-$3.01
$4.28
bear -$3.01
base -$0.50
bull $4.28
margin of safety $2.00
market price* $3.98
Business condition attached to the margin of safety
Two consecutive reported quarters in which operating cash flow, less stock-based compensation and capital spending (including capitalized software), is positive, AND the OrbiMed term loan's rising minimum-revenue covenant has been amended, refinanced or repaid without a share issuance at a price below the margin of safety price.
Myriad Genetics runs a U.S. molecular-diagnostics laboratory business. Its tests cover hereditary cancer risk (MyRisk, BRACAnalysis CDx), cancer treatment selection and prognosis (MyChoice CDx, Prolaris, Precise Tumor, the newer Precise MRD residual-disease test), prenatal and carrier screening (Foresight, Prequel, SneakPeek, FirstGene) and pharmacogenomic testing for depression treatment (GeneSight). Revenue was $824.5 million in 2025 and $391.1 million in the first half of 2026, with gross margins near 68-70%. The economics are simple to state and hard to fix: the tests are profitable at the gross-margin line, but sales, marketing, research and overhead have consumed all of that and more for six straight years. What the business earns. We measure owner earnings as operating cash flow, less stock-based compensation (a real cost paid in shares), less the swings in working capital, less the capital spending needed to keep the labs running. On that basis Myriad has not produced positive owner earnings since its fiscal year ended June 2019, when it was a different company that still owned its pharmaceutical-services unit and before reimbursement pressure on its core tests intensified. Calendar 2021-2023 were distorted by a one-time tax refund, the repayment of deferred payroll taxes and a large laboratory build-out, so we measure the business from 2024, after the build-out and divestitures. Owner earnings were about -$30 million in 2024 and -$42 million in 2025 (-$0.33 and -$0.45 per share). Adding back cash interest to see what the business itself produces before its lender gives roughly -$28 million and -$33 million, a starting level of about -$0.34 per share. The large 2025 write-down of $234.7 million of goodwill and $82.0 million of intangible assets in the Women's Health and Mental Health units is a non-cash accounting charge for past acquisitions and does not reduce these cash figures; the real information in it is that those lines are earning less than when they were bought, and that is carried in the cases below. What changed since our previous analysis. The first half of 2026 was worse, not better. Revenue fell 4% year over year, and second-quarter revenue fell about 10% ($190.7 million versus $213.1 million), as the loss of UnitedHealthcare coverage for GeneSight continued to bite and EndoPredict was withdrawn from the U.S. market. Our estimate of owner earnings for the half was about -$51 million (operating cash outflow of $24.0 million, stock compensation of $13.7 million and capital spending of $7.5 million) against about -$20 million in the first half of 2025. Some things improved: stock-based compensation fell by a third and cost per test has been falling since the automated labs were completed. But our previous estimate assumed the business would turn cash-positive within about two years and build toward a 5% margin; nothing in the actual results yet supports crediting that recovery, so we no longer include it. The earlier figures are not comparable to these. The debt. In July 2025 Myriad borrowed $125 million from OrbiMed under a secured term loan (with up to $75 million more available until June 2027), secured by substantially all of its assets, maturing July 31, 2030, at SOFR (floor 2.5%) plus 6.5% - 10.4% at year-end 2025 - with an exit fee and prepayment premium. The loan also requires trailing twelve-month revenue of at least $615 million at the end of 2025, rising every quarter to $974 million from December 2029. Myriad's highest annual revenue ever was $837.6 million (2024), and trailing revenue at mid-2026 was about $807 million and falling. On a straight-line reading of that schedule the requirement passes current revenue around 2027-2028. That covenant, not the 2030 maturity, is the date that matters for shareholders: unless revenue grows well beyond anything the business has done, Myriad will need its lender's agreement, new money or a refinancing within roughly two years. At June 30, 2026 it held $115.2 million of cash. Counting the $125 million loan plus its exit fee, setting aside about $30 million of cash as working money, and adding about $19 million for the cost of paying roughly 10.4% rather than a normal long-term rate through 2030, we treat the company as owing about $64 million net, or about $0.67 per share. What growth we credited. We credited only what is in the bag: falling cost per test from the completed laboratory automation, the lower stock-compensation run rate already visible in 2026, and continued hereditary-cancer volume growth (volume rose 7% in 2025) at prices that fell 5%. We did not credit a return of GeneSight coverage, Precise MRD becoming a meaningful product, new payer wins, or a broad cost-cutting program, because none of these has happened yet. Base case. Losses narrow slowly as lab costs and stock compensation fall and hereditary-cancer volume offsets GeneSight and EndoPredict: the business burns about $60 million in 2026, $35 million in 2027 and $15 million in 2028, breaks even in 2029 and earns about $15 million a year thereafter, fading over the following two decades as competition in hereditary testing continues to erode price. That business is worth only about $17 million today. With about $64 million owed net, the base case comes to about -$0.50 per share: on the median path the company owes slightly more than its business is worth, and the cash burn means it will likely need new equity or lender concessions before the 2027-2028 covenant step-ups, which would dilute today's shareholders. Bear case. Revenue keeps falling at the second-quarter 2026 pace as GeneSight reimbursement erodes further and hereditary-cancer prices keep falling - the pattern of the years after the Supreme Court's June 2013 ruling opened BRCA testing to competitors and payers cut prices, and of the 2020 shutdown, when Myriad's owner earnings swung from positive to -$0.87 per share. The business burns about $80 million a year for three years, breaches the revenue covenant around 2027, and the secured lender forces asset sales, a highly dilutive raise or a restructuring. The business is worth about -$214 million on that path; adding the $64 million net owed plus about $10 million of exit and prepayment costs gives about -$3.01 per share. In plain terms, in the bear case the company owes roughly $290 million more than it is worth, and shareholders would be left with little or nothing. Bull case. The levers Myriad has already built work well: the automated labs keep cutting cost per test, hereditary-cancer volume keeps growing high-single digits, and prenatal pricing holds. Losses shrink to about $40 million in 2026 and $10 million in 2027, the business earns $20 million in 2028, $45 million in 2029 and about $60 million a year from 2030 to 2035, then fades over twenty years. Revenue growth of roughly 5% a year would come close to satisfying the covenant schedule without new equity. That business is worth about $473 million; less $64 million owed, about $4.28 per share. How we set the margin of safety price. We are not confident in the earnings level: it has been negative for six years and got worse in the first half of 2026. The worst plausible outcome - a covenant breach in 2027-2028 with a lender holding a claim on substantially all assets - would likely leave shareholders with very little, and the business has no product or contract that would prevent it; concentration in a few tests and dependence on payer coverage decisions (UnitedHealthcare's GeneSight decision is the live example) make it fragile. If we are wrong about the two things we are most likely to be wrong about - how fast cost per test falls and whether hereditary-cancer volume growth holds - the value moves between roughly zero and the bull case. Nothing is paid out over the holding period. So price alone is not sufficient. If the business shows two consecutive quarters of positive owner earnings and the covenant problem is resolved without a cheap share issue, the bull path becomes the realistic one and the shares would be worth something near $4; from $2.00 that would then imply about 20% a year over three to four years even if the improvement comes slower than in the bull case. Our margin of safety price is $2.00, and only once those conditions are met.