The highest-paid CEOs, and what they actually got

Every chief executive, by reported pay

Awards above $100 million went to 6 of them; the median was $25.3 million. These totals count stock at its value on the day it was granted, whether or not it ever pays out. The last two columns put each figure beside what the stock made of it, and beside the pay of the company’s median employee.

One number is not like the others. Elon Musk’s $158.4 billion is 252 times the next award on the list, Clayton M. Magouyrk’s $627.5 million at Oracle. Below the top few, the list settles into a long, crowded middle: half of these chief executives were awarded between $16.2 million and $36.7 million.

Ranked by what the stock made of it

Sort by what the awards were worth at year end and the list turns over. Alexander Karp of Palantir goes to the top at $11.1 billion, from $8.6 million reported. At the other end, 14 chief executives were “paid” less than nothing: Allan Thygesen of Docusign by $157.7 million, as the stock already in hand lost more than the year’s pay added.

The measure works like this. Each year’s new awards are valued at what they are worth on the last day of the year, not on the day they were granted, and the awards still waiting to vest from earlier years are revalued too. A chief executive with a modest new award but a large pile of unvested stock in a rising company can be “paid” many times the headline figure. That is the pattern at the top of this ranking: Tony Xu of DoorDash ($431,864 reported, $238.5 million actually paid), Vladimir Tenev of Robinhood ($3 million reported, $1 billion actually paid) and David V. Goeckeler of Sandisk ($10.5 million reported, $2.7 billion actually paid).

The same arithmetic runs in reverse. Brian Chesky of Airbnb, William R. McDermott of ServiceNow and Marc Benioff of Salesforce each finished the year with a negative “actually paid” figure: the stock they already held lost more value than the year’s pay added. In all, 43 of the 104 chief executives with both figures got less than their reported pay. The headline number is what the board intended; the second number is what the shareholders’ year made of it.

1. Alexander Karp Palantir$11.1 billion$8.6 million reported
2. Elon Musk Tesla$10.4 billion$158.4 billion reported
3. David V. Goeckeler Sandisk$2.7 billion$10.5 million reported
4. Hock E. Tan Broadcom$2.3 billion$205.3 million reported
5. Vladimir Tenev Robinhood$1 billion$3 million reported
6. Clayton M. Magouyrk Oracle$413 million$627.5 million reported
7. George Kurtz CrowdStrike$330.3 million$247.6 million reported
8. David Solomon Goldman Sachs$309.8 million$118.9 million reported
9. Steven L. Huffman Reddit$307.4 million$3.1 million reported
10. Safra A. Catz Oracle$257 million$10.1 million reported

266 times the median worker

Since 2018, most US-listed companies have had to put the chief executive’s pay beside the pay of the company’s median employee. Across the 68 companies here with one chief executive and a stated ratio, the median ratio was 266 to 1. The highest were Starbucks at 1,794 to 1, Coca-Cola at 1,739 to 1, CrowdStrike at 1,391 to 1; the lowest, Atlassian at 0.23 to 1 and Airbnb at about 1 to 1.

The ratio measures the payroll as much as the boss. A company whose median employee works a counter or a kitchen will post a high ratio for a chief executive whose pay is ordinary on this list; a software company of well-paid engineers will post a low one. Read two proxies side by side:

“Based on an annual total compensation of our median employee for 2025 of $17,446, and the annual total compensation of $15.46 million for Scott Boatwright, our CEO, as reported in the 2025 Summary Compensation Table, the ratio of our CEO's annual total compensation to our median employee's annual total compensation is 886 to 1.” Chipotle proxy statement

“The median of the annual total compensation of all of our employees (except for our CEO) for fiscal 2026 was $165,164, resulting in a CEO to median employee pay ratio of approximately 7.8:1.” Zoom proxy statement

So a high ratio tells you who the company employs at least as much as what it pays at the top. Compare ratios between companies that hire the same kinds of people, or follow one company’s ratio from year to year.

Paid in a losing year

15 of these chief executives, at 11 companies, were paid for a fiscal year in which their company reported a net loss. The largest awards among them went to Lip-Bu Tan of Intel ($93 million, in a year the company lost $267 million), Niccolo M. de Masi of IonQ ($89.6 million, in a year the company lost $510.4 million) and Matthew Prince of Cloudflare ($60.6 million, in a year the company lost $102.3 million). The deepest loss was at Ford, $8.2 billion, where James D. Farley, Jr.’s pay was $27.5 million.

That is less of a contradiction than it looks. These totals count stock awards at their value on the day they were granted, and a stock award is a bet on what the shares do next; a board can believe in a company’s future through a bad year. The figure that answers whether the bet worked is, again, the second one.

Pay and job cuts are decided in different rooms, but they can land in the same year. 10 of the 92 companies filed a restructuring notice with the SEC, the filing a company makes when it commits to a plan to close or shrink operations, during the fiscal year their chief executive’s pay covers, among them CrowdStrike, Intel, Cisco and Nike. Every cut, with the reason given, is on the Layoff Tracker.