Sustainable Finance

Five stylized facts of firm-sustainability investment

The organizing spine

Remember:

1. Some consumers pay more for green products (but not all, not in every market).
2. Consumers punish firms after environmental scandals.
3. Greenwashing incentives exist (consumers respond to unverified greenness signals).
4. Labor markets reward sustainability (you attract more and higher-quality applicants).
5. Capital costs respond to emissions (more sustainable firms can borrow slightly cheaper).

Why: This list is the lecture's backbone — a "which did we see evidence for?" question maps straight onto it. Two methodological meta-lessons run through every case: counterfactuals and stated vs. revealed preference.

Consumers pay for green: product-level evidence

Organic price premia (descriptive)

Remember: The same product costs more in its "bio" version yet still sells (e.g. per-gram price 2.79 vs. 2.27), proving some WTP exists. Organics ≈ 12% of Swiss grocery spend (≈28% for eggs); Switzerland ranks ~2nd in Europe in per-capita organic spend, behind Denmark. Caveat: organic WTP bundles environmental and health motives — hard to separate. A price premium alone isn't enough to set strategy; you need the demand curve, which requires price/quantity variation.

Forest Stewardship Council (FSC) wood cutting boards (experiment)

What: A wood-cutting-board company experimentally varied the price premium of the FSC-certified board vs. the non-certified one on its website and tracked the share choosing green.

Why: Traces out a real, price-elastic demand curve for certified-sustainable wood. (This is the exact setup behind the sample-question "FSC lamp manufacturer.")

Remember: 20% premium → ~22% of customers chose the certified board; 50% premium → only ~10%. Green demand is real but price-elastic — a downward-sloping demand curve.

Dolphin-Safe tuna (counterfactual flip)

What: In the late 1980s a Pacific tuna-fishing dolphin-bycatch scandal led some firms to adopt Dolphin-Safe labels (changing fishing methods at cost). Using grocery-expenditure data (share of spend on canned tuna): the dashed line is the overall declining trend; adopters' sales fell after labeling.

Why: A naïve CEO sees own sales fall after adopting the label and wrongly concludes "the label cost us money — remove it." Wrong for both the dolphins and the company.

Remember: The correct counterfactual flips the conclusion: non-adopters fell even more (a general shift away from canned tuna toward other proteins), so against that backdrop the label helped — adopters outperformed the counterfactual. Always think actual − counterfactual.

Consumers punish scandals: BP / Deepwater Horizon

The punishment, and the counterfactual trap

What: BP — perceived as the greenest oil company after years of "Beyond Petroleum" branding — caused the largest accidental marine oil release in history in 2010 (rig explosion also killed 11 workers; BP pleaded guilty to "environmental crimes"). Chart: solid line = average retail price at BP-branded stations, dashed line = nearby non-BP competitor stations.

Why: Before the spill BP enjoyed a brand price premium; right after, that premium collapsed (consumers punished BP); in the long run the premium returned (consumers forget — seen again and again).

Remember (the counterfactual number): A naïve before/after on BP stations alone (e.g. $2.80 → $2.65) suggests a ~$0.20/gallon loss. But gas prices fell everywhere (wholesale oil price drop, seasonal/regulatory fuel-formulation changes), so the correct counterfactual is the non-BP control line: the true brand-premium loss is only ~$0.05/gallon — the naïve estimate overstates ~4×. BP-branded stations were also more likely to switch brands after the spill.

Greenwashing: the BP green-advertising IV study (Barrage et al.)

The question

What: "Beyond Petroleum" was an ~8-year campaign (won marketing awards; BP rated greenest oil company for years). Did pre-spill green advertising change how harshly consumers punished BP after the spill — hold it to a higher standard, or sway them that the spill was "bad luck, not negligence"?

Why: Identification problem — BP didn't choose ad locations randomly. If BP concentrated ads where it had a strong/sympathetic brand, you'd see a muted spill response there for reasons unrelated to the ads.

Remember (the instrument): Use US election advertising schedules as an Instrumental Variable (IV) for BP's ad spend. Election calendars are set far in advance; elections spike demand for TV/radio/billboard slots, raising ad prices, so BP advertised less where there were more elections — a shifter of ad spend plausibly unrelated to the 8-years-later spill response.

Result: +1 SD of pre-spill green ads (~$1.6M) ⇒ the demand/markup loss from the spill was ~50% smaller in the initial period before the investigation. Implication: generic green branding lets firms dodge the market penalty for real harm ⇒ problematic incentives ⇒ need certification/standards (anti-greenwashing).

