13F Window Dressing: When Institutional Holdings Lie

Every quarter, thousands of institutional managers file a Form 13F disclosing what they own, and retail investors treat those filings as a window into “smart money” conviction. Some of it is. But a well-documented slice of what shows up in a 13F is 13F window dressing: managers temporarily buying recent winners and dumping recent losers right before the reporting date, so the portfolio that becomes public looks better than the one that actually existed all quarter.

This isn’t a fringe theory. It’s been measured in pension funds, mutual funds, and, as of a 2025 paper, ESG-labeled funds gaming their own sustainability scores. This piece walks through that research, shows a live example from the June 30, 2026 quarter-end, and gives you a concrete way to test whether any given 13F holding is conviction or decoration.

What 13F Window Dressing Actually Is

Under SEC staff guidance, any institutional investment manager exercising discretion over $100 million or more in Section 13(f) securities must file Form 13F within 45 calendar daysof each quarter’s end. That single design choice — a public, dated snapshot filed only four times a year — is what creates the incentive. Managers know a moment is coming when their entire portfolio becomes visible to clients, competitors, and financial media at once, and unlike day-to-day trading, that snapshot gets scrutinized in isolation.

MarketPeel has already covered the basics of reading a single 13F filing. Window dressing is a different problem: it’s not about learning what the form contains, it’s about recognizing when the form is telling you something cosmetic. Per one of the earliest studies to document the practice, the underlying behavior is simple: a manager embarrassed by a stock’s recent performance sells it before the reporting date, and buys stocks that already went up so the portfolio looks correctly positioned all along — even if it wasn’t.

Four Decades of Academic Evidence

This isn’t speculation — it’s one of the more consistently replicated findings in institutional-investor research, spanning three separate eras and manager types.

Pension funds (1991). The original study, published as an NBER working paper and later in the American Economic Review, examined 769 pension funds managing a combined $129 billion in assets. It found that fund managers disproportionately sold underperforming stocks relative to their holdings, and that this selling accelerated in the fourth quarter — precisely when sponsors scrutinize portfolios most closely.

Mutual funds (2014). A study of 59,060 quarterly reports from 2,623 equity mutual funds spanning September 1998 to December 2008, later published in The Review of Financial Studies, found that window-dressing funds disclose disproportionately higher holdings in recent winners and lower holdings in recent losers at quarter-end — and that these funds go on to show worse future performance, both short- and long-term, along with higher turnover and trading costs. As the Harvard Law School Forum on Corporate Governance summarized it, window dressers receive higher investor inflows when reporting-delay performance happens to be good, and lower inflows when it’s bad — a “risky bet” that explains why managers do it despite the long-run downside.

Price-level evidence (1997).Professor David Musto’s analysis of mutual funds from 1985 to 1997, covered by Knowledge at Wharton, found funds outperformed the S&P 500 by about 0.5 percentage points on the final trading day of the year — then underperformed by about 0.6 percentage points on the first trading day after. Critically, the pattern showed up only at quarter-ends and year-ends: “other month-ends show nothing, despite many more observations,” which rules out routine explanations like paycheck-driven inflows.

The Quarter-End Price Effect: What the Data Shows

Musto’s research ties this behavior to a specific SEC concept: “marking the close,” defined as attempting to influence a stock’s closing price by executing purchase or sale orders at or near the close of the market. It’s a related but distinct tactic — the goal is to nudge the reported value of a holding right before the number gets locked in for the quarter, not just to change what’s held.

You don’t need a two-decade-old dataset to see the pattern in action. On June 30, 2026, the last trading day of Q2 2026, the Nasdaq Composite rose about 1.1% and the S&P 500 rose about 0.6%, with chip stocks and Apple leading gains. The Motley Fool’s coverage explicitly attributed part of the move to “typical quarter-end dynamics,” noting institutional investors rebalanced portfolios and “engaged in window dressing ahead of their mid-year reports.” None of this proves any single manager gamed their book that day, but it’s a live illustration of the same incentive researchers have measured since 1991.

How Window Dressing Shows Up Inside a 13F Filing

The concrete pattern to watch for is straightforward. Per an explainer on institutional ownership data, holdings that appear as a brand-new position in one quarter’s 13F and then disappear entirely from the very next filing are frequently the product of window dressing rather than a genuine investment decision. A real conviction position tends to survive into the following quarter, even if the size changes; a cosmetic one often doesn’t survive at all.

A structural quirk of the form itself worsens this. Per the University of Miami Business Law Review, Form 13F only requires disclosing certain long equity positions — it does not require disclosure of short positions, cash holdings, or many derivatives. That means a filing showing an increased equity stake can be misleading even without any window dressing at all, since the same manager might simultaneously hold options or a short position that isn’t required to appear anywhere on the form, reversing the apparent bullish signal entirely.

The Persistence Test: A Practical Way to Filter Signal from Decoration

You don’t need proprietary data to run this check. It takes two or three consecutive quarters of the same manager’s 13F filings, all free on EDGAR.

Green Window Dressing: The 2026 Twist on an Old Trick

The newest wrinkle on this decades-old behavior applies it to ESG ratings instead of raw performance. A paper by Gianpaolo Parise and Mirco Rubin, published in The Journal of Financein 2025, finds that mutual funds strategically time trades of ESG stocks around disclosure dates to inflate their sustainability ratings. Funds’ ESG exposure increases shortly before disclosure and decreases shortly afterward, and ESG stock prices themselves temporarily rise before disclosure and decline afterward.

The mechanism is identical to classical window dressing — only the dimension being gamed has changed. Funds have an incentive to look good on whatever attribute becomes visible the moment their holdings are disclosed, and the end of each calendar quarter is exactly when Form 13F requires that disclosure. The same 45-day clock that enables performance-chasing window dressing creates an identical incentive to game an ESG score instead.

Why Regulators Rarely Police This Directly

Window dressing sits in a genuine gray zone. It isn’t explicitly banned in the way insider trading or outright market manipulation are. The Form 13F rules require accurate disclosure of what a manager holds as of the reporting date — they don’t require the manager to have held the same portfolio all quarter, and they don’t police why a trade happened near quarter-end.

That’s also why enforcement is hard without direct evidence of intent. A manager who sells a losing position on the last day of the quarter can always say the trade reflected a genuine change in view, and proving otherwise requires internal communications or trading-pattern analysis regulators reserve for higher-priority cases. As the Miami Business Law Review notes, Form 13F’s structure (the 45-day lag, the ability to file late to avoid tipping off competitors, the exclusion of shorts and cash) already limits how much any single filing reveals, independent of any deliberate dressing.

What This Means for Reading 13F Data

None of this means 13F data is useless — it means a single quarter’s snapshot deserves less weight than a pattern across quarters. A position that shows up new, survives, and grows across two or three filings is meaningfully more informative than one that appears once and vanishes. And because most managers file as late as possible to avoid tipping off competitors, even a “fresh” 13F is already describing a portfolio that may no longer exist in that exact form.

This is also why 13F analysis pairs well with other institutional signals rather than standing alone. MarketPeel has covered how to measure crowding across many managers’ 13F filings at once, a related but separate question: crowding asks how many managers independently arrive at the same stock, while window dressing asks whether a single manager’s disclosed portfolio reflects what they actually held all quarter. The persistence test above, paired with a crowding check, gives a fuller read on whether a 13F holding is telling you something real.