A 90% revision on beef exports. A million jobs that never were. A hunger report shut down. When the same White House keeps producing the same “errors,” it stops being a pattern of incompetence and starts looking like something worse — a leadership that cannot govern the country whose economy it is supposed to measure.
On July 2, the United States Department of Agriculture published a weekly report claiming that American exporters had sold a mind-bending 126,062 metric tons of beef to foreign buyers in a single week — nearly 500% more than the week before. The number was so absurd that grain traders, brokers, and analysts began questioning it within hours. Italy and Chile, the report claimed, had suddenly become massive buyers of American beef, ordering tens of thousands of metric tons out of nowhere. It did not require a career in commodities to see that this was not a market surge. This was a mistake.
What happened next tells you almost everything you need to know about the state of federal data reporting under Donald Trump’s second administration. When contacted by Reuters after the release, a USDA export sales reporting specialist named Amy Harding told the outlet the agency had “confirmed numbers with an exporting company”. The agency, in other words, publicly defended the numbers before quietly disowning them.
Then, on July 9, USDA revised the figure down by roughly 90 percent, to 12,064 metric tons. Chile’s phantom purchase of 38,434 metric tons became 367. Italy’s 32,274 became 350. Fourteen other countries were also quietly re-tabulated. The agency called it a reporting error and offered no further explanation. Commodity analyst Austin Schroeder of Brugler Marketing & Management gave the assessment that has come to define this administration’s relationship with its own data: “Should the USDA have caught it? Probably. They may have just overlooked it.”
They may have just overlooked it. That is not a sentence any American should have to read about the world’s most closely watched agricultural statistics agency.
I. A Pattern, Not An Accident
The beef debacle would be embarrassing on its own. What makes it a scandal is that it is the third significant collapse of USDA data credibility in a year. And that is only within one agency.
Last autumn, USDA projected that American farmers would harvest 86.8 million acres of corn. By January, after successive upward revisions, the agency concluded the real number was 91.3 million acres — an undercount larger than the entire state of Delaware. Seth Meyer, who served as USDA’s chief economist until his departure in December, did not soften the assessment: “It was a miss. No other way to call it.” A five percent error may sound tolerable to a layperson. To grain traders, food manufacturers, and every farmer who makes planting and pricing decisions from these reports, it was the worst projection in recent memory.
The same agency has also delayed a quarterly agricultural trade report and, according to analysts, removed explanatory text linking Trump’s tariffs to a projected widening of the agricultural trade deficit — a scrub that critics said raised direct questions about the agency’s objectivity. And in September, USDA announced it would terminate the annual Household Food Security Report — the nation’s single longest-running measurement of Americans’ ability to feed their families — calling three decades of nonpartisan hunger data “redundant, costly, politicized, and extraneous.”
Two and a half months earlier, the president had signed legislation that the Congressional Budget Office estimated would push roughly three million Americans off SNAP. It does not require cynicism to notice the timing. It requires only literacy.
“The data is rife with inaccuracies slanted to create a narrative that is not representative of what is actually happening in the countryside.”
— USDA press release announcing the end of America’s 30-year hunger measurement
Read that USDA statement twice. It is a federal agency accusing its own long-running dataset — one collected through the Census Bureau’s Current Population Survey — of being “slanted.” The claim is not that the numbers are wrong. The claim is that the numbers are inconvenient. And when the numbers are inconvenient, this administration’s answer is not to fix them. It is to end them.
II. The 911,000 Jobs That Weren’t
The USDA is not the only agency whose reporting has cratered. In September, the Bureau of Labor Statistics released the preliminary benchmark revision for its nonfarm payroll data covering the twelve months ending March 2025. The revision was the largest markdown on record: 911,000 fewer jobs than the country had originally been told existed. In February, BLS finalized similar reductions and reported that the U.S. economy had actually added only 181,000 jobs in the entirety of 2025 — an average of roughly 15,000 a month. The country was told for months it was living through a job boom. It was not.
The economic damage is measurable. Prior to the revamp, the pre-revision figures suggested average job growth of 75,000 a month. After the revisions, that figure collapsed to 44,000. The Federal Reserve, which had been calibrating interest rate decisions against inflated employment data, has been forced into a corner. Markets, mortgage lenders, homebuilders, and small businesses that depend on accurate signals have been operating on numbers that were, quietly, fiction.
