Amsterdam's Top-Tier Score Was Never Checked — And the Same Review Credited the City for Speaking Up

A federal lawsuit filed September 14 by Mahmoud Khalil and a student group, and a second lawsuit filed in August by other students and faculty, both claim Columbia failed to protect them from harassment. Neither claim was used in this measurement — both are pending in court, and Columbia has not been found liable in either. What was used is already decided. On May 23, 2025, the U.S. Department of Health and Human Services and the Department of Education jointly found that Columbia showed "deliberate indifference" toward harassment of Jewish students for more than 19 months starting in October 2023, including that the university had no effective way to report antisemitism until the summer of 2024. Columbia then agreed to pay 221 million dollars combined to the U.S. Treasury and the Equal Employment Opportunity Commission — the largest public settlement the commission has reached in almost 20 years, now being paid out — and accepted outside-set rules on student discipline in exchange for restoring federal funding. Columbia publishes the settlement on its own website. Columbia's score would land in the same tier even if both pending lawsuits were thrown out tomorrow — the measurement does not depend on them. Three of the government web pages behind the federal finding blocked automated access, so that finding is recorded as a claim tied to the government's own web address rather than as a word-for-word quote.

A Seoul city official used a contractor's corporate card 51 times over about a year to buy roughly 21 million won (about 15,000 dollars) of personal goods, including a laptop for his child, after signing a 1.04-billion-won contract with that same contractor. That is Seoul's own failure: its own employee, its own contract decision, its own internal check that missed the misuse for a year. The catch is not Seoul's. South Korea's national Board of Audit and Inspection, a body separate from the city, found the misuse, filed a criminal complaint, and recommended penalties. Seoul's own documented role so far is that it was told after the fact; the official had already retired in June. The measurement credits Seoul for being subject to a national audit system that works, but does not credit the city's own record for a correction the city has not yet been shown to have made itself. Seoul's real strength is its welfare spending: 451.3 billion won committed over five years against loneliness and isolation deaths, and disability welfare spending raised 8.1% to 1.77 trillion won in 2025 — real, well-documented programs that keep the score from falling further.

All five measured well below their published scores — Zoetis measured lowest, 34.1 points below its published value; Travelers measured highest of the five, still 12.8 points below its published value. The reason is not five separate declines. All five companies' published scores were never individually checked; they came from one of two starting values assigned identically, digit for digit, to unrelated companies across unrelated industries. Waste Management and Zoetis share one starting score with nine other companies (11 total). SEI Investments, The Hartford and Travelers share a second starting score with sixteen other companies (19 total). A score that is identical across 11 or 19 unrelated companies cannot reflect real measurement of any single one of them. Because the pattern points at the benchmark's own unchecked starting data rather than at any one company's conduct, no score change was proposed for any of the five this cycle. Instead, the finding is being treated as one shared question about whether and how the remaining 25 companies sharing these two starting scores should be checked. SEI Investments' reading carries the least confidence of the five: about a third of its measured behaviors rest on the absence of any public disclosure — its most recent sustainability report is from 2023 — rather than on documented conduct either way.

• Amsterdam (94.4 published, 72 measured): the city's score rose for its own homelessness advocacy letter, and fell on an independent watchdog finding and a discriminatory algorithm the city itself let outside researchers expose.
• Seoul (60.9 published, 49 measured): a year of contractor-card misuse by a city official was caught by South Korea's national auditor, not the city, so Seoul's own record does not get the credit.
• Both retain real strengths — Amsterdam's openness about its own failures, Seoul's welfare spending — that keep both scores well above the benchmark's lower tiers.

• Columbia University (44.5 published, 31 measured): a May 2025 federal deliberate-indifference finding and a 221-million-dollar settlement Columbia itself publishes drive the score; two pending lawsuits are left out as unproven claims.

• El Salvador (15.0 published, held): the expiry of U.S. protected status for about 200,000 Salvadorans is a United States decision and does not count against the country.
• Thailand (32.8 published, 29.4 measured): an exam-fraud scandal's seizures and irregularities are treated as facts; who is responsible has not yet been decided.