The Bookkeeping: Sources & Methodology

Estimates here are reasonable but they are estimates. We have made concessions to produce a picture. The picture is not sharp. It describes everything in it with accuracy, not precision: the asymmetry between Ukraine's legible-supplemental shape and Israel's structural-aggregate shape is real, and that asymmetry is what the picture is for. Single-cell precision is not.

A reader who cares about a specific dollar figure should read the channel section it comes from, look at the confidence flag, and read the source. Any reader who wants to dispute a figure can do so directly against the cited primary source. Where the source does not exist, that absence is itself the claim.

Every cell in the charts is one of three things.

HARD - statutory or line-item, ±0. The figure exists in a primary source as a single number. Examples: MOU FMF baselines, supplemental appropriations.

PARTIAL - published but not granular by year or by channel. The aggregate exists in a primary source; the per-year or per-channel split is reconstructed from secondary documents (DoD budget submissions, NDAA texts, think-tank analyses). Error bar typically ±15-30%. Examples: missile-defence co-development annual splits; CENTCOM operations attribution post-October 2023.

ESTIMATE - no published source aggregates the channel. The figure is computed from a stated methodology against named inputs. Error bar typically ±50% or wider. Examples: sub-threshold transfers (Washington Post counted but did not aggregate); pre-2023 CENTCOM operations attributable to Israel defence.

The chart UI flags ESTIMATE cells with a hatched fill and an "(est)" suffix on segment labels. PARTIAL cells render with a solid fill and a smaller "(±N%)" indicator. HARD cells render plain. A reader looking at the Israel donut sees at a glance which slices are which.

Cumulative figures cover 1999-2024. The window opens on the Clinton MOU (19 July 1999), the first of the three ten-year floor commitments that anchor the architecture argument. It closes at the most recent full calendar year for which the supplemental and DoD-budget data are settled. Pre-1999 attribution is out of scope: the architecture being analysed begins with the first MOU.

The Israel side has continuous flows back to 1949. The CRS RL33222 inflation-adjusted figure of $317.9bn through 2022 is the canonical cumulative-since-1951 number and is cited in the rumination as such; the per-channel reconstruction here does not attempt to extend that aggregate further into the pre-MOU era.

The Ukraine side has effectively no flows of consequence before 2014; meaningful figures begin then, with the supplemental peak 2022-2024.

The per-channel sourcing pages are populated as the research subagents complete each channel. Where a channel is not yet sourced, this page shows the channel header and the planned sources only; the figures in the chart are flagged accordingly.

A. MOU FMF Baseline (Israel) - HARD

Definition: Statutory annual Foreign Military Financing appropriated to Israel under the three security-assistance MOUs (Clinton 1999, Bush 2007, Obama 2016), FY1999-FY2024. Excludes: Clinton-era ESF (separate phase-out, not military), the $1.2bn Wye River supplemental in FY2000 (separate Israeli-Palestinian peace-process agreement), the FY2003 P.L. 108-11 emergency supplemental ($1.0bn military grant - Channel F), and the $0.5bn/yr missile-defence carve-out under the Obama MOU (Channel B). Figures are appropriated FMF in current (nominal) USD millions.

Per-year table:

FY	MOU	Channel A FMF ($m)	Notes
1999	Clinton	1,860.0	Glide path year 1 (+$60m/yr ramp)
2000	Clinton	1,920.0	Excludes $1,200m Wye River supplemental (separate agreement)
2001	Clinton	1,975.6
2002	Clinton	2,040.0
2003	Clinton	2,086.4	$3,086.4m gross minus $1,000m P.L. 108-11 supplemental (Channel F); reflects 0.65% mandated rescission
2004	Clinton	2,147.3	0.59% rescission
2005	Clinton	2,202.2
2006	Clinton	2,257.0
2007	Clinton	2,340.0
2008	Clinton	2,380.0	Final year; appropriated above $1.8bn glide-path target as Bush MOU phase-in began early
2009	Bush	2,550.0	Bush MOU year 1; $30bn/10yr envelope, front-loaded ramp
2010	Bush	2,775.0
2011	Bush	3,000.0	First year at headline $3bn/yr
2012	Bush	3,075.0
2013	Bush	2,943.2	$3,100.0m appropriated; BCA sequester (P.L. 112-25) reduced by ~$155m (~5.0%)
2014	Bush	3,100.0
2015	Bush	3,100.0
2016	Bush	3,100.0
2017	Bush	3,175.0
2018	Bush	3,100.0	Bush MOU final year
2019	Obama	3,300.0	Obama MOU year 1; $500m missile defence (Channel B) excluded
2020	Obama	3,300.0
2021	Obama	3,300.0
2022	Obama	3,300.0
2023	Obama	3,300.0
2024	Obama	3,300.0	Channel A baseline only; $3.5bn FY2024 ISSAA supplemental (P.L. 118-50) is Channel F

Cumulative FY1999-FY2024 Channel A FMF: $70,926.7m ≈ $70.9bn (nominal).

Decomposition: Clinton MOU envelope (FY1999-FY2008) $21.21bn vs. $21.3bn pledged; Bush MOU envelope (FY2009-FY2018) $29.92bn vs. $30bn pledged (sequester accounts for the ~$80m shortfall); Obama MOU first six years (FY2019-FY2024) $19.80bn vs. $19.8bn on the $3.3bn/yr schedule (6/10 years of $33bn FMF total).

Methodology:

• Clinton MOU ESF phase-out: The 1999 Glide Path Agreement bundled an FMF ramp (+$60m/yr) with a parallel ESF wind-down (-$120m/yr). Channel A treats only the FMF leg. ESF figures (FY1999 $1,080m, zeroed by FY2008) are excluded.
• FY2000 Wye River: CRS Table B-1 lists FY2000 "Military Grant" as $3,120m, comprising $1.92bn baseline FMF plus $1.2bn for Wye Agreement implementation. The Wye supplemental was a separate Israeli-Palestinian peace-process commitment (1998 Wye River Memorandum) appropriated through P.L. 106-113 Title VI, not baseline MOU FMF. Channel A uses $1,920m; Wye River $1.2bn goes to Channel F.
• FY2003 P.L. 108-11 supplemental: The FY2003 Emergency Supplemental Appropriations Act provided $1.0bn in military grant aid on top of the MOU baseline. CRS gross figure $3,086.4m (post 0.65% rescission); Channel A is gross minus $1.0bn supplemental ≈ $2,086.4m.
• FY2013 BCA sequester: The Budget Control Act sequester reduced FMF by ~5.0% in FY2013 (~$155m). CRS Table B-1 shows appropriated $3,100.0m and "After Sequestration" $2,943.2m. Channel A uses the post-sequester figure as the actual flow. Alternative methodology (nominal $3,100.0m) raises cumulative by $155m to $71.08bn.
• Obama MOU $3.8bn → $3.3bn taxonomy correction: The 2016 MOU's headline "$38bn over 10 years" combines $33bn FMF (Channel A) with $5bn DoD missile-defence appropriations (Iron Dome / David's Sling / Arrow II / Arrow III - Channel B). The MOU joint statement (14 Sep 2016) and CRS 2019 update separate these explicitly. Channel A uses only the $3.3bn FMF leg.

Primary sources:

• CRS RL33222 (Sharp, recurring) - U.S. Foreign Aid to Israel, canonical per-year FMF table
• CRS RL33222 (Apr 2014, Appendix B Table B-1) - canonical FY1999-FY2014 per-year
• CRS RL33222 (Nov 2020, Table A-1) - FY2015-FY2018 Greenbook reconciliation
• CRS RL33222 (2024 update) - FY2019-FY2024
• 1999 Clinton MOU joint statement (Clinton-Barak, 19 Jul 1999)
• 2007 Bush MOU (Burns-Abramovich, 16 Aug 2007)
• 2016 Obama MOU fact sheet (Obama White House, 14 Sep 2016) - disaggregates $33bn FMF + $5bn missile defence
• USAID Greenbook (Overseas Loans and Grants)

Confidence: HARD. Per-year FMF figures are appropriated dollar amounts traceable to public laws and reproduced verbatim in successive CRS RL33222 updates. Greenbook obligations reconcile to appropriated figures within $20m for most years.

Status: sourced 5 May 2026.

B. Missile-Defence Co-Development (Israel) - PARTIAL

Definition: All annual DoD missile-defence appropriations for U.S.-Israeli cooperative programs (Iron Dome, David's Sling, Arrow II/ASIP, Arrow III, Iron Beam) - DoD R&D and procurement lines outside the foreign-aid topline. The Obama MOU $0.5bn/yr missile-defence carve-out (FY2019 onward) is captured here as the FY2019-FY2024 base-row totals (each row = $500m, which IS the carve-out - the carve-out and the underlying appropriations are the same money).

