What S&P 500 Earnings Estimates Signal, and Where Yardeni Adds Value

What S&P 500 Earnings Estimates Actually Tell You (And What They Don't)

S&P 500 earnings estimates are the single most consequential input to equity valuations, yet most commentary on them is written for people who don't already own $5M+ in equities. This piece covers what the estimates actually show, where Ed Yardeni's framework adds value, where it doesn't, and how to use estimate revision cycles to make concrete decisions in large taxable accounts.

How S&P 500 Earnings Estimates Are Built

S&P 500 earnings estimates aggregate the projected earnings per share (EPS) across all 500 index constituents. Two methodologies dominate: bottom-up and top-down.

Bottom-up estimates start at the company level. Analysts model individual income statements, roll up to sector totals, and aggregate to an index figure. This approach captures company-specific dynamics but tends toward optimism because sell-side analysts have structural incentives to maintain access to management teams.

Top-down estimates start with macroeconomic variables, including GDP growth, profit margins, and interest rates, and work down to an implied EPS figure. Yardeni Research uses a version of this approach, which is why his estimates often diverge from the bottom-up consensus when macro conditions are shifting.

According to research published in the Journal of Finance, sell-side consensus forecasts have historically exceeded realized earnings by a meaningful margin, particularly at the start of a fiscal year. The optimism bias is not random noise. It is systematic, and it compounds when you are managing a large taxable portfolio and making allocation decisions based on earnings trajectory.

For a deeper look at how earnings per share trends have evolved across market cycles, the historical data from FRED provides a useful baseline against which to stress-test any forward estimate you are currently using.

What Ed Yardeni's Framework Actually Offers

Yardeni Research publishes regularly updated forward earnings estimates and valuation metrics for the S&P 500, including 52-week forward EPS consensus data used widely by institutional investors. The firm's value is not in producing a single annual number. It is in the ongoing revision process and the analytical framework behind it.

Yardeni introduced the "Fed Model," which compares the S&P 500 earnings yield (the inverse of the forward P/E ratio) to the 10-year Treasury yield as a valuation signal. The logic is straightforward: when the earnings yield exceeds the Treasury yield by a wide margin, equities look cheap relative to bonds. When the spread compresses or inverts, the risk-reward shifts.

The model has real utility as a relative valuation tool. It also has documented limitations. Cliff Asness of AQR has criticized it for lacking predictive validity across interest rate regimes, particularly in environments where both yields and equity valuations are elevated simultaneously. That criticism is worth taking seriously. A framework built on a spread relationship can break down when both sides of the spread are distorted by central bank policy.

The practical takeaway: use the earnings yield as one input when assessing fair value estimates, not as a standalone signal. Cross-reference it with historical P/E ratio valuations to understand where current multiples sit relative to long-run averages.

Ed Yardeni's Current S&P 500 Earnings Estimates for 2025 and 2026

As of early 2025, consensus 2025 EPS estimates for the S&P 500 sit in the range of $270 to $280 per share, according to S&P Global's official earnings data. Goldman Sachs Global Investment Research publishes its own multi-year EPS forecasts that serve as a major institutional benchmark, and those figures have broadly tracked the consensus range while reflecting somewhat more conservative assumptions about margin sustainability.

Yardeni's estimates have generally sat at the higher end of the forecast distribution, consistent with his long-standing view that American corporate earnings are more resilient than consensus gives them credit for during periods of macro uncertainty.

The forward 12-month P/E ratio has traded above its 10-year historical average of approximately 18x for much of this period. That premium implies the market is pricing in continued double-digit earnings growth with limited tolerance for estimate cuts.

Here is where the math gets uncomfortable for large portfolios. A 10% downward revision to 2025 EPS estimates, without a corresponding P/E expansion to offset it, produces a meaningful drawdown in absolute dollar terms. On a $10M equity allocation, a 10% index decline is $1M. That is not a theoretical risk. It is a planning input.

Estimates as of early 2025. Forward P/E calculated using approximate index level. Figures subject to revision.

How Accurate Are S&P 500 Earnings Estimates Historically?

The honest answer is: directionally useful, quantitatively unreliable, and systematically biased upward.

