this figure misses entirely (Part 15)
  • Nearly all of it is grants. The US abandoned sovereign lending decades ago; the 2024 Ukraine loan structure was the first USAID loan to a foreign government in ~30 years [VERIFIED via CRS]

PART 3: THE RECIPIENTS, RANKED

Top recipients across three very different years [VERIFIED via foreignassistance.gov/Pew]:

Rank FY2023 (peak) FY2024 FY2025 (collapsed)
1 Ukraine $16.6B Ukraine $10.1B Ukraine $6.7B
2 Israel $3.3B Israel $6.8B (supplemental) Israel $3.3B (99.8% military)
3 Ethiopia $1.8B Jordan $1.7B Jordan $1.7B
4 Egypt $1.4B Egypt ~$1.4B Ethiopia $919M
5 Jordan $1.2B+ Ethiopia DRC $749M
  • The top 10 recipients take ~62% of the total; ~175 countries receive something. Full per-country lookup: the federal foreign assistance database [VERIFIED]
  • The 2025 pattern worth naming: military-heavy recipients (Israel, Egypt, Jordan) were spared the collapse almost entirely; humanitarian-heavy recipients (Ethiopia, Yemen, Somalia, Sudan, Afghanistan) absorbed the deepest cuts [DOCUMENTED via Pew/CGD analyses]

PART 4: THE CONDITIONS AND USAGE RULES

  • Leahy Laws: no assistance to foreign security-force units credibly implicated in gross human-rights violations, with individual-level vetting — enforcement rigor is contested by analysts in both directions [VERIFIED for the statutes; enforcement PARTIALLY DISPUTED]
  • Recipient-specific strings: a tranche of Egypt's FMF (recently $320M) is conditioned on democracy certifications — waived on national-security grounds, including fully in 2024, which critics argue nullifies the conditions [VERIFIED for the mechanism and waiver; the nullification argument DOCUMENTED]. The Taylor Force Act (2018) bars economic aid benefiting the Palestinian Authority while "martyr payments" continue — that channel remains closed [VERIFIED]. Lebanon's aid is conditioned on army independence from Hezbollah [DOCUMENTED]
  • The Buy-American architecture — load-bearing for Part 5: FMF must by statute be spent on US defense articles (Israel's unique carve-out to spend ~26% at home phases to zero by 2028); Food for Peace commodities must be US-grown and 50% must ship on US-flagged vessels [VERIFIED via the Arms Export Control Act, appropriations language, and cargo-preference law]
  • Appropriations earmarks lock in country minimums and sector directives — the 2025–26 constitutional fight is precisely over an executive declining to spend what Congress directed [VERIFIED; outcome pending]

PART 5: WHAT RETURNS ECONOMICALLY — THE CATEGORY TABLE

There is no single honest number. There is an honest table:

Category Share returning to US firms/workers Tag
FMF military aid (~$6B/yr) ~100% by statute — purchase credits at US defense plants VERIFIED
Ukraine military packages ~86–90% — drawdowns from US stocks, replenishment at US factories (155mm shells in Scranton, PA; HIMARS in Camden, AR; Javelins in Troy, AL) DOCUMENTED
Ukraine packages in total ~60–77% — the viral "90% stays home" is true only of the military subset PARTIALLY DISPUTED as commonly stated
Food aid US commodities + 50% US shipping mandated; GAO found ~65% of emergency food-aid budgets historically went to transport/handling/admin, and cargo preference adds ~$50M/yr in pure cost VERIFIED via GAO
USAID contracts/grants Majority awarded to US-based organizations (10 firms took >50% of contract dollars) — but much is then spent in-country; the viral "only 10% helps people" inverts a localization statistic and is wrong DOCUMENTED
Multilateral contributions / humanitarian cash Near-zero direct US return DOCUMENTED
  • Macro scale: total aid ≈ 0.15–0.25% of GDP — economy-wide effects are rounding error; the real effects concentrate in specific factories and farm districts. (And the jobs marketing inflates: the F-35's "jobs in 48 states" claim was found roughly double plausible estimates) [VERIFIED/DOCUMENTED]
  • The coupled truth this page exists to teach [ANALYTICAL FRAMEWORK]: "the money comes back to us" and "tied aid is wasteful" are the same verified fact from opposite sides. The OECD finds tying aid raises procurement costs 15–30% — over 40% for food aid. High domestic recapture is precisely why each dollar delivers less abroad. Any argument citing one half without the other is compressing
  • The waste ledger, stated plainly: of ~$146B in Afghanistan reconstruction, SIGAR documented $26–29B in waste, fraud, and abuse (93% waste) and concluded the mission model itself must confront "the real possibility of failure" [VERIFIED via SIGAR]

