← All studies · askamerica.ai
Original story: https://www.fb.org/news-release/farmer-losses-projected-to-deepen

AFBF's "$34.6 billion" farm-loss figure is real, but it's from November 2025, not the July 16, 2026 report — and it measures row-crop margins, not total farm-sector losses

American Farm Bureau Federation, Market Intel / News Release, checked against USDA ERS Farm Income and Wealth Statistics (ag.ers_farm_income)

AFBF's $34.6B Farm-Loss Figure: Dated to Nov 2025, Not the July 16, 2026 Rep… USDA ERS Farm Income and Wealth Statistics (ag.ers_farm_income), US total, 2018-2026 (2025-26 are ERS forecasts) AFBF figure actually in the July 16, 2026 report $31B (2026) / $32B (2027) Nine principal row crops, returns over total cost, before support Where $34.6B actually comes from AFBF Nov 24, 2025 Market Intel 2025/26 crop-year estimate, since revised by the July 2026 report's $31B figure AFBF's own row-crop "returns over total cost" estimates by crop year -36 -34 -32 -30 -28 -26 -24 -22 -20 Crop year $ billions 2023/24 2024/25 2025/26 (Nov'25 est.) 2026 (Jul'26 est.) 2027 (Jul'26 proj.) Same AFBF metric, three publications: Nov 2025 estimated 2025/26 at -$34.6B; the Jul 16 2026 report re-estimated the same 2026 crop year… US whole farm-sector net income stayed positive throughout 0 50 100 150 200 Year $ billions 2018 2019 2020 2021 2022 2023 2024 2025 2026 ag.ers_farm_income, category=Net income, subcategory=Farm income. Row-crop-only losses coexist with a sector-wide surplus, driven by… AFBF's per-crop "returns over total cost" arithmetic (USDA ERS Commodity Costs and Returns) is not loaded in this corpus and could not be independently recomputed; filed as a sourcing gap (govdata-ops, type:sourcing/kind:gap/schema:agriculture). AskAmerica · askamerica.ai
SVG

Summary

Verdict: mostly false, as attributed. $34.6 billion is a real American Farm Bureau Federation (AFBF) figure — but it comes from AFBF's November 24, 2025 Market Intel piece, "Farmers Urgently Need Economic Assistance," where it is the projected national-average returns-over-total-cost shortfall for nine principal row crops (corn, soybeans, wheat, cotton, rice, barley, oats, peanuts, sorghum) in the 2025/26 crop year, before crop insurance indemnities or any government support. AFBF's actual July 16, 2026 report — "Farmer Losses Projected to Deepen" / "Persistent Losses Leave Farmers Needing Economic Support," fetched and read directly — headlines $31 billion for 2026 and $32 billion for 2027, not $34.6 billion. AFBF revised its own estimate downward between the two reports. Separately, this narrower row-crop metric should not be read as "combined farm losses" for the whole farm sector: USDA ERS Farm Income and Wealth Statistics (loaded in this corpus) show the total U.S. farm sector's net income was positive every year from 2023 through the 2026 forecast, helped by profitable livestock and a large jump in ad hoc government payments.

Claim-by-claim verdicts

🤥🤥🤥○ Fidelity: 3 of 4 Pinocchios

The news claim under test dates a real AFBF figure to the wrong report and treats it as AFBF's current position. $34.6 billion is a genuine AFBF number, but it appeared in AFBF's Nov 24, 2025 Market Intel as the projected 2025/26 crop-year loss for nine principal row crops — not in AFBF's actual July 16, 2026 publication, which instead headlines $31 billion (2026) and $32 billion (2027), a materially different, lower and more current figure for the overlapping crop year. AFBF itself revised the number down between the two reports, so presenting the stale Nov 2025 figure as what AFBF's July 2026 report says misrepresents AFBF's own most recent analysis, not just its wording.