FRED's historical S&P 500 reported EPS data allows direct comparison of realized earnings against prior consensus forecasts. The pattern is consistent. In years that include a recession, sell-side consensus estimates made at the start of the calendar year have overestimated actual earnings by roughly 5 to 10%. In strong recovery years, they have underestimated. Yardeni has written about this dynamic extensively under the concept of "earnings resilience," arguing that analysts systematically underweight the ability of large-cap companies to protect margins through cost cuts and pricing power.

FactSet's weekly Earnings Insight report provides the most granular ongoing data on this. Their tracking shows S&P 500 companies have beaten consensus EPS estimates approximately 70 to 75% of quarters over the past decade. That sounds like analysts are too conservative. They are not. The beat rate reflects the well-documented practice of companies guiding analysts toward beatable numbers in the weeks before earnings.

The real benchmark institutional investors trade against is the "whisper number," which sits above the published consensus. A company that beats the consensus but misses the whisper number often sells off on the news. If you hold concentrated large-cap positions and you are reading a consensus beat as automatically bullish, you are using the wrong benchmark.

Sources: FRED, S&P Global, FactSet. 2024 realized figure is estimated pending final reporting.

How Earnings Estimates Affect S&P 500 Valuations

Earnings estimates are the denominator in every forward P/E calculation. That relationship is mechanical. What is less mechanical is how the market prices estimate revisions in real time.

When estimates are revised upward, the P/E ratio falls at a given index level, making the market look cheaper. When estimates are revised downward, the P/E rises, and the market looks more expensive unless prices fall proportionally. The CFA Institute's foundational equity valuation framework establishes that estimate revision momentum is a documented return predictor: stocks with rising estimate revisions tend to outperform, and stocks with falling revisions tend to underperform, even after controlling for other factors.

At the index level, this dynamic plays out through sector rotation. When technology sector EPS estimates are rising faster than the broader index, the market's aggregate P/E gets pulled upward by the sector's weight. That is part of what has driven the elevated forward P/E readings in 2024 and early 2025.

Morningstar's equity research team publishes aggregate market valuation assessments that incorporate forward earnings estimates, providing an independent cross-check to forecasters like Yardeni. Their fair value methodology tends to be more conservative than pure forward P/E analysis because it discounts the optimism bias in consensus estimates.

For context on where current valuations sit against long-run history, historical S&P 500 returns and long-term market performance data both underscore that periods of elevated forward P/E ratios have historically compressed subsequent 10-year returns.

Using Earnings Estimate Revisions for Tax-Loss Harvesting

This is where earnings estimate analysis becomes directly actionable for FATFIRE investors, and where most coverage fails to go.

Sectors experiencing sharp downward EPS revisions often create tax-loss harvesting windows before the broader market fully reprices. Energy in 2020 and regional banks in 2023 are clean examples. In both cases, estimate cuts preceded the full price decline by several weeks, giving investors in those sectors an opportunity to realize losses while maintaining market exposure through similar-but-not-identical ETFs.

The mechanics matter here. Under IRC Section 1091, the wash-sale rule prohibits claiming a loss if you repurchase a "substantially identical" security within 30 days before or after the sale. Swapping from a regional bank ETF to a broad financial sector ETF, or from an energy producer to an energy infrastructure fund, typically satisfies the "not substantially identical" standard while preserving sector exposure. Confirm the specific swap with your tax attorney before executing.

The timing signal to watch: when FactSet's weekly Earnings Insight shows a sector's forward EPS estimate declining for three or more consecutive weeks, that is a meaningful revision trend, not noise. Pair that with a position that is sitting at a loss in your taxable account, and you have a concrete harvesting trigger.

For investors tracking dividend payout ratios alongside earnings estimates, downward EPS revisions in dividend-heavy sectors also create a secondary signal: when payout ratios rise above 80% because earnings are falling faster than dividends, dividend cuts often follow, which accelerates the price decline and extends the harvesting window.

Bottom-Up vs. Top-Down S&P 500 Earnings Forecasts: Which One to Use

The distinction matters more than most investors realize.

