Finance Gazette analysis of executive remuneration disclosures at ten of the FTSE 100’s largest companies finds that headline CEO shareholding multiples are not always measuring the same thing. Differences in the treatment of deferred shares, vested awards and other equity interests can materially affect what investors infer about an executive’s personal exposure to the company’s share price.
A chief executive reported as holding shares worth 37 times salary might appear to have substantially more personal capital at risk than one reported at nine times salary. The comparison looks straightforward, but the underlying figures can represent quite different things.
Finance Gazette examined executive shareholding disclosures across ten of the largest companies in the FTSE 100 and found significant differences in the equity that companies count towards CEO ownership requirements. Some calculations are closely aligned with shares already beneficially held by the executive, while others can include deferred shares, vested awards or other qualifying equity interests.
This creates an important distinction for investors. A company's ownership requirement principally measures whether an executive complies with its remuneration policy. Assessing how much of an executive's personal wealth is exposed to movements in the company's share price is a related, but not necessarily identical, exercise.
The differences are visible in the companies' own remuneration disclosures.
| Company | CEO | Reported ownership position / requirement | What the disclosure shows |
|---|---|---|---|
| Rolls-Royce | Tufan Erginbilgic | 3,729% of salary / 400% requirement | Beneficial shares plus qualifying equity interests under the policy |
| HSBC | Georges Elhedery | 792% / 600% | Share interests; unvested interests excluded from the reported calculation |
| AstraZeneca | Pascal Soriot | 1,051% / 1,150% minimum requirement | Beneficial interests include certain shares in holding or deferral periods |
| Glencore | Gary Nagle | 946% / 525% | Beneficial ownership; unvested awards excluded |
| Unilever | Fernando Fernandez | 861% of fixed pay / 500% guideline | Beneficial ownership under Unilever's policy definition, with other awards disclosed separately |
| BAT | Tadeu Marroco | 867% / 600% | Eligible holding used for the policy calculation exceeds ordinary shares separately shown as held |
| Shell | Wael Sawan | 676% / 700% at 31 Dec 2025 | Position subsequently changed materially as incentive shares were delivered |
| Rio Tinto | Simon Trott | 43% of fixed-share requirement at 2025 year-end | Recently appointed CEO building towards a fixed number of required shares |
| GSK | Luke Miels | 725% requirement | New CEO; GSK applies its own share ownership requirement methodology |
| BP | Meg O'Neill | New CEO building towards requirement | Significant replacement and performance awards are conditional interests rather than simply outright shares |
Source: Finance Gazette analysis of company annual reports, remuneration reports and subsequent regulatory shareholding disclosures. Reporting dates and definitions differ, so the percentages above should not be interpreted as a like-for-like ranking.
A figure described as 900% of salary at one company can therefore represent a different combination of economic interests from the same percentage reported elsewhere. The differences become particularly apparent when the headline policy figures are compared with the underlying share disclosures.
Rolls-Royce provides the most striking example. Its 2025 annual report records chief executive Tufan Erginbilgic's shareholding at 3,729% of base salary at 31 December 2025, equivalent to 37.29 times salary, while the same remuneration disclosure separately records 115,883 shares as beneficially held.
Alongside those shares were 9,696,462 conditional shares not subject to performance conditions, 1,779,980 conditional shares subject to performance conditions and 42,037 salary-deferred shares.
There is no inconsistency in Rolls-Royce reporting the two figures because they answer different questions. The company's remuneration policy establishes which equity interests qualify when determining whether the CEO has satisfied the ownership requirement; the beneficial holding figure describes a narrower category of interests.
The distinction matters when the 3,729% figure is compared externally with another company's CEO ownership multiple. What appears to be a comparison of personal shareholdings can instead become a comparison between different remuneration-policy definitions.
British American Tobacco illustrates the same issue on a smaller scale. BAT's 2025 disclosure shows Tadeu Marroco with 197,613 ordinary shares held, while its ownership-requirement calculation uses 292,071 eligible ordinary shares, valued at £12.308 million. That produces an official position equivalent to 867% of salary, against a 600% requirement.
Neither figure is inherently more correct: 197,613 ordinary shares held and 292,071 shares eligible for a remuneration-policy calculation are simply different measures.
At the end of 2025, Glencore chief executive Gary Nagle beneficially owned 3,452,919 shares, which the company calculated at 946% of salary, compared with a requirement of 525%. Importantly for cross-company comparisons, Glencore explicitly states that unvested awards do not count towards satisfaction of its shareholding guidelines.