🤥🤥○○ Claims Accuracy: 2 of 4 Pinocchios

AFBF's underlying methodology is real and traceable (USDA ERS Commodity Costs and Returns, WASDE, NASS acreage, FAPRI projections), but the label 'combined farm losses' overstates what the figure measures even setting the dating error aside. It is specifically national-average returns over total production cost for nine principal ROW CROPS ONLY, before crop insurance indemnities and any government support — not a measure of total farm-sector profit or loss. Independently checked against USDA ERS Farm Income and Wealth Statistics in this corpus: the whole U.S. farm sector's net farm income was positive in every year 2023-2026 ($128-163 billion), buoyed by livestock profitability and a swing to large ad hoc/disaster government payments ($9.8B in 2025, $28.4B in 2026) that AFBF's own reporting describes elsewhere. Row-crop margins genuinely have been squeezed and AFBF's per-crop arithmetic is not fabricated, but 'combined farm losses' is a narrower and more negative framing than the farm sector's actual aggregate financial position for the period.

1 mostly false · 1 assertions checked

#Assertion (verbatim)Verdict
1The American Farm Bureau Federation, in a July 16, 2026 publication, claims combined farm losses of $34.6 billion.mostly false

Verdicts: true, mostly true, partially true, mostly false, false, not checkable here (no table carries the measure), stale vintage (the article cites a release this corpus has not loaded — a freshness gap, not a falsehood).

Claim detail

#1 · mostly false — The American Farm Bureau Federation, in a July 16, 2026 publication, claims combined farm losses of $34.6 billion.
Article says
$34.6 billion, attributed to AFBF's July 16, 2026 report
Warehouse says
US net farm income (all commodities + govt transfers): $146.8B (2023), $128.3B (2024), $162.6B (2025 fcst), $158.4B (2026 fcst) — positive every year ag.ers_farm_income
Independent evidence
AFBF's actual July 16, 2026 report states $31B (2026) and $32B (2027) projected row-crop losses, not $34.6B. The $34.6B figure is AFBF's own Nov 24, 2025 estimate for the 2025/26 crop year, since revised by AFBF's own later report to a lower figure.
Overrode Jev's score because
score_claim's candidate_verdicts list ('accurate'/'misleading'/'false') does not map 1:1 onto publish_report's fixed enum, which has no 'misleading' option. score_claim's chosen value, 'misleading', sits between this report's 'partially true' and 'mostly false' on the same underlying finding (a real AFBF number, wrong report/date, and a metric narrower than its 'combined farm losses' framing) -- this is a schema/vocabulary mismatch between the two tools, not a substantive disagreement about the facts. 'Mostly false' is used here because the specific, falsifiable factual assertion under test -- that AFBF's July 16, 2026 publication itself states $34.6B -- is false (it states $31B/$32B); only the looser, unattributed 'AFBF has published a $34.6B estimate at some point' reading is true, which the report attributes in the fidelity Pinocchio explanation.
Show SQL
SELECT "year", category, subcategory, amount FROM ag.ers_farm_income WHERE state='US' AND category='Net income' AND subcategory='Farm income' AND "year">=2018 ORDER BY "year"

What AFBF's reports actually say

Nov 24, 2025 ("Farmers Urgently Need Economic Assistance," by John Newton): "combined annual returns below total costs (for nine principal crops) from the 2023/24 to 2025/26 crop years at -$20.2 billion, -$34.8 billion and -$34.6 billion, respectively, and before crop insurance indemnities and other support." This is where $34.6B originates — the 2025/26 crop-year estimate.

Jan 21, 2026 ("Significant Farm Losses Persist, Despite Federal Assistance"): restates the framework and puts accumulated three-crop-year losses "in excess of $50 billion," but does not restate $34.6B as a standalone figure.

Jul 16, 2026 ("Persistent Losses Leave Farmers Needing Economic Support" / news release "Farmer Losses Projected to Deepen," by Faith Parum and Daniel Munch) — the report the claim under test attributes $34.6B to: "AFBF estimates national average returns over total costs, without federal assistance, at a $32 billion loss across nine principal crops in 2027, deepening from a $31 billion loss in 2026 ... Combined, losses across the nine principal crops are projected to reach $41.4 billion in 2027." It separately cites "more than $7 billion" in 2025 specialty-crop losses (six representative crops) and says cumulative uncovered losses across the farm economy "exceed $12 billion." None of these figures is $34.6 billion.

