The Role of Filing Clocks in Insider Trading Dynamics

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 filing clock is not a footnote, it is a trading variable.

The comparison between SEC Section 16 and the EU Market Abuse Regulation is often presented as a matter of legal style. That is too polite. For anyone trading around insider disclosures, the real issue is market microstructure disguised as compliance.

The timing gap, two business days in the United States versus three under MAR for persons discharging managerial responsibilities, is only the visible part. Underneath sit different definitions of who reports, what gets reported, when trading is allowed, how pre-arranged plans are treated, and how fast the information reaches a machine-readable feed that a desk can actually trade on.

This matters because insider filings still move prices, even after decades of academic demystification. The effect is uneven, context-sensitive, and often strongest in smaller names or when a trade looks informative rather than administrative. A one-day disclosure gap is not a licence to print money. It is, however, enough to matter in names with thin analyst coverage, dual listings, or investor bases that do not watch foreign regulatory plumbing.

Two clocks, two philosophies

Section 16 is a personal reporting regime with a short fuse

Section 16 of the US Securities Exchange Act applies to directors, officers, and beneficial owners of more than 10 percent of a registered class of equity securities. The practical filing workhorse is Form 4. Most reportable changes in beneficial ownership must be disclosed before the end of the second business day following the execution date.

That sounds straightforward. It is not, but it is at least concentrated. The SEC regime is built around a relatively crisp category of insiders, a standard form, and a filing system, EDGAR, that is centralised and machine-consumable. The market has had years to wire this into alerts, parsers, and event studies.

The result is a disclosure environment that is both noisy and efficient. Noisy, because Form 4 captures many transactions that carry little informational content, such as tax withholding on vesting or option exercises paired with sales. Efficient, because the filing path is standardised enough that the market can sort the useful from the administrative quickly.

MAR is broader in concept and more fragmented in practice

Under Article 19 of MAR, persons discharging managerial responsibilities, PDMRs, and persons closely associated with them, PCAs, must notify the issuer and the competent authority of transactions conducted on their own account relating to the issuer’s shares, debt instruments, derivatives, or other linked instruments. The notification must occur promptly and no later than three business days after the transaction.

The issuer must then make that information public, also promptly and no later than three business days after the transaction. In theory this is harmonised across the EU. In practice, implementation sits on top of national competent authorities, local filing channels, and issuer disclosure habits. The legal rule is European. The operational experience is still partly national.

For traders, this distinction is not academic. A US watcher can rely on EDGAR as the primary collection point. A MAR watcher often needs a mix of regulator sites, issuer announcements, exchange feeds, and local formatting quirks. Harmonisation is a noble aspiration. Data engineering remains a less romantic reality.

The one-day gap is real, but the usable gap is smaller

A clean statement would be that the US gives the market the information one business day sooner. Sometimes yes. Often not exactly.

Form 4 is due within two business days, but many filings arrive after market close. MAR notifications may also appear outside trading hours. Some issuers publish very quickly, effectively compressing the difference. Others use most of the allowed window. Weekends and local holidays add further distortion. A Thursday trade can produce a different practical lag profile than a Monday trade, especially across jurisdictions.

So the arbitrage window is not “US fast, EU slow”. It is “US usually faster and more standardised, EU often slower and more dispersed”. That is less catchy, but more useful.

What each regime says a trade means

Form 4 often tells you more about the transaction mechanics

US Form 4 has a reputation for over-reporting trivia. That reputation is deserved, but incomplete. The form can also be rich in detail. It distinguishes transaction codes, direct versus indirect ownership, derivative and non-derivative securities, and footnotes that explain whether a sale was discretionary, tax-related, or plan-based.

For a trader, this is a gift and a nuisance. The gift is granularity. The nuisance is that granularity requires interpretation. A chief executive selling shares to cover taxes after vesting is not the same signal as an open-market purchase funded with cash. Yet both can appear under the same broad umbrella of insider activity until someone reads the footnotes.

This is why simple “insider bought” screens age badly. The edge is in classifying the transaction properly before everyone else does.