PART 6: WHAT RETURNS STRATEGICALLY

Security aid is purchased for strategy, not development. The documented returns:

  • Egypt — the aid-for-peace architecture: the $1.3B/yr flows from the Camp David framework (a political undertaking — the treaty text itself contains no aid obligation, a precision most summaries miss). The return: 47 years of unbroken Egypt-Israel peace, expedited Suez Canal transit for US warships (762 expedited transits during Desert Storm; ~35,000 overflights permitted), and broad access since [VERIFIED via treaty texts and GAO/CRS records; DOCUMENTED for the transit figures]. The cost: the same money underwrites the most repressive era in Egypt's modern history — the canonical aid-props-autocrats case [DOCUMENTED via published scholarship]
  • Jordan — underwriting the 1994 treaty: $1.45B/yr under the fourth MOU; the 2021 Defense Cooperation Agreement grants US forces access to twelve facilities including five air bases; ~3,800 US personnel; one of the region's closest intelligence partnerships [VERIFIED via the DCA and CRS]
  • Bases and access elsewhere: Djibouti's Camp Lemonnier ($63M/yr rent), the Philippines' nine EDCA sites, and the cautionary Manas tale — access purchased for 13 years, then lost to Russian counter-pressure [VERIFIED/DOCUMENTED]
  • Aid buys UN votes — peer-reviewed: panel analysis of 143 countries over three decades found US aid, specifically budget support and grants, measurably increases recipient voting compliance with the US in the General Assembly — an effect not found for other G7 donors (Dreher, Nunnenkamp & Thiele). Read alongside Document 22's bloc arithmetic: the machine runs on more than ideology [DOCUMENTED via Public Choice (2008)]

PART 7: THE ALLIANCE LEDGER — RECIPROCITY THAT HAS NO PRICE

The Israel relationship is the sharpest case of value that flows back in a currency no audit can convert:

  • The documented in-kind flow: Mossad delivered Khrushchev's 1956 secret speech to the CIA; Operation Diamond (1966) delivered a defected MiG-21 — the West's first access to the Soviets' frontline fighter; captured Soviet systems from the 1967/1973 wars fed US countermeasure development; the 2018 Iranian nuclear archive was shared with Washington; co-developed missile defense (Iron Dome, David's Sling, Arrow) is battle-tested under fire on Israel's budget — the Arrow-3 sale to Germany generated $3.5B with up to half the work performed by US contractors; the Israeli-developed Trophy system protects US Abrams tanks; Israel pioneered the modern military drone the US later adopted [VERIFIED for the historical episodes and co-production; DOCUMENTED for the valuations, which come largely from pro-alliance sources]
  • The honest calibration [ANALYTICAL FRAMEWORK]: the reciprocal flow is verified in kind, documented in episodes, and unquantifiable in dollars. The famous valuations — Gen. George Keegan's claim that Israeli intelligence was "worth five CIAs," Alexander Haig's "unsinkable aircraft carrier that doesn't carry a single American soldier" — are attributed statements by senior officials, not audits. No audited replacement-cost figure exists or can exist: the counterfactual question is not what equivalent collection would cost the US intelligence budget (~$100B/yr), but whether native-fluency regional penetration could be bought at any price
  • The symmetry column an honest ledger must carry: the Pollard espionage case (an ally running an agent inside US intelligence); the cancelled Phalcon radar sale to China (2000) and related tech-transfer friction; and the published "strategic liability" school (Mearsheimer and successors: the US had to keep Israel out of the 1991 coalition; domestic politics, not strategy, sustains the relationship) — countered by the asset school and partially undercut by the Abraham Accords, in which Arab states allied with the US and Israel simultaneously [VERIFIED for the cases; the asset-vs-liability question PARTIALLY DISPUTED — both positions live in legitimate scholarship]
  • The upshot for every "is Israel worth it" argument: the people citing "$310 billion since 1948" and the people citing "worth five CIAs" are both pointing at real things measured in incommensurable units. This page presents both columns and declines to fake the conversion