Per-fiscal-year table (current $ millions) (sourced from CRS RL33222 May 2025 update Tables 4 and 5):

FY	Iron Dome	David's Sling	Arrow II + Arrow III	Iron Beam	Total Channel B
1999	-	-	46.9	-	46.9
2000	-	-	81.7	-	81.7
2001	-	-	95.2	-	95.2
2002	-	-	131.7	-	131.7
2003	-	-	135.7	-	135.7
2004	-	-	144.8	-	144.8
2005	-	-	155.3	-	155.3
2006	-	10.0	122.9	-	132.9
2007	-	20.4	117.5	-	137.9
2008	-	37.0	118.6	-	155.6
2009	-	72.9	104.3	-	177.2
2010	-	80.1	122.3	-	202.4
2011	205.0	84.7	125.4	-	415.1
2012	70.0ª	110.5	125.2	-	305.7
2013	194.0	137.5	115.5	-	447.0
2014	460.3ᵇ	149.7	119.1	-	729.1
2015	351.0	137.9	130.9	-	619.8
2016	55.0	286.5	146.1	-	487.6
2017	62.0	266.5	272.2	-	600.7
2018	92.0	221.5	392.3	-	705.8
2019	70.0	187.0	243.0	-	500.0
2020	95.0	191.0	214.0	-	500.0
2021	73.0	177.0	250.0	-	500.0
2022	1,108.0ᶜ	157.0	235.0	-	1,500.0
2023	80.0	167.0	253.0	-	500.0
2024 (base)	80.0	167.0	253.0	-	500.0
2024 (Supp PL 118-50)	3,000.0	1,000.0	-	1,200.0	5,200.0

Footnotes:

• a: FY2012 Iron Dome $70m was reprogrammed by the Obama Administration from other DoD accounts, not appropriated by Congress.
• b: FY2014 Iron Dome $460.3m includes the $225m P.L. 113-145 emergency supplemental, exempt from the March 2014 co-production requirement. Folded into Channel B per CRS Table 5; Channel F does not separately count P.L. 113-145 to avoid double-count.
• c: FY2022 Iron Dome $1,108m comprises $108m base appropriation plus $1bn supplemental aid for Iron Dome covering FY2022-FY2024 under P.L. 117-103 §8142. Folded into Channel B per CRS Table 5; Channel F does count P.L. 117-103 omnibus's Iron Dome top-up - see Cumulative Summary deduction.

Cumulative FY1999-FY2024 Channel B: ~$14.3bn (Iron Dome ~$6.08bn, David's Sling ~$3.83bn, Arrow II+III ~$4.21bn, Iron Beam $1.20bn).

Disaggregated cumulatives validate against rumination's anchors:

• Iron Dome cumulative US contribution between $3bn (CRS through 2022) and over $6bn after October 2023 - CRS May 2025 Table 5 = $6,075m ✓
• David's Sling has drawn over $3.8bn through 2024 - CRS Table 5 = $3,828m ✓
• Arrow US funding "just under half" of annual development - CRS confirms ✓

Methodology:

• In: All DoD missile-defence appropriations for cooperative programs FY1999-FY2024 (CRS RL33222 Tables 4 and 5). The $0.5bn/yr MOU carve-out FY2019-FY2024 is captured as the base-row totals (each ≈ $500m).
• Out: FMF (Channel A); Israeli national contribution; anti-tunnel cooperation (~$170m FY2016-FY2019); C-UAS / counter-drone cooperation; U.S. Army's $373m Iron Dome battery purchases for U.S. domestic use; U.S.-deployed THAAD batteries to Israel and U.S. Navy SM-3/SM-2 expended in defence of Israel during 2024 Iranian salvos (these are Channel E or operational expenditures).
• PL 118-50 gray zone: The FY2024 Supp row ($5.2bn = Iron Dome $3.0bn + David's Sling $1.0bn + Iron Beam $1.2bn) is folded into Channel B per CRS Table 5's authoritative classification. The same money appears in Channel F's PL 118-50 Israel-direct breakdown ($4.0bn Procurement + $1.35bn RDT&E ≈ $5.35bn). The Cumulative Summary deducts $5.2bn once to avoid double-counting. Per-channel sections show the canonical sourced figure unchanged.
• PL 117-103 gray zone: The $1bn FY2022 Iron Dome supplemental is folded into Channel B (CRS Table 5 places it there). Channel F also lists the same $1bn under P.L. 117-103. Cumulative Summary deducts $1bn once.
• PL 113-145 gray zone: $225m FY2014 Iron Dome emergency supplemental folded into Channel B per CRS Table 5. Channel F lists the same $225m. Cumulative Summary deducts $225m once.

Confidence: PARTIAL. HIGH on FY2006-FY2024 (CRS Table 5 canonical, MDA-sourced, repeated verbatim across multiple Sharp updates 2020 / 2022 / 2023 / 2025). MEDIUM on FY1999-FY2005 (Arrow only; CRS Table 4 cites MDA but does not disaggregate Arrow II vs Arrow III prep).

Error bar: ±5-10% on FY2006-FY2024 (canonical CRS table); ±15-20% on FY1999-FY2005 stub.

Primary sources:

• CRS RL33222 (Sharp, May 2025 update) - Tables 4 (Arrow FY1990-FY2025) and 5 (FY2006-FY2024+Supp+FY2025 disaggregation by program). Canonical.
• CRS RL33222 (Mar 2023, Nov 2020) - earlier baseline updates
• U.S. Missile Defense Agency budget justifications - source for CRS Table 4 Arrow figures
• P.L. 118-50 Israel Security Supplemental Appropriations Act (24 Apr 2024)
• P.L. 117-103 §8142 (FY2022 Consolidated Appropriations Act)
• P.L. 113-145 (FY2014 Emergency Supplemental Appropriations Resolution)

Status: sourced 5 May 2026.

C. WRSA-I Prepositioned Stockpile (Israel) - PARTIAL post-2023, ESTIMATE pre-2023

Definition: War Reserve Stockpile Allies-Israel. US-titled, DoD-owned defense articles physically prepositioned in Israel since the early 1980s. Israel maintains storage facilities and bears transportation costs; the US retains title until transfer. The stockpile is dispersed across multiple warehouses, has no public inventory, and operates under classified maintenance procedures. Channel C captures (a) US-paid additions to the stockpile under Section 514, (b) Section 506 Presidential Drawdown Authority (PDA) drawdowns from WRSA-I to Israeli use, with the FY2024 replenishment line cross-referenced to Channel F to avoid double-counting.

Architecture and regime distinction:

The pre-October 2023 regime was one of low utilisation. Section 514 of the Foreign Assistance Act capped annual additions to WRSA-I at $200m. Section 506 PDA carried a global $100m annual ceiling for non-emergency drawdowns. Documented drawdowns to Israel were episodic: precision-guided munitions during the 2006 Lebanon war; 120mm tank rounds and 40mm illumination rounds during the 2014 Gaza war; a January 2023 transfer of 300,000 155mm shells out of WRSA-I to Ukraine, with US-promised replenishment to the Israeli stockpile.

The post-October 2023 regime is structurally different. P.L. 118-50 Division A (Israel Security Supplemental, 24 April 2024) contained two WRSA-I provisions. Section 506(a)(1) was amended to substitute "$2,500,000,000" for "$100,000,000" - raising the annual Israel-specific PDA ceiling 25-fold for FY2024. Section 306 waived all FY2024 limits on the value of defense articles DoD may transfer into WRSA-I, removing the Section 514 $200m cap for the duration of the fiscal year. (The earlier rumination referenced this as "FY2024 NDAA Section 1262"; the enacted vehicle was Title III of the Israel Security Supplemental, not the NDAA - note this for reconciliation.)

Per-year table (millions of dollars; values are drawdown estimates, not ceilings):

FY	Sec 514 cap ($m)	PDA ceiling ($m)	Drawdowns to Israel ($m est)	Notes
1999-2005	200	100	~0 each	No documented drawdowns
2006	200	100	50-150	PGM drawdown during Lebanon war (value undisclosed)
2007-2013	200	100	~0 each	Stockpile valuation reaches $800m by 2010
2014	200	100	50-150	120mm tank + 40mm illumination drawdown (Gaza); USIPSA raises authorised stockpile ceiling to $1.8bn
2015-2022	200	100	~0 each
2023	200	100	200-500	Jan 2023: 300,000 155mm shells transferred to Ukraine (out of WRSA-I), replenishment promised. Post-Oct 7: no formal Section 506 PDA notifications, but stockpile actively drawn through accelerated FMS/below-threshold transfers
2024	waived	2,500	(memo-only - see Channel F)	$4.4bn O&M Defense-Wide replenishment in P.L. 118-50 already counted as Channel F line item; cross-reference to avoid double-count

Cumulative FY1999-FY2024 Channel C (excluding the FY2024 P.L. 118-50 line that lives in Channel F): ~$0.3-1.0bn (mid ~$0.6bn) nominal.

