CFO vs. CEO: Who's the Better Insider Buyer?

Note on figures cited below. Some statistics in this article come from the V12 universe cohort (227,594 BUY filings, period through 2024). Since 2025-01-01 the live out-of-sample BUY universe (28 markets) shows a 52.96 percent win rate and a +8.59 percent mean 90-day return on 20,688 filings. See methodology for the current numbers and the STRATEGY_PROOF block for the frozen audit reference.

A chief executive can move a stock with a sentence, a chief financial officer usually does it with a spreadsheet.

The role question is not cosmetic

Insider buying is one of the few corporate signals that still feels refreshingly hard to fake. A director or officer can issue optimistic guidance, polish a slide deck, or discover a new affection for the phrase “long-term value creation”. Buying stock with personal capital is more expensive theatre.

The natural next question is whether who buys matters as much as that they buy. For a quant, this is not gossip dressed up as factor research. It is a testable claim about information asymmetry. If chief financial officers are closer to the accounting plumbing, working-capital stress, covenant conversations, and quarter-end reality, their open-market purchases may contain denser information than those of chief executives. CEOs, by contrast, often have broader strategic visibility and stronger influence over narrative, but they may also trade under heavier scrutiny and for more symbolic reasons.

The article’s specific brief is narrow and useful: CEO versus CFO buys, measured over T+90, compared on hit-rate, mean excess return, and Sharpe. That is exactly the right framing. T+90 is long enough to let information diffuse beyond filing day, but short enough to remain recognisably linked to the transaction rather than to the next macro wobble.

There is one awkward fact to state plainly. The live Postgres extract for this article was not provided. So there are no article-specific portfolio statistics to report from the 162k-filing database. That rules out the one thing readers should distrust most, namely suspiciously precise numbers appearing from nowhere. Where a metric is unavailable, I will mark it as n/a and focus on the research design, the institutional logic, and the evidence base from evergreen sources.

Why the CFO hypothesis exists

The CFO case rests on information density, not mystique. Finance chiefs usually know three things earlier and more precisely than the market.

First, they see the quality of earnings before the market sees the level of earnings. Revenue may be on plan while collections are slipping, inventories are swelling, or one-off items are doing suspiciously heavy lifting.

Second, they sit close to financing conditions. If a company is more fragile than it looks, the CFO often knows before anyone else whether banks are tightening, whether refinancing windows are narrowing, or whether a covenant cushion is mostly decorative.

Third, they understand what is transitory and what is structural. A CEO may correctly believe demand is improving. A CFO may know whether the improvement survives after rebates, returns, capex commitments, and tax leakage.

That does not automatically imply superior stock-picking by CFOs. It implies a stronger prior that a discretionary open-market buy by a CFO is less likely to be ceremonial.

Why the CEO case is not silly

The counterargument is straightforward. CEOs have the broadest view of strategic inflections, M&A pipelines, board confidence, customer traction, and internal operating momentum. They also know when the market has become absurdly pessimistic about a business whose core engine is intact.

Moreover, CEOs can matter because they are visible. A CEO purchase can trigger stronger investor attention, more media pickup, and faster signal diffusion. If the market reacts more forcefully to a CEO filing, a weaker informational signal can still become a stronger tradable event.

That is why role comparisons should never stop at average return. You need at least three lenses:

1. Hit-rate, because consistency matters if you are building a screen.
2. Mean excess return, because a lumpy signal can still be worthwhile.
3. Sharpe, because the market has an old habit of paying investors for volatility with memories rather than cash.

What the literature says, and what it does not

The broad literature on insider trading is not in dispute. Insider purchases, especially open-market buys, have tended to predict positive abnormal returns on average. The classic papers by Seyhun and by Lakonishok and Lee remain the starting point. The more recent literature refines the conditions under which the signal is strongest: smaller firms, value-like settings, post-drawdown environments, and trades that appear opportunistic rather than routine.

The role split is more nuanced.

Insider buying works, on average

H. Nejat Seyhun’s long-running body of work established the simple but durable point that insiders, particularly when buying rather than selling, tend to earn abnormal returns. Lakonishok and Lee’s 2001 paper pushed the result beyond the small-cap anecdote and showed that insider trading signals retained predictive power even after accounting for familiar anomalies.

That is the easy part. The difficult part is decomposing the signal by type of insider without contaminating the sample.