PART 8: THE GENEROSITY LEDGER BEYOND GOVERNMENT

  • Remittances: the US is the world's largest source — outflows estimated at $80–110B/yr (official balance-of-payments methods run lower than World Bank estimates; the gap is methodological and flagged as such). That exceeds the entire official aid budget. Mexico alone received a record $64.7B in 2024, ~96% from the US [VERIFIED via World Bank/Banxico; the total-outflow range DOCUMENTED]. A 1% excise tax on cash-based remittances took effect January 2026 [VERIFIED via IRS]
  • Private philanthropy: US cross-border giving ~$50B/yr — foundations, congregations, corporations, universities — the largest of 47 countries measured; the Gates Foundation alone (moving to a $9B annual payout) became WHO's largest donor after the US withdrawal [DOCUMENTED via the Global Philanthropy Tracker and WHO reporting]
  • The "total engagement" argument and its critique, both stated: advocates sum official + private + remittances to ~$200B+ and argue the "stingy America" charge misses the US private-flow model. Critics answer that counting migrants' own wages as American generosity is indefensible — "nothing about the money involves any donation by any American." Both framings are real; the components are verified; the framing war is tagged as what it is [DOCUMENTED both directions]

PART 9: THE US AGAINST THE WEST

  • Both viral claims are half-true, and 2025 broke both: the US was the largest donor in absolute dollars for three decades (~$63–66B, ~30% of all Western aid) while ranking in the bottom quartile as a share of national income (0.24% vs the UN's 0.7% target; Norway ~1%, Germany 0.79%; per-capita: US ~$200 vs Germany ~$430, Norway ~$1,000) [VERIFIED via OECD]
  • 2025: US aid fell 56.9% — the largest single-donor cut ever recorded — dropping the US to #2 behind Germany for the first time in history and to dead last among developed donors (33rd of 33) at 0.09% of GNI. Global aid fell 23.1% — the largest contraction on record — with the US driving roughly three-quarters of it; the UK (funding defense), France, and Germany cut too; the OECD projects further decline in 2026 — the first simultaneous multi-year retreat by essentially all major Western donors [VERIFIED via OECD preliminary 2025 data]
  • The Marshall Plan yardstick: ~$13.3B in 1948–52 ≈ ~2% of US GDP in its first year. A proportionally equal effort today would exceed $1 trillion. The 2025 US level: 0.09% [VERIFIED/DOCUMENTED via the standard conversions]

PART 10: THE RIVALS — WHO ELSE IS AT THE TABLE

  • China's model is the mirror image of America's: ~$1.34 trillion across 20,985 projects in 165 countries (2000–2021), running ~$80B/yr — the world's largest single official development financier — at a 31-to-1 loan-to-grant ratio (the US portfolio is overwhelmingly grants). Developing-world debt to China: $1.1–1.5T. The 2020s pivot: "small and beautiful" projects and $240B in rescue lending to its own distressed borrowers — Beijing has become debt collector as much as banker [DOCUMENTED via AidData, the best independent dataset; China publishes no comparable figures]
  • The "debt trap" story, held to the evidence: the viral Hambantota narrative (China lent, then seized the port) is contradicted by scholarship — the port was a decades-old Sri Lankan project the US and India declined to fund; Chinese loans were ~9% of Sri Lanka's debt; the lease payment shored up reserves against mostly Western market-rate debt; Sri Lanka retains ownership. What remains genuinely disputed is whether the lending was imprudent and strategically advantageous to Beijing — Sri Lanka still defaulted [DOCUMENTED via Chatham House/Johns Hopkins CARI; the residual dispute tagged as such]
  • The Gulf model: Saudi Arabia ($5.6B ODA in 2024), the UAE, Qatar, and Kuwait give heavily in the region — but their signature instrument is the central-bank bailout deposit, exemplified by the UAE's $35B Ras El Hekma deal for Egypt (which unlocked IMF, EU, and World Bank packages). Qatar: $1.3B to Gaza 2012–2021 [VERIFIED via OECD; DOCUMENTED for the deal mechanics]
  • The vacuum question, answered with documentation rather than fear: after the US collapse (aid to Africa fell $12.1B → $7.9B in one year), China's fills have been real but token — $4M to Africa CDC, a $500M five-year WHO pledge, a $51B FOCAC package of loans and investment. The expert consensus: China's instruments don't map onto what USAID did — HIV treatment, famine response, election support. As the Center for Global Development put it: "Chinese assistance won't replace USAID — that's the problem" [DOCUMENTED]

