Researchbuybacksstock repurchasescapital allocation
Part of the Earnings Quality Analysis Hub series

The Buyback Illusion: We Analyzed $2.9 Trillion in Stock Repurchases - Here's Which Companies Are Wasting Shareholder Money

Companies spent $2.9 trillion on buybacks since 2019. But are they actually reducing shares? Our analysis reveals the uncomfortable truth that 95% of programs are ineffective.

15 min read

The Buyback Illusion: We Analyzed $2.9 Trillion in Stock Repurchases

Last Updated: December 8, 2025 Data Currency: SEC EDGAR 10-K and 10-Q filings through Q3 2025

TL;DR:

  • We analyzed $2.93 trillion in buybacks across 65 S&P 500 companies (2019-2025)
  • 94.9% of buyback programs are ineffective (less than 50% of spending actually retires shares)
  • Microsoft spent $213B on buybacks but shares only decreased 2.7% - $163B was lost to stock-based compensation dilution
  • Only Apple and Wells Fargo have truly effective programs (70%+ effectiveness)
  • The median company has 0% effectiveness - their share count didn't decrease at all

The $2.9 Trillion Question

In 2024, S&P 500 companies spent over $1 trillion on stock buybacks - more than they paid in dividends. Corporate executives tout these programs as "returning value to shareholders."

But here's what they don't tell you: most of that money never actually benefits shareholders.

We analyzed buyback programs at 65 of the largest S&P 500 companies, totaling $2.93 trillion in repurchases since 2019.

What Is Buyback Effectiveness?

Before diving into the data, let's define what we're measuring.

Buyback Effectiveness measures how much of a company's buyback spending actually translates into share reduction.

The formula is simple:

Effectiveness = (Actual Share Reduction %) / (Expected Share Reduction %) × 100

Where:
- Expected Reduction % = Total Buybacks / Average Market Cap × 100
- Actual Reduction % = Starting Shares - Ending Shares / Starting Shares × 100

Example: If a company with a $100B market cap spends $10B on buybacks, we'd expect a 10% reduction in shares. If shares only decreased by 2.5%, the effectiveness is 25% (2.5 / 10 × 100).

Where did the other 75% go? Stock-based compensation. Companies issue new shares to employees and executives, diluting existing shareholders. The buybacks just offset this dilution - they don't actually reduce the share count.

The Shocking Results

MetricValue
Total buybacks analyzed$2.93 trillion
Companies analyzed65
Average effectiveness11.2%
Median effectiveness0%
Companies with less than 50% effectiveness56 (94.9%)
Companies with over 70% effectiveness3 (5.1%)

Let that sink in: the median company achieved 0% effectiveness. They spent billions on buybacks and their share count didn't decrease at all.

The Top 10: Companies Actually Reducing Shares

Only a handful of companies have truly effective buyback programs:

RankCompanyEffectivenessBuybacksShare Change
1Apple (AAPL)100%$648.5B-17.8%
2GE100%*$13.7B-87.8%
3Wells Fargo (WFC)70.2%$109.4B-41.4%
4Qualcomm (QCOM)39.5%$27.2B-8.0%
5American Express (AXP)32.9%$25.7B-15.7%
6Cisco (CSCO)32.5%$53.4B-8.1%
7GE Vernova (GEV)28.3%$2.3B-1.2%
8Alphabet (GOOGL)24.9%$347.6B-12.6%
9Home Depot (HD)23.6%$41.8B-9.4%
10Microsoft (MSFT)23.6%$213.2B-2.7%

*GE's massive share reduction is primarily from spin-offs, not buybacks.

Apple stands alone with 100% effectiveness. They spent $648.5B and reduced shares by 17.8%. That's real capital return.

Wells Fargo at 70.2% is the only other company with genuinely effective buybacks among large-cap financials.

But look at Microsoft - they're in the top 10, yet only 23.6% of their $213B in buybacks actually retired shares. That means $163 billion was consumed by stock-based compensation dilution.

The Bottom 10: Billions Wasted

These companies spent enormous sums on buybacks while their share counts increased:

RankCompanyBuybacksShare ChangeProblem
1JPMorgan (JPM)$121.3B+6.6%Shares increased
2NVIDIA (NVDA)$89.5B+1.0%SBC dilution
3Exxon (XOM)$69.0B+0.7%No reduction
4Procter & Gamble (PG)$59.8B+40.0%Acquisition dilution
5Chevron (CVX)$60.5B0%No reduction
6Intel (INTC)$50.3B+2.6%Shares increased
7Booking Holdings (BKNG)$44.3B+2.1%No reduction
8Johnson & Johnson (JNJ)$43.6B0%No reduction
9Broadcom (AVGO)$38.8B+1085%VMware acquisition
10McDonald's (MCD)$28.6B0%No reduction

JPMorgan is the most egregious case among large banks. They spent $121.3 billion on buybacks, yet their share count increased by 6.6%. Every dollar was swallowed by compensation and acquisitions.

