The Big Bad Wolf

What the Rules Actually Say

The order signed on 4 September 2026 was published six days later in the Federal Register. Read it and the striking thing is how little it does.

It directs two departments to evaluate whether the gray wolf and the Mexican wolf should come off the endangered species list. It directs them to improve the standards by which ranchers are compensated for animals killed by predators. It directs them to make lethal removal easier where depredation is occurring.

Those are instructions to go and look at something. An evaluation is not a delisting. A direction to review compensation standards is not a cheque. The order changes no animal's legal status on the day it was signed, and it does not create any right for a private person to shoot a wolf.

There is no open season on the Mexican wolf. There was no open season on the Mexican wolf before 4 September, and there is none after it.

What confuses this is that killing a Mexican wolf has been lawful, in specific circumstances, for as long as there have been Mexican wolves in the wild to kill.

When the animal was returned to the Southwest in 1998 it was not released as a fully protected endangered species. It was released under a provision of the Endangered Species Act that creates what the statute calls a nonessential experimental population. The entire purpose of that provision is to let managers handle a reintroduced animal with a freer hand than the Act would otherwise allow — including killing it when it keeps eating livestock. That has been the legal position continuously for twenty-eight years, through every change of government in that period, and it is the provision every removal since has been carried out under.

The government has used it, steadily and not very often.

(US Fish and Wildlife Service, Causes and Outcomes of Mexican Wolf Management Removals 1998–2025, published 27 February 2026)

Livestock was the stated reason in 110 of those 283 removals. Over twenty-eight years, then, the federal government has shot fewer than one Mexican wolf a year on average, and moved about five.

So if the killing was already legal and the order signed in September only asks for a review, what actually changed?

One thing did, and it is smaller and stranger than either side has described.

Who Holds the Rifle

On 19 February 2026, the Fish and Wildlife Service issued a permit to two private individuals in Catron County, New Mexico — a sitting county commissioner and a neighbouring cattleman, operating adjacent ranches. It authorised them, personally, to kill one Mexican wolf on their own land, or one caught in the act of attacking livestock on the public grazing allotments next door.

It ran from 19 February to 4 April 2026. No wolf was taken under it.

Historically the government has done its own shooting. When an animal has to be removed, federal agents or USDA Wildlife Services carry it out, and the decision is written down in a memorandum that records what the animal did and what was tried first. Conservation organisations who track these things say permits issued directly to individual ranchers are a recent development. Reporting on this one notes that it carried no documented depredation record — where such authorisations normally set out the confirmed incidents that justify them.

We are going to be precise about what that is and is not. A permit issued without a written record of the losses behind it is a documented fact and we report it as one. It is not evidence that anybody acted in bad faith, and this piece makes no such allegation. Nor do we name the two individuals; they are named in public reporting and do not need to be named again here. The point is procedural, not personal. When the reason for a decision is not written down, nobody outside the room can check the decision — and that is as bad for the rancher who needs the removal to stick as it is for anyone else.

The rest of 2026 has run at a faster pace than usual. On 14 May, Wildlife Services carried out a lethal removal on an Arizona allotment. On 24 July, a decision memorandum authorised moving a wolf; on 4 August, another authorised killing one. The July memorandum notes, in the government's own words, that the May shooting temporarily reduced depredations, but incidents have resumed.

And across 2025 the totals rose: seventeen removals, three of them lethal — the heaviest year since 2020.

That is the true shape of it. Not a season. A quickening.

What Actually Kills Mexican Wolves

Before leaving the question of who shoots what, one number belongs on the record, because it reframes the entire argument about lawful removal.

The Service has documented 343 Mexican wolf deaths since 1998.

(US Fish and Wildlife Service, Causes of documented Mexican wolf mortalities 1998–2025, published 27 February 2026)

Just over half of every Mexican wolf whose death has been documented was killed illegally. In 2025 alone: twelve illegal killings, one vehicle strike, six natural deaths, three other, five unknown, two still awaiting necropsy.

Poaching kills roughly seven Mexican wolves for every one the government has ever lawfully removed.

Which tells you that the argument about whether the rules should permit more shooting is, to some extent, an argument about paperwork. A good deal of the shooting has been happening without it.

The Arithmetic Nobody Quotes

Now the question the whole thing rests on. How much livestock does this animal actually kill?

The answer is published four times a year and almost never repeated, so here it is.

