Buyback Announcement Drift Strategy

The thesis

When a US-listed company announces a new open-market share repurchase program, the stock tends to drift higher for months — far longer than the immediate announcement-day reaction. The market treats buyback authorizations as soft commitments (firms can buy fewer shares than authorized, or none at all) and as a result under-reacts to the implicit signal that management views the stock as undervalued and the balance sheet as strong enough to support the repurchase.

Ikenberry, Lakonishok, and Vermaelen (1995) documented roughly +12% abnormal returns over four years on a value-weighted basis after open-market repurchase announcements, with the effect strongly concentrated in value stocks (low book-to-market). Peyer and Vermaelen (2009) confirmed that the anomaly persists post-publication, particularly when filtering for high-quality announcements (new programs in value names with recent price weakness).

Academic basis

The 1995 Journal of Financial Economics paper by Ikenberry, Lakonishok, and Vermaelen remains the canonical reference. Their long-horizon event study found significant positive abnormal returns over the four years following a buyback announcement, with most of the effect concentrated in the smallest size deciles and value-tilted firms. Their 2000 follow-up replicated the result in Canadian markets, ruling out a US-only anomaly.

Peyer and Vermaelen's 2009 Review of Financial Studies paper revisited the strategy two decades after its initial documentation. They found that the simple version had decayed somewhat — consistent with McLean-Pontiff (2016) post-publication decay estimates — but a refined version that combined high book-to-market with recent price weakness still produced roughly +6% per year of abnormal returns over a four-year holding period.

The behavioral interpretation: management has private information about the firm's intrinsic value and its near-term cash-flow generation. A buyback announcement reveals that information indirectly, but the market under-reacts because (a) buyback authorizations are not legally binding, (b) firms can quietly repurchase fewer shares than announced, and (c) outside investors discount the announcement as cheap signaling. The drift represents the gradual correction of that under-reaction as actual repurchase activity is observed in subsequent 10-Q filings.

How Alpha Suite implements it

• EDGAR 8-K full-text search — the scanner queries the SEC EDGAR full-text search API for 8-K filings in the last 7 days containing repurchase-program language ("share repurchase program", "stock repurchase program", "authorized the repurchase").
• New-program filter — the filing text is parsed for language indicating a brand-new program (board authorization, "new repurchase program") versus an expansion or amendment of an existing program. New programs score higher; expansions still qualify but at a discount.
• Disqualification keywords — filings that announce the suspension, termination, or completion of an existing program are filtered out entirely.
• Program size extraction — dollar amounts ("$1.5 billion", "$500 million") and share counts ("up to 10 million shares") are parsed from the filing text. Programs above 5% of market cap earn a meaningful score boost; programs above 10% are flagged as conviction signals.
• Value-stock filter — trailing P/E below 15 or P/B below 2.0 marks the name as a "value" candidate, mirroring the Ikenberry value tilt. Value names get a +5 to +8 point boost.
• Recent-weakness boost — firms announcing buybacks after a 5%+ drawdown in the prior 60 days receive an additional boost, consistent with Peyer-Vermaelen findings that post-weakness announcements have the strongest drift.
• Mid-cap sweet spot — the Ikenberry effect is strongest in small-and-mid-cap names. Sub-$2B market cap earns a modest score bonus; mega-caps earn no premium.
• 90-day horizon — the literature shows drift extends one to four years, but the sharpest portion is concentrated in the first three months. Alpha Suite targets the 90-day window to align with the rest of the suite's horizon.
• Long-only — the buyback anomaly is asymmetric. The short side (firms that explicitly do not buy back) does not produce a meaningful negative drift.
• Post-publication decay — a 30% haircut on expected take-profit, matching McLean-Pontiff (2016) post-publication decay estimates and the same haircut applied to the 52-week-high momentum strategy.

When it fires

Buyback announcement signals fire most heavily in the second through fourth weeks of January, April, July, and October — the windows around quarterly earnings releases when boards typically authorize new programs alongside earnings results. The cleanest fires combine: (a) a brand-new program (not an expansion), (b) a value-tilted balance sheet, (c) program size above 5% of market cap, (d) a recent 60-day pullback of more than 5%, and (e) a non-mega-cap name where the buyback is materially impactful relative to free float.

The strategy interacts naturally with the insider Form 4 engine: when buyback announcements coincide with C-suite open-market purchases on the same name, the confluence engine elevates the signal to Tier 1 (Prime). Both strategies encode the same underlying claim — management thinks the stock is cheap — from different sides of the disclosure regime.