MAR disclosures can be broader in scope but thinner in explanatory texture

MAR captures transactions in a wider conceptual perimeter, including debt instruments and linked derivatives. It also extends to persons closely associated with PDMRs, which can widen the net around family and controlled entities. That breadth is useful. It can surface activity that a narrower US-style officer-director lens might miss.

But MAR disclosures are often less standardised in the way global investors would prefer. Depending on the issuer and market, the announcement may provide enough detail to classify the trade cleanly, or just enough to satisfy the rule. The legal form is there. The analytical convenience is not always.

This creates an odd asymmetry. MAR may catch more kinds of economically relevant transactions, but the US often packages the information in a way that is easier for systematic traders to parse quickly.

Thresholds and exemptions shape what never reaches your screen

Another structural difference is not about timing at all. It is about what gets filtered out before disclosure.

Under MAR, notification obligations only apply once a total amount of transactions has been reached within a calendar year, with the default threshold set at EUR 5,000, though national competent authorities may raise it to EUR 20,000. This means some low-value activity may remain below the public radar until the threshold is crossed.

Section 16 does not use the same annual threshold logic for Form 4 reporting. The result is a different noise profile. US filings may include more small or routine reportable events. MAR may suppress some minor activity, then reveal it only once cumulative value trips the threshold.

For event traders, that means a MAR disclosure can sometimes represent not just today’s trade, but the point at which a year-to-date accumulation became reportable. The market may read it as a fresh signal. Legally, it may be partly an accounting of already-completed activity.

10b5-1 plans versus MAR closed periods

The US tolerates pre-arranged selling, then asks for better labelling

Rule 10b5-1 plans are one of the most misunderstood features of US insider trading regulation. They do not bless every insider sale with innocence. They create an affirmative defence when trades are executed under a binding plan adopted in good faith when the insider was not aware of material nonpublic information, subject to conditions that the SEC tightened in 2022.

Those amendments introduced cooling-off periods for directors and officers, certifications, restrictions on overlapping plans, and enhanced disclosure requirements. Form 4 now requires checkbox disclosure indicating whether a reported transaction was intended to satisfy Rule 10b5-1(c).

Mechanically, this matters because 10b5-1 plans convert some insider trading into a more scheduled flow. The market often discounts plan-based sales relative to discretionary open-market sales. Usually sensibly. Not always correctly.

A plan-based sale still contains information, just of a different kind. It may reveal an insider’s medium-term liquidity preferences, diversification needs, expected vesting cadence, or confidence in future price levels when the plan was adopted. The signal is weaker than a spontaneous purchase, but weaker is not the same as worthless.

MAR leans harder on closed periods and issuer consent

MAR does not have a direct equivalent to the US 10b5-1 architecture. Instead, Article 19 imposes a closed period, generally 30 calendar days before the announcement of an interim financial report or year-end report, during which a PDMR shall not conduct transactions on their own account or for the account of a third party, directly or indirectly, relating to the issuer’s securities or linked instruments.

There are exceptions, including severe financial difficulty and transactions made under, or related to, employee share or saving schemes, qualification or entitlement of shares, and certain transactions where the beneficial interest does not change. ESMA has also issued guidance on what may qualify and how issuers should assess exceptional circumstances.

The philosophical difference is clear. The US says, in effect, if you pre-commit under the right conditions, we can tolerate execution later. MAR says, more sternly, there are periods when you generally do not trade, unless a recognised exception applies.

For traders, the distinction changes the baseline

This difference has a practical consequence that is easy to miss. In the US, a cluster of insider sales may be mechanically generated by 10b5-1 plans and therefore less tied to current information conditions. In Europe, the absence of a broad 10b5-1 analogue means observed trades outside closed periods may contain a somewhat higher discretionary component, though this should not be overstated.

At the same time, MAR’s closed periods can bunch activity. If insiders delay permissible transactions until the window reopens, post-close-period filings may reflect calendar mechanics as much as conviction. A trader who treats every post-window sale as a fresh negative signal is volunteering to subsidise better-informed readers.

The right question is not “which regime is stricter?” It is “which regime creates more discretionary signal per disclosed trade?” The answer varies by transaction type, issuer policy, and how faithfully the market reads the surrounding context.

