How CF Industries Pays

Operations and professional ladder — CF-A1 through CF-A7

Leadership and executive tiers — CF-A8 through CF-A10

Board — governance, not an employee band

Disclosed executive layer

Track divergence

Hierarchy qualifications and legacy structures

• No reliable public source located an internal CF grade table, job-code ladder, salary-grade numbers, promotion matrix or formal individual-contributor versus management framework. Job postings reveal titles and salary ranges and SEC filings reveal officer titles, but neither establishes a company-wide band code.
• The CF-A8 vice president and senior vice president layer is genuinely blank. CF names senior vice presidents on its leadership page but files no compensation for them, so the row in this report is an analytical bridge and not a CF figure.
• Manufacturing progression is gated by things a title crosswalk cannot see: plant licensing, safety qualification, technical credentials, union classifications where they exist, business need and vacancy availability. Years of experience are a weak predictor at a nitrogen complex.
• No recent merger-created band harmonisation programme, inherited grade catalogue or predecessor band crosswalk was found. CF carries historical pension and legacy option obligations, but those do not establish a current job-level architecture.
• The onshore versus offshore lens does not apply. CF is an industrial manufacturer with corporate and site roles, not an offshore delivery model, and the material structural split is corporate versus manufacturing and distribution.
• The non-employee director row is a governance package rather than an employment band and is excluded from the range chart.

Peer-level mapping

Office and market catalogue — calibration factors versus Northbrook

CF publishes approximate headcounts for four sites only: Donaldsonville at about 485 permanent employees plus 500 contractors, Medicine Hat and Courtright at about 200 each, and Billingham at about 145 permanent employees plus 100 contractors. Ince in Cheshire is closed and carries no current compensation population. Point Lisas in Trinidad is a joint venture whose employees are not part of CF's consolidated payroll, and a possible Amsterdam commercial presence has no disclosed employee count, so none of those three appears in the priced table. No material India, Australia, New Zealand, Singapore or Middle East employee population is disclosed anywhere in the 2025 Form 10-K.

Northbrook anchor medians — the basis of every modeled cell

• The Northbrook column is the anchor and the whole United States table sits on it. CF-A1 through CF-A5 come from employer-posted ranges observed in August 2026 and third-party salary pages; CF-A6 and CF-A7 are analytical envelopes; CF-A9 and CF-A10 are proxy disclosures.
• The same United States table applies to Deerfield, Donaldsonville, Sergeant Bluff and Port Neal, Verdigris, Yazoo City, Woodward and Waggaman. No public source disclosed a city-by-city salary differential, geographic premium matrix or cost-of-living adjustment schedule for any CF site.
• City-level H-1B evidence exists for Deerfield and Northbrook only, and it describes corporate professional roles. It must not be applied to frontline plant jobs, which are the larger part of the workforce.
• Canada carries base at 0.90 and total at 0.78 of the anchor. The strongest Canadian anchor is an employer-posted rotating-engineer range at Courtright of C$121,700 to C$152,700, whose midpoint converts to about 1.02 times the United States professional midpoint.
• The United Kingdom carries base at 0.86 and total at 0.76 of the anchor, drawn from CF Fertilisers salary-site estimates at Billingham. No London premium was found, so London carries the same factors.
• Blue Point One in Modeste, Louisiana is priced at the anchor because it is a United States site, but it has no current employee population. The official project page projects 103 permanent full-time jobs at about $110,000 average salary once operational, which is a forward-looking project average and not a band.