Carbon offsets: definitions and failure modes

What offsets are

What: Paying for emissions reductions elsewhere — ranges from "send money and hope" to verifiable engineering. Examples: planting trees; paying a firm not to cut a tree; Clean Development Mechanism (CDM, under the Kyoto Protocol) projects like switching refrigerant gases in China or installing end-of-pipe nitrous-oxide treatment. Distinct from carbon removal (e.g. Climeworks physically pulling CO₂ from the air — more certain).

Why: The professor is actually a fan of good offsets (e.g. end-of-pipe N₂O treatment with no other economic incentive) — they get a bad rap from bad eggs; not all offsets are equal.

Three failure modes

Remember:

• Additionality: paying for something that would have happened anyway = no real abatement. Example: ~52% of Indian wind CDM projects would have happened regardless (Calel et al.). "Pay me not to cut my tree" risks this; helping an emerging-market firm switch refrigerants it had no incentive/funds to switch is genuinely additional.
• Rebound: offset framing makes people consume more — told their emissions are offset, people take hotter/longer showers (youth-hostel experiment, Guenther et al.; effect exists but not huge). Shower energy +5–16%.
• Leakage: one country's reduction displaces emissions elsewhere. Eisenbarth et al. (Uni St. Gallen) modeled UK tree-planting + agriculture + global trade: taking UK farmland out of use ⇒ more imported meat ⇒ more Amazon deforestation for cattle.

Stated vs. revealed offset gap (the quotable numbers)

Remember (Carattini et al.): Survey WTP averages ~€30/tCO₂. In ~64,000 real bookings at an unnamed Swiss airline where 82% of passengers could have offset for < €30/t, only ~4.5% actually did (median revealed WTP = €0). Offset rates rise as cost falls: grocery delivery ~14% (at ~€0.24/t), bus ~30% (at ~€0.20). Higher stakes (flights) ⇒ lower take-up. The starkest stated-vs-revealed gap in the whole chapter is carbon offsets.

Labor markets reward sustainability

The Catho field experiment (Colonnelli et al.)

What: Field experiment on Catho, Brazil's largest job-matching platform. Job seekers saw hypothetical postings varying salary, work-from-home, ESG, multinational status, etc.; their choices fed the real matching algorithm, so answers had real stakes.

Why: Experimental control isolates the ESG effect from confounders (sustainable firms otherwise differ in many ways).

Remember: An ESG signal raised applicant interest by as much as a ~10% salary increase. Work-from-home beat ESG; within ESG, the environmental dimension had one of the bigger effects. Effect stronger among college-educated, white, politically liberal/moderate respondents.

Virgin Atlantic pilot experiment (Gosnell et al.)

Remember: Pilots were randomized into receiving information and personalized fuel-efficiency targets (e.g. single-engine taxi). Big fuel/cost savings — and job satisfaction rose when pilots were rewarded for efficiency with donations to a charity of their choice. Plus emerging evidence that firms aiding employees during natural disasters see lower turnover.

Capital costs: the carbon risk premium (Bolton & Kacperczyk 2021)

The study

What: Regress firms' monthly stock returns on carbon emissions, controlling for the kitchen sink (log firm market cap, book-to-market, leverage, return on equity, beta, 12-month return mean and SD, industry/quarter fixed effects, etc.) to fight omitted-variable bias.

Why: Asks whether greener firms borrow more cheaply — i.e. whether markets price carbon risk.

Remember (the carbon risk premium — levels vs. intensity):

• Higher emissions levels ⇒ higher required returns (a carbon risk premium). +1 SD Scope-1 total emissions ⇒ ~13 basis points (bp)/month ≈ ~1.5%/year higher stock return. Significant for Scope 1, 2, AND 3 levels (cram figures: ~+13 / +23 / +30 bp per SD).
• Emissions intensity (per $ sales) is NOT significant. (Caveat: "absence of evidence ≠ evidence of absence," but here the estimate is small enough to treat as ~0.)
• The choice of emissions measure drives the result — the single most-tested wrinkle.
• Cross-country: significant in North America, Asia, Europe; less clear in Africa, South America, Australia.
• Interpretation: the premium likely reflects expected future regulatory scrutiny, which targets big absolute emitters first (e.g. EU ETS covers firms above a capacity threshold) — consistent with levels mattering, not intensity.

The "1/3 of costs" reality check

Remember: Capital is only ~1/3 of a firm's costs (labor ~2/3), so even a real carbon risk premium can't, by itself, reallocate enough capital to meet climate goals — it nudges in the right direction but isn't sufficient (especially against large fixed transition costs). ESG can work via the market mechanism, but don't expect it to close the gap to climate targets.