Bottom-up consensus (aggregated from FactSet, Bloomberg, or S&P Global) reflects what company analysts expect from each individual business. It is granular and current, but it carries the systematic optimism bias described earlier. It is also slow to reprice during macro dislocations because individual analysts update models on their own schedules.

Top-down estimates, including Yardeni's, move faster during macro shifts because they start with GDP and margin assumptions rather than company-level models. When the Federal Reserve is tightening aggressively or a credit event is unfolding, top-down estimates tend to be revised downward earlier and more sharply than bottom-up consensus.

For portfolio construction purposes, the most useful approach is to track both and pay attention to the spread between them. When top-down estimates are materially below bottom-up consensus, that gap represents a risk that the market has not fully priced. When top-down estimates are above consensus, as Yardeni's often are, it signals a more optimistic macro view that you can either accept or fade based on your own assessment.

Tracking S&P 500 market forecasts from multiple sources simultaneously gives you a cleaner read on where the genuine disagreement lies, which is where the actionable information lives.

Earnings Estimate Scenarios FATFIRE Investors Should Plan For

Scenario planning around earnings estimates is more useful than point forecasts, particularly for investors managing large taxable accounts where the sequencing of gains and losses has real tax consequences.

Three scenarios worth stress-testing against your current allocation:

Base case (consensus holds): 2025 EPS comes in around $270 to $280. The market trades at 19 to 20x forward earnings. Modest single-digit index returns. No major rebalancing trigger. The primary action here is monitoring sector performance trends for rotation opportunities within your equity allocation.

Upside case (Yardeni's scenario): 2025 EPS reaches $285 or above, driven by AI-related productivity gains and margin expansion in technology. The market re-rates to 21x or higher. If you are underweight technology relative to the index, this scenario penalizes you. The tax consideration: if you have embedded gains in tech positions, a strong earnings beat cycle is not the time to harvest them.

Downside case (10%+ estimate cut): A recession, credit event, or significant policy shock drives 2025 EPS toward $245 or below. At a 19x multiple, that implies an index level around 4,650, roughly 15% below early 2025 levels. At a 17x multiple, the implied level drops further. On a $10M equity portfolio, that is a $1.5M drawdown in the base downside case. This scenario is when tax-loss harvesting windows open across multiple sectors simultaneously, and when earnings yield as a valuation metric becomes the most useful reentry signal.

Illustrative scenarios based on early 2025 consensus data. Not a forecast.

How to Read Yardeni's Estimates Without Over-Weighting Them

Yardeni's track record is strong relative to sell-side consensus, particularly in identifying earnings resilience during periods when consensus was too pessimistic. His framework is transparent, consistently published, and grounded in macro variables that are independently verifiable.

That said, no single forecaster should anchor your portfolio decisions. The documented optimism in Yardeni's estimates during certain periods, combined with the inherent limitations of the Fed Model in non-standard rate environments, means his figures are best used as one input in a broader framework.

The most productive use of Yardeni Research's output for a FATFIRE investor is not to take the EPS number at face value. It is to understand the assumptions behind it, specifically the GDP growth rate, the net profit margin assumption, and the revenue growth rate embedded in the estimate. When you disagree with one of those inputs, you have a specific, testable thesis rather than a vague sense that the market is overvalued.

For investors who want to cross-reference Yardeni's revenue assumptions against actual index-level data, revenue analysis of top companies provides a useful ground-truth check on whether top-line growth is tracking the assumptions embedded in current EPS estimates.

References

• Yardeni Research -- "S&P 500 Earnings, Revenues & Valuation" (2025)
• FactSet -- "Earnings Insight: S&P 500" (2025)
• Federal Reserve Bank of St. Louis (FRED) -- "S&P 500 Earnings Per Share" (2025)
• CFA Institute -- "Equity Asset Valuation (CFA Institute Investment Series)" (2020)
• Journal of Finance -- "Analyst Forecast Accuracy and the Role of Earnings Guidance" (2010)
• S&P Global -- "S&P 500 Index Earnings and Estimate Data" (2025)
• Morningstar -- "Market Fair Value and Earnings Outlook Reports" (2025)
• Goldman Sachs Global Investment Research -- "US Equity Outlook: S&P 500 EPS Forecasts" (2025)