A confirmed depredation is not a rancher's report. It is an incident investigated on the ground by USDA Wildlife Services, where the physical evidence — the bite pattern, the haemorrhaging, the tracks, the carcass itself — establishes that a wolf did it. The investigators sort every case into confirmed, probable, unknown or other. Producers are compensated only for the confirmed ones. It is a demanding standard, deliberately, and we will come back to what it misses.

Counting confirmed incidents alone, and dividing by the wolf population so that a growing number of wolves does not flatter the result:

(Mexican Wolf Recovery Program quarterly updates, US Fish and Wildlife Service and the Interagency Field Team. Figures through 2024 are taken directly from the quarterly updates. The 2025 figures — 147 confirmed incidents, 46 per 100 wolves — are from the Second Quarter 2026 update as reported by the Gila Herald on 13 August 2026; we have not been able to obtain that update as a document and say so rather than imply we have read it.)

Over five years the rate of confirmed depredation has fallen by nearly half, while the wolf population has grown by about seventy percent. More wolves, fewer incidents per wolf, year after year.

And 2025 went up. One hundred and forty-seven confirmed incidents against one hundred and six the year before — the first increase in the series since 2020.

We are putting that in the third paragraph of this section rather than the last one, because a piece arguing that a problem is overstated, which then buries the one year that got worse, deserves to be disbelieved. Forty-six per hundred wolves is still well under where the series stood five years ago. One year is not a trend. But it is a real increase and we are not going to pretend otherwise, and if it happens again in 2026 the argument in this piece gets weaker and we will say so then too.

Now the denominator, which is the part that never appears anywhere.

(USDA National Agricultural Statistics Service, Cattle, released 30 January 2026)

That is one confirmed incident for roughly every fourteen thousand cattle in the two states.

Set against what actually kills cattle in America, the picture does not improve for the wolf's reputation as a menace. USDA's death-loss work puts non-predator causes behind about 98 percent of adult cattle deaths and 89 percent of calf deaths. Of the remainder, coyotes are the leading predator — 40.5 percent of cattle predator losses and 53.1 percent of calf predator losses. Respiratory disease alone kills more than a million head a year.

Every wolf in the United States, of every kind, kills a small fraction of what coyotes kill. And coyotes kill a small fraction of what pneumonia kills.

There is one more piece of evidence, and it is the most useful in this entire file because of who paid for it. In March 2025 three agricultural economists at the University of Arizona published a study of what Mexican wolf presence and depredation actually do to ranch returns. It was funded and assisted by pro-ranching organisations. It found the effect of wolves on county-level cattle inventories and sales was not detectable.

(Duval, Frisvold & Montanía, University of Arizona Cooperative Extension, March 2025. This is an extension publication rather than a peer-reviewed paper, and we flag that rather than oversell it.)

And Now the Part That Cuts the Other Way

Everything above is a statewide average, and a statewide average is a bloodless way to describe something that is not happening to a state. It is happening to a handful of families.

On 24 July 2026 the Fish and Wildlife Service signed a decision memorandum authorising the removal of a wolf from a grazing allotment. The allotment's name is redacted in the copy that has been made public. The memorandum records what had happened there.

Nine incidents in four months on one piece of ground. A lethal removal had already been carried out there in May, and the memorandum states plainly that it worked only for a while.

That is not a statistic. That is somebody's year.

The same University of Arizona study that found no detectable county-level effect also found that the distribution of depredation across ranches is heavily skewed — most affected producers see the occasional incident, and a small number are hit hard and repeatedly. Catron County, which has declared a disaster over this and been joined by Sierra County, peaked in 2019 at around 121 probable and confirmed losses. As a share of the cattle in the area that is still under one percent. As a share of the income of the particular outfits carrying it, it is not.

The Mexican wolf is not a threat to the American cattle industry. It is a serious problem for perhaps a dozen operations. Both of those sentences are true, and every fight about this subject happens because people insist on only one of them.

There is a second thing we will not do, and it is the harder discipline. Confirmed kills are a floor, not a count. A calf that dies on open range in the Gila and is not found within a day or two cannot be confirmed, and every party to this argument agrees the real number is higher than the counted one. What we will not do is put a multiplier on it. Figures circulate — that for every confirmed kill five go unfound — and the ones we could trace lead back to an anonymous advocacy site with no methodology and no author. Wyoming pays seven times market value for a confirmed wolf kill and Washington pays double, but those are policy choices written into statute, not measurements of anything. The one study designed for these conditions — 930 radio-collared calves in the Mexican wolf recovery area, published in Biological Conservation in 2011 — found producers detected 79 percent of predator-killed calves, which cuts against large multipliers, and its wolf-killed sample was three animals, which is far too small to draw a rate from. So: the confirmed number understates the real one, nobody in this country has measured by how much, and we are not going to invent it. That admission is worth more than a number we cannot defend.