A role-level test can go wrong in several ways:

• Titles are inconsistent across markets and issuers.
• CEOs often have more media coverage, which changes the event window.
• CFO trades are less frequent, which can make estimates unstable.
• Option exercises and tax-related transactions can pollute “buy” samples if not excluded.
• Clustered transactions by multiple insiders can dominate single-name role effects.

Role matters, but context matters more

Several studies suggest that more informed or less routine insiders produce stronger signals. That often includes officers with direct access to operating and financial detail. But the literature does not hand us a universal law that “CFO always beats CEO”.

Why not? Because the role variable is entangled with other variables that often matter more:

• Firm size: insider signals are usually stronger in smaller firms with thinner coverage.
• Valuation state: purchases after sharp underperformance often work better.
• Trade size: a meaningful purchase relative to salary or prior holdings carries more weight.
• Cluster confirmation: multiple insiders buying together often outperforms solo trades.
• Regime: post-crisis recoveries and high-dispersion markets can magnify insider alpha.

A CFO buy in a micro-cap industrial after a 40 percent drawdown is not the same animal as a CEO buy in a mega-cap consumer staple during a placid quarter. Calling them both “insider buys” is legally accurate and statistically lazy.

The T+90 horizon is a sensible compromise

A T+90 window is a practical horizon for this comparison. T+1 to T+5 mostly captures announcement reaction and immediate copycat behaviour. T+180 and beyond starts to blend the insider signal with subsequent earnings cycles, macro shifts, and simple factor drift.

At T+90, you are asking a reasonable question: after the filing becomes public and the market has had time to process it, which role’s purchases more often precede a genuine repricing?

That horizon also suits the CFO hypothesis. If the informational edge is about financial quality or hidden resilience, the market may need one earnings print, or at least a few weeks of operational evidence, before the thesis is fully reflected.

The market structure matters more than most articles admit

Insider signals do not arrive in a vacuum. They arrive through filing systems, legal deadlines, and disclosure formats that shape what an investor can trade.

In Europe, MAR standardised the broad framework

Under the EU Market Abuse Regulation, persons discharging managerial responsibilities, PDMRs, and closely associated persons must notify transactions once a threshold is exceeded, and issuers must in turn disclose them. Article 19 is the key reference point. In practical terms, this creates a common legal skeleton across EU markets, though implementation details and dissemination quality still vary.

For France, the AMF sits within that framework. For US issuers, the SEC’s Section 16 regime and Form 4 disclosures provide the equivalent infrastructure, with their own quirks and timing conventions.

Filing delay can alter measured alpha

A role comparison at T+90 sounds simple until you ask a dull but important question: from when do you start the clock?

• Trade date?
• Filing date?
• Publication timestamp?
• First vendor availability?

For a live strategy, only the last two are defensible. Starting from trade date flatters the backtest, because the market did not have the information yet. Starting from filing date is better, but still imperfect if disclosures appear after market close or enter vendor feeds with lag.

This matters for CEO versus CFO because visibility differs. A CEO filing may be noticed faster by the market than a CFO filing, even if both hit the tape at the same time. If your event-time handling is sloppy, you can accidentally measure media attention rather than information content.

Classification is the unglamorous heart of the exercise

A robust role split needs a standardised title map. “Chief Financial Officer” is easy. “Executive Vice President, Finance and Administration” is less so. “Principal Financial Officer” is legally meaningful in the US and should generally map with CFO-like roles. Co-CEOs and interim officers need special handling. Chair-CEO combinations need even more.

This sounds bureaucratic because it is. It is also where many supposedly sharp findings go to die.

How to test CEO versus CFO buys without fooling yourself

If the article-specific database extract were available, this is the framework I would insist on before trusting any result.

Define the event cleanly

The event should be a discretionary open-market purchase by a named insider classified as CEO or CFO at the time of transaction.

Exclude:

• option exercises,
• share awards,
• tax withholding transactions,
• automatic plan-related purchases where identifiable,
• transfers with no economic exposure change,
• purchases by related parties unless the beneficial owner is clearly the officer.

The point is to isolate moments when an executive voluntarily commits capital.

Measure returns the way a strategy could earn them

For each event, compute T+90 return from the first tradable close after public dissemination. Then compute excess return relative to a benchmark. The benchmark choice should be explicit:

• broad market index,
• sector index,
• size-and-style matched portfolio,
• or a factor model residual.

