The Best Earnings Quarter in Five Years, and Why the Market Barely Cared

The Numbers Were Extraordinary

Start with the headline.With 97% of S&P 500 companies reporting first-quarter 2026 results the combined earnings growth rate was 28.6% year-over-year. This is the highest quarterly figure since the 32.0% in Q4 2021 and marks six consecutive quarters of double-digit earnings growth.The beat rate matched the growth rate of the drama: the84% of companies reported actual EPS above estimates the highest share since the second quarter of 2021. Companies didn't just beat the bar. They far exceeded it beating estimates by an average of 20.7% almost triple the historical average of 7%. Revenue told almost the same story: 81% of companies also beat revenue expectations

The margin line might be the most important number in the entire publication. The S&P 500's combined net profit margin hit 14.8% an all-time high in FactSet records dating back to 2009. That beats the previous record of 13.2% set just a quarter earlier in the fourth quarter of 2025. Companies aren't simply selling more. They're keeping more of every dollar they sell at a rate the index had never before.registered.If you want the one-line version of the AI productivity thesis it's this: Technology spending that looked like pure cost in 2023 and 2024 now shows up as operating leverage in 2026

Where the Growth Actually Came From

Information technology led all sectors with 54.3% earnings growth and that number is based on two more names than you might imagine. If you exclude NVIDIA and Micron the IT sector's growth drops from 54.3% to about 30%. It's still extraordinary. But it shows how concentrated the AI hardware boom really is at the top of the sector.S&P 500's first quarter. Communication Services shows the same pattern on a smaller scale: 48.9% overall growth turns into a 4.1% decline once Alphabet and Meta are removed. The AI-driven earnings boom is real. It's big. And it's concentrated on a small number of balance sheets. For the other 493 companies in the index the story is quieter and I would argue more durable: genuine earnings growth thanks to expanding earnings.margins and modest revenue increases without the fireworks

Goldman Sachs had the best quarter among financials. Net income for the first quarter of 2026 was $17.23 billion up 14% year-over-year and the second-highest quarterly total on record for the company. Stock trading set a new record of $5.33 billion up 27% beating Goldman's own record from the previous quarter by about $1 billion. Investment banking fees increased by48% and advisory fees rose 89% as the M&A market rebounded strongly from the geopolitical freeze of 2025. Annualized return on equity came in at 19.8% comfortably above the company's long-term cost of capital and above what most banking analysts had modeled at the outset.JPMorgan set the tone in mid-April with record net income of $16.5 billion. Goldman confirmed the pattern a day later

Expectations, Not Just Results, Move Prices

Here's the part that should bother you if you only read the previous two sections. It was the best earnings quarter in five years and the index basically shrugged. This is not a contradiction once you understand what the price of a stock really is. The price of a stock is not a report card for the quarter that just ended. It is an ongoing bet on each future quarter discounted to today. When a company reports the market has already been calculating for months what that report will say

This is the most useful idea in stock research and it's also the one that trips up almost everyone the first time they trade an earnings release: markets value what's anticipated. A result that matches what was already expected shouldn't move the stock much at all because the price already reflects it. What moves the stock is the surprise the gap between what actually happened and what the market had priced in.A fantastic quarter that was fully expected is in market terms close to a non-event. A mediocre quarter that beats a depressed expectation can send a stock up double digits. This seemed really counterintuitive to me the first time it happened to me with a stock I owned. The headline was good. My position was down 8% by midday

How a Consensus Estimate Gets Built

So what is this "expectation" that everyone continues to value? It has a name: consensusEach company with actual analyst coverage has anywhere from a handful to several dozen sell-side analysts researchers employed by banks and brokerages each of whom publishes their own model of what that company will earn next quarter. Each analyst creates a spreadsheet: revenue by segment margin assumptions share count tax rate all the way to a single earnings-per-share number

Data providers like FactSet Bloomberg and Visible Alpha collect each of those individual estimates and average them sometimes a mean sometimes a median until they get the number that the financial media calls a consensus. It's not static. Analysts constantly review their models in the weeks leading up to a report reacting to industry data competitor results management commentary and sometimes even shipping or web traffic data when it exists. As of press time the consensusIt's really the current best guess on the market built by dozens of people who do this for a living and get paid based on how accurate their guesses are

