California GOAT Herder Wage Odds

On July 1, the cost of employing one California goat herder went from about $4,938 a month to roughly $20,212 a month. Nobody voted for that. A sentence in a three-year-old budget bill simply expired, and a rule written for people who watch livestock around the clock stopped applying to the people who watch goats.

Six weeks later, the world learned that a rancher in Red Bluff had done something no rancher had ever done about a state legislature. Tim Arrowsmith, who runs roughly 4,000 grazing goats at Western Grazers and employs eight herders, bought a contract that pays him up to $500,000 if Sacramento fails to fix the rule by the end of September. A startup called Castle structured the coverage, Kalshi listed the market, and Susquehanna, one of the largest trading firms in the world, took the other side and quoted the price.

For four weeks that price was the most interesting number in California labor policy. Traders said relief was coming. Eight AI models, handed the statute, the calendar and the state's own study, and never shown the price, said something close to the opposite. The gap between them was 56 points, and the whole argument came down to one question about a legislative calendar that had already run most of the way out.

The calendar answered. This page was built on August 24 with the market at 84 cents and the panel at 27.5%. On September 18 the Governor signed the fix, and on September 19 Kalshi settled the market YES. The traders were right and the panel was wrong, and the record of exactly how is the first thing below. Everything after it is the argument as it stood on August 24, kept intact because a graded miss is only worth something if you can still read the reasoning that produced it.

The Record

	Panel (Aug. 24)	Market (Aug. 24)	Result (Sept. 19)
YES Relief Before Oct 1	27.5%	84¢	YES, paid $1.00
NO Qualifying Action	72.5%	16¢	NO, paid $0.00

The panel missed. Its median said relief was a roughly one-in-four event, and every one of its eight seats sat below 50%; the highest, Claude Sonnet and Kimi, stopped at 40%. The market said 84 cents at the panel's stamp, and the last print of the following day, August 25, was 85 cents. Kalshi finalized the contract YES on September 19 at 11:41 a.m. ET, well ahead of its October 7 expiration, with 655,737 contracts traded over its life and 519,095 still open at the finish.

What settled it was the door the panel itself named. On August 28, AB 187, a Committee on Budget bill titled simply "Employment," was amended in the Senate to add a new Section 2695.4 to the Labor Code, with the stated intent "to codify certain labor protections that should be afforded to goat herders." The operative line lets an employer of a goat herder working a regularly scheduled 24-hour, seven-day on-call shift pay "no less than the monthly minimum wage specified in Section 4(E) of Wage Order No. 14-2001," which is the sheepherder monthly figure, $4,938.21 for 2026. The section "shall remain in effect only until January 1, 2029." The bill passed the Senate Budget and Fiscal Review Committee 16-0 on August 31, cleared the Senate floor 40-0 the same day, and the Assembly concurred 76-0 before adjournment. It reached the Governor's desk on September 8 and was approved on September 18 as Chapter 256, Statutes of 2026, one of 71 bills signed that day. As a budget trailer bill it took effect immediately on the Governor's signature, and because the new wage equals the sheepherder wage, it clears the contract's Net Monthly Cost test with the full 15% margin to spare. Melissa Hurtado, whose SB 801 had died in committee in February, claimed the fix on September 2: "I refused to stand by while grazing operators go out of business, workers lose their livelihoods, and communities face greater wildfire risk." She also said the work was not finished, and the sunset says why.

That is the same vehicle, the same clause and nearly the same date this page laid out four weeks ago. The first item under "What Would Have Changed The Panel's Mind" was goat-herder wage language appearing in print in a live vehicle "before roughly August 28." It appeared on August 28. Claude Fable had said that one event alone would flip its estimate to about 80%. So the panel's model of the mechanism held; what it got wrong was the prior. Nearly every seat reasoned from the same two facts, that no public vehicle existed six days out and that labor opposition had killed AB 1099 and SB 801, and Claude Opus added a third, that a completed budget meant a closed budget cycle. All three were true facts about the world and none of them was the deciding one, which was simply that a trailer bill is invisible until it is printed, and leadership prints it when it decides to.