If the $4.4bn FY2024 replenishment is folded in (instead of left in Channel F), Channel C cumulative rises to ~$5.0bn but Channel F drops by $4.4bn - net Israel total is unchanged. The companion uses Channel F as the home for that line because P.L. 118-50 is the appropriation event; Channel C holds only the residual stockpile-as-channel flows separate from that supplemental.

Methodology decisions:

1. Valuation basis: replacement value (the dollar amount Congress appropriated to backfill drawn stocks) rather than equipment-as-valued at transfer. Parity with Channel I's PDA accounting choice for Ukraine.
2. Pre-2023 floor: near-zero default. The Section 514 cap is not used as a proxy ceiling - most years saw little or no addition activity because the stockpile was already capitalised toward its authorised ceiling.
3. Section 514 cap as proxy: rejected. The cap is an upper bound on additions, not a measure of actual flows. Treating the cap as "value transferred per year" would systematically overstate Channel C by an order of magnitude (~$5bn over the window with no factual basis).
4. Expansion vs drawdown: WRSA-I expansion authorisations (recent NDAAs allocating funds to add to the stockpile) are Pentagon procurement that lands in Israel-stored stocks. Channel C counts these only when subsequently drawn for Israeli use; otherwise they are US-titled inventory that would double-count against DoD baseline procurement.

Double-counting risks:

• Against Channel B: Iron Dome interceptor inventory pre-positioned in Israel is in scope for missile-defence co-development accounting, not Channel C. Channel C explicitly excludes interceptors funded under the $500m/yr Obama MOU missile-defence carve-out and any Iron Dome procurement in P.L. 118-50 Division A.
• Against Channel F: The $4.4bn O&M Defense-Wide replenishment line in P.L. 118-50 is the load-bearing FY2024 figure for the stockpile, but it is also the flagship Channel F line item. Channel C reports it as memo-only with a cross-reference; Channel F is the home of record. The chart UI flags any cell where this cross-reference is in play.
• Against Channel E: maintenance and security of WRSA-I stocks fall under EUCOM/CENTCOM force-protection lines pre-2023. Not separately costed in Channel C.

Confidence: PARTIAL post-FY2024 (the ceiling raise and replenishment appropriation are visible; actual drawdown values remain unnotified to Congress per Just Security's January 2024 finding). ESTIMATE pre-2023 (operating policy mostly classified; episodic drawdowns documented in news reporting but not aggregated).

Error bar: ±50% pre-2023 (the $0.3-1.0bn cumulative band reflects this); ±25% post-2023 once cross-reference to Channel F is settled.

Primary sources:

• Chappell, John Ramming, and Sarah Harrison. The "War Reserve Stockpile Allies-Israel" Explained & Why Congress Should Not Expand It. Just Security, 16 January 2024
• CRS RL33222 (Sharp, recurring) - stockpile valuation history ($100m initial → $800m 2010 → $1.8bn USIPSA 2014 → up to $4.4bn current, unadjusted)
• P.L. 118-50, Division A - Section 306 (Section 514 waiver); Section 506(a)(1) amendment ($100m → $2.5bn PDA ceiling); $4.4bn O&M Defense-Wide replenishment line
• Section 514 of the Foreign Assistance Act of 1961 (22 U.S.C. §2321h)
• Section 506 of the Foreign Assistance Act (22 U.S.C. §2318)
• United States-Israel Strategic Partnership Act of 2014 (P.L. 113-296)
• Marcellus Policy Analysis (John Quincy Adams Society), The Hidden Arsenal: Evaluating WRSA-I's Legal Framework and Oversight Mechanisms
• Responsible Statecraft (Carney), Who's minding the stockpile of US weapons going to Israel?
• The Intercept (Grim, 25 November 2023), Joe Biden Moves to Lift Nearly Every Restriction on Israel's Access to U.S. Weapons Stockpile

Status: sourced 5 May 2026.

D. Sub-Threshold Transfers (Israel) - ESTIMATE

Definition: Arms Export Control Act transfers to Israel falling below the Congressional notification thresholds set under 22 U.S.C. §2776(b) and §2776(c). Israel receives "NATO-equivalent" treatment under AECA, so the prior-notice thresholds are $25m for major defense equipment and $100m for defense articles and services (vs the standard non-NATO $14m / $50m), applied to both Foreign Military Sales (FMS) and Direct Commercial Sales (DCS). Transfers under these thresholds proceed without Congressional notification, do not appear in DSCA Major Arms Sale notices, and surface only in DoD's annual Section 655 report and DDTC's commercial-licensing report - both lagged, both aggregating without per-transfer detail.

Why this is an ESTIMATE: No standing publication aggregates sub-threshold transfers to Israel as a single line. The Washington Post's March 2024 reporting (John Hudson lead) established the count - "more than 100 separate foreign military sales" since 7 October 2023, of which "only two had been publicly notified" - by reconstructing a classified Congressional briefing. The Post did not aggregate a dollar value. Forum on the Arms Trade has identified specific cases retrospectively where sub-threshold originals were forced into disclosure by later expansions (M933A1 mortar $0.41m original; Heavy Duty Tank Trailers $40.1m; Tactical Vehicles $62.4m). These three documented originals span two orders of magnitude per transfer - confirming the right-skewed distribution any aggregate must assume.

Methodology:

1. Pre-October 2023 baseline (FY1999-FY2023): Israel's procurement profile is dominated by items the thresholds were written to catch - F-35 lots, F-15 packages, Apache buys, missile-defence interceptors, Tamir/Stunner co-production. Sub-threshold flow exists (consumables, spares, training, small munitions) but is structurally low. Default assumption: $50-150m/yr, midpoint $100m/yr. No source aggregates this for the period; the figure is a structural inference.

2. Post-October 2023 surge (Oct 2023 onward): WaPo's "100+ transfers in 5 months" is the only anchor. Per-transfer valuation must assume right-skewed distribution: most transfers small (consumables, spares, small munitions, training) with a tail approaching the $25m / $100m caps. Forum on the Arms Trade originals span $0.4m to $62m. Defensible mid-point: $15-20m per transfer (median ~$5m, mean pulled up by the tail).

• Low: 100 × $5m = $0.5bn (5-month) → $1.2bn annualized
• Mid: 100 × $20m = $2.0bn (5-month) → $4.8bn annualized
• High: 100 × $50m = $5.0bn (5-month) → $12.0bn annualized

3. Annualization: The pace did not slow. Forum on the Arms Trade documents continued sub-threshold cases through August-September 2024 surfacing only via expansion. State Department's April 2025 disclosure - 751 active FMS cases with Israel valued at $39.2bn - implies a deep tail of below-threshold cases that never trip notification. Conservative assumption: tempo holds at 80% of WaPo pace through end-CY2024.

4. Independent cross-check via DCS aggregate: DDTC's Section 655 report logs commercial sales authorisations separate from FMS. The Biden administration authorised $8.7bn in DCS to Israel across 2021-2024 (Forum on the Arms Trade). If 15-25% ran below threshold, that contributes ~$1.3-2.2bn over four years, or $300-550m/yr - independent of the FMS sub-threshold flow.

5. Double-counting check: Sub-threshold transfers are distinct from WRSA-I drawdowns (Channel C, separate statutory mechanism) and from emergency supplemental procurement (Channel F, Congressionally appropriated). Some FMS sub-threshold cases are paid for with FMF dollars from the MOU baseline (Channel A); the FMF appropriation appears in the foreign-aid topline, the resulting weapons transfer does not. No double-count with C or F. Partial overlap with A on the funding-source side, no overlap on the delivery-tracking side.

Per-year estimates (mid case, $m):

FY	Estimate	Note
1999-2010	75/yr	Pre-Iran-deal era; baseline structural floor
2011-2017	100/yr	Iron Dome era; sub-threshold component growth (Tamir consumables)
2018-2022	125/yr	OSP-era growth; F-35 sustainment ramp
FY2023	200	Includes pre-Oct 2023 partial year baseline
FY2024	3,500	Oct 2023 - Sep 2024. WaPo 5-month → annualized at 80%. Combined FMS sub-threshold (~$2.8bn) + DCS sub-threshold share (~$0.7bn)

Cumulative FY1999-FY2024 mid-point: ~$6.0bn, error band $2.5-15bn.

Calendar-year 2024 chart validation: chart uses $3bn. This sits inside the defensible band (low ~$1bn, mid ~$3.5bn, high ~$10bn). $3bn is conservative against the WaPo-anchored mid case but defensible: it credits only the FMS sub-threshold flow at the mid per-transfer assumption ($20m × ~150 transfers across 12 months) and excludes the DCS sub-threshold contribution. A reader who pushes back that $3bn is too high can be answered by the WaPo count alone.