The White House response was not to explain how or why the country had been given inaccurate figures. It was to attack the messenger. Trump fired the BLS commissioner, Erika McEntarfer, after an earlier weak jobs report, accusing her without evidence of manipulating data. Press Secretary Karoline Leavitt declared: “The BLS is broken. This is exactly why we need new leadership to restore trust and confidence in the BLS’s data.” Labor Secretary Lori Chavez-DeRemer, in an official statement, said: “There is no room for such a significant and consistent amount of error.”
She was correct about that. She was not correct about the cause. The BLS revision process — which reconciles monthly survey samples against near-universal state unemployment insurance tax records through the Quarterly Census of Employment and Wages — has been a routine feature of American economic statistics since 1949. It is not evidence of political manipulation. It is the correction mechanism. What is new is not the revision. What is new is a White House that publicly demolishes the integrity of the agencies producing the numbers, then blames those agencies for the fact that the numbers were flattering to the White House’s political story.
USDA cut its late-June beef export figures from 126,062 metric tons to 12,064 after admitting the initial report was “reported in error” — with no further explanation.
The USDA’s initial projection missed real harvested corn acreage by more than the size of Delaware — its worst projection in recent memory, per its own former chief economist.
The largest BLS benchmark revision on record erased nearly a million jobs from the official ledger — and Trump responded by firing the agency’s commissioner.
Right after Trump signed law cutting SNAP for an estimated 3 million people, USDA killed the 30-year survey that would have measured the resulting hunger.
III. Housing Numbers That Cannot Be Trusted, Either
In housing, the administration’s problem is not a single statistical scandal. It is a running credibility gap between what the White House claims about the market and what the underlying data actually show. In July, Trump told reporters aboard Air Force One that his predecessor had driven up rental costs “by 40 and 50%” by “housing illegal aliens in all of those empty units.” He cited a Federal Reserve working paper to support the claim. PolitiFact went to the paper: it attributed a 4.3 percent increase in rents to unauthorized immigration — not forty, not fifty. The president was off by roughly an order of magnitude, and there was no evidence for the empty-units claim at all.
Meanwhile, the actual Census Bureau data released in June show May 2026 housing starts falling 15.4 percent from April — the lowest level since May 2020. Multi-family starts collapsed by 41.6 percent. The White House’s April Economic Report of the President had proclaimed the beginning of a housing renaissance. The building permit data, the housing start data, and the completions data all point the other way. And when Congress attempted to pass the bipartisan 21st Century ROAD to Housing Act — a cross-party measure to boost supply — the president reportedly halted it in an attempt to force Republicans to pass a separate voter-restriction measure. His housing approval numbers, according to Newsweek’s reporting on Focaldata polling, cratered from -30 to -32 net after the maneuver.
What connects the housing story to the USDA and BLS story is not statistical methodology. It is a governing philosophy in which economic reality is treated as a public relations problem, and the appropriate response to bad numbers is to attack, disband, or replace the people producing them.
IV. A Chronology of Numbers That Broke
Get Involved Today
Contribute to our mission and turn your concerns into action.
V. What Bad Numbers Do to a Real Economy
It is tempting to treat all of this as bureaucratic embarrassment — spreadsheet errors, staffing shortages, a bad week for the press office. It is not. Federal statistics are load-bearing infrastructure. Grain traders in Chicago and Kansas City price contracts off USDA reports. The Federal Reserve calibrates monetary policy off BLS payrolls, CPI, and PPI. Mortgage lenders, homebuilders, insurers, foreign governments, and every business making a hiring or capital decision runs on this data. When the numbers are wrong — and when the country cannot tell whether the numbers were wrong by accident or by design — the entire pricing mechanism of a $28 trillion economy begins to fray at the edges.
At least a dozen federal datasets have been axed or altered under the Trump administration, spanning climate, food security, and health outcomes. The Center on Budget and Policy Priorities has documented in detail the systematic elimination of advisory panels and data streams — the 2030 Census Advisory Committee, the Bureau of the Census Scientific Advisory Committee, the Technical Advisory Committee at BLS, all disbanded. Roughly 20,300 employees left the Department of Agriculture in the first five months of Trump’s second term, according to the agency’s own inspector general. The Foreign Agricultural Service, the unit whose beef export figure just imploded, lost about 21 percent of its workforce.
