Is September Bad for the Stock Market? Why It Doesn't Change a Business

Every year, right on schedule, the same story comes back. Summer ends, the calendar turns, and the financial press reaches for the headline it ran last year and the year before: the "September curse," the "scariest month for stocks," the warnings that the major indexes are "testing support." The coverage is dependable enough that you could set a watch by it — and every year a fresh wave of intelligent, recently liquid people wonder whether they ought to be doing something.

This piece is about that pull: where it comes from, and why it deserves far less attention than the volume of coverage implies. Not because the seasonal pattern is invented — there is a real number underneath it — but because that number describes something entirely different from what actually determines whether capital compounds over decades. For founders and family offices deciding where to put serious, long-horizon money, learning to tell those two things apart is one of the more freeing shifts available.

What the "September effect" actually is — and isn't

Start with the honest version of the claim, because the story does rest on something real.

Take a long history of monthly returns for the major U.S. stock indexes and average them, and September tends to come out near the bottom — frequently cited as the weakest month of the year on average.¹ This is a genuine statistical footnote, and it has shown up across a long stretch of market history, which is exactly why the story has legs: it isn't pure invention, and a journalist can point to an actual figure.

But it's worth looking closely at what that figure is and isn't telling anyone.

First, it's an average of averages, and the average hides enormous variation. Plenty of individual Septembers have been perfectly ordinary, and some have been excellent. "September is historically the weakest month" does not mean this September will be weak. It means that if you lined up many decades of Septembers, the typical one landed slightly below the others — and a slightly soft long-run average is compatible with any single September doing almost anything. It is a statement about a distribution, not a forecast for a date on the calendar.

Second, whatever tilt exists is small in magnitude and, more to the point, unreliable as a thing to act on. If the pattern were both large and dependable, sophisticated traders would long ago have sold in late August and bought back in October until the pattern was arbitraged away — traded out of existence. The fact that a faint seasonal lean persists at all is itself evidence that it's too weak and too noisy to build a strategy around. Those who have tried to systematically dodge September have, over full cycles, generally fared worse than those who simply stayed put, because dodging also means missing the recoveries and the strong years, all while paying taxes and transaction costs to churn in and out.

Third — and this is the part that matters most for an owner of businesses — the effect is a fact about how people trade shares, not about the businesses themselves. No factory produces less cash because the month is September. No software company loses its customers, no consumer brand surrenders its pricing power, no toll-road-like operator stops collecting its tolls because a page of the calendar turned.

The proposed explanations for the seasonal tilt make this obvious. Portfolio managers returning from summer and repositioning. Funds beginning to sell losing positions for tax reasons ahead of year-end. Thinner summer trading volumes that let small shifts move prices more than usual. And the simple self-fulfilling anticipation that comes from everyone having read the same article — participants expect weakness, sell to avoid it, and thereby create the very dip they feared. Every one of those forces lives in the behavior of the crowd. Not one of them lives inside the productive engines the crowd is trading.

That distinction — between the price of a thing and the thing itself — is the whole game.

Price is a vote. Cash flow is a weight.

There's an old line, attributed to Benjamin Graham — the investor and teacher whose thinking shaped much of modern value investing² — that the market is a voting machine in the short run and a weighing machine in the long run.² It rewards slowing down on, because it isn't just a tidy aphorism. It's a precise description of two systems running on two entirely different clocks.

In the short run, a stock price is a vote. It reflects the mood, the fear, the positioning, and the attention of everyone trading that day. Votes are emotional, contagious, and easily swayed by a headline. A "September is scary" story is, quite literally, an attempt to move the vote — to change how a crowd feels this week — and it can succeed. Prices do wobble in September, sometimes sharply. To anyone whose mental picture of their wealth is the number on the screen, a September wobble feels like something happening to them.

Over the long run, though, the market becomes a weighing machine. Across five, ten, twenty years, the price of owning a business converges on something far less emotional: the cash that business throws off and can hand back to its owners.

Free cash flow is the money a business generates after paying for everything it needs to keep running and to grow — wages, taxes, equipment, reinvestment. It's the cash that is genuinely left over and belongs to the owners. A business that grows that leftover cash, year after year, becomes worth more over time for reasons that have nothing to do with the calendar and everything to do with the enterprise.

Here is the mechanism, because it deserves to be understood rather than asserted. Picture a share of a business that generates one dollar of free cash flow per share this year, and suppose that dollar grows to two dollars over a decade as the business gets larger, more efficient, and better at charging fairly for what it makes. The weight of what an owner holds has doubled. Along the way, the vote — the daily price — will have swung around wildly, soaring in exuberant stretches, sagging in frightening ones, dipping in more than one September. But the two dollars of cash flow is a fact about the business. Eventually the weighing machine catches up to it. Owners who kept holding through the votes captured the full weight. Owners who sold during a scary September to escape a wobble sometimes sold the weight to avoid the vote — and then had to decide, uncomfortably, when to buy it back.

