Insider trading: what is prohibited, what is allowed and what directors' dealings reveal

In depth12 min read · 24 sources

Insider trading means trading on inside information, i.e. on knowledge that others do not have and that could move the price significantly. It is prohibited and a criminal offence in the EU and in the US. What is allowed, on the other hand, is for management and supervisory board members to trade shares in their own company, as long as they do not use inside information; they must report these directors' dealings. Here you will learn where the line runs, how supervisors track down insiders and what research has found about reported insider purchases. An overview, not legal advice.

Key terms: what the Market Abuse Regulation prohibits

At the heart of European law is the Market Abuse Regulation (MAR) of 2014, most recently amended in 2024 by the so-called Listing Act; it applies directly in all EU member states (MAR 2014). There are four terms you should know:

• Inside information (Art. 7): precise information that has not been made public, relating to an issuer or a financial instrument, which would be likely to have a significant effect on the price, i.e. information a reasonable investor would probably use as part of the basis of an investment decision. An intermediate step in a protracted process, such as ongoing takeover talks, can also qualify.
• Insider dealing (Art. 8): buying or selling financial instruments on the basis of inside information, for yourself or for others, directly or indirectly. Cancelling or amending an order placed earlier also counts as using it.
• Recommending and inducing (Art. 8(2)): advising a third party to trade on the basis of inside information, or getting them to do so. Anyone who follows the recommendation and knows, or ought to know, what it is based on is engaging in insider dealing themselves.
• Unlawful disclosure (Art. 10): passing on inside information, except in the normal exercise of an employment, a profession or duties.

Art. 14 prohibits all of this, and in the case of insider dealing even the attempt. It is not only board members who are affected: Art. 8(4) lists members of governing bodies, shareholders, anyone who obtains the information through their work, criminals and anyone who knows or ought to know that it is inside information (MAR 2014). According to the Court of Justice of the European Union, information only becomes public once it has been published under Art. 17 MAR (CJEU 2026). Analyses of publicly available data, by contrast, are not inside information per se (BaFin 2026). More under alternative data.

Issuers' obligations: ad hoc disclosure and insider lists

So that inside knowledge does not persist for long, an issuer must publish inside information that directly concerns it without delay. This is the ad hoc disclosure under Art. 17 MAR. It may delay publication only if immediate disclosure would be likely to prejudice its legitimate interests, the information withheld does not contradict its latest public statement on the matter, and it can ensure confidentiality. Since 5 June 2026, for protracted processes such as takeovers, only the final event has to be published; the intermediate steps can still be inside information, however, and the prohibitions continue to apply to them unchanged (Gleiss Lutz 2026).

Under Art. 18, issuers and their service providers keep insider lists: who had access to which inside information and when. They must be kept for at least five years (MAR 2014).

Directors' dealings: allowed, but reportable

Persons discharging managerial responsibilities, such as members of the management and supervisory boards, and persons closely associated with them report their own transactions, the directors' dealings, to the issuer and the supervisor under Art. 19 MAR no later than three business days after the transaction; the issuer publishes them within two business days (BaFin, managers' transactions 2025). Since the Listing Act, MAR sets a threshold of €20,000 per calendar year, which national supervisors may raise to €50,000 or lower to €10,000 (MAR 2014). BaFin raised it to €50,000 with effect from 1 January 2026; all transactions in a year are added up, without netting purchases against sales (BaFin, managers' transactions 2025).

On top of this comes a trading ban before financial reports: in the 30 calendar days before the announcement of an interim or year-end report, managers may not trade in the issuer's shares, debt instruments or derivatives, not even on behalf of third parties. The issuer may permit exceptions only in narrow cases, such as severe financial difficulty or employee share schemes (MAR 2014).

Penalties in Germany and how supervisors find insiders

Anyone who breaches Art. 14 MAR faces imprisonment of up to five years or a fine under Section 119(3) WpHG (Securities Trading Act); even the attempt is a criminal offence (WpHG § 119 1998). Breaches of the obligations relating to ad hoc disclosure, insider lists and managers' transactions, and of the trading ban before financial reports, are administrative offences under Section 120(15) WpHG if they are committed intentionally or recklessly (WpHG § 120 1998).

BaFin continuously analyses the transaction reports of banks and investment firms, monitors ad hoc disclosures and follows up on tip-offs. It compares unusual price and volume movements ahead of an announcement with what was known at the time and identifies who traded; if the suspicion is substantiated, it reports the offence to the public prosecutor's office (BaFin, insider surveillance 2017).

USA: Rule 10b-5, Form 4 and trading plans

The legal basis in the US is Rule 10b-5, the general anti-fraud rule of the securities regulator, the SEC. Officers, directors and shareholders with more than 10% of a class of shares report their trades on Form 4 (SEC, Investor Bulletin 2021).

Many executives trade through a Rule 10b5-1 plan: they fix trades in advance while they have no material non-public information, and can later rely on this. The SEC tightened the rule on 14 December 2022 (SEC, Rule 10b5-1 2022): for directors and officers, trading can begin no earlier than 90 days after the plan is adopted or modified, or two business days after the quarterly results, whichever is later, but at most after 120 days. For other persons, the period is 30 days. In addition, there is a good-faith certification, limits on overlapping plans and a checkbox on Form 4 for filings from 1 April 2023.