HSBC also uses a comparatively narrow calculation. Georges Elhedery had 1,109,810 share interests at 31 December 2025, producing a reported holding equivalent to 792% of salary, above HSBC's 600% guideline. The bank states that the calculation does not include unvested interests, while deferred and performance-related awards are disclosed separately.
A CEO reported at 946% under one company's rules and another reported at a similar percentage elsewhere cannot therefore automatically be assumed to have equivalent forms of personal equity exposure.
Restricting the comparison to figures labelled beneficial ownership initially appears to solve the problem, but the underlying reports show that the terminology itself requires careful reading.
AstraZeneca reported that Pascal Soriot had beneficial interests in 235,607 ordinary shares at 31 December 2025. Its footnote explains that, for executive directors, beneficial interests include shares in holding and deferral periods where those shares are no longer subject to performance measures or continued employment. The company calculated Soriot's position against its minimum shareholding requirement at 1,051%, compared with a requirement of 1,150%.
His position subsequently changed in a particularly revealing way. On 14 September 2026, Soriot personally purchased another 60,000 AstraZeneca shares at £121.02 each, committing approximately £7.26 million of capital.
That transaction introduces a distinction conventional ownership multiples largely ignore: where the shares came from.
Shares purchased using an executive's existing personal capital are not economically identical to shares accumulated through annual bonuses or long-term incentive plans. Vested remuneration shares still expose an executive to subsequent movements in the share price if they continue to be held, but the route by which that exposure was created is different.
Unilever provides another example. Fernando Fernandez had 283,529 beneficially owned shares at the end of 2025, while performance-conditioned and other awards were disclosed separately. Under Unilever's methodology his actual ownership represented 861% of fixed pay, compared with a 500% guideline, although the company's disclosures explain that certain deferred annual-bonus shares without performance conditions can form part of beneficial ownership.
Even restricting a comparison to figures explicitly described as beneficial holdings would therefore not necessarily create a perfectly standardised measure.
CEO ownership rules are intended to align executives with shareholders, addressing the longstanding corporate-governance problem that the people making decisions about a company do not necessarily bear the same economic consequences as its owners. The disclosures examined by Finance Gazette, however, contain several different forms of alignment within the broad concept of executive share ownership.
An executive who spends £5 million of existing personal wealth buying shares has deliberately exchanged cash or another asset for exposure to the company. An executive who receives £5 million of shares as remuneration and retains them also has substantial exposure to subsequent share-price movements, but arrived at that position differently. An executive holding £5 million of conditional performance awards has potential equity exposure again, although the ultimate value may depend on vesting, continued employment or performance conditions.
The distinction is particularly relevant on the downside. If a company's shares fall 30%, ordinary shares already owned by an executive decline in value alongside those of other shareholders. Conditional awards can also lose substantial potential value, but their economic characteristics depend on the terms of the award and whether the interests have vested.
“Skin in the game” consequently encompasses several different financial relationships that a single headline percentage cannot fully describe.
There is a second measurement issue in the denominator. Executive ownership requirements are commonly expressed as a multiple of base salary, even though salary can represent only part of a FTSE 100 chief executive's annual remuneration once bonuses, long-term incentives and other awards are included.
A requirement equal to 600% of a £1.5 million salary, for example, represents £9 million of qualifying equity. Described as six times salary, that is substantial, but if the same executive's annual total remuneration is several times the base salary, the ownership requirement represents a much smaller multiple of one year's overall compensation.
Salary is not necessarily an inappropriate denominator. It is comparatively stable, transparent and avoids the volatility created by annual incentive outcomes. The distinction does mean that investors should understand what a headline figure such as “600% of salary” represents before interpreting it as the scale of an executive's personal financial exposure.
It also provides a useful counterpart to conventional CEO-pay analysis. The amount an executive receives and the amount of equity subsequently retained are related measures of remuneration and alignment, but neither can be inferred reliably from the other.
Executive ownership is also path-dependent. A chief executive who has spent a decade in the position has had years in which bonuses and long-term incentive awards can vest, dividends can accrue and shares can be retained; someone appointed six months ago has not.
Shell demonstrates how quickly the position can change. Wael Sawan's reported ownership stood at 676% of base salary at 31 December 2025, slightly below Shell's then 700% target. By 4 March 2026, Shell reported the figure at 1,073% of salary, while the target under its new arrangements was 900%, following changes that included the delivery of shares under incentive arrangements.