Executive incentive targets — percent of salary

Named-executive outcomes

• At 31 December 2025 approximately 6.3 million of the 10,615,515 authorised plan shares remained available, or about 59.3 percent of the original maximum. The complementary 40.7 percent, or 4.316 million shares, is a rough not-available-versus-original-cap indicator and is not gross historical grant utilisation.
• Plan accounting can add back forfeitures or otherwise recycle shares, so a true utilisation rate cannot be derived from a cap and a remaining balance alone. Any figure quoted as CF's equity pool utilisation should be treated with that caveat attached.
• Stock-based compensation expense was $45 million in 2025, $36 million in 2024 and $37 million in 2023, against a defined-contribution retirement expense of $41 million in 2025. On CF's cost base, retirement contributions are close in size to the entire equity programme, which is unusual and reflects how narrowly equity is distributed.
• No standard share-count matrix for manager, director, vice president or senior vice president levels was disclosed in any source, so there is no way to estimate a grant size below named-executive level from public information.

Equity plan capacity and grant activity

Vesting — common reported employee schedule

Standard restricted stock units vest one third annually over three years subject to continued service. Performance units do not follow this schedule at all: they cliff after a three-year performance period and settle only once the committee certifies return on net assets with the relative total-shareholder-return adjustment, anywhere between 0 and 240 percent of target. Special retention awards use an award-specific cliff, and the residual legacy options are already fully exercisable.

Eligibility by hierarchy level

• CF-A1 and CF-A2 employees are legally eligible under the broad plan language but no routine grant evidence exists for them. There is no public threshold or recurring grant programme for frontline employees.
• CF-A3 and CF-A4 employees may be considered as selected key contributors on a discretionary, role, performance and potential basis. No band floor, grant matrix or typical value is published.
• CF-A5 to CF-A8 covers selected leaders and key contributors, and the evidence supports discretionary annual awards rather than universal eligibility. Generic grant values for this whole span are not publicly disclosed.
• Regular annual long-term incentive is disclosed only for named executives. The legal plan language is broader than actual grant practice, and being eligible under the plan does not mean receiving an annual grant.
• Grant cadence is annual, with award agreements generally communicated around January and administration through Charles Schwab. The 60 percent performance unit and 40 percent restricted stock unit split has applied to named executives since 2018.

Indicative annual grant value by band — Northbrook

Annualized planning value (±25%), not the face value of every new-hire grant. Share equivalents use the $127.26 reference price at 25 August 2026 market data and ignore plan valuation rules and PSU performance. Highlighted rows are disclosed grant-date stock awards.

Named executive target equity

The other column above combines the disclosed change in pension value and all other compensation. For W. Anthony Will those components are $57,700 and $328,000 as disclosed, shown here as a $381,938 residual so that the row reconciles to the exactly reported $13,711,938 total against otherwise rounded component disclosures. Susan Menzel's 2025 stock award of $3.91 million is the largest non-chief-executive figure in the table and reflects a special retention award cliffing on 2 January 2028 layered on top of her annual grant, not a change in her ongoing target. Andrew Scribner's disclosed $140,000 sign-on payment takes his first-year package to about $3.355 million against a $3.21 million steady-state target.

Board compensation framework

Non-employee directors are a governance population and not part of the employee band structure. The employee chief executive receives no separate director pay. The absence of meeting fees means the disclosed retainers are close to the full cash story for a standard director seat, with committee service reflected in the individual totals rather than in a separate published fee schedule.

Not publicly disclosed

Modeled promotion planning timeline

Cadence is the one part of the cycle with real visibility. The corporate annual incentive is settled once, in the first quarter after the performance year. Manufacturing and distribution employees see the company half in the first quarter and the site half quarterly. Long-term incentive award agreements are generally communicated around January and administered through Charles Schwab. What is missing is everything about the individual: no rating, no distribution, no merit matrix.

Dataset summary

City-level H-1B wage history

All-years aggregates. P25/P75 are rounded reconstruction values marked modeled; medians and record counts are the stronger fields.