Types of insider trading

• Classic insiders: board members, employees, advisers, bankers or auditors who trade ahead of results, takeovers or regulatory approval decisions.
• Tippee: someone who receives a tip from an insider and trades on it (tippee). In the EU, it is enough that they know or ought to know that it is inside information.
• Misappropriation: in the US, liability can also arise for anyone who misuses confidential information from their source (such as their employer or a client) for their own trades, without working for the company whose shares they trade.
• Shadow trading: trading in a comparable company instead of your own (shadow trading). In SEC v. Panuwat, an employee of Medivation learned of the planned takeover by Pfizer and within minutes bought short-dated out-of-the-money calls on the comparable company Incyte. After the announcement, Incyte rose by around 8%; according to the SEC, his profit was more than US$100,000. The SEC sued in 2021, and on 5 April 2024 a jury found in its favour (SEC, Panuwat 2024).
• Front-running: trading ahead of your own clients' known orders (front-running). For persons who execute client orders, even the knowledge of orders not yet executed can be inside information (Art. 7(1)(d) MAR).
• Politicians: since 3 July 2012, under the STOCK Act, US senators have reported transactions in stocks, bonds and other securities above US$1,000, including those of spouses and dependent children, within 30 days of becoming aware of them and no later than 45 days after the transaction (Senate Ethics Committee 2012). Whether politicians trade better is disputed (see below).

Worked example: abnormal return after an insider purchase

Whether insider trades reveal anything is measured in research with event studies: how does the stock perform after the event compared with what the market alone would have led you to expect? The abnormal return is the actual return minus the normal return, in simplified form beta times the market return. Added up over a few days, this gives the cumulative abnormal return (CAR).

What the research shows

Insiders know more; copycats barely benefit. Using US data, Seyhun (1986) showed that insiders earn excess returns with their trades; for outsiders who imitated the reported trades, little was left after trading costs. Lakonishok and Lee (2001) examined all insider trades on the NYSE, AMEX and Nasdaq from 1975 to 1995: the market barely moves on the day of the trade or of the filing, insiders trade contrarian, and the predictive power lies in the purchases, especially at small companies. Sales reveal little. That is plausible, because insiders also sell when they need money or want to diversify more broadly.

Routine or opportunity. Cohen et al. (2012) separate insiders whose trades follow a predictable pattern from the opportunistic ones. A value-weighted portfolio based on the trades of the opportunistic insiders earned an abnormal return of 82 basis points a month, while that of the routine traders earned practically nothing.

Germany. Betzer and Theissen (2009) found significant abnormal returns after insider trades, larger ahead of earnings announcements; the insider's position in the company made no difference. Dymke and Walter (2008) linked insider trades to subsequent ad hoc disclosures: insiders as a group appear to use inside information, with supervisory board members the most active. Both studies date from before MAR.

Politicians: conflicting findings. According to Ziobrowski et al. (2004), a portfolio replicating the stock purchases of US senators from 1993 to 1998 beat the market by 85 basis points a month, while their sales lagged by 12 basis points. Eggers and Hainmueller (2013) re-examined earlier studies for 1985 to 2001 and analysed the years 2004 to 2008. They found no informational advantage, and the average member of Congress would have earned more with a passive index fund.

Options ahead of takeovers. Augustin et al. (2019) examined 1,859 US takeovers from 1996 to 2012: in about 25% of them there was unusually high trading volume beforehand in options on the target company, especially in short-dated out-of-the-money calls; more than half of this can hardly be explained by speculation, news or rumours. The SEC brought cases to court for only about 8% of the takeovers in the sample.

Enforcement matters. According to Bhattacharya and Daouk (2002), of 103 countries with stock markets, 87 had an insider trading law, but only 38 had ever enforced it; before 1990, the figures were 34 and 9. The cost of equity did not fall with the law, but only after the first prosecution.

For advanced readers: reading event studies correctly

Event studies go back to work such as Fama et al. (1969), which used this method to study price adjustment around stock splits. The normal return is usually estimated with a market model R_t = α + β × R_M,t from a window before the event; the example above sets α = 0. Across N events, you average the CAR and divide by its standard error:

• Filing date instead of trade date: outsiders only learn of the trade when it is filed; nobody could have copied returns from the trade date.
• Costs: especially for small caps, where the signal is strongest, the bid-ask spread is often wide.
• Known factors: insiders trade contrarian. Without controlling for size, value and momentum, you may only be measuring known factor premiums.
• Many tests: if you check enough subgroups, you will always find one with an excess return. This is the problem of multiple testing.

In theory, Kyle (1985) describes how an insider exploits their advantage: they spread their trades out so that their knowledge enters the price only gradually through the order flow. For the efficient market hypothesis, this means: its strong form according to Fama (1970), under which not even private information yields profits, does not stand up to the evidence on insider profits. For the semi-strong form, what matters is whether excess returns remain after the filing. The studies above find this at most for subgroups such as opportunistic purchases, and before costs.

Limitations, misconceptions and common mistakes

• “An insider purchase means buy”: a single reported purchase says little; research finds average effects across many trades. Sales had hardly any predictive power.
• Overlooking the threshold: since 2026, only those who reach €50,000 in a year report in Germany. If you compare time series before and after 2026, you are comparing different samples.
• Passing on tips: anyone who passes on inside information or trades on a tip can be committing a criminal offence themselves.
• Confusing insider trading with manipulation: anyone who drives prices with false signals, as in a pump and dump, is engaging in market manipulation, which is also prohibited. And not every financial scandal has to do with an informational advantage: cum-ex was about tax refunds.

How Portfolio Lab calculates it

Portfolio Lab has no insider tool: no data on directors' dealings, Form 4 or insider lists, and no event studies. The beta for a calculation like the one above is provided by the Backtest: with a benchmark, it shows beta, alpha and R² from the same regression of excess returns, from six common periods onwards (metrics methodology). The Factors page regresses monthly returns on market, size, value and other factors and shows alpha with t-statistics, optionally with Newey-West standard errors (factors methodology). Reactions over a few days after a filing cannot be measured with monthly data, however. More under beta and alpha.

This article explains a method; it is not investment or tax advice. Worked examples are simplified and rounded. All terms in the glossary.