Where the arbitrage windows actually appear

Window one, US information reaches Europe before Europe catches up

The simplest case is a dual-listed or globally followed issuer where a US insider files a Form 4 and the European investor base is slower to digest it. Because EDGAR is centralised and heavily monitored, US desks may react within minutes. European price adjustment can lag if local coverage is thinner or if the issuer’s European line is less liquid.

This is not a guaranteed free lunch. In large-cap names with active arbitrage desks, the lag may be measured in seconds. In smaller cross-listed names, ADRs, or issuers with fragmented ownership, the lag can be longer. The practical edge depends on whether your workflow captures the filing, classifies the trade correctly, and maps it to the right instruments before everyone else.

Window two, MAR disclosures can be under-watched by US screens

The reverse case is less discussed and often more interesting. Many US-centric insider monitors are built around EDGAR and US officer-director logic. They do not systematically ingest MAR PDMR disclosures from AMF, BaFin, or issuer websites. That creates blind spots.

A meaningful PDMR purchase in France or Germany may be public, lawful, and still underpriced in the attention market because the disclosure did not pass through the US data supply chain that many quant and event-driven shops rely on. This is not regulatory arbitrage in the illicit sense. It is attention arbitrage caused by different disclosure plumbing.

The irony is excellent. Markets that pride themselves on efficiency can still be tripped by file formats and jurisdictional habits.

Window three, amendments and late filings

Both regimes generate amendments, corrections, and occasional late filings. These are fertile ground for false positives.

A Form 4/A may correct share counts, transaction codes, or ownership characterisation. A MAR correction may clarify transaction nature or aggregate volume. If your system treats every amended filing as a fresh event, your backtest will look better than your PnL. Briefly.

The more subtle issue is that late filings themselves can carry information about governance quality or compliance discipline. The market does not always price that separately, but in smaller issuers repeated tardiness can be a legitimate red flag. It is just not the same trade as “insider bought stock”.

Window four, transaction taxonomy mismatch

Cross-regime strategies fail quietly when they assume categories line up neatly. They do not.

A US “sale” may be a tax-withholding event. A MAR “acquisition” may arise from a grant or vesting-related mechanism. Derivative treatment differs. Closely associated persons add another layer. If you compare raw buys and sells across Section 16 and MAR without normalising transaction origin, you are not doing comparative regulation analysis. You are sorting legal leftovers into a spreadsheet.

How to build a watcher that survives contact with reality

Start with timestamps, not headlines

The first discipline is timestamp integrity. Record at least three moments where possible: execution time, filing or notification time, and public dissemination time. The legal deadline uses one clock. Market reaction uses another. Your strategy lives or dies on the second.

For MAR, this is especially important because the chain can involve PDMR to issuer, issuer to market, and regulator publication. The legally relevant deadline may be met even if your chosen data source lags. If you backtest on regulator page timestamps alone, you may be measuring your scraper’s punctuality rather than the market’s information set.

Build a transaction ontology before you run regressions

A useful insider-monitoring system needs a taxonomy. At minimum:

• open-market purchase
• open-market sale
• option exercise without sale
• exercise-and-sell
• tax withholding or sell-to-cover
• grant or award
• transfer between ownership forms
• plan-based execution
• PCA-related trade
• derivative-linked adjustment

Without this, the model will overstate the value of “insider activity” because it will mix high-signal discretionary buys with low-signal administrative noise.

In the US, transaction codes and footnotes help. Under MAR, the work is often more manual or issuer-specific. That is annoying, but also where some of the remaining edge lives.

Treat purchases and sales asymmetrically

This is one of the oldest findings in the literature and still one of the most ignored in production systems. Insider purchases tend to be more informative than insider sales. Sales are contaminated by diversification, tax, liquidity, and planning motives. Purchases require an insider to put fresh capital at risk.

The rule survives both Section 16 and MAR, though the exact magnitude varies by market and sample. If you are forced to be simple, be simple in the right direction. Weight purchases more heavily. Discount sales unless they are unusually large, clustered, discretionary-looking, and unsupported by obvious plan or compensation mechanics.