Attribution across all three reports is consistent: the losses are attributed primarily to a structural squeeze between elevated production input costs (fertilizer, fuel, interest, labor — up 25-71% since 2020 per AFBF's own cost breakdown) and low/declining commodity prices, not primarily to tariffs. Trade/export losses (particularly to China) are cited as a compounding factor in the Nov 2025 piece, and the Jul 2026 piece separately notes the Iran conflict pressuring fertilizer/fuel markets — but tariffs are not the headline driver AFBF names for the $31-41B figures.

What this corpus can and cannot check

search_catalog and list_tables on the ag schema confirm there is no USDA ERS Commodity Costs and Returns table — the specific per-crop, per-acre cost-vs-revenue series AFBF's $34.6B/$31B/$32B figures are actually built from. ag.ers_farm_income (USDA ERS Farm Income and Wealth Statistics) is the closest table, but it is a whole-farm-SECTOR aggregate (cash receipts, production expenses, net income), not a per-crop cost-of-production breakdown, so AFBF's exact row-crop arithmetic cannot be independently reproduced here. This gap has been filed (type:sourcing, kind:gap, schema:agriculture, status:open).

What the corpus can check is the broader financial-health question the $34.6B figure is implicitly used to support ("farmers are in a loss position"). Querying ag.ers_farm_income for the US total, category='Net income', subcategory='Farm income' (USDA ERS's headline Net Farm Income series, which nets ALL commodities, all expenses, and all government payments) shows:

2023$146.8B
2024$128.3B
2025 (ERS forecast)$162.6B
2026 (ERS forecast)$158.4B
2026 (ERS forecast)$158.4B
2025 (ERS forecast)$162.6B
2024$128.3B
2023$146.8B
2025 (ERS forecast)$162.6B
2026 (ERS forecast)$158.4B
2023$146.8B
2024$128.3B

Net farm income was positive and in the $128-163 billion range every year 2023-2026 — the sector as a whole did not post a $34.6 billion (or any) net loss. The 'Government transactions' subcategory swung from -$4.6B (2023) to +$9.8B (2025) to +$28.4B (2026), consistent with AFBF's own description of the ECAP, Farmer Bridge Assistance, and ASCF disbursements ($23B+ combined per AFBF's July report) propping up the sector. Crop cash receipts alone (category='Crops') fell from $282.7B (2022) to $238.4B (2025) before an estimated rebound to $253.0B (2026) — directionally consistent with AFBF's story of pressured crop margins, even though this is revenue, not the cost-adjusted return AFBF's $31-41B figures measure.

Verdict

Mostly false, as attributed. The $34.6 billion number is a genuine AFBF estimate, not fabricated — but it belongs to AFBF's November 2025 analysis of the 2025/26 crop year, not the July 16, 2026 report the claim cites, and AFBF's own subsequent (July 2026) analysis revised the comparable 2026 figure down to $31 billion, a real and material discrepancy of over $3.5 billion against AFBF's own current position. Separately, "combined farm losses" overstates the scope of what AFBF's nine-crop, pre-support, returns-over-cost metric measures: the U.S. farm sector's actual net income, per USDA ERS data loaded in this corpus, remained positive throughout 2023-2026.

Every query behind this report

6 warehouse calls ran in this session, in order. Each is reproducible against the same snapshot.

query — 1207 ms
SELECT "year", category, subcategory, amount, unit_desc FROM ag.ers_farm_income WHERE state = 'US' AND category ILIKE '%net farm income%' AND "year" >= 2018 ORDER BY "year"
query — 100 rows — 1393 ms
SELECT DISTINCT category FROM ag.ers_farm_income WHERE state = 'US' ORDER BY category LIMIT 100
query — 6 rows — 1222 ms
SELECT DISTINCT category FROM ag.ers_farm_income WHERE state = 'US' AND category ILIKE '%income%'
query — 45 rows — 1378 ms
SELECT "year", category, subcategory, amount, unit_desc FROM ag.ers_farm_income WHERE state = 'US' AND category = 'Net income' AND "year" >= 2018 ORDER BY "year"
query — 18 rows — 917 ms
SELECT "year", category, subcategory, amount FROM ag.ers_farm_income WHERE state='US' AND category='Crops' AND "year">=2018 ORDER BY "year"
query — 25 rows — 1239 ms
SELECT DISTINCT category, subcategory FROM ag.ers_farm_income WHERE state='US' AND (category ILIKE '%expense%' OR category ILIKE '%production expense%')