A Calendar, Without Comment

We are going to set down four dates and then stop, because the only honest thing to do with them is to print them.

Nine days separate the beef proclamation from the wolf order. Four days separate the opening of the first duty-free tranche from it.

For context on the scale of the first of those: imports of beef ran to 3,297 million pounds in the first half of 2026, up twelve percent on a year that was itself a record. Imported lean trim reached 38.7 percent of all the trim going into American ground beef in 2025, against a twenty-year average nearer 25 percent. The American cattle herd stands at 86.2 million head, the smallest since 1951, with the smallest calf crop since 1941. On the day the 300,000-tonne plan was announced, live and feeder cattle futures fell to eight-month lows at the opening bell.

The American Farm Bureau Federation puts the customs duty foregone on that 300,000 tonnes at roughly $650 million.

Here is what we are not saying, and it matters more than what we are. We searched the public record for any source connecting these two decisions. There is none. No agency document links them. The order of 4 September contains no import, tariff or quota provision — we checked the text. Three cattle producer organisations raised beef imports and wolf policy in the same statements on the day the order was signed, but as two separate grievances listed side by side, with no claim that one was payment for the other. Conservation groups have characterised the wolf order as political cover, and that is their characterisation, offered as advocacy rather than as evidence. We have no evidence of motive and we are not going to assert one. We have printed the dates. Readers can hold two facts next to each other without our help.

What It Would Cost to Just Pay Them

Here is the part of this story that has no constituency, which is presumably why nobody tells it.

The Fish and Wildlife Service publishes what is paid out each year to ranchers in Arizona and New Mexico for livestock confirmed killed by Mexican wolves. The figures are in the programme's annual reports.

Twelve years of paying ranchers for every animal Mexican wolves killed, across two states, costs less than one decent ranch in either of them.

Now put the two halves of this piece beside each other.

The duty given up on imported beef in ninety days is roughly sixteen hundred times the annual cost of making every wolf claim in both states whole.

Whatever this dispute is about, it is not about money. There is no budget in which $400,000 a year is a difficult number. It is smaller than the rounding error on the other decision taken that same week.

The Thing That Worked, and Was Allowed to Stop

The obvious objection to compensation is that it does not cover what the rancher actually loses. That objection is correct, and it has a documented answer that was built in this country, by ranchers, and then quietly defunded.

In 2011 a body called the Mexican Wolf/Livestock Coexistence Council was formed, with eleven members, five of them ranchers. Alongside ordinary compensation for confirmed kills, it did something different. It paid operators for wolf presence on their ground — not for carcasses.

The formula was mechanical and required no investigator and no body. Two points for each wolf core area overlapping the operation. One point for each territory. One point for each pup that survived to the end of the year. The whole score doubled where the operator had conflict-avoidance measures in place, then weighted by the number of livestock exposed. Each qualifying operation took its share of the year's pot.

It paid $85,500 to 28 operators in 2014, and the same amount to 35 operators in 2015.

Then Arizona stopped in 2018. New Mexico stopped after 2020. Across eight years the two states paid about $697,300 in presence payments, and since 2020 they have paid nothing. Arizona's statute still expressly authorises it. The state has simply not used the power in eight years.

We know what that model finds, because another state ran it with real money and published the accounts. California's Wolf-Livestock Compensation Pilot spent $3 million across 109 grants to 36 producers before the fund was exhausted in March 2024.

(California Department of Fish and Wildlife, Wolf-Livestock Compensation Pilot Program final accounting)

Presence paid seven times what carcasses did. When you stop making a rancher produce a body, the real size of the burden appears.

That is the single most important number in this piece after the $650 million, and it is a state government's own accounting rather than anybody's advocacy.

What Good Would Look Like

Other places have solved parts of this and it is worth saying what they did, because none of it is exotic.

Speed. In Majella National Park in Italy, with something near a hundred wolves in ten packs, a farmer reports a kill, a park vet performs the necropsy, a compensation proposal follows the next day, and the money is paid the same day. Twenty-four to forty-eight hours, start to finish.

Benefit of the doubt. In France, a death of undetermined cause is indemnified where the local predation context supports it. Wolf responsibility does not have to be proved, only not excluded. The scheme also pays for the indirect damage — abortions, lost milk, weight that did not go on.

