CONVEX
Glossary/Equity Markets/Earnings Per Share
Equity Markets
6 min readUpdated Apr 12, 2026

Earnings Per Share

ByConvex Research Desk·Edited byBen Bleier·
EPSearningsnet income per sharediluted EPSquarterly earningsearnings season

A company's net profit divided by its outstanding shares, the most fundamental measure of corporate profitability, the primary driver of long-run equity returns, and the basis for P/E ratio valuation.

Current Macro RegimeSTAGFLATIONDEEPENING

The macro regime is unambiguously STAGFLATION DEEPENING. The hot CPI print (pending event, 24h ago) is not a surprise — it is a CONFIRMATION of the pipeline signals that have been building for weeks: PPI accelerating faster than CPI, Cleveland nowcast at 5.28%, breakevens rising +10bp 1M across the …

Analysis from May 14, 2026

What Is Earnings Per Share?

Earnings Per Share (EPS) is the most fundamental measure of corporate profitability, the amount of net profit attributable to each share of common stock:

EPS = Net Income ÷ Weighted Average Shares Outstanding

For diluted EPS (the standard measure used by Wall Street), the denominator includes all shares that could potentially be created through stock options, RSUs, convertible securities, and warrants.

EPS is the single most important number in equity markets. Over any multi-year horizon, stock prices track earnings. A company that grows EPS at 12% per year will, on average, see its stock price compound at approximately the same rate. Everything else, sentiment, momentum, flows, narrative, creates noise around this earnings signal.

The S&P 500 EPS History

Period S&P 500 Operating EPS Avg Annual Growth Key Driver
2000-2002 $57 → $47 -9%/yr Dot-com bust; earnings recession
2003-2007 $47 → $92 +14%/yr Housing boom, financial leverage
2008-2009 $92 → $57 -22%/yr GFC; worst earnings collapse since 1930s
2010-2019 $57 → $163 +11%/yr Recovery + buybacks + tax cuts (2018)
2020 $163 → $140 -14% COVID recession
2021 $140 → $208 +49% Stimulus-fueled recovery; base effect
2022-2024 $208 → $240+ +5%/yr Normalisation; AI-driven tech earnings

The long-term trend: S&P 500 EPS has grown at approximately 6-7% annually since 1950, compounding from $2.84 in 1960 to $240+ in 2024. This ~7% earnings growth plus a ~2% dividend yield equals the ~9-10% annualized total return of the S&P 500 over the long run.

Decomposing Equity Returns

Stock market returns have three components, and understanding which is driving performance is essential:

Component Definition Typical Contribution 2020-2021 2022
EPS growth Change in underlying earnings 6-7%/yr (long-run avg) +49% (recovery) +5%
Dividend yield Cash returned to shareholders 1.5-2.0%/yr 1.3% 1.6%
Multiple change Change in P/E ratio 0%/yr (long-run) +20% (expansion to 21x) -25% (compression to 15x)

The critical insight: multiple expansion is borrowed from the future. When the S&P 500 P/E expands from 17x to 22x, it boosts returns by ~30%, but the P/E cannot keep expanding indefinitely. At some point, multiples revert toward the long-run average, and future returns depend entirely on EPS growth. Investors who buy at 22x forward P/E are implicitly accepting lower forward returns than investors who buy at 15x.

The EPS Revision Cycle

Why Revisions Matter More Than Reported Numbers

The direction of analyst EPS estimate revisions is one of the most powerful short-term equity trading signals:

Revision Pattern Typical Stock Performance Why
Estimates rising (3+ analysts raising) Outperforms by 5-10% annualized Business momentum is improving; estimates are still catching up
Estimates stable In-line with market No new information being priced
Estimates falling (3+ analysts cutting) Underperforms by 5-10% annualized Business deteriorating; estimates haven't fully adjusted
Large upward revision by top-ranked analyst Immediate rally (1-3% typical) High-credibility signal; institutions reposition

The "Estimate Staircase" Pattern

Analyst revisions are inherently serial and conservative: an analyst who raises estimates once is statistically likely to raise them again, because the business momentum that caused the first revision typically continues. This creates a staircase of upward revisions over multiple quarters that the market prices in gradually, and the stock continues to outperform for 3-6 months after each revision.

Earnings Season: The Quarterly Event

Schedule and Structure

Quarter Reporting Period Peak Season Key Reporters
Q4 January-February Mid-Jan to mid-Feb Banks lead (JPM, GS, BAC), then tech
Q1 April-May Mid-Apr to mid-May Same structure
Q2 July-August Mid-Jul to mid-Aug Same structure
Q3 October-November Mid-Oct to mid-Nov Same structure

Beat Rate and Expectations

On average, approximately 75% of S&P 500 companies beat consensus EPS estimates each quarter. This high beat rate exists because: (1) companies sandbag guidance (setting low bars), (2) analysts calibrate estimates to be beatable, and (3) the "whisper" number (what the market really expects) is typically higher than the published consensus.