Sensitivity (CY2024):

Scenario	Per-transfer mean	Transfers	Annual ($bn)
Low	$5m	150	0.75
Mid (chart)	$20m	150	3.0
High	$35m	200	7.0
WaPo-anchored extrapolation	$20m	240	4.8

What this channel does to the headline: Sub-threshold transfers do not change the order of magnitude of total Israel cost - the MOU baseline, supplementals, and the F-15IA package dominate. They matter for the bookkeeping argument: a delivery channel sized around $3bn in a single year, captured in no DSCA notification and aggregated in no public document until the WaPo investigation forced it into view, is the cleanest illustration of the illegibility the piece's structural claim depends on. The methodology paragraph IS the data; the number is the band.

Confidence: ESTIMATE. Error bar wide; honest range is $1-10bn for calendar 2024 with $3bn as a conservative mid.

Primary sources:

• Washington Post (Hudson lead, 6 March 2024) - U.S. floods arms into Israel despite mounting alarm over war's conduct - 100+ transfers, only 2 notified
• CRS RL31675 Arms Sales: Congressional Review Process - threshold tiers; Israel's $25m / $100m
• Forum on the Arms Trade, Under-Threshold Arms Sales - only public catalogue of identified sub-threshold cases (M933A1 mortar $0.41m; HDTT $40.1m; Tactical Vehicles $62.4m)
• Forum on the Arms Trade, Biden Administration Arms Sales and Transfers to Israel - $8.7bn DCS aggregate 2021-2024
• State Department, U.S. Security Cooperation with Israel - 751 active FMS cases / $39.2bn (April 2025)
• 22 U.S.C. §2776(b)-(c) - statutory threshold authority

Status: sourced 5 May 2026.

E. CENTCOM Operations Attribution (Israel) - PARTIAL post-Oct 2023, ESTIMATE pre-2023

Definition: US Central Command operations and maintenance spending attributable to Israel-defence work: standing posture in the AOR (carrier rotations, air-defence forward deployment, exercises supporting the Israel relationship) plus marginal-cost increments tied to specific Israel-related contingencies (Iranian missile shootdowns April and October 2024, THAAD deployment to Israel October 2024, anti-Houthi operations Oct 2023 onward, Eastern Mediterranean surge post-Oct 2023). Distinct from FMF (Channel A), missile-defence co-development (Channel B), supplementals (Channel F), and DoD munitions replenishment.

Methodology - two-tier estimate:

Post-October 2023 (active period). Brown Costs of War (Bilmes, Hartung, Semler, October 2025) is the anchor: $9.65-12.07bn for the 26-month window October 7, 2023 through end-2025. The number is marginal cost in Bilmes's tradition (incremental DoD operations and personnel costs above pre-conflict baseline, including munitions expended, deployment surcharges, and operations against the Houthis "sparked by or in support of Israeli military operations"). The Y1 (Oct 7, 2023 - Sep 30, 2024) operations component was $4.86bn (Brown Y1 first-year report, October 2024). Y2 (Oct 2024 - Sep 2025) and Q1 FY2026 = the residual ~$4.79-7.21bn.

Pre-October 2023 (baseline period). No published aggregate exists. Israel's transfer from EUCOM to CENTCOM AOR completed September 2021 formalised an attribution that had been operational for decades through Sixth Fleet, exercises (Juniper Cobra, Juniper Oak), bilateral air-defence integration, and Iran-deterrence posture. Project estimate at fractional standing-baseline: ~$0.2-1.5bn/yr, scaled by AOR contingency tempo. The rumination's own language ("attribute the increment, not the standing baseline") implies the pre-2023 floor could legitimately be zero in the strictest reading; the table provides a low column near zero for that reading and a mid/high carrying the steady-state attribution.

Calendar 2024 mapping. Brown's $9.65-12.07bn is a 26-month figure. Roughly 12 months land in CY2024 → ~46% → $4.4-5.6bn calendar-2024. April 2024 (300+ Iranian projectiles, first-ever SM-3 combat use, ~$50-200m interceptors alone) and October 2024 (180-200 Iranian ballistic missiles plus THAAD deployment) fall inside this window. FY2024 (Oct 2023 - Sep 2024) maps almost exactly onto Brown's Y1 → use $4.86bn as the FY2024 mid.

Per-fiscal-year table:

FY	Low ($bn)	Mid ($bn)	High ($bn)	Basis
1999-2005	0.0	0.2	0.5	Fractional baseline; pre-CENTCOM-AOR; Sixth Fleet posture
2006	0.1	0.3	0.7	Fractional baseline; 2006 Lebanon war Israel-defence posture
2007-2008	0.1	0.3	0.7	Iran nuclear deterrence ramp
2009	0.1	0.3	0.8	Cast Lead posture
2010-2011	0.1	0.4	0.8	Arab Spring AOR strain
2012	0.2	0.5	1.0	Pillar of Defense; Juniper Cobra
2013	0.2	0.5	1.0
2014	0.2	0.5	1.0	Protective Edge posture
2015-2017	0.2	0.5	1.0	JCPOA-period Iran-deterrence posture
2018	0.3	0.6	1.2	JCPOA exit; Iran-deterrence ramp
2019	0.3	0.7	1.3	Soleimani period buildup begins
2020	0.4	0.8	1.5	Post-Soleimani CENTCOM surge
2021	0.3	0.7	1.4	Israel-CENTCOM AOR transfer (Sep 2021)
2022	0.3	0.7	1.4	First full-year CENTCOM-AOR Israel
2023	0.5	1.0	1.8	Fractional baseline through Oct 6 + ~3 months active-period (Eisenhower CSG, Ford CSG extension, anti-Houthi posture)
2024	4.0	4.86	6.0	Brown range Y1 ($4.86bn) anchor; FY2024 ≈ Brown Oct 2023 - Sep 2024

Cumulative FY1999-FY2024: low $7.6bn / mid $15.0bn / high $27.6bn. Of which:

• Active period (FY2024 + ~3-month FY2023 tail): $4.5-7.8bn (mid $5.9bn)
• FY1999-FY2023 standing-baseline: $3.1-19.8bn (mid $9.1bn)

Houthi attribution: Bilmes treats the Houthi targeting as "sparked by" the Israel-Hamas war, so Operation Prosperity Guardian and the spring 2025 Operation Rough Rider campaign are on the Israel-defence ledger. The Houthi campaign lands in CY2025, outside the 1999-2024 cumulative.

THAAD deployment cost: NOT separately added - already inside Brown's range. The battery deployed October 13, 2024 lands in FY2025 (out of FY1999-FY2024 window by two weeks). Worth a footnote if the rumination ever runs through FY2025.

Confidence: PARTIAL post-October 2023 (Brown's $9.65-12.07bn is the published error bar; the table uses the lower Y1 figure $4.86bn for FY2024 conservatively). ESTIMATE pre-October 2023 (every cell is project judgment without published aggregate support).

Error bar: ±20% on the active-period anchor; ±50% on the pre-2023 fractional baseline.

Structural Attribution (Wider Reading)

The figures above are the marginal-cost reading: dollars the United States would not have spent at the same intensity, in the same locations, with the same forces, absent specifically Israel-related contingencies. That is the methodology Brown's Bilmes uses for the post-October-2023 anchor, and it is the methodology this companion has otherwise adopted throughout. It is also the methodology that systematically excludes the standing posture itself.

The wider reading attributes a fraction of the standing posture to the Israel relationship: the Fifth Fleet at Manama, Al Udeid in Qatar, Al Dhafra in the UAE, Camp Arifjan in Kuwait, the rotating CSGs in the Gulf and Eastern Mediterranean, the Iran-deterrence air-and-missile-defence architecture across the AOR - and a fraction of the post-9/11 wars that were, in part, prosecuted on Israel's strategic account. The argument that this fraction is non-zero is not fringe; it is articulated explicitly by Mearsheimer and Walt (2007), Bacevich (How We Became Israel, 2012; America's War for the Greater Middle East, 2016), the Quincy Institute paradigm paper (Pillar/Bacevich/Sheline/Parsi, 2020), and Crawford/Bilmes inside the Brown Costs of War project itself when the framing widens past the 26-month window. The argument that this fraction is zero - that CENTCOM would exist at its current size and tempo across FY1999-FY2024 in a counterfactual where Israel did not - is harder to defend than the fractional reading, but it is what the marginal-cost methodology silently assumes.

The companion reports both readings. The wider reading is canonical for the chart and the rumination; the marginal-cost reading is documented as the auditable floor. The chart UI flags any cell using the wider reading.