Selected title / worksite records

Evidence classification

Explicit “Not publicly disclosed” index

Known gaps and diligence before relying on a cell

1. 1Only one research bundle exists for CF Industries, so there are no cross-bundle disagreements to reconcile. The trade-off is that no figure here has independent corroboration from a second research pass, and the thin Canadian and United Kingdom rows have no second opinion at all.
2. 2The CF-A8 vice president and senior vice president row carries modeled numbers against a not-publicly-disclosed evidence class. CF files nothing for that layer, so the row is a bridge between the CF-A7 envelope and the disclosed CF-A9 targets and should not be read as a CF figure of any kind.
3. 3CF-A9 and CF-A10 are kept in the band table but marked chart-excluded, and their seniority is set at 11 and 12 so that the employee ladder occupies seniority 1 through 9. Because CF discloses no Canadian or United Kingdom executive pay at all, the country factors must not be read against those rows: there is no evidence of a local executive package anywhere outside the United States.
4. 4Canada and the United Kingdom carry flat country factors with no junior-to-senior interpolation. The Canadian evidence points both ways, with entry roles at about 0.64 of the United States total and an employer-posted professional range at about 1.02 of United States base, and applying a level-interpolated factor would misstate one end of that ladder. The United Kingdom evidence is three Glassdoor role estimates and one maintenance-manager estimate reused for two levels.
5. 5CF-A9 uses the midpoint of the three continuing executive-vice-president 2026 target packages rather than a published band, and the base component is the midpoint of the disclosed $675,000 to $750,000 span rather than an average of the three actual salaries.
6. 6The neoOutcomes targets are derived by halving each disclosed 2025 payout, because CF discloses the uniform 200 percent attainment and the paid amounts but not a published target table for the named executives.
7. 7The other compensation column for W. Anthony Will is shown as a $381,938 residual so that the row reconciles to the exactly reported $13,711,938 total. The underlying disclosures are a $57,700 pension value change and $328,000 of all other compensation, which sum to $385,700 against otherwise rounded components.
8. 8No employer benefit load is included anywhere in the total compensation figures. CF's 401(k) can deliver 10 to 13 percent of eligible pay in employer contributions, but that is employer cost rather than pay, and adding it would inflate every United States total by roughly a tenth.
9. 9Ince in Cheshire, Point Lisas in Trinidad and a possible Amsterdam commercial presence are excluded from the priced location table. Ince is closed, Point Lisas is a joint venture whose employees are not part of CF's consolidated payroll, and Amsterdam has no disclosed employee count. Pricing any of them would have been invention rather than evidence.
10. 10India, Australia, New Zealand, Singapore and the Middle East appear in the original research schema as selector placeholders but have no material CF employee population disclosed in the 2025 Form 10-K, so no compensation is estimated for them and they are omitted from this report entirely.
11. 11CF discloses no geographic revenue split in any reviewed source, so no regional revenue panel is shown.
12. 12Nine of the thirteen insider transactions were surfaced through an insider-feed summary rather than transcribed from a Form 4. Only the 27 February and 2 March 2026 W. Anthony Will rows came from the cited filing, and the value-weighted average prices shown for those two multi-tranche sales are calculated rather than filed.
13. 13No company-wide 2025 or 2026 merit-increase percentage, pay freeze, broad salary cut, off-cycle market correction programme or campus compensation revision was located anywhere in the 2024 to 2026 search window. Absence of evidence is not evidence that none occurred.

FX rates used — 1 USD equals, snapshot 2026-08-26

Single-date conversion layer for comparability; it ignores payroll-date FX, tax, purchasing power, benefits valuation and hedging. Compensation intelligence, not legal, tax, investment, immigration or employment advice.

Source register — 16 sources

Year-end growth has been steady at roughly 100 employees in each of 2024 and 2025, with the United States share creeping up from 78 to 80 percent. The percent-derived country counts are approximations because the filing reports rounded percentages, which is why they are shown as approximately 2,320 United States, 406 Canadian and 174 United Kingdom employees for 2025 rather than as exact figures. CF publishes no attrition rate at any frequency, so the trend line here is a stock measure with no flow measure behind it.