What actually drives an against recommendation, how to simulate the result while there is still time to act, and what a credible response to a low vote looks like.
Say-on-pay is advisory. That is exactly why a weak result hurts — there is nothing to comply with, only a public statement that shareholders do not accept how the committee is paying its executives.
By the time a company is reacting to a bad vote, the decisions that caused it are eighteen months old. The grant was made, the metrics were set, the disclosure was drafted, and the proxy advisers formed their view weeks before the meeting. The work that changes a vote happens long before the ballot.
Our team has run this cycle from inside large public companies — drafting the CD&A, briefing the committee chair, handling adviser engagement — and advised on it from outside. The two vantage points produce different advice, and the in-house one is usually the more practical.
Proxy advisers publish their methodologies and update them annually, so the specific tests and thresholds should always be read from the current-year policy rather than from memory. The underlying concerns, though, have been stable for years:
Metric selection and target setting are the say-on-pay decisions, whether or not anyone calls them that. A target that is obviously achievable at grant is a disclosure problem at payout. Model the payout curve against plausible performance ranges and ask what the resulting realised pay would look like in a table next to TSR.
Run the quantitative screens yourself, using the current-year policy and the adviser's likely peer group rather than the committee's. If the result is uncomfortable, there is still time to act on it. Discovering it in the adviser's report is discovering it too late.
The CD&A is the only place the committee gets to explain itself. Lead with the rationale rather than the mechanics, show the link between performance and payout explicitly, and address any discretion head-on. Committees consistently under-invest here relative to how much the document determines.
Shareholder outreach conducted only in years with a problem reads as damage control. Outreach conducted routinely builds the record that makes responsiveness credible when it matters.
A low vote creates an obligation to respond that compounds if unmet — a second consecutive weak result attracts board-level scrutiny, and in some markets adverse recommendations against remuneration committee members.
What a credible response looks like: identify what shareholders objected to specifically, through direct engagement rather than inference; make a change that is legible in the next proxy; and say plainly what you heard and what you did about it. Committees that survive a low vote well are the ones that treat it as information. Committees that do it badly explain why the shareholders misunderstood.
Related: peer group construction is where most pay-for-performance problems originate, and dilution and burn rate modelling is where equity plan proposals are won or lost.
In the United States it is advisory — the vote does not compel the board to change anything. In the United Kingdom the position differs: quoted companies must put their directors' remuneration policy to a binding shareholder vote at least every three years under section 439A of the Companies Act 2006, with an annual advisory vote on the implementation report.
There is no statutory threshold. In practice, support materially below the market norm draws attention, and a result in the region of 70 percent or lower is generally treated by boards and advisers as requiring an explicit response in the following year's disclosure. The more useful question is not the number but whether the company can identify what specific concern drove it.
The decisions that determine the vote are made twelve months earlier, at metric selection and target setting. Simulation against the current-year proxy adviser policy is best done around six months out, while there is still room to act. Work that begins when the proxy is being drafted is limited to explaining decisions rather than shaping them.
Yes, and against their likely peer group rather than the committee's. The most common surprise in this process is a company being measured against a comparison it never chose. Because the policies are republished annually, always run the current year's tests rather than last year's.
Committee cycle support, pay-for-performance simulation, and CD&A drafting — at partner level throughout.
Tell us what you’re working through and we’ll follow up directly.