A defense supplier with a record backlog and intact cash guidance, priced as if its 2028 targets will be badly missed.
| The pushback | The answer, from the report |
|---|---|
| “The CEO was just removed. Why trust the guidance?” | The change followed a conduct violation, not a business miss, and came after results that raised guidance. The model works from cash flow and the company’s own targets; the risk table lists the open investigations, and a cut to cash guidance is named as what would change the value. |
| “Your beta of 0.80 is higher than the measured 0.61. Are you inflating the discount rate?” | It raises the discount rate, which lowers the value: the conservative choice. 0.80 matches defense peers; the measured figure understates risk after this year. Even at 9.0% the value is $346, still well above the price. |
| “GAAP margins are only about 10%. Isn’t the stock fully priced?” | GAAP profit is held down by amortization from the Aerojet purchase. Free cash flow, which adds that back, has grown every year since 2023. The price needs long-run cash flow of about 9% of revenue, against 14.6% in guidance. |
| “The debt is high.” | About $11.0 billion against $1.5 billion of cash, with $560 million of interest a year. The model deducts it, along with the Pentagon’s preferred stock, treated as debt. |
What would change the call: A cut to the $3.0 billion cash-flow guidance, new findings from the conduct investigation, or the bear case (growth of 3–4% and cash flow of 12.5% of revenue) becoming the base case.
The highest-returning large bank. On its own it looks fully priced; against an expensive market it is still cheaper than average.
| The pushback | The answer, from the report |
|---|---|
| “Your value is below the price. How is this a Buy?” | The rating is relative. On the same assumptions the S&P 500 is worth about 68% of its price; JPMorgan is worth about 86% of its price. That is 27% better than the market, so it clears the Buy line. On its own it is labelled overvalued, and the report says so. |
| “Why an excess-return model and not a DCF?” | Debt is a bank’s raw material, so free cash flow has no clear meaning. A bank is worth what it earns on its equity: price to book = (ROE − g) ÷ (cost of equity − g). A bank earning exactly its cost of equity is worth book value. |
| “Returns are 20% now. Why only 16%?” | Today’s returns come with high rates and strong markets revenue. The bank’s own through-cycle target is 17% on tangible equity, about 14.5% on all equity. 16% sits between that and today. |
| “Does the rating survive a different market view?” | Not fully: it would be Buy if the index grows 6% a year and Hold at 10%. That sensitivity is printed in the report. |
What would change the call: Credit losses rising faster than expected, rate cuts squeezing net interest income, or a change to the 17% through-cycle target.
A great business in an expensive building phase: AWS is accelerating, but capital spending has turned free cash flow negative and the price assumes margins Amazon has never earned.
| The pushback | The answer, from the report |
|---|---|
| “Net income was $62.6 billion last quarter. How is cash flow negative?” | Most of it was a $53.4 billion gain from revaluing the Anthropic stake. Capital spending rose $66 billion in a year to $169 billion net of proceeds, more than operating cash flow. |
| “You leave out the Anthropic stake.” | Yes, deliberately, and the report says so: it is upside to the value. It covers part of the gap to the price, not all of it. |
| “The market expects AI to pay off. Why don’t you?” | The base case lets margins rise from 13% to 18% by 2031, well above anything Amazon has earned. The price needs about 29%, or growth of about 23% a year for two years. |
| “Hold, with the value 40% below the price?” | Against the market it is only 12% dearer, inside the Hold band. It would be Sell if the index grew 10% a year. |
What would change the call: AWS growth staying above 30% with margins holding, operating income against guidance, or any change to the $220 billion capex plan.
Margins are recovering after the 2025 collapse; the price assumes they stop at about 6%, against more than 8% before 2025.
| The pushback | The answer, from the report |
|---|---|
| “$610 million of the improvement was reserve releases.” | Agreed, and it is listed as a risk. The base case does not need it: the margin starts at 5.5% and rises to 7%, still below the 8–9% before 2025. |
| “There are federal investigations.” | Into Medicare billing and Optum’s ties to the insurer. They are in the risk table, and their outcome is one of the next inputs the report names. The bear case, $282, is margins stuck near today’s level. |
| “Revenue was flat.” | Deliberately: the company is shedding unprofitable members. Earnings from operations rose 52%. The model assumes 4% then 7% growth. |
| “The Street is at $481. Why higher?” | Analysts are close to this model’s base case, and both assume margins keep recovering. The range is lopsided: the bull case is $323 above the base and the bear case $234 below, which lifts the expected value. At a 9.0% discount rate the value is $417. |
What would change the call: The 2027 outlook around the turn of the year, Medicare Advantage rates for 2028, and the outcome of the investigations.
An excellent business with no margin of safety: the cash flows support $831.50 a share against a price of $970.31.
| The pushback | The answer, from the report |
|---|---|
| “Costco always looks expensive. You’d never have bought it.” | The fundamentals are excellent and the report says so. A Hold is not a Sell: it is a call that the price leaves no room for error. |
| “Why give peer multiples no weight?” | They cannot capture the membership fees and the stability of growth, and they imply only $300–500. A 70/30 blend gives about $716, which makes the call stronger, not weaker. |
| “The liquidity ratios look weak.” | Negative working capital is the model working: members pay before suppliers are due. The cash conversion cycle is near zero, interest is covered more than 80 times and the Altman Z-score stayed above 3 every year. |
| “Analysts average $1,072.” | Their targets build in premium multiples continuing. This analysis puts price discipline first; even the best companies can be poor investments at the wrong price. |
What would change the call: A price below the cash-flow value, or growth that holds above the 8.45%-to-5.07% path the model assumes.