You cannot cut a fifth of the people who count things and expect the counts to stay accurate. That is not a partisan claim. It is arithmetic.
What America has been handed, over the past year, is not a run of bad luck. It is what happens when a government treats measurement itself as a threat — when the administration in power responds to unflattering statistics by attacking the statistician instead of the problem. That is a governance failure that is orthogonal to ordinary partisanship. It is the failure that makes a modern economy impossible to steer.
“Unable to discharge the powers and duties of his office.”
The 25th Amendment, drafted principally by Senator Birch Bayh of Indiana and ratified in 1967, empowers the Vice President and a majority of the Cabinet to declare a president “unable to discharge the powers and duties of his office.” The amendment never defines “unable” or “inability” — and the drafters did that on purpose. As the Congressional Research Service records, Bayh understood the terms to refer to an “impairment of the President’s faculties” affecting his capacity “to make or communicate his decisions.” Rep. Richard Poff, a Republican co-architect, was even more explicit: Section 4 applies not only when a president is unconscious but also when, “by reason of mental debility,” the president is “unable or unwilling to make any rational decision, including particularly the decision to stand aside.”
The drafters chose deliberately open-ended language because they could not, in 1965, foresee every form incapacity might take in a nuclear age. They left the definition to the judgment of a future Vice President and Cabinet acting in good faith.
The Case Being Made
Since January 2026, when Sen. Ed Markey of Massachusetts first publicly called for invoking the amendment, more than seventy lawmakers — including Sens. Chris Murphy of Connecticut and Ron Wyden of Oregon, and Reps. Ro Khanna, Alexandria Ocasio-Cortez, Yassamin Ansari, Rashida Tlaib, Maxwell Frost, and Sydney Kamlager-Dove — have argued that the 25th Amendment applies to Trump’s second term. Rep. Raja Krishnamoorthi told his constituents in April that Trump’s conduct reflects “a dangerous pattern of reckless escalation, erratic decision-making, and general conduct that raises grave questions about his fitness to discharge the duties of the presidency.” Rep. Mike Quigley of Illinois said plainly that “every day, Donald Trump commits acts unfit for the office of the President.”
The data crisis is one such act, quietly compounding. A presidency that fires statistical agency chiefs for producing accurate numbers, that terminates hunger measurements days after cutting food aid, that publicly defends impossible export figures and then walks them back without explanation, is a presidency that has lost the capacity to govern the country’s economy honestly. The stewardship of the nation’s official statistics is not a peripheral duty. It is central to the “powers and duties of the office” — because without it, no other economic power of the presidency can be exercised responsibly.
The Practical Barriers
Invocation requires Vice President JD Vance and a majority of the Cabinet. As PBS NewsHour has reported, there are no public indications that either the Vice President or the Cabinet is prepared to act. Removal must then be sustained by two-thirds of both chambers of Congress. Given the current Republican majorities, this is politically improbable. That is honest. It is also beside the constitutional point.
Why the Barriers Do Not Negate the Case
The 25th Amendment exists precisely because its drafters understood that political inconvenience is not the standard for its invocation. The standard is capacity to govern. Bayh insisted the amendment be broad enough to reach a president whose faculties are impaired in ways the drafters could not enumerate. A pattern of reality-denying behavior across the government’s economic reporting apparatus — one so pronounced that traders, farmers, economists, and even the White House’s own Labor Secretary now speak openly of eroded trust — is precisely the kind of impairment the Amendment was written in open-ended language to reach. The question is not whether Cabinet officials will act. The question is whether the constitutional standard has been met. On the evidence, the answer is not obvious the other way.
VI. What This Reveals About Priorities
Every administration prefers flattering numbers. What separates a functioning presidency from a failing one is what it does when the numbers refuse to flatter. A functioning presidency invests in the measurement, staffs the agencies, listens to the professionals, adjusts the policy. A failing presidency fires the messengers, dissolves the advisory panels, cancels the surveys, cuts the workforce that produces the data, and then insists that the resulting collapse of quality proves the numbers were rigged all along.
The Trump administration has, quite consistently, chosen the second path. The pattern is not confined to jobs, or crops, or hunger. It is showing up in the missing CPI data for October 2025, in the termination of food security measurement, in the disbanding of the census advisory committees, in the vanishing quarterly agricultural trade report. This is a leadership that treats accountability as an adversary, and evidence as an inconvenience. That is not sustainable in an economy that depends on trust in what the federal government tells the world.