The timing difference also illustrates why applying a later share price to an earlier shareholding without accounting for intervening transactions could create an apparently precise but economically inaccurate estimate.
Rio Tinto appointed Simon Trott chief executive on 25 August 2025. At the end of that year he had reached 43% of the company's fixed-share ownership requirement, with Rio noting that he had only recently become CEO and would continue building his position. GSK's Luke Miels similarly became chief executive on 1 January 2026, with a salary of £1.375 million and an ownership requirement of 725% of salary.
A long-serving CEO's 15-times-salary position cannot therefore be interpreted automatically as evidence of greater personal conviction than a recently appointed CEO's two-times-salary holding. Part of the difference may simply reflect the accumulated history of remuneration and tenure.
Meg O'Neill's appointment at BP provides perhaps the clearest example of why different forms of equity cannot simply be aggregated.
After becoming chief executive on 1 April 2026, O'Neill received 1,901,277 restricted replacement shares and 809,232 performance shares, replacing remuneration forfeited when she left her previous employer. BP subsequently awarded another 1,415,701 performance share units under its 2026 incentive arrangements.
Those awards are potentially valuable and clearly give O'Neill an economic interest in BP's future performance, but they vest over time and, in some cases, remain subject to performance conditions. Adding more than four million conditional interests together and describing the result simply as O'Neill's personal BP shareholding would therefore answer a different question from how many BP shares she currently owns outright.
For investors, the disclosures point to at least three distinct measures: shares already owned or beneficially held; equity qualifying towards the company's formal ownership requirement; and additional potential exposure through deferred, restricted or performance-linked awards.
A further distinction is the source of the investment. Soriot's approximately £7.26 million AstraZeneca purchase, for example, represents a direct allocation of personal capital that an ownership multiple alone does not identify.
The appropriate measure depends on what an investor wants to understand. A remuneration committee assessing compliance with its ownership policy may reasonably include interests that an investor analysing current personal capital at risk would prefer to separate.
Finance Gazette initially examined whether the ten executives could be placed into a single ranking based on the value of their holdings relative to salary. Reviewing the underlying disclosures showed that such a ranking would imply a level of comparability that the available data do not support.
Definitions differ between companies, reporting dates vary, some executives have only recently taken their positions, holdings can change substantially between annual reports, and businesses treat deferred, vested and conditional interests differently. Rio Tinto adds the complication of holdings across its plc and Limited share structures.
A genuinely standardised ranking would require reconstructing each executive's outright ownership transaction by transaction at a common cut-off date, applying consistent treatment to connected-person holdings and employee share plans, converting foreign holdings using the same exchange-rate methodology and adopting a single definition of ownership.
Where those inputs cannot be established consistently, an exact league table would create false precision. The difficulty in constructing one is itself informative: the figures look standardised until theirunderlying definitions are examined.
Finance Gazette reviewed annual reports, directors' remuneration reports and subsequent regulatory shareholding disclosures for the chief executives of ten of the largest companies in the FTSE 100.
The analysis distinguishes between shares described by companies as beneficially owned or held, equity interests qualifying under company share-ownership policies, and conditional, deferred or performance-related awards. Company-reported ownership percentages have been retained as reported rather than recalculated into a purported common ranking because the purpose of the comparison is to examine differences between the underlying methodologies.
Subsequent transactions were considered where they materially affected interpretation of year-end figures, including Soriot's September 2026 AstraZeneca purchase and the change in Sawan's reported Shell ownership position during 2026.
The analysis does not assume that personally purchased shares necessarily produce better management decisions, nor that remuneration-derived shares provide ineffective alignment. Its purpose is narrower: to establish what economic interests sit behind figures that can otherwise appear directly comparable.
Executive ownership requirements are designed to give senior management meaningful exposure to the fortunes of the businesses they run. The Finance Gazette analysis does not establish that one company's approach is superior to another; different remuneration structures can legitimately pursue the same objective in different ways.
What the comparison does show is that headline CEO shareholding multiples cannot automatically be treated as like-for-like measures of personal financial exposure. Two executives reported at nine times salary may have substantially different combinations of outright shares, retained remuneration and qualifying awards, while a larger headline multiple can reflect tenure and the structure of previous incentive awards as well as shares already owned.
For investors assessing executive alignment, the headline percentage is therefore the beginning of the analysis rather than the end. What matters is not simply the ownership multiple a company reports, but the economic exposure represented by the interests counted within it.