Two durable examples illustrate how cash economics operate on their own schedule, entirely independent of the market calendar. A mission-critical enterprise software provider with high switching costs and multi-year recurring contracts continues to bill and collect on the same schedule whether the broader index is rising or falling. Its free cash flow trajectory depends on customer retention, expansion within existing accounts, and disciplined product development — not on September trading volumes. Similarly, a dominant industrial supplier with entrenched customer relationships and pricing discipline tied to essential inputs sees demand governed by its customers' long-term production plans. Seasonal market sentiment may move the quoted price of its shares, but it does not change the volume of orders or the margins earned on those orders. The calendar is simply irrelevant to the operating rhythm of either business.

This is why the instinct of a founder is often better calibrated than that of a trader. Anyone who has actually built and run a company knows the difference in their bones. An operator with a business held outright doesn't receive a price quote every morning, and certainly doesn't panic in September. They watch the things that matter: whether customers are renewing, whether margins are holding, whether demand is growing, whether cash is building in the account. They value the company the way the weighing machine does — by its capacity to generate cash over time.

The strange thing that happens to many people after an exit is that they stop thinking like the owner they were and start thinking like a trader they never were, simply because now there's a screen with a number that moves every day. The seasonal-scare headline is aimed squarely at that borrowed trader mindset.

Why the story keeps coming back — and who it's really for

It helps to be clear-eyed about why this coverage exists, because once the machinery is visible, the pull loses most of its grip.

Financial media runs on a daily and weekly clock. It has to produce something every single day, whether or not anything of lasting consequence has actually happened. Most days in markets, nothing of long-term consequence has happened — the businesses underneath are quietly doing their work. But "nothing important happened today" is not a publishable headline. So the industry has built a deep repertoire of stories that manufacture significance out of ordinary noise, and the calendar is one of its most reliable tools. September's supposed curse is a story that writes itself, needs no new reporting, and dependably draws clicks from anxious readers. It will run next September, and the one after that, regardless of what any actual business anywhere is doing.

None of this is a conspiracy — it's an incentive structure. The media clock and the compounding clock are simply different clocks. The people producing daily market coverage are, for the most part, doing their jobs honestly on a horizon of hours and days. The trouble starts only when someone whose real horizon is decades — someone stewarding capital meant for the next generation or a foundation — borrows the emotional clock of someone whose horizon is the next headline. That mismatch is where costly decisions get made.

Consider the behaviors the September story tends to encourage:

• Selling to "get out before the drop" requires being right twice — once about when to leave and again about when to return — and usually just converts a paper wobble into a realized tax bill and a lingering question of when to climb back in.
• Sitting in cash "until things calm down" feels prudent and quietly isn't, because the businesses one would otherwise own keep compounding their cash flow while the waiting continues — and the media has no incentive to ever announce that calm has arrived.
• Trading activity in general is where fees, taxes, spreads, and mistakes live. Every round trip is a small tax on compounding, and seasonal scares are engineered to prompt round trips.

The common thread is that none of these behaviors touches the actual source of long-term return — the growth of free cash flow across many years. They all operate purely on the vote, while the weighing machine keeps running underneath.

Where long-horizon owners direct their attention instead

The practical difference appears in what questions get asked when the seasonal commentary arrives.

Rather than asking what the aggregate index will do over the next thirty days — an inherently noisy question with little bearing on any single business — the more useful questions concern the specific companies under consideration or already held. Has the durability of demand changed? Are margins stable or expanding? Is management allocating the cash the business generates in ways that increase future free cash flow per unit of ownership? These questions are answerable through operating results reported on company schedules, not market calendars.

Owners who have run businesses themselves recognize the difference in cadence. While operating a company, attention naturally fell on customer metrics, cost discipline, and cash accumulation — data points that arrived at the pace of real operations. The same discipline translates directly to capital allocation: periodic review of the fundamentals that drive free cash flow, with less weight given to daily price discovery that largely reflects the collective mood of other participants.

Free cash flow compounds multiplicatively, and that matters enormously over long horizons. A business that steadily grows its annual free cash flow per ownership unit creates far more value than one whose cash flow merely fluctuates with sentiment. The calendar does not accelerate or retard that growth rate. Acting as though it does substitutes a short-term participation pattern for the slower, more durable process of enterprise improvement — and that substitution has a cost that shows up not in any single September but across the full arc of a compounding period.

The September pattern will likely continue to appear in future data for the same institutional and behavioral reasons that produced it historically. That persistence will remain a fact about how shares change hands — about the vote. It will not be a fact about the cash a well-run business produces over the next decade. For owners whose horizon matches the compounding horizon of durable enterprises, keeping that distinction clear is less a matter of discipline than of simply remembering what they already know from having built something themselves.