The ESG Rating Agencies: How MSCI, Sustainalytics, and S&P Disagree on the Same Companies
Three agencies rating the same company often reach opposite conclusions. What they're measuring, how they diverge, and what it means for the $35 trillion in capital allocated using their scores.
ESG rating agencies rate the same company using the same publicly available data and routinely reach opposite conclusions. A 2022 MIT Sloan study by Berg, Koelbel, and Rigobon found that the correlation between ESG ratings from major providers is approximately 0.54 — roughly the same as the correlation between two strangers rating a film. Thirty-five trillion dollars in capital is allocated using scores this inconsistent. The agencies are not measuring the same thing. They are not transparent about what they are measuring. And the companies being rated have discovered they can influence the scores.
Key Findings
- The correlation between MSCI and Sustainalytics ESG scores for the same company is approximately 0.54 — the 2022 Berg/Koelbel/Rigobon MIT study found divergence is driven primarily by methodological differences, not data differences
- Tesla received an MSCI ESG rating of “A” (above average) while S&P Global removed it from the S&P 500 ESG Index in 2022 over workplace safety and working conditions concerns
- Exxon Mobil received higher ESG scores than Tesla from some agencies in some periods — because ESG ratings measure risk management practices relative to industry peers, not absolute environmental impact
- The ESG ratings industry generates approximately $1 billion annually in revenue from companies that pay to be rated
- Tariq Fancy, former BlackRock global CIO for sustainable investing, published a direct critique in 2021 stating ESG investing “is a giant financial industry marketing that is a distraction from the genuine, much harder policy work”
How the Agencies Differ
The three major agencies — MSCI ESG Ratings, Sustainalytics (owned by Morningstar since 2020), and S&P Global ESG Scores — each use different methodologies, different data sources, and different weighting systems.
| Dimension | MSCI | Sustainalytics | S&P Global |
|---|---|---|---|
| Rating scale | AAA to CCC | 0-100 risk score (lower = better) | 0-100 score (higher = better) |
| Industry adjustment | Yes — scores relative to industry | Yes | Yes |
| Data sources | Public disclosure + MSCI research | Public disclosure + stakeholder data | SAM Corporate Sustainability Assessment |
| Company engagement | Yes — companies can dispute scores | Yes | Yes — companies submit questionnaire |
The industry-relative adjustment is the critical piece most discussions miss. An oil company rated “AA” by MSCI is not rated favorably in absolute environmental terms. It is rated as a well-managed risk, relative to other oil companies. This is methodologically defensible but routinely misrepresented to retail investors.
The Tesla Contradiction
Tesla is the case that made the divergence visible to a broad audience. In May 2022, S&P Global removed Tesla from the S&P 500 ESG Index. The stated reasons included racial discrimination and poor working conditions claims at Tesla’s Fremont factory, the NHTSA autopilot investigations, and gaps in low-carbon strategy for the non-EV parts of Tesla’s business.
At the same time, Tesla maintained a “BBB” rating from MSCI (later upgraded to “A” in some periods), reflecting the agency’s view that Tesla was managing industry-relative ESG risks adequately.
Elon Musk’s response on Twitter/X characterized the situation accurately: “ESG is a scam.” He noted that Exxon was rated in the top 10 on ESG by S&P while Tesla was excluded. He was correct about the Exxon ranking. Exxon’s score reflected strong governance and risk management practices by the agency’s methodology, not absolute environmental performance.
The agencies are measuring different things. None of them adequately disclosed this to retail investors or the financial press.
The Influence Problem
Berg, Koelbel, and Rigobon’s 2022 MIT study found that divergence between ratings is driven by three factors: scope (what attributes are included), measurement (how attributes are measured), and weights (how attributes are combined into a final score).
The more significant finding, which received less coverage: companies can influence their scores. The major agencies allow companies to review draft ratings and submit additional information before final publication. The agencies frame this as a data quality measure. It is also a mechanism by which companies can dispute negative assessments or provide context that improves their score.
Companies with ESG teams — which larger companies consistently have — systematically produce better ESG disclosure. Better disclosure produces better ESG scores. This is not the same as better ESG performance.
Tariq Fancy’s Critique
Fancy’s 2021 Medium essay, published after he left BlackRock’s sustainable investing division, is the most direct internal critique of the ESG industry from a former practitioner. His core argument: ESG investing doesn’t reduce emissions, improve labor practices, or produce the social outcomes its marketing claims. It prices positive corporate disclosures into asset values without changing corporate behavior.
“The financial services industry is profiting from the ESG craze,” Fancy wrote, “by selling ESG products on the basis of unsubstantiated claims.”
BlackRock, where Fancy worked until 2019, was the largest seller of ESG funds in the world at the time of his critique.
What $35 Trillion Buys
Capital allocated through ESG frameworks has grown from roughly $10 trillion in 2015 to an estimated $35-40 trillion as of 2024, depending on how “ESG-aligned” is defined. Much of this growth came through index inclusion — when companies are included in ESG indices, passive funds tracking those indices buy the stock.
The result: companies with high ESG scores receive cheaper capital than companies with low ESG scores, all else equal. This is the intended mechanism — ESG advocates argue it creates a financial incentive for better corporate behavior.
The Berg study undermines this logic. If two major rating agencies disagree on whether a company has a “high” ESG score with a correlation of 0.54, then the signal companies are responding to is primarily about what each agency’s methodology rewards — not about underlying environmental or social outcomes.
Companies have optimized for ESG rating methodology. That is not the same as improving ESG performance.
Sources
- Berg, F., Koelbel, J., Rigobon, R. “Aggregate Confusion: The Divergence of ESG Ratings.” MIT Sloan Management Review, 2022 — papers.ssrn.com (verified 2026-05-08)
- Tariq Fancy, “The Secret Diary of a Sustainable Investor,” Parts 1-3, Medium, August 2021 (verified 2026-05-08)
- MSCI ESG Ratings Methodology — msci.com (verified 2026-05-08)
- S&P Global ESG Scores Methodology — spglobal.com (verified 2026-05-08)
- “S&P 500 ESG Index Removes Tesla, Adds Exxon,” Reuters, May 18, 2022 (verified 2026-05-08)