Decision point · Mar 2015
Ackman & Valeant — the "platform" that wasn't
Pershing Square took a $4B Valeant stake at the peak — the leverage and accounting flags were already loud.
Ackman framed Valeant as a "modern Berkshire" — a serial acquirer with disciplined capital allocation, growing pharma cash flows via roll-up, and a low-tax-rate structural advantage. The pitch was 25%+ IRR through compounded acquisitions.
“Pershing Square took a $4B Valeant stake at the peak — the leverage and accounting flags were already loud.”
Citron Research and others exposed channel-stuffing through pharmacy intermediary Philidor. The DOJ investigation, debt covenants, and a collapsing acquisition pipeline drove the stock from ~$260 in 2015 to under $10 in 2017. Pershing Square reported a roughly $4B realised loss.
A Prism memo at the 2015 decision point would have triggered leverage_trap (D/E > 5×, interest coverage thin) and accounting_red_flag (cash conversion ratio < 0.4). The behavioural signal would have read negative on insider selling. Three of the most quality-strict frameworks (Munger, Klarman, Schloss) would have rejected — yielding a "quality vs valuation" disagreement classification with the bear camp dominating.
Patterns the engine would have flagged
- leverage_trap
- accounting_red_flag
- quality_trap
Negative signals at the time
- Balance Sheet (D/E > 5)
- Quality (ROC inflation via M&A, not organic)
- Behavioral (insider selling at peak)
“Multiple frameworks rejecting in unison is a stronger signal than one charismatic thesis.”
The same multi-framework engine running on every memo today would have surfaced this as a majority-bearish disagreement at the decision point, not in retrospect.
Charlie Munger
Munger (Quality)
Munger publicly compared Valeant to "ITT in the 1960s" — debt-fuelled M&A masking organic decay.
Seth Klarman
Klarman (Margin of safety)
No margin of safety with 5×+ debt and goodwill > tangible book.
Walter Schloss
Schloss (Net-net)
Tangible-book deeply negative — disqualifies on the most basic Schloss filter.
When the case for a "compounder" depends on debt-financed M&A rather than organic returns, the leverage-trap pattern is doing the work, not the moat. Multiple frameworks rejecting in unison is a stronger signal than one charismatic thesis.
Primary sources
Want to find these patterns before they cost you?
For research and educational purposes only. Not financial advice. Past performance does not guarantee future results. Conduct independent due diligence before making any investment decision.