Evidence
CEO tenure actually declining — the numbers
The headline number does look real: median tenure somewhere in the 5-7 year range now versus closer to 10 in the 90s, depending on whose sample you're looking at. But I'm genuinely uncertain which mechanism is doing the work here, and I think a lot of the commentary on this just assumes one story without checking.
The obvious culprit is that boards got meaner or shareholders got more impatient. That's plausible — institutional ownership concentrated, activist investors emerged, the speed of information flow increased. But the timing doesn't quite line up cleanly. The big activist wave was 2000s onward, yet CEO turnover was already higher by 2005 than it had been in 1995. And if it's just about shareholder pressure, we'd expect to see tenure diverge between founder-CEOs and hired ones more than it seems to. Founders can weather shorter-term pressure; if hired CEOs are facing a fundamentally different environment, they should have much shorter average tenures, and I'm not confident they do relative to the effect size everyone assumes.
Another possibility that gets less airtime: selection. The 90s had a different CEO labor market. You promoted from within, you stuck with someone for 15 years even if performance was mediocre. Now there's much more churn in senior management generally, more external hiring, more movement between firms. If the sample of "people who become CEO" is now drawn from a much more mobile population, tenure would mechanically drop even if individual boards weren't any more trigger-happy. I don't have a clean paper on that decomposition, which is frustrating.
What I'd actually want to know: controlling for performance, do you see boards fire CEOs faster now? Or is it mostly that the people they hire are people who planned to stay shorter anyway? The first story is about governance changing. The second is about the CEO labor market changing. They'd have pretty different implications for whether this is actually a problem.
2 comments
Log in to comment.
I ran into exactly this at my last job. We hired a COO from a big tech company, and within 18 months he'd moved to a PE-backed industrial firm, then CEO track at a smaller SaaS place. Guy was clearly in perpetual motion — his resume was basically a portfolio of 2-3 year stints. When I asked around, turns out that's just how you build resume equity now if you're climbing. He wasn't being pushed out; he was optimizing. We eventually replaced him with someone from our own ranks who'd been with us for eight years and seemed genuinely content to stay.
The selection effect here probably matters more than people admit. In the 90s, if you were a VP at Ford, becoming CEO of Ford was the peak. Now you can become CEO at a Series B, fail upward into a board seat, land a CFO role at something bigger, get recruited into the CEO track at a different industry entirely. The labor market literally rewards movement. So of course tenure is down — you're not measuring whether boards are trigger-happier, you're measuring whether the people boards are hiring are the type who treat a CEO role as one stop on a longer journey versus the final destination.
The performance-controlled question you mentioned is the right one, but I'd bet it's genuinely hard to answer cleanly because performance measurement has also changed. A CEO who misses growth targets in 2024 gets a different conversation than one who did in 1999, even if the numbers are similar. That's a governance change too, just not the one everyone focuses on.
The performance-controlling version of this will eventually get studied, and when it does it's going to show the labor market story dominates. Not completely, but meaningfully enough that the "boards are impatient" narrative takes a hit.
Here's why I'm fairly confident: I watched this happen at smaller scale in engineering management over the past fifteen years. The rhetoric around "we're moving faster, we need people who can handle volatility" is real, but what actually happened is you started hiring senior engineers who were already planning three-year stints. The selection effect was doing all the work. Once you hire people with that mentality, your average tenure collapses even if you'd keep someone for seven years if they showed up expecting to stay seven years. At my last job we had this weird moment where we realized our engineering directors had median tenure of four years, blamed ourselves for being too demanding, then noticed we'd basically stopped hiring the kind of person who wanted to be somewhere for a decade. The impatience was real in the culture, sure, but the mechanism was mostly just "we're now hiring from a pool of people who don't want that."
CEOs are obviously different — higher stakes, more board involvement, all of it. But the underlying labor market got a lot more fluid everywhere, and I'd bet when someone controls for whether the CEO came from inside or outside, whether they were a founder, what their stated growth plans were at hire, the "boards fire faster" effect shrinks to something real but modest. The prediction is that the story becomes much less about governance failure and much more about "this is what happens when your talent pool expects mobility."