A pump-and-dump is a securities-fraud scheme in which manipulators first quietly accumulate a thinly traded asset, then artificially inflate its price ("pump") through coordinated buying and the spread of false or misleading hype, and finally sell their position into the demand they manufactured ("dump"), leaving later buyers holding a collapsing price. Its defining tension is informational asymmetry weaponized: the people promoting the asset are the same people exiting it, so the very enthusiasm a victim acts on is the signal that the insiders are leaving. It is explicitly illegal under U.S. securities law (Section 9(a) and 10(b) of the Securities Exchange Act, Rule 10b-5) and is among the SEC's most-prosecuted retail-fraud archetypes.

How it's formed (mechanics)

The classic cycle has four stages:

1. Accumulation. Operators buy a low-float, low-liquidity asset — historically OTC microcap/penny stocks, more recently low-cap cryptocurrencies — where modest dollar volume moves price sharply. 2. The pump. Demand is manufactured via promotional spam emails, paid "research" newsletters, social media, chat rooms (Telegram/Discord for crypto), boiler-room cold-calls, and sometimes coordinated buying to create the appearance of momentum. Fake "inside information," fabricated revenue, and unrealistic price targets are typical. 3. The dump. Once price and volume spike, insiders sell into the inflow. In crypto this can happen within seconds; the asset's chart shows a near-vertical spike. 4. Collapse. Promotion stops, buying dries up, and price reverts — often below the pre-pump level — stranding retail buyers.

Common variants: the classic microcap promotion (paid touts on shell companies, often post-reverse-merger); the boiler room (high-pressure phone sales); the hack-and-dump (compromised brokerage/social accounts used to buy or hype); and the coordinated crypto group pump (a Telegram channel announcing a coin/time, sometimes a "tiered" structure where channel admins and paying members buy ahead of free members).

How it's used in practice (and how it's detected)

For an analyst the practical value is recognizing it, not running it. Recognition red flags, drawn from SEC/Investor.gov guidance and the empirical literature:

• A price/volume spike with no corroborating fundamental news (filings, earnings, legitimate catalyst).
• Unsolicited promotion — spam, DMs, "hot tip" newsletters, guaranteed/extraordinary returns.
• A microcap or low-cap asset with limited disclosure, low float, and a recent reverse merger or name change.
• Disclaimers in promotional material admitting the promoter was paid (often in tiny print) and may sell.
• A vertical run-up on extreme volume followed by an equally fast reversal — the signature "spike-and-revert."

Automated detection (used by exchanges, regulators, and researchers) keys on anomalous deviations in volume, return, and trade-count over short windows, often with ML classifiers. Crypto research has shown such patterns are detectable in near-real time.

Adoption, debate & evidence

That these schemes work mechanically is well documented; the contested question is magnitude and who profits. Key measured evidence:

• Stock spam (Frieder & Zittrain, 2007, "Spam Works"). Studying touted U.S. microcaps, they found a touted stock shows a significant short-run price bump and reversal: an investor who buys on the most-heavily-touted day and sells two days after touting ends loses roughly 5.5%, while a spammer buying the day before the heaviest touting and selling at the peak earns about 4.3% before transaction costs. This is the canonical demonstration that disclosure-based regulation alone doesn't protect retail.
• Crypto, Kamps & Kleinberg (2018, "To the moon," Crime Science). Defined detectable criteria from sudden price/volume anomalies; established that organized crypto P&Ds are frequent and identifiable in trading data, clustered on specific exchanges/coins.
• Crypto, Xu & Livshits (USENIX Security 2019, "The Anatomy of a Cryptocurrency Pump-and-Dump Scheme"). Studied 412 Telegram-organized pumps (June 2018–Feb 2019); documented the second-scale spike-and-dump dynamic and built a predictive model. Reported median price increases of roughly 5% for top-75 coins versus ~19% for coins ranked below 500 on Telegram (small, illiquid coins move far more), and a model-driven strategy achieving a return reported "as high as 60%" over ~2.5 months — figures specific to low-cap crypto, not generalizable to equities.
• Crypto, La Morgia et al. (2023, "The Doge of Wall Street," ACM TOIT; arXiv:2105.00733). Followed ~100 groups and ~900 events (2017–2021); computing conservative lower bounds on insider profit under different liquidation strategies, the authors report median insider returns above ~100% and upper-quartile returns exceeding ~2000% — again low-cap-crypto-specific and far larger than anything documented in equities.

The honest synthesis: the average late participant loses money, the edge accrues almost entirely to early insiders, and "profitable for the operator" does not mean "profitable for a retail follower." Claims that one can reliably front-run pumps as an outsider are largely folklore — by the time a signal is public, the asymmetry has usually already played out, and the activity is itself often illegal.

Strengths & limitations

There is no legitimate "strength" — this is fraud. As an analytical concept, its value is defensive: the spike-and-revert-on-promotion signature is one of the more recognizable manipulation patterns, and detection methods have measurable accuracy. Limitations of detection: legitimate news-driven momentum and small-cap squeezes can look similar, producing false positives; sophisticated operators stagger buying and use multiple venues to blur the anomaly; and in equities, surveillance plus SEC trading suspensions have pushed activity toward less-regulated crypto venues. The #1 misuse is mistaking a pump for a tradable opportunity ("I'll ride it and get out first"): the data show the asymmetry favors insiders, retail followers lose on average, and participation can expose a trader to liability.

Sources

• SEC / Investor.gov — Pump-and-Dump Schemes and Investor Alert: Don't Trade on Pump-and-Dump Stock Emails (sec.gov/answers/pumpdump.htm; investor.gov). Definition, distribution channels, microcap vulnerability, red flags.
• Cornell Legal Information Institute — Investor Protection Guide: Micro-cap Stock Fraud ("Pump and Dump"). Promoter mechanics, boiler rooms.
• Frieder, L. & Zittrain, J. (2007), Spam Works: Evidence from Stock Touts and Corresponding Market Activity, Hastings Comm. & Ent. L.J. — measured −5.5% late-buyer / +4.3% spammer returns.
• Kamps, J. & Kleinberg, B. (2018), To the moon: defining and detecting cryptocurrency pump-and-dumps, Crime Science 7:18.
• Xu, J. & Livshits, B. (2019), The Anatomy of a Cryptocurrency Pump-and-Dump Scheme, USENIX Security (arXiv:1811.10109). — 412 pumps; median price increases ~5% (top-75 coins) to ~19% (rank >500) on Telegram.
• La Morgia, M., Mei, A., Sassi, F. & Stefa, J. (2023), The Doge of Wall Street: Analysis and Detection of Pump and Dump Cryptocurrency Manipulations, ACM TOIT (arXiv:2105.00733). — ~900 events; insider-profit lower bounds with median >~100% / upper quartile >~2000%.
• Wikipedia — Pump and dump / Microcap stock fraud (corroborating overview only).

Flagged disputes: Profitability figures are regime- and asset-specific — equity-spam returns (Frieder-Zittrain) are modest single digits; the much larger crypto insider returns (La Morgia et al.: median >~100%) are lower-bound estimates under assumed liquidation strategies, apply only to low-cap crypto on specific Telegram-organized pumps, and should not be generalized to equities. Note that Xu & Livshits' median price-increase figures (~5–19%) and La Morgia et al.'s insider-profit figures (>100%) measure different things and are not directly comparable. "Front-running a pump as an outsider" is unproven and likely unprofitable post-signal, in addition to being potentially illegal.