American markets do not run on optimism. They run on numbers. Take away the numbers — or worse, corrupt them — and everything downstream begins to fail. Bond markets. Wheat futures. Municipal budgets. Corporate hiring. The Federal Reserve’s interest-rate signals. Every farmer trying to decide what to plant and every family trying to decide whether to buy a house. All of it depends on a government that can, at minimum, be trusted to tell the country what it sees.
What we are watching, in real time, is a government that can no longer be trusted to do that. And the reason it can no longer be trusted is not a technical problem. It is a leadership problem.
Editorial Conclusion
A president who cannot tolerate accurate measurement cannot govern a modern economy. A White House that responds to bad data by firing statisticians, dissolving expert panels, and canceling the surveys that produce inconvenient truths has not merely mishandled its accountability duty — it has abandoned it.
The 25th Amendment does not require unanimous political will. It requires only that the country be honest about what it sees. What it sees, right now, is a leadership incapable of the basic constitutional task of stewarding the public’s economic instruments. That is not a partisan judgment. It is a factual one, written in the ledgers of a Department of Agriculture that has publicly defended and then disowned its own numbers within a week.
Democracy does not survive on flattering data. It survives on true data — and on leaders who can bear to hear it. This one cannot. The country deserves better, and the Constitution offers a path.
Sources & References
- Mediaite — “Trump Administration Walks Back Key Export Figure With Stunning 90% Revision.” mediaite.com
- GV Wire / NYT — “Crop Undercount Raises Questions About Reliability of USDA Data.” gvwire.com
- GV Wire / Reuters — “USDA Lowers Beef Export Sales by 90% Amid Growing Doubts Over Data.” gvwire.com
- Farm Policy News (U. Illinois) — “USDA Crop Data Reliability Remains Under Scrutiny.” farmpolicynews.illinois.edu
- The Hill — “Trump administration halts government hunger report.” thehill.com
- NBC News / AP — “After cuts to food stamps, Trump administration ends government’s annual report on hunger in America.” nbcnews.com
- Center on Budget and Policy Priorities — “Federal Data Are Disappearing as Statistical Agencies Face Budget Cuts and Political Pressure.” cbpp.org
- WUSF Public Media — “The Trump administration’s ‘war on measurement’ is scrubbing consequential data for Floridians.” wusf.org
- Detroit News / Reuters — “Trump cut to food security survey could make measuring US hunger harder.” detroitnews.com
- CBS News — “The U.S. labor market added 911,000 fewer jobs than earlier reported, BLS says.” cbsnews.com
- NPR — “BLS revision shows annual hiring was overstated by 911,000 jobs.” npr.org
- NBC News — “U.S. gained 130,000 jobs last month, but labor market growth stalled in 2025.” nbcnews.com
- U.S. Department of Labor — Secretary Chavez-DeRemer statement on BLS preliminary benchmark revision. dol.gov
- Economic Policy Institute — “Today’s BLS preliminary benchmark revisions are necessary for timely and accurate data—not fodder for Trump’s attacks.” epi.org
- PolitiFact — “Trump exaggerates about how much illegal immigration raised rental housing prices.” politifact.com
- U.S. Census Bureau / HUD — New Residential Construction Press Release, June 2026. census.gov
- Newsweek — “How Trump Approval Rating Has Changed Since Blocking Housing Bill.” newsweek.com
- NBC News — “Dozens of Democrats call for Trump’s removal after his Iran threats.” nbcnews.com
- Newsweek — “Lawmakers Demand 25th Amendment Be Invoked Against Donald Trump: Full List.” newsweek.com
- Congress.gov / CRS — “Presidential Disability Under the Twenty-Fifth Amendment.” congress.gov
- PBS NewsHour — “Could the 25th Amendment be invoked against Trump? Here’s how it works.” pbs.org
- Office of Rep. Raja Krishnamoorthi — “Krishnamoorthi Calls for President Trump’s Removal Under 25th Amendment.” krishnamoorthi.house.gov
- Office of Rep. Mike Quigley — “Quigley Calls for Trump’s Removal Under 25th Amendment.” quigley.house.gov
- U.S. Bureau of Labor Statistics — “2025 Federal Government Shutdown Impact on Consumer Price Index.” bls.gov
Related News