Component table (cumulative FY1999-FY2024 unless noted):

Component	Underlying total ($bn)	Share Low	Share Mid	Share High	Israel Low ($bn)	Israel Mid ($bn)	Israel High ($bn)
Gulf base posture (Bahrain, Qatar, UAE, Kuwait, Saudi, Oman) - ~$15bn/yr × 26 yr forward-presence O&M + MILCON	~390	10%	20%	35%	39	78	137
Iraq war (FY2003-FY2024 budgetary, Brown Costs of War, Crawford 2023 to-date)	~1,790	5%	15%	30%	90	269	537
Afghanistan war (FY2002-FY2021 budgetary, Brown 2021, includes Pakistan war zone)	~2,313	1%	3%	5%	23	69	116
Iran-deterrence standing posture increment (carrier rotations, Strait MCM, USAF presence over pre-1999 baseline) - ~$5bn/yr × 26 yr	~130	25%	40%	60%	33	52	78
Structural-attribution cumulative					~185	~468	~868

The Channel E marginal-cost figures shown in the per-fiscal-year table above (mid ~$15bn) are subsumed in the wider reading rather than additive: the FY2024 active-period $4.86bn anchor is an operational-tempo spike on top of the standing-posture share already attributed in row 1. Counting both would double-book the contingency tempo against the basing infrastructure that hosts it. The wider-reading cell replaces the marginal-cost cell; it does not stack on it.

Cumulative range under structural attribution: low ~$185-250bn / mid ~$470-575bn / high ~$870bn-1.2T.

Substituting this for the marginal-cost mid (~$15bn) in the cumulative summary moves Israel cumulative FY1999-FY2024 from ~$118.9bn to ~$575bn (mid), exceeding Ukraine's $198bn cumulative by roughly 3×. This is the magnitude the rumination's "permanent and structural drain" framing requires; the marginal-cost reading does not produce it.

The counterfactual logic the share columns operationalise - Iran-as-threat-state coherence, Iraq-war framing, Gulf basing waves all loading onto an Israel-coupled rationale - is argued in the rumination itself; this companion does not re-prove it. A reader who finds the counterfactual unconvincing should drop to the minimal-structural reading below.

Minimal-structural alternative (auditable floor for counterfactual-skeptics)

Component	Underlying total ($bn)	Minimal share	Israel-attributed ($bn)
Gulf base posture	~390	13%	51
Iraq war	~1,790	8%	143
Afghanistan war	~2,313	1.5%	35
Iran-deterrence increment	~130	28%	36
Minimal-structural cumulative			~265

Substituting the minimal cumulative for canonical Channel E produces Israel ~$365bn against Ukraine ~$215bn - Israel ×1.7. The rumination's structural-drain thesis survives even under hostile attribution.

Confidence on the structural subsection: ESTIMATE-aggressive. Every share column is argued from named scholarly literature (Mearsheimer-Walt for Iraq, Bacevich and Quincy for CENTCOM/basing, Gholz for oil-flow share), but no single source aggregates the four components into the wider Israel-attribution figure produced here. The cumulative-range band (low ~$185bn → high ~$870bn) is the honest expression of how much room the methodology choice covers.

Primary sources (marginal-cost reading + structural-attribution reading):

• Brown Costs of War (Bilmes), Costs of United States Military Activities in the Wider Middle East Since October 7, 2023 (October 2025) - $9.65-12.07bn marginal cost for Oct 2023 through end-2025
• Brown Costs of War (Hartung), U.S. Military Aid and Arms Transfers to Israel, October 2023 - September 2025 (October 2025)
• Brown Costs of War (Bilmes/Hartung/Semler), Y1 first-year report (October 2024) - $4.86bn FY2024 operations anchor
• Crawford, N., Blood and Treasure: United States Budgetary Costs and Human Costs of 20 Years of War in Iraq and Syria, 2003-2023, Brown Costs of War (March 2023) - Iraq war cumulative ~$1.79T to-date
• Brown Costs of War, Costs of the 20-Year War on Terror: $8 Trillion and 900,000 Deaths (September 2021) - Afghanistan ~$2.3T
• Mearsheimer, J. and Walt, S., The Israel Lobby and U.S. Foreign Policy (Farrar, Straus and Giroux, 2007) - Iraq war attribution thesis
• Bacevich, A., America's War for the Greater Middle East (Random House, 2016) - CENTCOM regional posture rationale
• Bacevich, A., How We Became Israel, The American Conservative (2012)
• Pillar, Bacevich, Sheline, Parsi, A New U.S. Paradigm for the Middle East, Quincy Institute Paper No. 2 (July 2020)
• Gholz, E., Nothing Much to Do, Quincy Institute (June 2021)
• Stiglitz, J. and Bilmes, L., The Three Trillion Dollar War (W.W. Norton, 2008) - full-cost methodology precedent
• Stieb, J., Why Did the United States Invade Iraq? The Debate at 20 Years, Texas National Security Review 6:3 (Summer 2023)
• CRS IF12645 The Terminal High Altitude Area Defense (THAAD) System
• GAO-14-440 - CENTCOM HQ cost growth FY2001-FY2013
• DoD Comptroller, OCO/Overseas Operations Cost appendices, FY1999-FY2024

Status: sourced 5 May 2026.

F. Emergency Supplementals (Israel) - HARD

Definition: Every Israel-specific emergency supplemental appropriation, FY1999-FY2024, with line-item breakdown. Excludes the standing MOU baseline (Channel A) and missile-defence co-development carried in regular DoD R&D (Channel B).

Per-supplemental table:

FY	Public Law	Enacted	Israel total ($bn)	Component breakdown
1999	P.L. 106-31	21 May 1999	0.90	FMF (FY1999 portion of Wye package)
2000	P.L. 106-113 (Div. B / Title VI)	29 Nov 1999	1.20	FMF (Wye remainder, FY2000-2001 spread)
2003	P.L. 108-11 (Emergency Wartime Supplemental)	16 Apr 2003	1.00	FMF (one-quarter cash, three-quarters US procurement). $9bn loan guarantees not counted - guarantee not grant
2014	P.L. 113-145 (Iron Dome Emergency Supplemental)	4 Aug 2014	0.225	Iron Dome procurement (transfer to GoI)
2022	P.L. 117-103 (Consolidated Appropriations Act)	15 Mar 2022	1.00	Iron Dome interceptor replenishment (omnibus vehicle after standalone H.R. 5323 stalled in Senate)
2024	P.L. 118-50 Div. A (Israel Security Supplemental)	24 Apr 2024	14.30	See breakdown below

P.L. 118-50 Division A line items (Israel-direct):

Account	$bn	Purpose
O&M, Defense-Wide	4.40	Replenish defense articles/services drawn from DoD stocks for Israel
Procurement, Defense-Wide	4.00	Iron Dome and David's Sling, transfer to GoI
RDT&E, Defense-Wide	1.35	Of which $1.20bn Iron Beam
Defense Production Act / industrial base	1.00	Artillery and critical munitions production capacity
Foreign Military Financing	3.50	FMF, fully usable inside Israel (procurement-in-Israel waiver)
State Dept embassy security / evac	0.05	Embassy security and emergency evacuation
Israel-direct subtotal	~14.30

Out-of-channel components in P.L. 118-50 Div. A package (counted toward the popular "$26.4bn" headline but not Channel F):

Component	$bn	Channel attribution
CENTCOM O&M (force protection / Houthi response / Iran deterrence)	2.40	Channel E (operations) - dual-use, not Israel-direct
Humanitarian assistance (IDA $5.66bn + MRA $3.49bn)	9.15	Not US-to-Israel; Gaza/West Bank/regional humanitarian
Nonprofit Security Grant Program (US synagogues/churches)	0.40	US domestic spending
Full package total	26.38

Cumulative:

• Israel-direct, FY1999-FY2024: $18.43bn (0.90 + 1.20 + 1.00 + 0.225 + 1.00 + 14.10 + ~0.005)
• Full-package (including PL 118-50 humanitarian + CENTCOM + nonprofit security): $30.71bn

77% of the Israel-direct cumulative landed in a single bill in April 2024. The supplemental architecture is bimodal: nothing for sixteen years, then most of it in one act.

Methodology choice:

Channel F counts the $14.3bn Israel-direct portion of P.L. 118-50, not the $26.4bn full package. Three reasons:

1. The $9.15bn in IDA and MRA is not US-to-Israel transfer - US-funded humanitarian relief flowing through UN agencies and NGOs, predominantly to Gaza and the West Bank. Israel actively opposed disbursing through UNRWA. Counting it as "aid to Israel" double-charges the line.
2. The $2.4bn CENTCOM O&M is dual-use and belongs in Channel E (operations attribution). Bilmes's $9.65-12.07bn marginal-cost figure already attempts the Israel-attributable share of CENTCOM; double-booking inflates the total.
3. The $0.4bn Nonprofit Security Grant Program is US homeland spending. Not Israel-direct.

The chart reads Israel-direct cumulative. The full-package number is documented as a footnote so a reader who prefers the AIPAC headline ("$26.4bn for Israel") can see where the difference goes.

