Why a $112 Billion Profit Can Be an Illusion: How Unrealized Gains Distort Earnings
Why a $112 Billion Profit Can Be an Illusion: How Unrealized Gains Distort Earnings
Record net income, falling stock price — what did the market see that the headline missed?
Read operating income and free cash flow, not headline net income. Last week Alphabet and Tesla reported minutes apart. Both printed enormous profits. Both stocks fell hard. The reason is simple: most of each company's reported profit was a paper markup on a stake neither one sold.
The numbers: roughly $99 billion of Alphabet's $112 billion quarterly profit came from a one-time investment gain. Tesla's headline "over a billion dollars" was likewise mostly a roughly $1 billion paper gain on its SpaceX investment.
Read only the headlines and you would conclude both companies crushed it. Underneath, Tesla's operating margin collapsed to 1.4% and Alphabet burned cash for the first time in its history.
The headline and the numbers were saying opposite things.
What the accounting rules changed
This illusion starts with an accounting standard, not with the market.
Under current rules, changes in the fair value of a company's equity investments flow through that period's income statement even if nothing was sold. These unrealized gains and losses used to accumulate quietly in equity. Now they land in the middle of net income.
The result is that the bottom line mixes two completely different kinds of money.
| Operating income | Equity investment gains | |
|---|---|---|
| Source | Selling products and services | Market price moves on holdings |
| Cash received | Yes | No |
| Repeatability | Recurring | Non-recurring, direction unpredictable |
| Management control | Yes | Essentially none |
| Next quarter | Broadly comparable | Can flip to the opposite sign |
That last row is the whole point. This line item can reverse direction entirely next quarter.
What happens next quarter: the illusion in reverse
SpaceX stock is down a lot right now. The rule is symmetric, so next quarter a large unrealized loss could drag both companies' net income down.
Here is what matters. That loss will not be caused by the business getting worse. Not by Google selling less search advertising. Not by Tesla delivering fewer cars. It will happen for one reason only: the market price of a holding went down.
And the headline will be written just as simply as before: "Google profit collapses." Anyone who got excited by taking this quarter's headline at face value is very likely to get scared by taking next quarter's headline at face value.
It is the same mistake made twice, with the sign flipped.
A five-step checklist for stripping out the illusion
This is the order I actually work through when I read a report.
1. Start with operating income, not net income. Operating income is what the business earned by running itself, and equity gains do not touch it. Tesla is the clean example: net income over $1 billion, operating income $398 million, down 57% year over year.
2. Convert it to a margin. $398 million on $28 billion of revenue is 1.4%. That single line is the honest summary of the quarter.
3. Move to the cash flow statement. Take cash from operations minus capital expenditures. Paper gains are not cash, so this step filters them out automatically. The fact that Alphabet burned cash for the first time ever is visible only here, never on the income statement.
4. Explicitly back out the one-time items and recalculate. Remove roughly $100 billion of investment gains from Alphabet's $244 billion of net income and the 16.5x P/E does not survive. Meanwhile the 76x price-to-free-cash-flow is overstated because of the capex surge. Neither multiple is trustworthy right now.
5. Check against multi-year averages. One quarter is mostly noise. Alphabet's underlying net margin, excluding investment gains, has walked from 30% over ten years to 34% over five to 36–37% recently. A single headline will never show you that trend.
Why this is a money question, not a bookkeeping one
This one habit prevents two opposite mistakes.
First, overpaying for a company that looks cheap. When headline net income is inflated, the P/E looks lower than reality. You buy thinking it is cheap.
Second, missing a good company — which can hurt more. A business like Alphabet, whose free cash flow temporarily shrank because it ramped capex, screens as expensive. But if that spending is building a segment growing over 80% a year, the ugly number is not a problem; it is evidence of investment. You cannot tell those two situations apart from the headline.
I keep coming back to this because most people never look past the headline. That is exactly how you end up overpaying and losing money, or missing the moment when a real growth engine starts to compound.
For the company-specific versions of this, see Tesla's 1.4% Operating Margin: The Number Hiding Behind Record Revenue and Why Alphabet Fell 7% Despite an 82% Cloud Surge.
FAQ
Q: Is running unrealized gains through net income bad accounting? A: It is not a violation — it is the rule. Companies are required to do it, and the intent is to show asset values transparently. The problem is not the accounting; it is the interpretation. The error starts the moment you read a non-recurring item as if it recurs.
Q: Should I just ignore net income entirely? A: No, but never read it without checking its composition. Find the investment-related gains and losses in the earnings materials or the other-income line, strip them out, and then look again.