The market reaction depends not on whether a company beats, but by how much and what it says about the future:

Outcome Typical Market Reaction
Beat EPS + raise guidance +3-10% (strongest positive)
Beat EPS + maintain guidance +1-3% (modest positive)
Beat EPS + lower guidance -2-8% (guidance overrides beat)
Miss EPS + maintain guidance -3-8%
Miss EPS + lower guidance -5-15% (worst case)

GAAP vs Non-GAAP EPS

Measure Includes Excludes Best Use
GAAP EPS All revenues and expenses per US accounting standards Nothing, comprehensive Regulatory filings; long-term analysis
Non-GAAP EPS "Core" or "adjusted" operating earnings SBC, restructuring, amortization, impairments Peer comparison; trend analysis
Free Cash Flow per Share Cash generated after capex Non-cash charges entirely True cash earnings power

The GAAP-to-non-GAAP gap for the S&P 500 is approximately 15-25% and has been widening, particularly in the tech sector where stock-based compensation can represent 15-30% of operating expenses. Critics argue that non-GAAP has become "earnings without the bad stuff", a way to flatter results by excluding real economic costs.

EPS Across the Business Cycle

Cycle Phase EPS Growth Trend Typical Sectors Leading Investment Implication
Early recovery Trough to strong growth (base effect) Financials, Consumer Discretionary Buy cyclicals; earnings revision momentum
Mid-cycle expansion Steady 8-12% growth Technology, Industrials Broad equity exposure; rising tide
Late cycle Decelerating; margin pressure Energy, Healthcare Rotate to quality; earnings growth scarce
Recession Negative growth (-15 to -30% typical) Utilities, Staples (defensive) Avoid cyclicals; EPS estimates still too high

The earnings recession signal: When S&P 500 EPS estimates for the next 12 months start declining (not just slowing growth, but actual declines in the estimate), the economy is approaching or entering a recession approximately 80% of the time. This signal led by 2-4 months in 2001, 2008, and 2020.

What to Watch

  1. S&P 500 forward EPS estimate, the single best real-time gauge of corporate America's earnings trajectory. Track weekly on FactSet or Yardeni Research.
  2. Estimate revision breadth, the percentage of S&P 500 companies with rising vs falling estimates. Above 50% = positive; below 40% = recession risk.
  3. GAAP vs non-GAAP gap, if the gap is widening for a company or sector, investigate what is being excluded and whether it is truly non-recurring.
  4. Earnings yield vs Treasury yield, S&P 500 earnings yield (1/P/E) minus 10Y Treasury yield. When this spread narrows below 1%, equities are "expensive" relative to bonds.
  5. Buyback-adjusted EPS growth, some companies grow EPS primarily through share count reduction rather than revenue growth. Separating organic EPS growth from buyback-driven EPS growth reveals true business momentum.