Edge cases:

• FY2011 Iron Dome ($205m) and FY2012-2013 top-ups are not Channel F. They flowed through DoD authorisation/appropriation (FY2011 NDAA and successor DoD bills), not emergency supplementals. They belong in Channel B.
• P.L. 117-103 (FY2022 omnibus) is a hybrid case. It is not formally an emergency supplemental; it is the regular consolidated appropriations act. The $1bn Iron Dome top-up was added because the standalone Iron Dome Supplemental Appropriations Act (H.R. 5323) passed the House 420-9 in September 2021 but stalled in the Senate. Treatment as Channel F is defensible because the appropriation is identifiably emergency-style top-up appended to the omnibus rather than MOU baseline.
• P.L. 108-11 (2003) carried the $1bn FMF emergency grant plus $9bn in loan guarantees. Loan guarantees are contingent liabilities; the US incurs cost only on default. CRS treats them as a separate instrument. Channel F counts the $1bn grant only.
• Wye package vehicle split: the $1.2bn Wye FMF was enacted across two vehicles - $0.9bn in P.L. 106-31 (May 1999) and the residual in FY2000 plus FY2001 within P.L. 106-113 (Nov 1999). The request, justification, and accounting all framed it as Wye-implementation extra-MOU. Conservative alternative: treat only the $0.9bn portion as Channel F. Documented either way.
• Empty window 2003-2014 and 2014-2022 is the "MOU is doing the work" period. When Channel A absorbs routine top-ups, Channel F goes quiet. When existential war forces a step-change, Channel F reactivates.

Primary sources:

• P.L. 118-50 enrolled text
• CBO H.R. 8034 estimate
• CRS RL33222 (Sharp, recurring; May 2025 update)
• CRS IN12274 FY2024 National Security Supplemental: Defense
• CRS IN10158 Iron Dome assistance and coproduction
• CRFB What's in the Latest House National Security Supplementals (19 Apr 2024)
• AIPAC $14.3 Billion for Israel's Security one-pager
• P.L. 113-145 (Iron Dome 2014)
• P.L. 108-11 (Emergency Wartime Supplemental 2003)
• P.L. 106-31 (FY1999 supplemental, Wye Israel/Jordan/Palestinian portion)
• State Department Indyk testimony, 25 Mar 1999 (Wye package design)
• Stephens Semler aggregation (cross-check)

Confidence: HARD. All figures are appropriated dollar amounts traceable to public laws. The $14.3bn vs $26.4bn split is a methodology choice (documented) not a data dispute.

Status: sourced 5 May 2026.

G. Standing FMF (Ukraine) - HARD

Definition: State Department-administered Foreign Military Financing (FMF, 22 U.S.C. §2763) and International Military Education and Training (IMET) allocations to Ukraine, FY1999-FY2024. Excludes all emergency supplemental appropriations (those flow to Channel I) and USAI (DoD, supplemental-funded throughout). Channel G captures only the regular State-Foreign-Operations appropriations stream.

Methodology: Pre-2014 figures from State Department Country Assistance fact sheets (the 2001-2009 and 2009-2017 archives) and CRS RL33460. FY2003 ($3.0m FMF + $1.7m IMET) and FY2014 ($4.2m FMF + $1.9m IMET) are confirmed from primary sources. Intermediate-year FMF/IMET splits FY1999-FY2013 are inferred from CBJ patterns within $1-5m bands. FY2015-FY2021 figures from CRS IF12040 Table 2 (Welt et al., reproduced identically across versions). FY2022-FY2024 standing FMF for Ukraine is essentially zero - the regular State-Foreign-Operations bills did not break out a Ukraine FMF line; FMF flows in those years were supplemental (Channel I) and pooled with "countries impacted by the situation in Ukraine." The $1,317.6m FMF figure in CRS Table 2 for FY2022 is supplemental-appropriated and belongs to Channel I.

Per-year allocations (millions of dollars, nominal):

FY	FMF	IMET	Total	Notes
1999	<2	<2	~3	FSU programs context
2000	<2	<2	~3
2001	~2	~1.7	~3.7	Earmarked FSA recipient
2002	~2	~1.7	~3.7
2003	3.0	1.7	4.7	Confirmed line items
2004	~3	~1.7	~4.7	Per CBJ pattern
2005	~3	~1.7	~4.7
2006	~10	~1.8	~11.8	FMF rising as Ukraine pursued NATO MAP
2007	~10	~1.8	~11.8	"Other" category $16.51m
2008	~10	~1.9	~11.9	"Other" category $11.00m
2009	-	-	~5	Peace and Security $35.29m incl. NADR + others
2010	-	-	~5	Total foreign ops $117.93m
2011	-	-	~5	Cumulative FY1990-FY2011: $3.6bn all sources
2012	-	-	~5
2013	~5	~1.9	~6.9	Total FY2013 Ukraine $102.576m, mostly ESF/Health
2014	4.2	1.9	6.1	Pre-Crimea request; Crimea response via ERI/PDA/CTR
2015	47.0	~1.9	~48.9	First post-Crimea jump
2016	85.0	~2.4	~87.4
2017	99.0	~2.4	~101.4
2018	95.0	~2.5	~97.5
2019	115.0	~2.5	~117.5
2020	115.0	~2.0	~117.0
2021	115.0	3.0	118.0	Confirmed $115m FMF + $3m IMET
2022	0	0	0	Standing FMF zeroed; flows shift entirely to Channel I
2023	0	0	0
2024	0	0	0

Cumulative FY1999-FY2024: approximately $760m (nominal). Pre-2014 cumulative (~$80m), FY2014 ($6m), FY2015-FY2021 ($686m), FY2022-FY2024 ($0).

Shape: Three regimes. (1) FY1999-FY2013: ordinary post-Soviet IMET partner. Fifteen years of standing flows total roughly $80m. (2) FY2014-FY2021: Crimea triggers a step change. Standing FMF jumps from $4.2m (FY2014) to $47m (FY2015) to $115m by FY2019. Eight years account for ~$680m - more than 8x the prior fifteen years combined. (3) FY2022-FY2024: full-scale invasion. Standing FMF zeroed; the entire flow shifts to supplementals. The $115m/year FY2019-FY2021 ceiling was approximately what Israel received in three weeks of standing FMF.

Primary sources:

• CRS IF12040 U.S. Security Assistance to Ukraine (Welt et al.) - canonical Table 2 for FY2015-FY2024 FMF
• CRS RL33460 Ukraine: Current Issues and U.S. Policy (Woehrel, March 2014) - FY2014 request
• CRS R47275 SFOPS Supplemental Funding for Ukraine: In Brief
• State Department FY2003 Assistance to Ukraine fact sheet
• State Department Foreign Operations Appropriated Assistance: Ukraine (2001-2009 / 2009-2017 / 2017-2021 archives)
• State Department FY2024/FY2025 Congressional Budget Justification
• USAID Greenbook (cumulative obligations FY1990-FY2011)
• Stimson Center U.S. Security Assistance to Ukraine Breaks All Precedents (2022)
• ForeignAssistance.gov Ukraine country page

Confidence: HARD for FY2015-FY2024 (CRS IF12040 Table 2 is canonical). MEDIUM for FY2003 and FY2014 (specific fact-sheet line items recovered). SOFT for FY1999-FY2002 and FY2004-FY2013 (figures inferred from CBJ patterns within $1-5m bands; State Department archive pages render as corrupted base64 in current retrieval). Per-year softness does not change the cumulative shape - pre-2014 rounds to "trivial."

Status: sourced 5 May 2026.

H. Theater Operations Attribution (Ukraine) - ESTIMATE

Definition: The cost of EUCOM doing the work of supporting Ukraine and deterring Russian escalation, beyond what the standing peacetime baseline would have funded. ERI/EDI is the legible spine; the post-2022 forward-presence increment is the surge. Both are imperfectly attributable to Ukraine specifically rather than Eastern-flank deterrence broadly.

Methodology:

Marginal-cost framework. Channel H counts spending the United States would not have done at the same intensity, in the same locations, with the same forces, absent the Ukraine war. A standing EUCOM baseline that would have run anyway (ordinary stationing of US Army Europe and Africa, the Sixth Fleet's routine ops tempo, USAFE basing) is excluded. What is counted: the ERI/EDI line, attributed by Ukraine-share; the post-February 2022 troop increment from ~80,000 to ~105,000 forward-deployed personnel; ramp costs for Grafenwoehr and Wiesbaden training and intelligence operations not absorbed by USAI line items.

ERI/EDI Ukraine-share attribution. ERI/EDI was created in 2014 as a direct response to Russia's annexation of Crimea. The five-pillar architecture (Increased Presence, Exercises and Training, Enhanced Prepositioning, Improved Infrastructure, Building Partner Capacity) names Ukraine as a beneficiary alongside Baltic states, Poland, Romania. Default Ukraine-share: 30% (low 20%, high 40%) for FY2015-FY2021, rising to 50% (low 35%, high 65%) for FY2022-FY2024 once active war replaced post-Crimea posture as the binding constraint.

Post-2022 forward-presence increment. EUCOM grew from ~80,000 to ~105,000 troops post-February 2022. Marginal cost of ~25,000 additional rotational forces (additional ABCT in Poland, F-35 squadron rotations to RAF Lakenheath and Spangdahlem, naval surge to the Eastern Mediterranean and North Atlantic) ≈ $2.5-4bn/year, drawing on CSIS Jones (2025) and CRS Defense Primer cost-per-rotational-BCT-year of $1-1.5bn.