PART 11: THE MULTILATERAL EXIT

The channel most summaries forget — and the one the US just left:

  • The normal-year baseline: 22% of the UN regular budget; ~25% of peacekeeping; one-third of the Global Fund (statutory cap; $1.65B/yr); $4B pledged to the World Bank's IDA21; ~$300M/yr to Gavi; largest donor to WHO (~16–18% of its budget) [VERIFIED via CRS/KFF]
  • The withdrawal, 2025–26: WHO exit effective January 22, 2026 — WHO cut its budget ~21% and halved its management layers; UNRWA funding ended outright (February 2025 executive order, after the 2024 congressional ban); 44 UNFPA grants terminated (a $335M maternal-health gap); the pocket rescission cancelled $900M in UN contributions and $326M in peacekeeping; January 2026 brought withdrawal from 66 international organizations including the UN's climate treaty; the UN ended 2025 with record arrears and warned it could run out of cash by mid-2026 [VERIFIED via the executive orders, UN documents, and UNFPA statements]
  • Cross-reference: Document 22 for what the UN system is and does; this is the funding half of that story

PART 12: UKRAINE VS ISRAEL — THE THREE-YEAR COMPARISON

Ukraine Israel
2023 Drawing on $113B of 2022–23 supplementals; US the largest single donor $3.8B MOU baseline; emergency drawdowns begin Oct 2023
2024 +$61.7B supplemental (April); ~$9.5B of economic aid converted to forgivable loans — $4.65B forgiven that November $3.8B MOU + $14.3B supplemental military aid; ~$17.9B in year one post-Oct 7
2025 ~$0 new appropriations — allocations fell ~99%; pivot to sales: NATO allies buy US weapons for Kyiv ($6.7B in the PURL mechanism's first year) plus the minerals agreement MOU continues untouched; cumulative post-Oct 7 military aid $21.7B+ — an explicit floor (Brown Costs of War); $4B in FY2026 appropriations
Cumulative ~$175B appropriated since 2022 (the Kiel tracker's $135B counts allocations — different ledger, both correct); ~2/3 defense-related, much spent at US factories; Europe has now given more than the US ~$310B since 1948 — the largest recipient since WWII; grants throughout; no loan component
Structure Ad hoc, politically contested, partly self-spent, halted abruptly — no top-recipient program had ever been zeroed in a single year Standing MOU: predictable, treaty-like, survived the 2025 collapse untouched

[VERIFIED via CRS/CRFB/Kiel; the Costs of War figures DOCUMENTED]

PART 13: UKRAINE ACCOUNTABILITY — THE SAME RAZOR AS GAZA

The "Ukraine aid was stolen" claim conflates three records this page separates [the same three-way razor Document 16 applies to Gaza aid]:

  1. Ukrainian corruption with Ukrainian funds — real, documented, and prosecuted by Ukraine's own US/EU-backed institutions: the 2023 procurement scandals that removed the defense minister; the $40M mortar-round embezzlement (indicted); Operation Midas (Nov 2025) — a ~$100M energy-sector kickback scheme reaching Zelensky's inner circle, exposed by NABU wiretaps; ministers dismissed; and the July 2025 crisis in which Zelensky briefly subordinated the anti-corruption agencies, met mass protests and EU pressure, and reversed himself in nine days by a 331–0 vote [VERIFIED via NABU/prosecution records; individual guilt pending]
  2. US process lapses — real, and remediated: the DoD Inspector General found ~$1B (59%) of serial-number-tracked items had overdue inventories in mid-2023 — delinquent paperwork, not missing weapons — falling to 12% by May 2024 [VERIFIED via DoD IG reports]
  3. Mass diversion of US aid — not documented by anyone: no IG, GAO, or special-inspector report has substantiated diversion of US weapons to black markets; military aid was delivered in kind (Ukrainian officials never touch the cash), and budget support flowed as World Bank-verified reimbursements with third-party audit — the viral yacht-and-villa catalogue (22 debunked claims and counting) traces to fabricated "whistleblower" videos amplified by pro-Russia networks [VERIFIED via the oversight record and independent fact-checks]

