ESG and Sustainable ETFs

The families, from lightest to heaviest

Ordered by how far they move from the parent index:

1. Exclusion screens ("SRI-lite", the ESG Screened family). Remove controversial weapons, tobacco, thermal coal above thresholds, and UN Global Compact violators. Typical result: 95–99% overlap with the parent, tracking within tenths of a percent. This is the default of most "ESG" versions of mainstream indices, and honesty requires saying plainly: it is nearly the same fund.

2. Best-in-class tilting (ESG Leaders, ESG Enhanced). Keep every sector but overweight higher-rated companies within it. Sector-neutral by design, which preserves the parent's shape while shifting which oil major or bank you hold. Deviation: moderate; tracking difference versus parent: a percent or so per year, either direction.

3. Strict SRI (best-in-class with deep cuts). Keep only the top quartile or so by ESG rating, plus harder exclusions. Now you own a meaningfully different portfolio: often 100–400 stocks instead of 1,400, with real sector and factor tilts (structurally light on energy, heavy on tech and quality). Deviation: large, and the return story of the last decade is substantially this tilt's story.

4. Climate-specific: Paris-Aligned (PAB) and Climate Transition (CTB) benchmarks. EU-regulated categories with mandated properties: an immediate carbon-intensity reduction versus the parent (50% for PAB, 30% for CTB) plus a 7% per year decarbonization trajectory. These are rules-based decarbonizing machines, and the trajectory requirement means the portfolio must keep moving, year after year: a structural feature no other family has.

5. Thematic sustainability (clean energy, water, circular economy): not really this chapter; these are Chapter 47's thematic funds with a green subject, and they behave like concentrated sector bets, because they are.

Does it cost or add return? The honest answer

Three findings, each stated with its caveat:

1. The theoretical prior is a small cost. Constraining an opportunity set cannot increase the theoretical maximum; a screened universe is a subset. Whether that matters in practice at 95% overlap is another question: it plainly doesn't at rung 1.

2. The realized record is roughly a wash, with a large sector asterisk. ESG funds broadly kept pace with parents over the 2010s (many outperformed), then lagged notably in 2022. Both facts have one explanation: strict ESG portfolios are structurally underweight energy and overweight technology and quality. That combination won for a decade and lost badly in the 2022 energy year. Read carefully, the "ESG outperforms" literature of 2019 and the "ESG underperforms" literature of 2023 are largely the same finding sampled at different points in a sector cycle: which is why this book's line is that strict ESG is a factor and sector bet with an ethical rationale, and should be sized as one.

3. The risk-management claim is more defensible than the alpha claim. Governance quality has a respectable evidence base as a downside-risk indicator, and carbon-transition exposure is a genuine long-horizon risk for some business models. "This reduces certain tail risks" is a stronger claim than "this earns more", and it's the one serious practitioners make.

Greenwashing, ratings disagreement, and impact

Three inconvenient structural facts every honest ESG chapter must print:

• Rating agencies disagree profoundly. Correlations between major ESG raters' scores for the same companies have been measured around 0.4–0.6 (versus ~0.99 for credit ratings). Different frameworks weight different things: your fund's "ESG" is one vendor's opinion, disclosed in the methodology.
• Labels have been enforced against. European regulators (the SFDR Article 8/9 regime and ESMA's fund-naming rules) forced hundreds of funds to reclassify or rename after finding labels outrunning holdings. That enforcement is a good sign about the system and a bad sign about what was being sold.
• Buying shares on the secondary market funds nothing directly. Your purchase transfers existing shares between investors; the company receives nothing. The real-world impact channels are indirect and genuinely debated: cost of capital effects at scale, and stewardship (voting and engagement), which is why the serious impact question for an index investor is not "which fund excludes most" but "how does my provider vote its shares?": a question Chapter 87 takes up fully, and which most ESG marketing carefully avoids.

A worked example

Nils wants his world equity core to exclude weapons and coal, and asks whether to move to a strict SRI fund.

• Rung 1 check: the ESG Screened version of his index excludes exactly what he named, holds ~1,320 of 1,400 stocks, 98.6% overlap, TER +0.05%, five-year TD versus parent −0.1%. His stated values are satisfied; his portfolio is essentially unchanged.
• Rung 3 check: the strict SRI version holds 380 stocks, 24% tech versus the parent's 18%, near-zero energy, and its five-year path diverges from the parent by several percent in both directions. Nils would be taking a real sector bet.
• His decision: rung 1 for the core (values expressed, portfolio intact), and, having read the stewardship section, he notes the provider's voting record as the thing that would actually matter for impact. He declines rung 3, not because it's wrong but because he doesn't want an unlabeled sector bet inside his core: if he wanted the transition thesis, Chapter 61 says it belongs in a satellite, sized.

That decomposition: values at the core, bets in satellites, stewardship as the impact question, is this chapter's whole method.