Frequently Asked Questions

What is the difference between basic and diluted EPS?
Basic EPS uses the actual number of shares currently outstanding as the denominator. Diluted EPS adds all shares that could potentially be created through the exercise of stock options, vesting of restricted stock units (RSUs), conversion of convertible bonds, and exercise of warrants. For example, if a company has 1 billion basic shares outstanding and 50 million in stock options that could be exercised, diluted shares would be approximately 1.05 billion (the exact calculation uses the treasury stock method, which accounts for the exercise price). Diluted EPS is always equal to or lower than basic EPS and is the standard measure used by Wall Street analysts. The gap between basic and diluted EPS can be significant at tech companies with heavy stock-based compensation — at companies like Salesforce or Palantir, dilution from stock-based comp can reduce EPS by 10-20% relative to basic. This is one reason investors increasingly focus on free cash flow per share rather than EPS, since FCF is not affected by non-cash stock-based compensation charges.
Why do stocks sometimes drop on an EPS beat?
A stock dropping on an EPS beat is one of the most confusing events for new investors, but it happens frequently for several reasons: (1) Guidance matters more than the beat — if a company beats Q3 estimates by $0.05 but lowers Q4 guidance by $0.20, the net effect on forward earnings expectations is negative. The stock responds to the guidance, not the beat. This is why analysts say "it is a guidance game, not an earnings game." (2) Whisper numbers — Wall Street has official consensus estimates, but hedge funds and institutional traders often have "whisper" numbers that are higher than consensus. A company can beat the published consensus but miss the whisper number that sophisticated investors were actually positioning for. (3) Revenue quality — a company might beat EPS through cost cuts, one-time gains, or accounting adjustments while missing on revenue or reporting slowing growth rates. The market rewards sustainable revenue growth more than financial engineering. (4) Valuation already priced in — if a stock has rallied 30% into earnings on expectations of a blowout quarter, even a strong beat may not exceed the expectations already embedded in the price. The classic "buy the rumor, sell the news" pattern. (5) Margin trends — a company can beat EPS but show declining gross or operating margins, signaling future earnings pressure even if the current quarter looks good.
How do EPS estimate revisions predict stock performance?
EPS estimate revisions are one of the most powerful short-term equity signals, backed by decades of academic and practitioner research. The "earnings estimate revision momentum" factor (also called "SUE" — Standardized Unexpected Earnings) shows that stocks with rising consensus EPS estimates outperform stocks with falling estimates by approximately 5-10% annually, with the effect persisting for 3-6 months after the revision. Why it works: analyst revisions are inherently conservative and serial — analysts who raise estimates once tend to raise them again because the underlying business momentum that caused the first revision typically continues. The result is a staircase pattern of upward revisions that the market prices in gradually. How to use it: (1) Track the direction of the next-quarter and next-year consensus EPS estimates for your holdings weekly (available on Bloomberg, FactSet, or free on Yahoo Finance). (2) When 3+ analysts raise estimates in the same quarter, the stock typically has further upside. (3) The most powerful signal is a large revision by a historically accurate analyst (measured by past hit rate). (4) Sector-wide revisions can signal macro trends — if all energy analysts are raising estimates simultaneously, it reflects commodity price strength that may affect other asset classes.
What is earnings season and how should traders position for it?
Earnings season occurs four times per year, starting approximately 2 weeks after the end of each calendar quarter. The busiest period is typically the 3rd through 5th week after quarter-end, when the majority of S&P 500 companies report. Key dates: Q4 earnings season starts in mid-January; Q1 in mid-April; Q2 in mid-July; Q3 in mid-October. Banks (JPMorgan, Goldman Sachs) traditionally kick off each season. Mega-cap tech (Apple, Microsoft, Amazon, Google, Meta) usually reports in the 4th-5th week and is the most market-moving cluster. Positioning strategies: (1) Pre-earnings: implied volatility (IV) rises ahead of reports because the market knows a large move is coming but not the direction. Option sellers can sell premium (straddles or strangles) to profit from the IV crush after earnings. Option buyers can buy directional exposure if they have conviction. (2) Post-earnings: the initial market reaction is often wrong or exaggerated — "earnings drift" studies show that stocks that beat estimates continue to outperform for 60-90 days after the report, and stocks that miss continue to underperform. (3) Sector signals: when the first companies in a sector report, their results and guidance provide signals for the rest of the sector. If JPMorgan reports strong loan growth, other banks likely benefit from the same trend.
What is the difference between GAAP and non-GAAP (adjusted) EPS?
GAAP (Generally Accepted Accounting Principles) EPS includes all revenue and expenses as defined by US accounting standards — it is the official, audited number. Non-GAAP or "adjusted" EPS excludes items that management considers one-time, non-recurring, or non-cash — stock-based compensation (SBC), restructuring charges, amortization of acquired intangibles, litigation settlements, and impairment charges. The gap between GAAP and non-GAAP has widened dramatically, especially in the tech sector. For the S&P 500 as a whole, non-GAAP EPS typically exceeds GAAP EPS by 15-25%. For individual tech companies, the gap can be enormous: in 2023, some large tech companies reported non-GAAP EPS 50-100% higher than GAAP EPS, primarily because of massive stock-based compensation exclusions. The debate: companies argue that non-GAAP better reflects "underlying" business performance by removing noise. Critics argue that SBC is a real economic cost to shareholders (dilution), restructuring is a recurring cost for serial acquirers, and non-GAAP has become a tool for flattering results. Best practice: use non-GAAP for trend analysis and peer comparison (since GAAP can be distorted by one-time items), but always check the GAAP-to-non-GAAP reconciliation to understand what is being excluded and whether the exclusions are truly non-recurring.

Earnings Per Share is one of the signals monitored daily in the AI-driven macro analysis on Convex Trading. The platform synthesises data across monetary policy, credit, sentiment, and on-chain metrics to generate actionable trade recommendations. Create a free account to build your own signal layer and see how Earnings Per Share is influencing current positions.

ShareXRedditLinkedInHN

Macro briefings in your inbox

Daily analysis that explains which glossary signals are firing and why.