PART 14: THE PALESTINIANS — THE FULL ARC, AND GAZA

  • The complete arc, 1993–2026: more than $7.6B in US aid since 1993 (plus ~$4B in cumulative UNRWA contributions since 1950) through three channels — UNRWA, USAID projects (no cash to the PA), and the US Security Coordinator mission that trained PA security forces. Then: the 2018 Trump cutoff (UNRWA $360M → $0; projects wound down; security cooperation survived every swing — Israel itself valued it); the 2021 Biden restoration (~$1B+ over three years); the wartime response (~$1.2B humanitarian Oct 2023–Jan 2025 — the $9.3B figure in the April 2024 bill was global, a standard conflation); the 2024 UNRWA ban (statutory, renewed — now a continuing total ban); and 2025 near-zero — Gaza disbursements frozen, PA security assistance ended [VERIFIED via GAO/CRS/statute]
  • The pier: $230M, operational ~20–25 days, delivered roughly one day's worth of Gaza's assessed need, dismantled July 2024 [VERIFIED]
  • The $30M Gaza Humanitarian Foundation grant (June 2025): awarded after a State official waived nine mandatory counterterrorism and anti-fraud safeguards over reviewers' objections; the UN attributed 1,000+ deaths to shootings near its distribution sites (responsibility DISPUTED); GHF shut down permanently November 2025 [VERIFIED for the grant and closure; DOCUMENTED for the waivers]
  • The "US funds both sides" arithmetic, with definitions instead of slogans: post-Oct 7 military aid to Israel ~$21.7B (floor) vs humanitarian aid to Palestinians ~$1.23B — roughly 17:1, with the humanitarian side going to near-zero in 2025 while military aid continued. Reconstruction need: $53B (UN/EU/World Bank assessment); financing unresolved [VERIFIED components; DOCUMENTED assessment]
  • Aid diversion: individual incidents documented; a mid-2025 internal US review found no evidence of systematic diversion of US-funded aid to Hamas; Israel maintains the systematic claim [PARTIALLY DISPUTED — full treatment in Document 16]

PART 15: THE PENTAGON LEDGER — AND THE SALES CONFUSION

  • A parallel system the "foreign aid" totals never count: DoD-funded security cooperation — global train-and-equip (~$1.1B/yr), the counter-ISIS fund, Pakistan's $14.5B in Coalition Support reimbursements (2002–17), Taiwan's first-ever FMF and drawdowns (2023) — runs on the order of $10–20B/yr, and by design is not tracked country-by-country the way State's ledger is. ~95% of Ukraine security assistance ran through Pentagon accounts ($31.7B drawdowns + $33.5B USAI vs $3.7B State-ledger FMF) [VERIFIED via CRS/GAO; the annual total DOCUMENTED as a range]
  • Sales are not aid — and they're five times bigger: US government-brokered and licensed arms sales hit $318.7B (FY2024) and $331.2B (FY2025) — records, overwhelmingly partner cash (Saudi Arabia's ~$142B package; Israel's $18.8B F-15 purchase on top of its FMF grants). Foreign money buying US weapons is commerce; FMF grants are aid; public debate conflates them constantly, in both directions [VERIFIED via State Department arms-transfer fact sheets]