Theory of the case: how sustainable finance could decarbonize

The market

Remember: Sustainable assets under management (AUM) ≈ $35 trillion (some estimates: >35% of AUM in some "sustainable" category; +55% vs. 2016). "Sustainable" varies — ESG integration or norms-based screening (exclude weapons, tobacco, oil & gas). Green bonds are distinct: debt raised for a specific environmental project (vs. ESG equity investing); also social, sustainability-linked, and transition bonds.

Two effects of cheaper capital for green firms

What: If green investors accept lower returns, green firms face a lower cost of capital $r$, producing:

• Reform effect (more intuitive): make the interest rate depend on fossil energy use, $r(E)$. Then every firm has an incentive to green its operations because being dirtier raises financing cost — an implicit tax on fossil energy.
• Growth effect: capital flows to greener firms ⇒ they invest more, grow, and take a larger market share.

Why: Markets — not only governments — could drive decarbonization.

The micro model

What: Standard profit max: $\pi = p\,F(K,L,E) - rK - wL - p_E E$. First-order conditions equate each input's marginal product to its cost.

Capital: $\dfrac{\partial F}{\partial K} = r$ (buy pizza ovens until the marginal product of capital equals its cost).

Energy (no green finance): $\dfrac{\partial F}{\partial E} = p_E$ (order coal until marginal product equals price).

Now let $r = r(E)$. The energy first-order condition gains a new term:

$$\frac{\partial F}{\partial E} = p_E + \frac{\partial r}{\partial E}\cdot K$$

Remember: The reform effect is bigger than it looks because $(\partial r/\partial E)\cdot K$ applies to the entire capital stock, not just the marginal unit of energy — that's the implicit fossil tax. The growth effect appears because a lower $r$ shifts the optimal $K$ rightward along the (downward-sloping) marginal-product-of-capital curve, so green firms invest more.

Does sustainable finance actually change the real economy? (debated)

Capital allocation responds; real environmental effects are mixed

Remember:

• Berg et al. 2023: ESG rating upgrades ⇒ ESG ownership +17% two years later; downgrades ⇒ −13%. But environmental management scores are not significantly affected by ownership changes; governance is (course-correcting). So money moves, but the environmental channel is weak.
• Knaur 2024: tax-exempt pollution-control bonds do drive sales / investment / R&D growth — real-economy effects where the instrument is well-targeted.
• Bottom line: capital allocation clearly responds to sustainability signals; whether that translates into real environmental improvement is mixed and still being researched.

Key formulas & one-line takeaways

Key formulas

Green-finance energy first-order condition (reform effect / implicit fossil tax):

$$\frac{\partial F}{\partial E} = p_E + \frac{\partial r}{\partial E}\cdot K$$

Capital first-order condition (growth effect operates through this): $\dfrac{\partial F}{\partial K} = r(E)$.

Causal effect from a graph: effect = actual − counterfactual (never treated before − after).

One-line takeaways

• Five stylized facts: consumers pay for green, punish scandals, reward unverified greenness (greenwashing), labor rewards ESG, capital prices carbon.
• FSC wood: 20% premium → ~22% buy green; 50% premium → ~10% — green demand is real but price-elastic.
• Dolphin-Safe tuna and BP both teach that the counterfactual can flip or shrink the naïve before/after conclusion (BP: $0.20 naïve vs. ~$0.05 true loss).
• BP green-advertising IV (election schedules): +1 SD (~$1.6M) ads ⇒ ~50% smaller spill penalty — greenwashing pays, so we need certification.
• Offsets fail via additionality (~52% of Indian wind CDM non-additional), rebound (hotter showers), and leakage (UK trees → Amazon deforestation); revealed WTP (4.5% offset, median €0) ≪ stated (~€30/t).
• ESG ≈ a ~10% wage premium in labor markets (Catho); environmental dimension is a strong driver.
• Carbon risk premium (Bolton & Kacperczyk): +1 SD Scope-1 emissions ⇒ ~13 bp/mo ≈ 1.5%/yr — significant for levels, NOT intensity; choice of measure drives the result.
• Theory: cheaper capital for green firms ⇒ growth effect + reform effect (implicit fossil tax on the whole capital stock); evidence shows capital reallocates but real environmental impact is mixed.
• Capital is only ~1/3 of firm costs, so sustainable finance nudges but can't alone deliver climate goals.