NVIDIA spent $89.5 billion while shares went up 1%. Their stock-based compensation completely overwhelmed their buyback program.

Case Study: Microsoft's $163 Billion SBC Problem

Let's dive deeper into Microsoft, a company often praised for its capital return program.

YearShares OutstandingBuybacks
20197.64B$76.7B
20207.57B$26.1B
20217.52B$29.2B
20227.46B$28.6B
20237.43B$20.0B
20247.43B$17.5B
20257.43B$15.0B
Total-210M shares (-2.7%)$213.2B

Microsoft spent $213.2 billion over 6 years, but only retired 210 million shares (2.7% reduction).

At their average stock price of ~$300 during this period, 210 million shares would cost about $63 billion to repurchase.

Where did the other $150 billion go?

Stock-based compensation. Microsoft issues massive amounts of stock to employees:

  • Engineers receive RSUs worth hundreds of thousands of dollars
  • Executives get performance shares worth millions
  • Acquisitions often include stock as payment

Every share issued to employees is a share that buybacks must offset before they benefit existing shareholders.

The math is brutal:

  • Expected reduction: $213B / $1.83T avg market cap = 11.7%
  • Actual reduction: 2.7%
  • Effectiveness: 2.7 / 11.7 = 23.6%
  • Lost to SBC: $213B × (1 - 0.236) = $163 billion

Why This Matters for Investors

When you hear a company announce a "massive buyback program," here's what you should ask:

1. What's Their Historical Effectiveness?

If a company has 0% effectiveness over the past 5 years, their new buyback won't reduce shares either. It's just offsetting SBC.

2. How Much Stock-Based Compensation Do They Pay?

Companies with high SBC will have low buyback effectiveness. Tech companies are particularly notorious for this.

3. Are They Making Acquisitions?

Stock-funded acquisitions (like Broadcom's VMware deal) can completely overwhelm buyback programs.

How to Find Effective Buyback Programs

Based on our analysis, here are the characteristics of effective buyback programs:

Signs of an Effective Buyback Program:

  • Shares outstanding declining year-over-year consistently
  • SBC as a percentage of revenue below 5%
  • No major stock-funded acquisitions
  • Cash buybacks (not debt-funded)
  • Management owns significant stock (aligned incentives)

Companies that meet these criteria:

  • Apple - Minimal SBC, massive cash generation, consistent execution
  • Wells Fargo - Regulated bank with lower SBC
  • Qualcomm - Disciplined capital allocation

Sector Analysis: Where Buybacks Work (and Don't)

SectorAvg EffectivenessProblem Area
Technology15-25%High SBC dilution
Financials10-70%Varies by institution
Consumer Staples0-10%Acquisitions
Energy0%No reduction
Healthcare0-20%Mixed results
Industrials15-25%Moderate SBC

Technology has the worst effectiveness due to stock-based compensation culture. Even "good" tech companies like Microsoft only achieve 20-25%.

Financials have the widest range - Wells Fargo is highly effective, while JPMorgan is not.

Energy companies like Exxon and Chevron spent $130B combined with essentially no share reduction.

The Bottom Line: Stop Celebrating Buyback Announcements

When a company announces a $50 billion buyback program, the financial media celebrates. Analysts upgrade the stock. Investors pile in.

But our data shows that 95% of the time, those buybacks won't actually reduce shares.

Before cheering the next buyback announcement, ask yourself:

  1. What's this company's historical effectiveness?
  2. How much do they spend on stock-based compensation?
  3. Have their shares actually decreased over the past 5 years?

If you can't answer these questions, that "shareholder-friendly" buyback might just be an illusion.


Methodology

Data Source: SEC EDGAR 10-K and 10-Q filings via XBRL

Time Period: 2019-2025

Metrics Used:

  • stock_repurch - Payments for repurchase of common stock (cash flow statement)
  • shares_outstanding - Common shares outstanding (balance sheet)
  • market_cap - Market capitalization (price × shares)

Effectiveness Formula:

Expected Reduction % = Total Buybacks / Average Market Cap × 100
Actual Reduction % = (Start Shares - End Shares) / Start Shares × 100
Effectiveness = min(100%, Actual / Expected × 100)

Adjustments:

  • Stock splits adjusted using known split ratios (AAPL 4:1, GOOGL 20:1, NVDA 40:1, AMZN 20:1, TSLA 3:1)
  • Companies with major acquisitions flagged as "warning" data quality

Limitations:

  • Analysis covers 65 companies from top 100 by market cap (35 excluded due to insufficient data)
  • Effectiveness capped at 100% (some companies may repurchase at discounts)
  • Stock-based compensation not directly measured - inferred from gap between buybacks and share reduction

Explore More Earnings Quality Analysis

This article is part of our comprehensive Earnings Quality Hub, which covers cash flow verification, accounting red flags, and quality screening frameworks.

Related earnings quality research:

MetricDuck Research

CFA charterholders and former institutional equity analysts