The tape saw the print before the news did. The day the amendment landed, the contract opened at 88 cents, took a 4,700-contract buy at 88 at 12:41 p.m. ET, and then ran from 89 to 95 cents between 9:44 and 10:22 p.m. ET. The day's 10,784 contracts were almost all buyers of YES, 9,969 of them. It touched 98 the next morning, printed 98 on the evening of August 31 as both houses voted, and printed 99 cents on 16,814 contracts on September 1, more volume than any day since this page was built. From there to settlement it never traded below 98.

That opening block deserves its own sentence, because it is the reason this market existed. The first print on the tape, on August 4 at 3:37 p.m. ET, is a single fill of 500,000 contracts at 90 cents, taken on the NO side. NO at 90 cents on the YES price costs 10 cents a contract, so that fill bought a $500,000 payout for about $50,000 before fees. Kalshi's own write-up, published nine days later, said the Western Grazers contract "pays up to $500,000 if California does not fix the goat-herder wage rules by the end of September 2026"; the tape does not carry a name, but it carries a fill of exactly that size. If that was the hedge, it settled worthless on September 19, which is the outcome the hedge existed to make survivable. Arrowsmith kept his business, paid a premium for the insurance, and the Legislature did the thing he had bet against it doing.

Below is the page as it stood on August 24, unchanged except for dated lead-ins, so the reasoning that missed can be read next to the result that graded it.

The $4,938 Wage That Became $20,212

New to Kalshi event markets? Start here.

Range herding is a job with no shifts. A herder lives with the animals, moves with them, and is on call every hour of every day. California handled that for decades by paying sheepherders a monthly wage instead of an hourly one. The Department of Industrial Relations sets that figure each year. For 2026 it is $4,938.21 a month, built from a $3,004.50 base monthly minimum wage plus $1,933.71 in required overtime, computed off a regular rate of $4.13 an hour across a 168-hour week.

Goat herders had that same deal until this summer. DIR's own page states the change plainly: "Effective July 1, 2026, Labor Code sections 2695.3 and 2695.4 are repealed." From that date, a goat herder is an ordinary hourly employee, and a 24-hour on-call shift becomes 24 compensable hours.

The arithmetic that follows is brutal, and it is not an estimate. Kalshi's filed contract terms spell out exactly how to compute it: every workweek counts as 40 hours at the regular rate, 40 hours at time-and-a-half, and 88 hours at double time, with a month equal to 52/12 workweeks. That is 276 rate-hours a week and 1,196 rate-hours a month. California's minimum wage has been $16.90 an hour since January 1. Multiply it out and one herder costs $20,212.40 a month, or $242,548.80 a year, before housing, food and the cell service the job requires.

This did not sneak up on anyone. The state studied it. DIR's SB 143 Study, published March 30, said the quiet part in one line: "All stakeholders agreed that the minimum monthly wage should be the same for sheep and goat herders because the workers and work involved in herding sheep and goats are similar and often overlap." The same study noted why the stakes have grown. California has about 500,000 sheep and 125,000 goats, and while nobody tracks it precisely, stakeholders agree at least 20 percent now work in targeted grazing, eating brush in parks, orchards, solar farms and the neighborhoods that back onto open hillsides. That sector is growing while the meat-and-wool business shrinks.

So why did a rule everyone agreed on lapse? Because the fix was always temporary. Labor Code 2695.4 carried a sunset, and the sunset arrived. The two bills written to prevent it both died in labor committees without a vote. AB 1099, authored by Megan Dahle and titled "Goat herders: labor protections," died in Assembly Labor and Employment on January 31, 2024. SB 801, authored by Melissa Hurtado, was returned to the Secretary of the Senate on February 2, 2026 out of Senate Labor, Public Employment and Retirement, six months before the lapse it was meant to prevent.