Pre-2014 floor: zero. No Ukraine-specific EUCOM tasking existed.

PDA double-counting risk against Channel I. Channel I counts PDA replenishment at appropriated value. The actual EUCOM work of moving stocks (TRANSCOM, USAFE airlift, Ramstein hub operations, Joint Multinational Training Group-Ukraine at Grafenwoehr where it ran outside USAI) sits in Channel H. Where SIGOAR or LIG reports book O&M expenditures inside the supplemental, those are excluded from Channel H to avoid double-counting against Channel I.

Divergence from Brown. Brown's Costs of War Ukraine line (Hartung) does not break out an EUCOM operations attribution comparable to Bilmes's Israel-side regional-operations figure. Channel H reconstructs the attribution Brown does not publish.

Per-year table FY2014-FY2024 (USD bn; pre-2014 = $0 each year):

FY	ERI/EDI ($bn)	Low	Mid	High	Basis
2014	0	0.0	0.0	0.0	Crimea annexation Mar 2014; ERI announced Jun 2014, no FY14 enacted
2015	0.985	0.20	0.30	0.39	20-40% of ERI
2016	0.789	0.16	0.24	0.32	20-40% of ERI
2017	3.4	0.68	1.02	1.36	20-40% of ERI; presence pillar dominant
2018	4.8	0.96	1.44	1.92	20-40% of EDI (renamed)
2019	6.5	1.30	1.95	2.60	20-40% of EDI; peak year
2020	6.0	1.20	1.80	2.40	20-40% of EDI
2021	4.5	0.90	1.35	1.80	20-40% of EDI
2022	3.81	3.5	5.0	7.0	35-65% of EDI ($1.3-2.5bn) + $2-4.5bn forward-presence increment Feb 2022 onward (10 months)
2023	4.27	4.0	5.5	7.5	35-65% of EDI ($1.5-2.8bn) + $2.5-4.7bn full-year increment
2024	3.63	3.5	4.5	6.5	35-65% of EDI ($1.3-2.4bn) + $2.2-4.1bn increment

Cumulative FY1999-FY2024: low $16.4bn / mid $23.1bn / high $31.8bn. Of which:

• FY2015-FY2021 (post-Crimea baseline): $5.4-10.8bn (mid $8.1bn)
• FY2022-FY2024 (active-war surge): $11.0-21.0bn (mid $15.0bn)

Calendar-year 2024 chart validation: chart uses $4.5bn - exactly the FY2024 mid case. Defensible.

Confidence: ESTIMATE. Three judgment calls (ERI/EDI Ukraine-share, increment cost-per-troop-year, PDA double-counting boundary) each carry ~25% uncertainty; combined error bar ±40% per year. The ERI/EDI line is itself appropriated value, knowable to the dollar. The forward-presence increment is the soft layer.

Wider EUCOM reading (sensitivity case)

The companion does not adopt a parallel structural expansion for Ukraine because EUCOM's load-bearing rationale is NATO Article 5 - treaty commitments to member states (Baltics, Poland, Romania, Germany) - not Ukraine, a non-NATO partner. Russia threatens NATO members directly; EUCOM would exist at substantial scale even in a counterfactual where Ukraine had joined NATO in 2008 or been absorbed in 2014. By contrast, Iran-deterrence is functionally indistinguishable from Israel-defence on the Walt-Mearsheimer / Bacevich / Quincy reading. The asymmetry is on the merits, not methodologically convenient.

A non-zero Ukraine-attribution of EUCOM standing posture is still defensible: ERI/EDI was created in 2014 because of Russia's actions in Ukraine, and the post-February 2022 forward-presence surge is unambiguously Ukraine-driven.

Component	Underlying total ($bn)	Ukraine share Low	Mid	High	Ukraine attributed Low	Mid	High
EUCOM standing posture pre-2014	~$220 (~$15bn/yr × 15 yr base presence)	0%	0%	0%	0	0	0
EUCOM standing posture FY2014-FY2021	~$90 (~$11bn/yr × 8 yr)	5%	10%	15%	4.5	9.0	13.5
EUCOM standing posture FY2022-FY2024	~$33 (~$11bn/yr × 3 yr)	15%	20%	25%	5.0	6.6	8.3
Wider-reading addition					9.5	15.6	21.8

Adding the wider reading to Channel H's marginal-cost $23bn brings Ukraine Channel H to ~$39bn (mid), and Ukraine total cumulative to ~$214bn. Even under this generous reading, Ukraine cumulative remains below Israel's structural $575bn by ~2.7×. The underlying totals are asymmetric: Iraq alone (Brown Crawford 2023, $1.79T to-date) is larger than the entire EUCOM standing baseline FY1999-FY2024 by an order of magnitude, and even a small Iraq-Israel-attribution share dominates any plausible EUCOM-Ukraine-attribution share.

Primary sources:

• DoD Comptroller, FY2024 / FY2022 / FY2019 / FY2018 / FY2017 / FY2016 EDI / ERI Justification Books
• CRS IF10946 The European Deterrence Initiative: A Budgetary Overview
• CRS IF11130 Defense Primer: U.S. European Command (EUCOM)
• CRS IN12107 Department of Defense Supplemental Funding for Ukraine: A Summary
• GAO-23-105619 European Deterrence Initiative: DOD Should Establish Performance Goals
• CSIS Jones, Deterring Russia: U.S. Military Posture in Europe (Jan 2025)
• CSIS Bergmann, Europe and America (Oct 2025)
• CSBA Shevin-Coetzee, The European Deterrence Initiative
• SIGOAR Quarterly Report (cumulative $182.99bn OAR/Ukraine response through Q3 2024)
• Brown Costs of War, Hartung Ukraine pieces

Status: sourced 5 May 2026.

I. Emergency Supplementals (Ukraine) - HARD

Definition: Ukraine-specific emergency supplemental appropriations enacted FY2022-FY2024. Five public laws between March 2022 and April 2024 carry the entire Ukraine-direct supplemental load; pre-2022 supplemental flows are zero or negligible.

Per-supplemental table (DoD figures from CRS IN12107 Department of Defense Supplemental Funding for Ukraine: A Summary, October 2024 update; SFOPS figures from CRS R47275 and R47054; aggregate from CRFB $175 Billion, 10 May 2024):

PL #	Common name	Enacted	Total Ukraine ($bn)	DoD ($bn)	SFOPS / non-DoD ($bn)	Notes
P.L. 117-103, Div. N	March 2022 USAA (FY2022)	15 Mar 2022	13.60	6.53	7.07	Embedded in Consolidated Appropriations Act, 2022. DoD: USAI $0.30bn, $3.5bn PDA replenishment authority, $3.9bn EUCOM-related O&M. SFOPS: ESF $4.0bn, MRA/refugee ~$1.4bn, FMF $0.65bn (~$322m for Ukraine itself).
P.L. 117-128	Additional Ukraine Supplemental (May 2022)	21 May 2022	40.15	20.10	20.05	Standalone Ukraine bill. DoD: USAI $6.00bn, PDA replenishment $9.05bn, EUCOM $3.9bn. SFOPS: ESF $8.77bn, MRA/refugee $0.9bn. PDA cap raised to $11bn for FY22.
P.L. 117-180, Div. B	Sept 2022 CR / FY23 USAA	30 Sep 2022	12.35	7.81	4.54	First FY23 supplemental, attached to CR. DoD: USAI $3.0bn, $4.71bn O&M Defense-Wide. SFOPS: ESF $4.5bn for direct budget support.
P.L. 117-328, Div. M	Dec 2022 omnibus / FY23 AUSAA	29 Dec 2022	47.37	27.87	19.50	DoD: USAI $11.88bn, PDA replenishment $11.88bn (largest single replenishment tranche). SFOPS $16.57bn (ESF $9.91bn for budget support). HHS refugee assistance $3.3bn.
P.L. 118-50, Div. B	April 2024 Ukraine Security Supplemental	24 Apr 2024	60.84	48.43	12.35	Standalone bill via H.R. 815 vehicle. DoD: USAI $13.4bn, PDA replenishment $13.4bn, EUCOM $7.3bn. SFOPS: ESF $7.86bn for budget support, MRA/ERMA $0.48bn.
Total			~174.31	110.74	63.51

Cross-check vs CRFB aggregate: per-supplemental act sum $174.24bn (CRS IN12107). CRFB headline "$175 billion." Difference ~$0.76bn, within ±$1bn target. CRFB rounds and folds in residual FY24 CR mini-provisions (humanitarian carry-overs, IDA topups). The per-act sum is the auditable floor; the CRFB number is the reasonable ceiling.

