Science1 distinct publisher3 min readUpdated
USDA's revived trapping program cut corn damage claims from 70 acres to 10 per policy, but all the money committed since 2018 is worth less than a month of the damage hogs do.
The Scientist · Science desk

Compiled by The ScientistSomething wrong?How this is made
The federal feral swine program that began as a $75 million pilot in the 2018 Farm Bill has become a standing line item: Congress added $105 million to run it through 2029 in the budget and immigration package it passed in July 2025, and the Farm Bill now awaiting Senate debate would add $150 million more and extend the effort to 2031 [12][15][18]. A new evaluation by agricultural economists at the University of Tennessee and the University of Arkansas, writing in Scientific American, found the pilot cut damage claims sharply for one crop and did nothing measurable for four others [19][22][24]. That moves the decision from whether coordinated trapping works to how much containment is worth buying.
Start with the biology, because it sets the ceiling. USDA counts feral swine in 35 states and last estimated the national population at 7 million, in 2016 [1][2]. Hogs breed year round, producing up to two litters a year of four to 12 piglets each, and populations have been estimated to double in as little as four months [3][4]. A four-month doubling time implies up to eightfold annual growth where nothing checks it [8]. Against that, USDA puts annual property damage at $2.5 billion to $3.4 billion, of which at least $800 million, roughly a quarter to a third, is destroyed crops [5][6][7]. The remainder is disease transmission to livestock, wrecked fences and farm roads, damaged recreational parks, and degraded wildlife habitat, water quality and plant ecosystems [9].
The program's design targets an economics problem rather than a wildlife one. Hogs range across large tracts of private land, so if neighbors do not act together the animals relocate instead of being killed or contained; the authors' argument is that each landowner has an incentive to wait for someone else to pay, so nobody does [10][11]. Since 2020, USDA has funded landowners' purchases of trapping equipment, on-farm trapping and restoration of damaged ground in selected counties across ten states, from Texas and Oklahoma east to Florida and North Carolina [13][14].
The measurement used federal crop insurance claims to compare program and non-program counties, before and after the trial began [20]. Among counties that reported wildlife crop damage, non-program counties averaged 70 acres of corn damage per policy; program counties averaged 10, a decline of about 86 percent that the authors report as statistically significant [21][22][23]. Soybeans, wheat, cotton and peanuts showed no difference [24]. Corn is reportedly the crop most commonly damaged by feral swine, which the authors offer as a partial explanation for the single-crop result [25]. They also note the pilot launched into COVID-19 restrictions that limited the community meetings and outreach participation depends on [26].
Then the arithmetic that eradication talk does not survive. Federal money committed to the program since 2018 totals $180 million [27]. At the low end of USDA's damage range, that is less than a single month of losses [28]. If the House figure survives the Senate, the total reaches $330 million spread across 2018 to 2031, under 1 percent of cumulative damage over those years at the low-end estimate [29][30]. That buys suppression in chosen counties. It does not buy removal of an animal with an eightfold growth ceiling from 35 states.
Three things to watch. Whether the $35 million first tranche, one third of the 2025 allocation, is fully subscribed by the September 21, 2026 application deadline [16][17]. Whether the Senate keeps the $150 million [18]. And whether USDA publishes a population estimate newer than 2016, which will be a decade old by that deadline; without one, national containment cannot be judged on anything but insurance claims for corn [2][32].
Follow any of these and your For You feed starts watching them — no settings page required.
Ranked by verification strength, evidence, and original report placement.
Feral swine roam in 35 states, according to the U.S. Department of Agriculture.
The most recent population figures, from 2016, estimated that 7 million feral pigs were loose in the United States.
A research team of agricultural economists at the University of Tennessee and the University of Arkansas examined the program's performance.
The team used federal data on crop insurance claims to compare crop damage in counties where the program was active against counties where it was not, both before and after the federal trial began.
The authors write that the recently revived federal effort offers the potential for bringing feral hogs under control.
Feral hogs breed year-round and each year can produce up to two litters of four to 12 piglets per litter.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
Quantified but single-source and self-reported
The core effect size is specific and falsifiable (70 versus 10 damaged corn acres per crop insurance policy, drawn from federal claim data with before/after and program/non-program comparison), and the funding and damage figures are attributed to USDA and named legislation. But the cluster contains one item, written by the evaluating economists themselves as a Conversation op-ed, with no linked paper, sample sizes, or independent corroboration, and the authors concede their measurement excludes sub-claim damage and all non-crop damage.
Live federal program, geographically partial
This is a funded, operating program rather than a proposal: USDA has been paying landowners for trapping equipment, trapping labor and land restoration since 2020, it survived its 2023 sunset with $105 million more through 2029, and a $35 million application window is currently open. Uptake is bounded, though — only selected counties in 10 of the 35 hog-occupied states, with the authors reporting that pandemic-era outreach limits and pilot-status hesitancy held landowner participation down.
Mildly overstated toward the program's promise
The conclusion that the effort 'offers the potential for bringing the hogs under control' runs ahead of the evidence, which is one significant crop out of five, from self-reported analysis, with non-crop damage unmeasured. Working the other way, the piece is unusually candid about limitations and about eradication being nearly impossible, and the cluster's own arithmetic — $180M committed versus a low-end $2.5B of annual damage — undercuts any suggestion the money is close to sufficient, so the gap is small rather than severe.
Evaluators publishing while their program's funding is before Congress
The authors are academic agricultural economists assessing a USDA program whose reauthorization is live — a $35 million application window open through September 2026 and a House-passed Farm Bill adding $150 million awaiting the Senate — and their piece closes by recommending more funding and better design. Distribution via a Conversation op-ed reprinted by a subscription-funded outlet gives the argument reach without adversarial editing, and the damage figures that justify spending come from the agency running the program. No funding disclosure appears in the supplied text, so the alignment is inferred from the timing and the ask, not from documented payments.
Moderate: verifiable public facts, thin verification of the result
Legislative and budget facts (2018 Farm Bill $75M, July 2025 package $105M, open $35M tranche, House-passed $150M) are checkable public record and the derived arithmetic follows directly from the article's own figures, so those elements are solid. Confidence in the central effect size is materially lower: one self-reported, unlinked analysis, no independent replication in the cluster, and admitted exclusions from the measurement.
science
HIPAA Covers Less Than You Think, And "Anonymized" Is Not A Legal Shield1 distinct publisher
science
Claude's watermark is a compliance artefact, not a cheating detector1 distinct publisher
science
Psilocybin's 'oneness' now has a brain-network correlate, and it was found by sorting people on self-report1 distinct publisher
science
NASA's CADRE puts three rovers on the moon with nobody driving1 distinct publisher
Distinct publishers with included, body-backed reporting in this cluster.
1 article · August 16, 2026