That record is where the fight actually lives, and both sides have put it in writing. Reason ran the lapse under the headline "California's War on Goats," quoting Arrowsmith: "The goats grazing in your neighborhood will disappear, the goats grazing in your cities and municipalities and HOAs, and things like that, they'll be gone." California Globe attributed the failure of both bills to organized labor, reporting that "labor groups opposed permanent exemptions, arguing for consistent overtime and minimum wage 'protections' across agricultural workers rather than industry-specific carve-outs." The worker-advocate position is on the record in the state's own study too, and it is not a slogan: advocates want more enforcement, not less, and told DIR they oppose "no exemption from overtime wages for range herders who are on duty 24/7," warning that H-2A herders risk their jobs if they complain. The same study found that federal and state enforcement agencies turn up few violations, and that ranchers say high labor costs are pushing them out of the state. Two sides, one wage line, and a carve-out that has now been killed twice in committee before it ever reached a floor.

The Calendar Is The Whole Story

This section is the August 24 read, kept as written. Everything about this market is a clock, and the clock is public. From the 2026 Tentative Legislative Calendar, compiled by the Secretary of the Senate and the Assembly Chief Clerk:

• Aug. 14: last day for fiscal committees to meet and report bills to the Floor.
• Aug. 17 To 31: Floor Session only. No committee other than conference and Rules may meet for any purpose.
• Aug. 21: last day to amend on the Floor. This one has already passed.
• Aug. 31: last day for each house to pass bills. Final recess begins upon adjournment.
• Sept. 30: last day for the Governor to sign or veto.

Read that sequence again with the contract in hand. The market asks whether California authorizes relief before October 1, and the Governor's signing window runs to September 30, so a bill passed in the last week of August and signed in late September settles this YES. But the language has to exist somewhere first. Floor amendments closed on August 21. Committees other than Rules and conference cannot meet. And Proposition 54, which amended Article IV, Section 8(b) of the state constitution in 2016, requires a bill in its final form to be in print and published online for 72 hours before either house votes on it, waivable only by a gubernatorial emergency declaration plus a two-thirds vote of both houses. Passage by August 31 therefore means language visible around August 28.

There is one door that is still open, and it is the door the last fix walked through. SB 143, the 2023 measure that extended the goat-herder wage to begin with, was a Committee on Budget and Fiscal Review bill, a budget trailer bill, chaptered September 13, 2023 as Chapter 196. Kalshi's terms name budget bills and trailer bills explicitly as qualifying vehicles. Trailer bills are not bound by the ordinary policy-bill deadlines, and they surface with little or no warning.

But the 2023 analogy has a crack in it that matters more than it looks. 2023 was the first year of a two-year session, and the Legislature sat until September 14, which is why a September trailer bill was possible at all. 2026 is the second year. Final recess begins when the Assembly and Senate adjourn on August 31, and they do not come back. Whatever happens has to happen in about a week.

The non-legislative doors are narrower still. The Industrial Welfare Commission could amend Wage Order 14, except that DIR's own site states flatly: "The IWC is currently not in operation." A formal, binding statewide enforcement policy from the Labor Commissioner would qualify, but Kalshi's filed terms specifically list an informal FAQ saying enforcement is not a priority as something that would not settle YES. Arrowsmith has said publicly that Western Grazers will sue the state and seek an injunction if there is no resolution by August 31, and a court order does qualify, but only if it reaches goat-herder employers as a class or applies statewide, and only if it is un-stayed on the expiration date. A ruling that covers one ranch does not count.

What The Panel Said On August 24

Eight models were handed the live market payload, the filed contract terms, DIR's wage tables, the current status of all three bills, the legislative calendar and the dated news record. None of them saw the price. Then each read the other seven anonymized answers and was invited to revise.

The blend and the price it was scored against are in The Record above. Prices fetched August 24, 2026 at 11:58 PM ET. Bid 83 cents, ask 84 cents, 591,392 contracts traded lifetime and 512,967 in open interest, which for a state labor-code question is an enormous book.

Other boards the panel priced the same night, at their August 24 prices: 2026 Governor Races, where Xavier Becerra sat at 97¢ in California · Trump $250 Bill, another does-it-become-law market, 4¢ on August 24.

Here is every seat, before and after the revision round.

