A corporate action is any event initiated by a public company that changes its securities — the share count, the per-share price, the form of the claim, or the cash returned to holders. The category spans the routine (a quarterly dividend, a 2-for-1 split) to the existential (a merger that ends a stock's life, a rescue rights issue). The unifying tension for a trader is that most corporate actions are mechanically value-neutral at the instant they happen, yet they are not informationally neutral — the decision to split, repurchase, spin off, or issue equity is a signal about management's view and the firm's condition, and the empirical records for these signals diverge sharply by action and direction. A second, more practical tension runs through the whole cluster: many of these events produce a mechanical, non-economic price discontinuity (an ex-dividend gap, a split re-scaling, an ex-rights drop) that will masquerade as a technical breakout, breakdown, or stop trigger if the price history isn't properly adjusted. For a quant or systematic trader, getting corporate-action handling right is therefore first a data-hygiene problem and only second an alpha problem.

What the section covers

This section treats corporate actions from the holder's / trader's point of view — what happens to your shares, the calendar and key dates, the mechanical price effect, and what (if anything) the event signals — not the corporate-finance rationale of why boards choose them (that strategic/valuation logic lives in the Fundamentals branch). The six child nodes partition the space by event type:

• Stock Splits & Reverse Splits — arithmetically neutral re-scaling of share count and price. Forward splits are issued from strength (a well-replicated but modest, partly mechanical announcement pop; durable outperformance is contested); reverse splits are predominantly defensive (listing-compliance-driven) and are robustly associated with subsequent underperformance.
• Dividends & Ex-Dividend Mechanics — the four dates (declaration, ex, record, payment), the ~dividend-sized mechanical ex-date price drop, the T+1 change (since May 28, 2024 the ex-date is the same business day as the record date), and why "dividend capture" is not free money after spreads and taxes.
• Mergers & Acquisitions (Holder View) — what target shares convert into (cash / stock / mixed), the takeover premium and post-announcement gap, the merger-arbitrage spread, and binary deal-break risk. A stock under a definitive deal is effectively terminated as a technical instrument.
• Spin-Offs & Carve-Outs — separating a unit by distributing it to holders (spin-off, tax-free under IRC §355) vs. selling a stake for cash (carve-out). The forced-index-selling dislocation is real; the "spin-offs beat the market" edge is empirically supported but smaller, period-dependent, and concentrated in the subsidiary.
• Rights Offerings — raising equity by offering discounted new shares to existing holders pro rata (TERP, renounceable vs. non-renounceable rights, the "rights offer paradox" of why the cheapest method nearly vanished in the U.S.). Deeply discounted "rescue" issues are a distress flag.
• Tender Offers & Buybacks — the four repurchase mechanisms (open-market, fixed-price self-tender, Dutch auction, ASR), their ranked signaling strength, and the documented repurchase anomaly (strongest in undervalued value firms; concentrated in older samples).

The two organizing axes

Two classifications cut across the children and are worth holding in mind:

Mandatory vs. voluntary. Industry/custodian taxonomy (DTCC, CFI) sorts actions into three buckets: mandatory (execute regardless of holder action — splits, ordinary dividends paid in cash, spin-offs, the conversion in a completed merger); mandatory with options (the action happens but the holder picks a form — e.g. a dividend offered as cash or stock, with a default); and voluntary (the holder must elect to participate or the action doesn't touch them — tender offers, and renounceable rights, where doing nothing forfeits value). The practical point: voluntary and option-bearing actions carry a "do-nothing trap" — the single most common retail error in rights issues and tenders is passive inaction that destroys value the holder could have captured.

Cash-returning vs. structure-changing. Dividends and buybacks return capital; splits, spin-offs, mergers, and rights change the structure of the claim (share count, float, identity, or the security's continued existence). The first axis is mostly about signaling and tax; the second is mostly about data continuity and benchmark/float membership.

When it matters vs. when it doesn't

Corporate actions matter most at the boundaries of a position's life: when an event re-prices the stock overnight, changes what the shares are, or caps/terminates the upside. They matter least as standalone alpha — none of these events is a reliable push-button edge. The honest cross-cutting evidence picture is that the direction of most signals is more robust than the magnitude (reverse-split underperformance and acquirer value-destruction are directional averages, not per-name guarantees; announcement-effect percentages are study- and era-specific), and the cleaner anomalies (buyback under-reaction, spin-off subsidiary outperformance) are concentrated in specific subsamples and older data, with decay and implementation frictions both real concerns.

Sources

• Corporate action — Wikipedia and Corporate Finance Institute — Corporate Action — mandatory / mandatory-with-options / voluntary taxonomy; definition and examples.
• Broadridge / ShareholderEducation — Understanding Corporate Actions and DTCC corporate-action processing — custodian/holder-side mechanics.
• StockCharts — Price Data Adjustments and QuantConnect — Corporate Actions — back-adjustment of price history (split factor; dividend factor); why unadjusted series misstate returns.
• Child nodes (this section) — per-event formulas, dates, base rates, and the primary academic literature for each action type.