PART 16: DOES AID WORK? THE EVIDENCE LEDGER

  • The works column (mortality-grade evidence): smallpox eradication cost ~$300M total and repays the US contribution every 26 days in avoided costs; PEPFAR's mortality effect is peer-reviewed (~20% reduction in adult death odds in focus countries; the headline "26 million lives saved" is the program's own model — cited as such); Gavi's 18.8M deaths-averted estimate (also modeled); insecticide-treated bednets and direct cash transfers carry the strongest trial evidence in development economics; PL-480 food aid bridged India to its Green Revolution; South Korea went from food-aid recipient to OECD donor — the only member ever to make that journey [VERIFIED for the peer-reviewed anchors; DOCUMENTED for program models, labeled as models]
  • The doesn't-work column: peer-reviewed work finds no robust aid-growth correlation and documents Dutch-disease harm to recipients' export sectors (Rajan & Subramanian); mass deworming — a former development darling — is genuinely contested (the "worm wars"); Egypt is the canonical case of aid entrenching autocracy; and Afghanistan is the $146B proof that dollars can fail completely [DOCUMENTED via the published literature; SIGAR VERIFIED]
  • The honest synthesis [ANALYTICAL FRAMEWORK]: humanitarian and health aid has strong life-saving evidence; transformational development aid has weak macro evidence — the success stories had aid as one input among many; security aid buys strategy and must be judged on Part 6's ledger, not development metrics

PART 17: THE LEDGER — DOLLARS OUT VS VALUE BACK, PER NATION

Dollar columns VERIFIED; recapture rates are the category estimates of Part 5; strategic returns and costs are documented items deliberately not priced — a single "net profit/loss" number for an alliance is false precision by construction:

Nation US $ out Est. % recaptured by US economy Documented value back Documented costs & risks
Israel $3.8B/yr; ~$310B since 1948; +$21.7B post-Oct 7 ~74% → ~100% by 2028 The Part 7 intelligence/technology flow; co-developed missile defense; battle-test data; no US troops ever requested Pollard; Phalcon-China; the "liability" school's regional-standing argument [DISPUTED]
Ukraine ~$175B appropriated 2022–24; ~$0 new 2025 ~60–77% total; ~86–90% military Degraded the principal European adversary's military without one US soldier engaged; rearmed US production lines; allies now buying US weapons for Kyiv ~$29B undelivered at handoff; early tracking lapses (fixed); escalation risk
Egypt $1.3B/yr; >$50B cumulative ~100% 47 years of peace with Israel; Suez surge access; overflight Underwrites authoritarian entrenchment; conditions routinely waived
Jordan $1.45B/yr Mixed 1994 treaty; 12-facility US access; premier intelligence partnership Stability dependence
Taiwan First FMF + drawdowns (2023–) ~100% Deterrence at the highest-stakes flashpoint Direct entanglement risk
Humanitarian tier (Ethiopia, Yemen, Somalia, Sudan, DRC) ~$1–2B/yr each at peak; deepest 2025 cuts Low Mortality-grade returns; epidemic security; the soft-power channel rivals cannot fill Weak growth evidence; dependency critique; conflict-zone diversion risk
Palestinians $7.6B+ since 1993; near-zero today Low PA security coordination Israel itself valued; US leverage Taylor Force problem; UNRWA findings; the diversion contest (Doc 16)
Afghanistan (legacy) ~$146B over 20 years Low The cautionary row: $26–29B documented waste — proof that dollars out ≠ value anywhere Mission failure — the row that disciplines every other
Pakistan (legacy) $14.5B reimbursements 2002–17 Low Two decades of war logistics access — purchased, then lost The documented double game; suspended 2018

How to read this table [ANALYTICAL FRAMEWORK]: the third column is arithmetic; the fourth is real but unpriceable; the fifth is what advocacy versions omit. The pattern across rows: security aid buys strategy and mostly returns economically; humanitarian aid buys lives and mostly doesn't; the worst outcomes came not from theft but from mission failure — and the two nations most often accused of "robbing" America, Israel and Ukraine, are precisely the rows where documented dollar-recapture is highest.

THE READER'S TOOL — FIVE QUESTIONS FOR ANY FOREIGN-AID CLAIM [ANALYTICAL FRAMEWORK]

  1. What stage of spending? Appropriated, obligated, and disbursed differ by tens of billions for the same year.
  2. What category — and what's its return rate? A dollar of FMF and a dollar of famine relief have opposite economics.
  3. What's the denominator? Aid is ~1% of the budget; Americans guess 26%.
  4. Which ledger? State's aid budget, the Pentagon's parallel system, and $300B+/yr of arms sales are three different things wearing one vocabulary.
  5. Compared to what? The honest comparisons are other donors (absolute AND income-relative), the rival lending model, and the counterfactual — including the strategic goods no invoice can price.

VERIFICATION SOURCES