Seat	Round 1	Round 2	The core of its reasoning
Claude Sonnet	40%	40%	Trailer bills are deadline-exempt and surface late; that mechanism is real
Kimi	40%	40%	An unnamed trailer bill is a reporting gap, not evidence of absence
GPT	38%	38%	Leadership and the Governor still control a late vehicle
GLM	30%	30%	Letters are not floor votes; the executive-order path is thin but not zero
Gemini	35%	25%	Six days left, no visible vehicle, and demonstrated labor opposition
Claude Fable	22%	22%	2023 extended an existing sunset; 2026 asks for a new carve-out
DeepSeek	22%	22%	No leak, no signal, this late in session
Claude Opus	16%	18%	The budget cycle is closed and the lapse was allowed deliberately

Blend after revision: 27.5%, the seat median. The mean is 29.4%. The spread runs 18% to 40%, so this is not a panel that agrees with itself; it is a panel that agrees on direction and argues about magnitude.

Every seat on this panel is graded against real market settlements — records to date: Gemini 86% on 10,833 graded calls · GPT 85% on 11,590 graded calls · Claude Opus 86% on 4,673 graded calls · Claude Fable 86% on 4,489 graded calls · Claude Sonnet 85% on 4,649 graded calls · Kimi 85% on 5,849 graded calls · GLM 83% on 5,242 graded calls · DeepSeek 84% on 5,057 graded calls. Recomputed daily; the full scoreboard is public. This market now counts as a miss for all eight.

Model estimates generated August 24, 2026. Models are frequently wrong; the market price reflects real traders' money. Every number here gets graded in public once the market settles, and you can check the panel's homework on the full graded scoreboard.

That scoreboard was also the honest frame for a 56-point disagreement, and it held. When this panel and a liquid market disagree, the market has been right about two times out of three, and it was right here. A gap that size is a signal that one side is reasoning from something the other cannot see, and here the asymmetry is obvious: the panel can only read what is public, and a budget trailer bill is invisible until it is printed. Susquehanna is not quoting 84 cents out of optimism. Either somebody knows a vehicle exists, or the professional read is simply that Sacramento does not let a wildfire-prevention industry die over a drafting sunset in an election year.

Where The Panel Changed Its Mind

Only two seats moved in the August 24 revision round, and the arguments for holding were sharper than the revisions. Read them now knowing that the trailer bill Kimi refused to rule out was in print four days later.

"Peers highlighted the significance of the August 21 floor-amendment deadline passing without a visible vehicle; with only six days left until the August 31 session end, relying on a last-minute trailer bill or executive rescue against strong labor opposition warrants a lower probability." — Gemini, cutting from 35% to 25%

"Peers leaned on the 2023 SB 143 trailer-bill precedent and the 32-legislator letter; I already weighted both, and the disanalogy holds. 2023's fix rode an active budget cycle with a sunset that leadership chose to extend, while 2026 has a completed budget, a lapse labor deliberately allowed to happen, and an ask that is explicitly for a new carve-out rather than a continuation." — Claude Opus, holding near the floor at 18%

"Held at 0.40: the lower estimates restate factors already in my model, and the data card explicitly warns that an unnamed trailer bill is a reporting gap, not evidence of absence; the 2023 fix itself surfaced late as a budget trailer over the same opposition." — Kimi, refusing to move

GLM held too, but credited a peer for a path it had underweighted: an executive order, which it kept out of its number on the grounds that any such order "would face immediate legal challenge and must clear the contract's formal-policy bar, not mere forbearance."

What Would Have Changed The Panel's Mind

Every seat named the same decisive event, which almost never happens. Here is the August 24 list with what actually happened to each item.

• Goat-Herder Or Sheepherder Wage Language Appears In Print In Any Live Vehicle Before Roughly August 28. This was the one. Fable put it plainly: that alone flips the estimate to about 80%. It happened on August 28, in the Senate amendments to AB 187.
• A Budget Trailer Bill Or Rules Committee Waiver Reaches A Floor Calendar Before August 31. Pushes YES hard. It happened on the deadline itself, not before it: committee 16-0, Senate floor 40-0, Assembly concurrence 76-0, all on August 31.
• The Legislature Adjourns On August 31 With Nothing. Would have collapsed YES to whatever the court and executive paths were worth, which the panel priced in single digits. Did not happen.
• Western Grazers Files Suit In Early September And Wins Class-Wide Or Statewide Relief That Survives To October 7. A narrow path, and it was never needed.