Q: How can an individual investor check this quickly? A: The fastest tell is the gap between operating income and net income. When that gap is unusually wide, a one-time item is almost always involved. Confirming cash from operations on the cash flow statement usually settles it.
Q: What should I look at first in next quarter's reports? A: If a headline says profit collapsed, find out how much of the decline was an unrealized loss. If operating income and revenue held up, the business is fine and only the books moved. If operating income fell too, that is a real signal.
More in this Category
Meta Drops 10% on Earnings: 28% Revenue Growth Against a 91% Free Cash Flow Collapse
Meta Drops 10% on Earnings: 28% Revenue Growth Against a 91% Free Cash Flow Collapse
Meta posted $60.8 billion in revenue, up 28%, but EPS came in at $6.18 against a $7.15 consensus, and free cash flow fell from over $8.5 billion a year ago to $784 million. My ten-year model puts the midpoint fair value at $925.
Microsoft's Record Single-Day Market Cap Gain: What the $678 Billion Backlog Really Says
Microsoft's Record Single-Day Market Cap Gain: What the $678 Billion Backlog Really Says
Microsoft posted $90 billion in revenue (up 18%), 43% Azure growth, and a commercial backlog of $678 billion (up 84%), sending the stock up nearly 10%. My ten-year model produces a $363 to $883 value range with a $571 midpoint.
I Split My 65-Stock Watchlist Into Three Tiers: Why 9%, 12%, and 15% Are Completely Different Decisions
I Split My 65-Stock Watchlist Into Three Tiers: Why 9%, 12%, and 15% Are Completely Different Decisions
Of the 65 names on my watchlist, 37 have moved into my calculation range: 14 project 9-10% annually, 13 project 11-15%, and 10 project above 15%. What separates the tiers isn't business quality — it's today's price.
Next Posts
Is AI a Substitute for Adobe, or an Amplifier? I Followed Both Arguments All the Way Down
Is AI a Substitute for Adobe, or an Amplifier? I Followed Both Arguments All the Way Down
Adobe is down roughly 66% from its all-time high, yet last quarter's revenue grew 13% and annual recurring revenue passed $27 billion. I tested both the bear and bull case on whether AI replaces Adobe or strengthens it.
Adobe at 9x Free Cash Flow — Building the Fair Value From Assumptions, One at a Time
Adobe at 9x Free Cash Flow — Building the Fair Value From Assumptions, One at a Time
A $91 billion market cap against $10.3 billion of free cash flow last year. Running a 10-year model with 3/6/9% revenue growth, 37/40/43% FCF margins, and 18/21/24x exit multiples produces a fair value band of $400 to $890, with a midpoint of $595.
What It Takes to Buy the Most Hated Stock — Five Arguments and the One Condition That Would Prove Me Wrong
What It Takes to Buy the Most Hated Stock — Five Arguments and the One Condition That Would Prove Me Wrong
Terry Smith is worried about Adobe while Michael Burry, Oakmark, and Mohnish Pabrai are buying it. Here are the five tests I run when smart investors land on opposite sides, plus the falsification condition I write down before I buy.
Previous Posts
Meta Started Selling Its Spare Compute — What the $10 Billion Anthropic Talks Really Mean
Meta Started Selling Its Spare Compute — What the $10 Billion Anthropic Talks Really Mean
Meta raised its 2026 capex guidance to $125–145 billion and the stock fell 11% over the past year, but the company has begun leasing surplus AI compute to outside firms. Reports of a deal with Anthropic worth up to $10 billion over two years are the first hard evidence of that shift.
What Is Meta Actually Worth? Running a $1.6 Trillion Company Through a 10-Year Model
What Is Meta Actually Worth? Running a $1.6 Trillion Company Through a 10-Year Model
Meta trades at a $1.6 trillion market cap, 23x earnings and 33x free cash flow. Running a 10-year model with 7/10/14% revenue growth and 18/22/26x exit multiples produced a low of $540–560, a midpoint of $850–870 and a high near $1,400.
Three Risks Every Meta Shareholder Should Underwrite: Capex, Reality Labs, and a $1.4 Trillion Lawsuit
Three Risks Every Meta Shareholder Should Underwrite: Capex, Reality Labs, and a $1.4 Trillion Lawsuit
Meta guided 2026 capex to as much as $145 billion and has not repurchased shares since Q3 2025, Reality Labs posted more than $4 billion in operating losses in a single quarter, and four states are suing for roughly $1.4 trillion.