Use issuer context to filter the legal signal

The same filing means different things in different issuers. A founder-led small cap with concentrated ownership, sparse coverage, and one meaningful open-market buy is not the same setup as a mega-cap where executives routinely sell under standing plans.

Likewise, under MAR, a PDMR purchase in a lightly covered continental mid-cap can be more informational than a mechanically timed transaction in a large bank with dense governance procedures and regular compensation events.

The filing is the start of the work, not the end. Regulation gives you the event. Context gives you the trade.

What the academic and regulatory record supports

The broad evidence, yes, insider trades still matter

The academic literature has long found abnormal returns around insider transactions, especially purchases, though the effect has attenuated in some markets as disclosure improved and strategies became crowded. Studies in the US and Europe generally support the idea that legal insider trades contain information, with stronger effects in smaller firms and in transactions that look more discretionary.

That should temper both extremes. The romantic view, insiders always know and always win, is nonsense. The dismissive view, filings are old news and fully arbitraged, is also too neat. Markets are efficient enough to punish laziness, not efficient enough to eliminate every operational delay and classification error.

The regulatory record, disclosure quality has improved, comparability has not fully converged

SEC reforms around 10b5-1 and executive compensation disclosure have made US insider-sale interpretation somewhat cleaner than it used to be. ESMA guidance has improved consistency under MAR, especially around closed periods and managers’ transactions. But cross-jurisdiction comparability remains imperfect.

That is unlikely to disappear soon. The US has one federal securities regulator and one dominant filing system for this purpose. Europe has a harmonised regulation administered through national competent authorities and issuer-level publication channels. Same broad objective, different administrative metabolism.

The quote is regulator-safe and almost aggressively unobjectionable. It is also correct. Transparency helps. It just helps unevenly, depending on how quickly and cleanly the market can consume it.

The trader's checklist for Section 16 versus MAR

If you monitor both regimes, the mechanical checklist is not glamorous, but glamour is rarely a source of alpha.

1. Capture both channels natively. EDGAR for Section 16, plus regulator and issuer channels for MAR.
2. Normalise timestamps. Execution date is not dissemination time.
3. Classify transaction type before ranking signal strength.
4. Flag 10b5-1 indicators in US filings.
5. Flag MAR closed-period adjacency and exception language.
6. Track amendments separately from original events.
7. Weight purchases more than sales.
8. Adjust for issuer size, liquidity, and listing structure.
9. Watch PCAs under MAR, but with extra scepticism about motive attribution.
10. Do not assume a one-day legal deadline gap equals a one-day tradable edge.

That last point deserves repetition. The legal asymmetry creates the possibility of an arbitrage window. It does not guarantee one. The window narrows sharply in names with good data coverage and active cross-market arbitrage. It widens in neglected names, fragmented feeds, and situations where the filing’s true nature is hidden in footnotes or local disclosure conventions.

The practical payoff is simple. If you want to trade insider disclosures across the US and Europe, stop treating regulation as static background. It is part of the signal-generation process. The next concrete step is to build a cross-regime event sample with transaction-type labels and dissemination timestamps, then test whether the residual alpha survives after removing plan-based US sales and threshold-distorted MAR events. If it does, you have something worth staffing. If it does not, you have at least avoided confusing disclosure law with edge.

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 of insider trading through the lens of filing clocks reveals significant insights from our backtest data. In the US, we observed 187,150 priced trades within a 90-day horizon, yielding a mean return of -2.28% and a win rate of 40.37%. The annualised Sharpe ratio for this cohort stands at -0.09, indicating that the risk-adjusted returns are less than favorable. In a broader global context, encompassing 227,594 BUY filings across 17 markets, the aggregate mean return is slightly better at -1.70%, with an aggregate Sharpe ratio of -0.08. These figures underscore the challenges associated with capitalizing on insider trading signals, as the negative returns suggest that the timing and execution of trades in relation to filing clocks may not be as advantageous as previously assumed. It is important to note that these findings are based on a specific sample size and time horizon.

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