The Hedge That Made This Market Real

It is worth sitting with what Arrowsmith actually did, because it is the first clean example of a small business insuring itself against a bill, and because the insurance did what insurance is supposed to do: it cost him a premium and he never needed the payout. His herd is around 4,000 to 5,000 goats depending on the count, he employs eight herders, and at $242,548.80 apiece the new rule is an existential number. He told Action News Now on August 19: "As of right now if we were paying the wage we would file bankruptcy but we opted not to pay it." So he bought a payout that lands exactly when the bankruptcy would.

The people who built it were direct about what it is. "A rancher should be able to hedge a bill just like a farmer hedges wheat," said Lucas Cavalieri, CEO of Castle. Eric Passmore, a senior trader at Susquehanna, framed it as risk management rather than speculation. Nicolas Hull of Kalshi noted that before this, there was no practical way to hedge a specific legislative outcome. And Arrowsmith, who has lived this cycle once already, said the thing that explains why the market exists at all: "Another three-year extension isn't a fix, it's a countdown clock."

He is right about the pattern. In 2023 he told the Sacramento Bee that a fix meant "we get to stay in business for another two and a half years," and complained that lawmakers "kick the can down the road." Assemblyman Vince Fong's reaction to that year's trailer bill has aged well: "I didn't have goat herders on my bingo card for this budget year."

Settlement Timeline, As It Happened

Date	What the August 24 page expected	What happened
Aug 28	Last practical day for language to hit print under the 72-hour rule	AB 187 amended in the Senate to add Labor Code 2695.4; the contract opened at 88 cents and finished at 95 after a 9:44 to 10:22 p.m. ET sweep
Aug 31	Last day for each house to pass bills; final recess begins on adjournment	Committee 16-0, Senate 40-0, Assembly concurrence 76-0
Sept 1 To Sept 30	Governor's window to sign or veto	Enrolled and presented Sept. 8; approved Sept. 18 as Chapter 256, Statutes of 2026
Sept 19, 11:41 a.m. ET	(Expiration was set for Oct. 7)	Kalshi finalized the market YES; last trade 99 cents

The market settled on official California records, with California Legislative Information at the top of the hierarchy, and AB 187 becoming law on the Governor's signature on September 18, chaptered the same day as Chapter 256, was the qualifying action. Availability varies by state and you must be 18+ to trade on Kalshi, which is a CFTC-regulated exchange, not a sportsbook. Fees and thin books both matter on contracts like this one. This market is closed and this page is its graded record; the August 24 price stamp above marks the read the panel was scored on.

The Bottom Line

The market said 84 cents. The panel said 27.5%. The answer was YES, and this page said on August 24 exactly which sentence would decide it: "If a bill surfaces by roughly August 28, this settles YES and the panel was wrong." The bill surfaced on August 28.

The lesson is not that eight models cannot read a calendar; they read it correctly, down to the day. It is that a legislative calendar tells you when a thing can happen and nothing about whether leadership wants it to, and that second question was priced by people who had been watching Sacramento for longer than one afternoon. Susquehanna quoted 84 cents against a panel at 27.5%, and the difference between them was one unprinted bill. The fix lasts until January 1, 2029. Arrowsmith called the last one a countdown clock, and this one is too, so the same market will exist again in a little over two years, and the panel now has a graded case to learn from before it does.

Somewhere in Red Bluff, a man with 4,000 goats paid a premium for insurance he did not need, which is the best outcome insurance ever has.

More on this: The Money Spent A Day Catching Up To The Poll · Why Kurt Alme (R) Will Win the Montana Senate Race · Greenland Acquisition Odds: Will The US Take Any Part? · Why Ashley Moody (R) Will Win the Florida Senate Primary · Why Chris Pappas (D) Will Win the NH Senate Primary