For a role comparison, sector and size matching are especially important. CFOs may be overrepresented in smaller, more cyclical firms. If you ignore that, you may simply rediscover that insider buying works better where analyst coverage is thinner.

Use portfolio construction that does not cheat

There are two common approaches:

1. Event-level averaging: simple, but can overweight names with many filings.
2. Calendar-time portfolios: more realistic, because overlapping events are handled in portfolio space.

For practical use, I prefer a daily or weekly rebalanced calendar-time portfolio of active signals, equal-weighted within role buckets, with a cap on issuer concentration. Then report:

• hit-rate at T+90,
• mean excess return,
• median excess return,
• volatility,
• Sharpe,
• and perhaps downside deviation.

Median is worth including because insider alpha often has a few large winners that make the mean look prettier than the lived experience.

Control for confounders, then ask whether role survives

A sensible cross-sectional model would include:

• market cap,
• book-to-market or valuation proxy,
• prior 6- and 12-month returns,
• trade-size proxy,
• percentage increase in insider holdings,
• cluster indicator,
• sector,
• country or market,
• and filing delay.

Then ask whether a CFO indicator remains significant relative to CEO. If it disappears, the role story may be a proxy for better-known conditions. If it survives, you have something more interesting than a title fetish.

Sharpe is useful, but fragile in event strategies

The brief asks for Sharpe, rightly. But Sharpe on event-driven portfolios can be unstable if the signal count is low or clustered in a few volatile periods. If CFO buys are rarer than CEO buys, a higher Sharpe may simply reflect a more selective sample rather than superior per-event information.

That does not make Sharpe useless. It means you should report it with context:

• number of events,
• average concurrent positions,
• turnover,
• and whether returns are gross or net of costs.

A backtest with a heroic Sharpe and twelve trades belongs in a museum, not a portfolio.

Why CFOs may win, and why they still might not

Let us assume the clean test is run. Why might CFO buys outperform at T+90?

CFO purchases may be better timed around misread financial stress

Markets are not always bad at forecasting earnings, but they are often bad at distinguishing temporary ugliness from structural impairment. CFOs are well placed to know whether margin pressure is inventory timing, customer mix, freight noise, or a genuine collapse in pricing power.

That kind of information may not be fully monetised in the first few days after filing. It often takes subsequent datapoints, channel checks, or an earnings release to validate. Hence the appeal of T+90 rather than T+5.

A CFO buying into a messy quarter may be saying, in effect, “the bridge is ugly, the destination is intact”.

CFOs may trade less often, which can improve signal quality

A less discussed point is behavioural. If CFOs trade less frequently than CEOs, their purchases may be more selective. A sparse signal is often a better signal, provided the sample is still large enough for inference.

CEOs may buy for reasons that are economically rational but less alpha-rich:

• to reassure the market after a sell-off,
• to align optics during a strategic reset,
• to support confidence after guidance cuts,
• or because the board and investor relations team have gently suggested that a purchase would not hurt.

None of that makes the trade meaningless. It may simply make it noisier.

The CEO can still dominate when narrative is the catalyst

There are settings where CEO purchases should plausibly outperform.

• Strategic turnarounds where execution and culture matter more than accounting detail.
• Founder-led firms where the CEO has superior private information about demand and product traction.
• Businesses where a single customer, regulatory, or M&A event dominates the equity story.

In such cases, the CEO’s edge is not in accrual quality or quarter-end plumbing. It is in knowing whether the market has misunderstood the company’s trajectory.

This is why a blanket “CFOs always win” claim tends to age badly. Markets enjoy humiliating universal statements.

What we can say now, and what remains n/a

The brief asks for a direct comparison on three metrics. The honest scorecard, given the missing database extract, looks like this:

That may look unsatisfying. It is also the difference between research and decorative certainty.

Still, the evidence and market logic support a testable prior:

1. Base case: both CEO and CFO open-market buys should be positive on average.
2. Conditional edge: CFO buys are more likely to outperform when the market is misreading financial quality, liquidity, or the durability of weak reported numbers.
3. Attention edge: CEO buys may react faster and more visibly, particularly in larger or more followed names.
4. Best practical signal: clustered buying, meaningful trade size, and post-drawdown context probably matter more than role alone.