Cross-check on PL 118-50: CRFB and CRS R47868 both report $60.84bn as the Ukraine Security Supplemental Appropriations Act (Div. B of P.L. 118-50) total. CRS IN12274 gives $48.43bn DoD-only, which matches Div. B Title-by-Title. The $60.84bn figure includes $48.43bn DoD plus the SFOPS, Treasury, DFC, and other accounts in Div. B. Confirmed.

Methodology:

• In: every line item explicitly designated for Ukraine response inside one of the five enumerated public laws, regardless of agency. Includes USAI, PDA replenishment authority, ESF for direct budget support, MRA/ERMA refugee assistance flowing through US accounts, HHS refugee/entrant assistance for Ukrainian parolees, FMF for Ukraine, DoD-O&M for the EUCOM forward-presence increment funded inside the supplemental, RDT&E and procurement for Ukraine-related programs, classified/intel support, Defense Production Act funding directed at Ukraine-related industrial base.
• Out: standing Channel G FMF/IMET; pre-2022 minor supplementals (de minimis); EUCOM forward-presence costs that fall outside supplemental authorities (handled in Channel H as estimate); replacement equipment values exceeding appropriated replenishment.
• PDA accounting choice: Channel I uses replenishment value (the dollar amount Congress appropriated to DoD for back-filling drawn stocks), not the drawdown value (the equipment-as-valued at transfer). The two diverge: as of mid-2024, drawdown announcements totalled ~$24.9bn against ~$39.34bn in PDA replenishment authority. The replenishment number is the right one for Channel I - that is what Congress appropriated.
• "Ukraine-direct" boundary: the four pre-2024 acts and Div. B of P.L. 118-50 are titled and structured as Ukraine supplementals; we count the full act totals. Some FMF and ESF inside these acts is technically for "Ukraine and partner countries / NATO eastern flank" (CRS R47275 flags this in P.L. 117-103: $650m FMF total of which ~$322m for Ukraine itself), but the regional balance is overwhelmingly Ukraine-war-driven. The alternative ("Ukraine-only line items") would shave 5-8% off the total but produces a less honest picture of war-driven spending.
• Refugee figures: there is a CRFB internal discrepancy on PL 118-50 refugee assistance - the 19 April 2024 House-bill summary gave $1.4bn, the 10 May 2024 enacted-law summary gave $481m. The $481m is the enacted MRA line; we use the enacted figure.

Cumulative FY2022-FY2024: ~$174.3bn (per-act sum) to ~$175bn (CRFB rounded). Either is defensible; the chart should display $175bn with a footnote pointing to the $174.3bn auditable floor.

Primary sources:

• CRFB, Congressionally Approved Ukraine Aid Totals $175 Billion (10 May 2024)
• CRFB, What's in the Latest House National Security Supplementals? (19 Apr 2024)
• CRS IN12107, Department of Defense Supplemental Funding for Ukraine: A Summary (Oct 2024 update)
• CRS R47275, SFOPS Supplemental Funding for Ukraine: In Brief
• CRS IN12274, FY2024 National Security Supplemental Funding: Defense Appropriations
• CRS IF12040, U.S. Security Assistance to Ukraine (recurring)
• CRS IF12305, U.S. Direct Financial Support for Ukraine
• GAO-24-107232, Ukraine: Status and Use of Supplemental U.S. Funding, as of First Quarter, FY2024 (Jan 2024)
• Public Law texts: P.L. 117-103, P.L. 117-128, P.L. 117-180, P.L. 117-328, P.L. 118-50
• Special Inspector General for Operation Atlantic Resolve / Ukraine Oversight Interagency Working Group quarterly reports

Confidence: HARD. Error bar ±$1bn against the ~$175bn aggregate, almost entirely a function of which rounding/inclusion convention is used for FY24 CR residuals. Per-public-law totals are exact to the cent in primary law text.

Status: sourced 5 May 2026.

Per-year low/mid/high values for each channel, 1999-2024. Populated as channels A-I complete. The dataset shipped with the chart components imports from this section's tables once compiled.

[Pending compilation. The current chart values are LLM-estimated and clearly flagged in the UI as such until this section is filled.]

Total dollar share by confidence type, cumulative 1999-2024 (nominal).

The cumulative is reported under two methodologies (defined in Channel E): the marginal-cost reading (Channel E ~$15bn) is the auditable floor; the structural-attribution reading (Channel E ~$470-575bn) is canonical for the chart and rumination.

Israel

Channel	Marginal mid ($bn)	Structural mid ($bn)	Confidence	Notes
A. MOU FMF Baseline	70.93	70.93	HARD	CRS RL33222 per-year appropriated FMF
B. Missile-Defence Co-Development	14.32	14.32	PARTIAL	CRS RL33222 Tables 4 & 5
C. WRSA-I Prepositioned Stockpile	0.60	0.60	ESTIMATE pre-2023	FY2024 $4.4bn cross-references to F
D. Sub-Threshold Transfers	6.00	6.00	ESTIMATE	WaPo-anchored mid; band $2.5-15bn
E. CENTCOM / Standing-Posture Attribution	15.00	~470	ESTIMATE	Marginal: Brown Y1 anchor + fractional pre-2023 baseline. Structural: Gulf basing 20% + Iraq war 15% + Afghanistan 3% + Iran-deterrence 40%. Band low $185bn / high $870bn
F. Emergency Supplementals	18.43	18.43	HARD	Israel-direct cumulative; or $30.71bn full-package
Subtotal raw	125.28	~580
Less: B↔F double-count	(6.43)	(6.43)		PL 113-145 $0.225bn + PL 117-103 $1.0bn + PL 118-50 missile-defence $5.2bn
Israel net cumulative	~$119bn	~$575bn

Structural-reading band: low ~$285bn / mid ~$575bn / high ~$985bn. Dominated by Channel E methodology choice.

Marginal-reading band: low ~$107bn / mid ~$119bn / high ~$140bn.

Ukraine

Channel	Mid ($bn)	Confidence	Notes
G. Standing FMF	0.76	HARD	CRS IF12040 Table 2
H. Theater Operations Attribution	23.10	ESTIMATE	EDI Ukraine-share + forward-presence increment; band $16.4-31.8bn
I. Emergency Supplementals	174.31	HARD	CRS IN12107 per-act sum
Ukraine cumulative	~$198bn		No inter-channel double-counts

Ukraine range: low ~$191bn / mid ~$198bn / high ~$207bn.

There is no equivalent "structural reading" for Ukraine because EUCOM does not exist primarily because of Ukraine. The Ukraine number is what it is.

Comparison

Reading	Israel cumulative	Ukraine cumulative	Ratio
Marginal-cost (Brown narrow, counterfactual-rejecting)	~$119bn	~$198bn	Ukraine ×1.7
Minimal-structural (counterfactual-skeptic auditable floor)	~$369bn	~$215bn	Israel ×1.7
Structural attribution (canonical, counterfactual-accepting)	~$575bn	~$215bn	Israel ×2.7
Aggressive structural (Walt-Mearsheimer-aggressive shares)	~$985bn	~$215bn	Israel ×4.6

The mainstream "Ukraine costs more" framing is correct only under the marginal-cost reading. The rumination's structural-drain thesis survives the fork either way: Israel ×1.7 under minimal-structural, ×2.7 under canonical structural. Ukraine has no equivalent structural expansion (NATO Article 5, not Ukraine, is EUCOM's load-bearing rationale - see Channel H).

What the rumination's structural-shape argument depends on (independent of magnitude)

• Ukraine: 88% lives in supplementals (one instrument), 12% in operations attribution. Two channels carry 100%. The supplementals end and the architecture goes back to nothing.
• Israel: under the structural reading, 12% MOU baseline + 2% missile-defence + 1% stockpile + 1% sub-threshold + 82% standing-posture / regional-war attribution + 3% supplementals. Six channels across five Congressional committees plus the entire CENTCOM enterprise. None of them is going away.

The rumination's reference to the CRS RL33222 inflation-adjusted $317.9bn since 1951 is the cumulative-since-1951 frame and is orthogonal to the 1999-2024 reconstruction here. That figure remains the canonical pre-MOU-era anchor.

Channel research is ongoing. Each completed channel updates this page with sourced figures and dates the update. The chart components consume the data shape compiled here; the rumination prose stays stable unless a sourced figure invalidates a load-bearing claim.

• 5 May 2026: page created.
• 5 May 2026: HARD channels (A, F, G, I) sourced. Channel A taxonomy correction (Obama MOU $3.8bn split into $3.3bn FMF / $0.5bn missile-defence) applied to chart.
• 5 May 2026: PARTIAL/ESTIMATE channels (B, C, D, E, H) sourced under marginal-cost methodology.
• 5 May 2026: Channel E expanded with structural-attribution wider reading (canonical for chart and rumination; marginal reading retained as auditable floor).
• 5 May 2026: Channel H wider-EUCOM sensitivity case added.
• next: rebuild chart data from structural-reading figures; reconcile rumination prose; add visual confidence indicators to chart segments.