There is a second less official number that worries the operating tables so much: the whispered number.It is the informal unpublished number that active traders and portfolio managers actually expect sometimes higher than the published consensus sometimes lower. It tends to circulate through chat rooms trading floors and research notes rather than appearing on a data terminal. A stock can beat the published consensus and still fall because it did not reach the whisper number that the people actually trading the stock were using that morning. I can't give you a clear data source for what the whisper number was for any stock in acertain day.Almost by definition no one publishes it.But if you've ever seen a stock move in a direction that the headline doesn't explain the whispered number is usually part of the answer

Why Guidance Matters More Than the Quarter Just Reported

Once it is accepted that prices move by surprise relative to expectations the next question is which expectations matter more. The answer is almost never the quarter that just ended.of the business over the next few years

That's why orientation management's own forecast for the next period routinely moves a stock more than the reported number does. A company can beat the consensus this quarter and still fall sharply if it guides the next quarter below where the market was. A company can miss this quarter and still recover if its forecasts suggest that the failure was a blip and that the future trend is stronger than feared. Guidance is management talking about the only thing the market has not yet valued: the future. EverythingThe rest in the results call confirms in a certain sense a past that the market already knew

A Worked Example: Two Companies, Opposite Reactions

Let me make this clear with illustrative numbers only not with actual companies or actual operations. Call them Alpha Company and Beta Company

Company Alpha reports earnings and Street expects $2.00 in quarterly EPS. It reports $2.16. That's a pace of $0.16 or $0.16 divided by $2.00 a pace of 8.0%. A clean and healthy number by any normal reading. But on the call Alpha guides next quarter's EPS at $1.85 versus a market expectation of $2.10.That guidance is $0.25 below expectation or $0.25 divided by $2.10 about 11.9% light. Let's say Alpha stock started the day at $80. A 9% drop a plausible market reaction to a double-digit guidance error takes it to $80 minus $7.20 or $72.80. The headline said pace. The stock fell anyway because thefigure that was not yet discounted for the next quarter came out weak

Company Beta follows the opposite pattern. The Street expects earnings per share of $1.50. Beta reports $1.41 an error of $0.09 or $0.09 divided by $1.50 an error of 6.0%. The headline reads wrong. But Beta guides the next quarter at $1.70 versus an expectation of $1.55 which is $0.15 above expectations or $0.15.divided by $1.55 about 9.7% ahead. Let's say Beta stock started the day at $40. A 7% rally again a plausible reaction to a significant guidance beat takes it to $40 plus $2.80 or $42.80. The headline said miss. The stock rose anyway because the forward figure the only one still uncertain came out strong

If we put them side by side the lesson won't be subtle. Alpha beat and led weakly. Beta failed and led strongly. Read just the EPS top lines and you'll expect Alpha's stock to outperform Beta's that day. The opposite is true in this example because the market was never really rating the quarter that already happened. It was rating the one that hasn't happened yet

Case Study: Meta's February 2022 Reaction

The clearest real-world version of this that I know of is Meta's Q4 2021 earnings report released in early February 2022. The top-line results weren't a disaster. Revenue still grew. But two things broke the story the market had been discounting.Meta revealed its first quarterly drop in daily active users a metric that investors had always assumed would continue to rise. And management guided first-quarter 2022 revenue significantly below what analysts had modeled citing pressure from Apple's privacy changes and growing competition for attention from TikTok

The stock didn't go down. It plummeted dropping on the order of a quarter of its value in a single session wiping out roughly $230 billion in market capitalization at the time the largest one-day loss of market capitalization for any company in the history of the stock market.guidance cut told him that assumption was wrong. The stock was not falling by what Meta earned in the fourth quarter of 2021. It was falling by what the market now expected for each subsequent quarter

I think a lot about this one because it's close to a mirror image of the previous Q1 2026 story. Meta's actual quarter was fine and the stock tanked in the forward reading. The S&P 500's Q1 2026 was outstanding and the index barely moved as the forward reading was clouded by questions about capex and a hostile macro environment. Same mechanism opposite headline same lesson: look at what changed in the future not what alreadyit happened

Why the Market Barely Reacted

Now let's go back to the first quarter of 2026 with the framework in hand. The S&P 500 ended the earnings season more or less the same as where it started despite what should have been on a raw basis one of the best fundamental catalysts in years. A few things explain the gap

