5-Year CEO Comp Windows: What Boards Are Actually Buying

Comp committees are tying more CEO pay to longer and longer vesting windows. This is a bad idea, despite sounding intuitive.

In 2026, median public company CEO pay is up 14% YoY, with most long-term comp now vesting over 5 years instead of 3. Glass Lewis, NACD, and Harvard Corp Gov all blessed it. The intent was good. The math is unforgiving.

Take two packages.

Package A: $3M bonus, vests over 3 years. That's $1M/year, on paper.

Package B: $5M bonus, vests over 5 years. That's also $1M/year, on paper.

Same yearly comp. Very different risk.

Here's why. Each year's comp is roughly independent — some years the company does well and the CEO gets paid near the top of the band, some years it's a soft year and they get paid near the bottom. Stack 3 of those years into a 3-year total and you get one number. Stack 5 of those years into a 5-year total and you get a wider number, not because the average changed but because more time means more chances for the random ups and downs to land on different sides of average.

The math for how much wider is sqrt(5/3) ≈ 1.29. Square root of 5, divided by square root of 3. That ratio says a 5-year total is about 29% more spread out than a 3-year total, even when the expected value is identical.

Every dollar in the 5-year package is 29% more volatile than the same dollar in the 3-year package. Same expected pay. More risk per dollar.

They work less hard. The headline pay is the same, but the CEO's mental model shifts from "I'm earning $1M/year" to "I might earn $700k this year, I might earn $1.3M." That uncertainty isn't motivating. It's paralyzing.

They get risk-averse. When downside matters more, the CEO walks away from the big swing that could've added $5M to enterprise value, because the volatility on their comp would have been brutal in the down case. The expected value of taking the bet was positive. The realized outcome on the CEO's paycheck was a coin flip.

They leave at the first sign of bad news. This is the part most boards don't model. The CEO's next opportunity is anchored to that $1M/year expected number. If realized comp slips to $700k in a soft year, the gap to the next job market gets wide. So they walk — usually about 18 months into a soft cycle, framed as a "surprise."

Despite Glass Lewis and most governance shops endorsing these longer windows, it's going to cost shareholders quietly. To keep risk-adjusted pay neutral, nominal CEO pay will need to climb — a lot — to offset the embedded volatility. That'll be its own news cycle in a few years.

For a comp committee already squeamish about CEO pay levels, that's the deal they signed up for.