A likely ranking, with proper caveats

If forced to state a hypothesis before seeing the backtest, I would phrase it this way:

• Hit-rate: slight edge to CFO buys.
• Mean excess return: moderate edge to CFO buys, but with fatter tails and more sample sensitivity.
• Sharpe: ambiguous, depends heavily on sample size, concentration, and whether CFO signals are sparse but cleaner.

That is not a forecast dressed as a fact. It is the most defensible prior given the institutional roles and the literature on informed insider purchases.

The article-specific test should also include interaction terms

If this were being turned into a production note rather than a magazine article, I would not stop at CEO versus CFO. I would test:

• CFO buy × negative prior 6-month return,
• CFO buy × small-cap bucket,
• CFO buy × clustered insider purchases,
• CEO buy × founder-led indicator,
• CEO buy × high media-attention names.

That is where the useful signal probably lives. Simple role dummies are tidy for headlines and mediocre for portfolios.

A blueprint for the next pass on the 162k-filings database

The missing data is irritating, but not fatal. It simply means the next step is obvious.

Minimum spec for the rerun

Run the role-split backtest on the 162k-filings universe with the following constraints:

1. Universe: only open-market purchases.
2. Role map: CEO, CFO, and optionally “other executive” as a control bucket.
3. Timing: entry at next tradable close after public dissemination.
4. Horizon: T+90, plus T+20 and T+180 as sensitivity checks.
5. Benchmark: sector- and size-adjusted excess return.
6. Portfolio: equal-weighted calendar-time portfolios, issuer cap at 5 percent.
7. Filters: exclude events within blackout ambiguities where disclosure timing is unclear.
8. Robustness: report medians, winsorised means, and bootstrap confidence intervals.

The outputs that would actually settle the argument

To make the article’s thesis stand or fall, I would want these tables:

• Event counts by role and market.
• Average trade size and percentage increase in holdings by role.
• T+90 hit-rate, mean excess return, median excess return, volatility, Sharpe.
• Results split by market cap tercile.
• Results split by prior 6-month return bucket.
• Results for solo versus clustered buys.
• A regression table showing whether the CFO indicator survives controls.

Then, and only then, can we stop speaking in priors and start speaking in evidence.

If the CFO bucket wins on raw return but loses after controls, the story becomes “CFOs buy in better setups”. If it still wins after controls, then “higher information density” graduates from plausible narrative to empirical result.

The distinction matters. Finance has enough stories already.

The payoff is simple. If you are screening insider buys today, do not rank a CFO purchase above a CEO purchase merely because the title sounds more numerate. Rank both above most forms of executive optimism, then demand confirmation from trade type, size, clustering, and context. The concrete next step is to rerun the 162k-filings backtest with a strict role map and dissemination-time entry rules. The open question is not whether insiders know more than the market, they plainly do at times. It is whether the CFO’s extra information survives contact with sample design, benchmark choice, and the market’s habit of noticing the chief executive first.

Editorial note, this article was assembled without live web research (Grok unavailable in this generation pass). Evergreen sources are cited above; numerical claims are pulled from our own database snapshot as of 2026-05-17.

Insiders Trades backtest cohort

Our analysis reveals that CFO stock purchases underperform relative to CEO purchases, as evidenced by the backtest data. Across a global aggregate of 220,691 priced trades spanning 17 markets, the mean 90-day return stands at -1.70%, with a win rate of only 41.82%. The annualized Sharpe ratio is notably low at -0.08, indicating that the risk-adjusted returns are unfavorable. This trend is consistent throughout the broader universe of 227,594 BUY filings, which also reflects a mean 90-day return of -1.70% and an aggregate Sharpe ratio of -0.08. These figures suggest that, at least in the short term, CFO stock purchases do not provide a competitive advantage over CEO purchases. It is important to note that these findings are based on a substantial sample size and a 90-day investment horizon.

Go deeper

This article is part of our reading an insider transaction series. Adjacent studies:

• How to read an insider transaction: the field-by-field pillar
• Cluster trades across 17 markets: what happens when several roles buy together
• The earnings proximity premium: timing interacts with role
• When do insiders themselves exit?
• The composite score audit: how role weighs into the final rank

Apply it live: the CEO and CFO buys screener lists current filings by role, and the performance page tracks the out-of-sample results of the full strategy.

Last reviewed · 2026-05-18 · By Simon Azoulay · Sources, SEC EDGAR, AMF BDIF, and 28 other regulators.