First of all expectations were already high.A wave of upward estimate revisions in the weeks leading up to the start of the season meant that the bar had already been raised before a single company reported so surpassing it produced less surprise than the raw beat rate implies. Second the Magnificent 7 came with a real countervailing concern: Capital spending on AI is approaching $300 billion a year at hyperscalers and investors are increasingly wondering whether that spending isgenerating revenue fast enough to justify itself. Third and probably the most important factor the macroeconomic backdrop during earnings season was as hostile as they come. The energy shock of the Iran war. Oil above $100 a barrel. An aggressive Federal Reserve holding rates. Tariff uncertainty overlapped. All of that competed for investors' attention with the fundamental story and macroeconomics won

Zooming out this tension will define the second half of 2026: historically strong corporate earnings against a macro backdrop that deteriorates on several fronts at once. History is leaning toward earnings eventually winning as corporate profitability is the primary long-term driver of stock returns and macro headwinds that compress multiples tend to be temporary. What's really unclear is the timeline whether that reversal will occur over months or yearsThe June FOMC meeting is the best argument the bears have. Both are real and the expectations framework above is exactly why the bulls' argument hasn't been enough on its own yet

Where This Model Breaks

I've made expectations versus outcomes seem like a clean universal law. It's not and I want to honestly defend the other side

First the model assumes that markets are efficient enough that the consensus reflects genuinely available information. In illiquid or thinly covered stocks with two or three analysts instead of thirty the consensus can be outdated or simply wrong and a "surprise" could actually simply be the market catching up with information that one analyst already had. Second macroeconomics can simply overwhelm the framework which is a big part of what happened in this very article: even a big surpriseReal upside can get buried under an energy shock a Federal Reserve decision or a geopolitical event that has nothing to do with any individual company's quarter. When market-wide correlations rise during a macro scare stocks stop trading based on their own fundamentals and start trading as a single basket. Third one's own guidance can be fooled. Companies have every incentive to guide conservatively so they can beat their own numbers next quarter meaning a guidance beat is a no-brainer.Sometimes less impressive than it seems and a careful analyst has to adjust to how limited the guidance of a given management team usually is. Fourth at the extremes a quarter can be so catastrophic that it completely negates expectations regardless of what has been discounted.Enron's accounting collapse did not occur due to consensus error. It negotiated on a going concern issue

My honest opinion is that expectations versus results explain most of the variation in earnings day stock movements not all. It's a very good prior not a law of physics

How I Actually Read an Earnings Season

The way I would actually use this framework going through an earnings season like the first quarter of 2026 is to almost completely ignore the headline growth number on the first pass. My read is that the beat rate and the size of the beat 84% of the companies and an average outperformance of 20.7% here tell me more about how the season is landing relative to expectations than the 28.6% growth number alone. I would be a bit worried.high growth number with a low beat rate. A high growth number with a very high beat rate like this tells me that expectations have already been raised and that I shouldn't expect much greater appreciation from growth alone

Next I go straight to the call guidance language even before I finish reading the table of reported numbers. Words like "in line" "ahead" or "below our internal plan" tell me more about the stock's reaction next quarter than anything contained in the press release. I then try to separate the concentrated story from the overall one as this article does with NVIDIA Alphabet and Meta versus the other 493. A market cap-weighted index may seemfantastic and at the same time quietly disguising a much more common underlying economics and I think that gap is one of the most underestimated risks when reading a single combined number

I will say clearly that I find it really difficult to do well in real time. Knowing that expectations matter more than results is easy. Knowing in the moment exactly what the price was set before a specific print is much more difficult because the consensus is public but the whispered number for the most part is not and the honest answer is that I often reconstruct it after the fact from how the stock actually traded rather than predicting it beforehand. This is not investment advice and I am not tellingno one who trades for profit following this logic. It's a way of reading financial news that I think makes the headlines make a lot more sense

The Bottom Line

The first quarter of 2026 produced the S&P 500's best earnings quarter in five years: 28.6% combined growth an 84% beat rate margins at a record 14.8% and a Goldman Sachs quarter for the record books. None of that was enough to move the index because the market had already priced in most of it before the season began and what little was left to react to concerns about spendingcapital in AI and a genuinely hostile macroeconomic environment pointed in another direction. This is not a paradox once you separate the results from expectations. Consensus estimates are formed from dozens of analysts who review their models for weeks before a release and there is often a less visible whisper next to them. Guidance not the last quarter conveys the news that the market has not yet priced in. A stock can beat and fall or fail and recover and both are the model that workscorrectly it doesn't break. The mechanism is not universal. Macroeconomic shocks thin analyst coverage and limited guidance can distort it. But as a way to understand why a great quarter can end with a shrug expectations versus results is the most useful idea I know in equity research