No carcass at all. Sweden pays reindeer herders per verified carnivore reproduction in their grazing area. No body, no claim, no investigator. It also carries a warning: the payment has been frozen at the same figure since 2002, and the herders have lost faith in it. A scheme that is not indexed dies.

Multipliers for what is never found. Wyoming pays seven times fair market value for a confirmed wolf kill in trophy game areas, precisely to account for the animals nobody finds. Washington pays double on operations over a hundred acres, and also pays for reduced pregnancy rates and weight gain measured against a three-year average. Wisconsin pays for missing calves outright — $234,804 of missing-calf claims between 2021 and 2025.

And two warnings. Colorado appropriated $350,000 a year against $1.072 million of claims from 2025 alone, and ranchers who lost animals in that calendar year were still waiting on commission decisions in March and May of 2026. In Castilla y León in Spain, farm unions put €10 million of wolf compensation outstanding with delays of up to a year, as of this month. Underfund the promise and you have not built goodwill. You have built a grievance with paperwork attached.

One thing in this country has already improved and it deserves saying plainly, because a piece that will not credit a fix cannot be trusted on a failure. The federal Livestock Indemnity Programme historically paid 75 percent of a national average value — not what the calf was worth, and not what it was worth locally. A final rule published on 9 July 2026 changed predation losses to 100 percent of market value. That is real and it is an improvement.

What has not been fixed is that probable kills are still not federally paid, and probable is exactly where open-range losses live. New Mexico's County Livestock Loss Authority — a joint body of Catron, Sierra and Socorro counties, formed in 2022 to replace the old Coexistence Council — does pay full fair market value for confirmed and probable. The better model is already running in the very counties making the most noise about wolves, and it is starved.

Stewardship

So here is what we would put in place of an argument about whether to shoot three hundred and seventeen animals.

One. Full local market value for the animal that died. Not a national average. The price at the sale barn that week. Already done for confirmed kills as of July. Extend it to probable.

Two. Presence payments to the operations carrying the wolves. The formula already exists, it was written by a body that was five-elevenths ranchers, Arizona's statute still authorises it, and California's accounts show it is where most of the real cost actually sits. This is the part that pays for the burden rather than the corpse.

Three. Paid inside a fortnight. An Italian national park manages it in a day. A claim that takes a year is not compensation. It is an insult with a reference number.

Four. Indexed, and funded to the level of the claims. So it does not die the way Sweden's has, or fail the way Colorado's is failing.

At something around $400,000 a year for both states, this is not a budget decision. It is a decision about what kind of country this is.

We can pay the dozen families who happen to graze cattle where the last three hundred Mexican wolves live. Or we can shoot the wolves. Those are the options, and only one of them is cheap.

The word for the first one is stewardship, and it is a word that belongs to ranchers before it belongs to anybody else. The people who run cattle in the Gila and the Blue Range are not tourists there. They know that ground in a way no agency does, and a great deal of what is still wild about it is wild because they kept it unsubdivided. Ask the same families what they want for their grandchildren and you will not hear that they want an empty country. You will hear that they want to still be there, and for it to still be worth being there for.

That is the same answer the conservationist gives. It has been the same answer for thirty years. The two sides have been arguing about a wolf when the argument was always about who absorbs the cost of keeping one.

Nobody should have to go broke hosting the nation's wildlife on their own grass.

Pay them properly, pay them fast, pay them for the burden and not just the body, and the wolf question becomes what it always should have been: a question about how many wolves the country wants, asked by people who are not being asked to fund the answer out of their own calf cheque.

What We Are Not Saying

We are not saying the Mexican wolf is harmless. Nine confirmed incidents on one allotment in four months is not harmless to the operation holding that allotment, and any argument built on statewide averages is built on a number that describes nobody.

We are not saying ranchers exaggerate. The concentration is real, the confirmation standard genuinely does miss losses on open range, and the amount it misses has never been measured in this country. Someone who tells you they know that figure is guessing.

We are not saying the September order opened a season on wolves, because it did not, and the people saying so loudest are not helping the animal they are trying to defend.

We are not saying anyone traded wolves for beef imports. We printed four dates. We found no source linking them and we assert no connection between them.

And we are not saying compensation solves everything. It does not touch the stress of it, the nights out, the dog that will not go near the far pasture any more, or what it feels like to find what is left of a calf you have been watching since February. Money is not the whole of the injury. It is simply the part that can be fixed this year, for very little, and has not been.