The Analyst’s Course · Quant SQL
Case File: The Screen That Would Have Caught 2022
The 2021 setup
Late 2021: zero rates, CAPE ~38, every equity multiple stretched. The macro track told you the denominator risk; this track asks the mechanical question — which names, concretely, were most exposed? The rate-shock casualty profile is knowable from first principles: long-duration cash flows (profits far in the future), weak current cash generation, priced for perpetuity growth. In table terms: high or absent P/E, negative or thin FCF conversion, momentum_3m high (crowding), volatility elevated.
The query, and what it surfaced
SELECT symbol, pe, market_cap FROM stocks WHERE (pe > 60 OR pe IS NULL) AND roe IS NULL ORDER BY market_cap DESC — "expensive or unprofitable, ranked by size." Run against the 2021 universe it surfaced the exact cohort: unprofitable hyper-growth, profitless software, the SPAC complex. When rates repriced in 2022, that cohort fell 60–80% while cash-generative value fell 10–25%. The screen did not predict 2022 — it identified, mechanically, who was long the denominator. That identification was available to anyone with a WHERE clause.
The honest framing
What the exercise proves and does not: it proves that rate-sensitivity is screenable — duration is a characteristic, not a secret. It does not prove timing: the same screen in 2019 would have flagged winners-to-be (many unprofitable growth names rippled higher for two more years). The discipline is exposure management, not prophecy: knowing your book's duration profile lets you pre-decide sizing when the denominator moves — which it always eventually does. Combined with the macro panel's quadrant map (Track C, lesson 9), the screen becomes the bridge from regime to portfolio.
ARK and the duration tell, quantified
The most famous long-duration book of 2021 (ARKK) was dominated by names with negative FCF and decade-plus profit horizons. As the 10Y rose 2.7pp in 2022, ARKK fell ~65% — almost exactly what its holdings' duration profile implied. The fund's own fact sheet contained the inputs; the WHERE clause was the missing step. When commentators said "nobody could have seen it," the honest answer was: nobody with a screener looked.
What you'll practise
The 2022 casualty profile in table terms was…
3 graded checkpoints · certification exam at the end of the track
Sources
Fama & French (1992, 2015); Daniel & Moskowitz (2016); screener traps per standard practice
Learn content is for education only — not individualized financial advice, a recommendation, or a solicitation to buy or sell any security. Options involve substantial risk. Examples are simplified and historical patterns never guarantee future results.