Introduction to Sustainability

What sustainability means

The word and its history

What: From Latin sustinere. One of the earliest economic references is Hans Carl von Carlowitz's 1713 forestry book Sylvicultura oeconomica, which defined Nachhaltigkeit (sustainability) in a forestry context: deforestation should not exceed new growth — don't cut more trees than regrow, to keep the forest size constant.

Why: Shows the idea is old and operational, originally about a renewable natural-capital stock staying constant.

Remember: Modern notions are broader — they add the economic and human-development dimensions, not just conservation.

Brundtland definition (1987)

What: The UN Brundtland Commission defined sustainable development as development that meets the needs of the present generation without compromising the ability of future generations to meet their own needs.

Why: The canonical definition. It frames sustainability as an INTERTEMPORAL (across-generations) choice problem.

Remember: This is the one-line definition to quote.

Sustainability economics definition (Baumgärtner & Quaas 2010 flavor)

What: Economics is the study of how people use scarce resources and respond to incentives. Sustainability economics = understanding and managing the long-term relationships between humans and nature so that scarce environmental goods and their man-made substitutes/complements are used efficiently over time.

Why: It's about intertemporal choice under scarcity, with justice/ethics dimensions layered on.

Remember: Two big challenges follow directly: (1) how do we MEASURE sustainability, and (2) how do we ACHIEVE it.

Outcome- vs. opportunity-oriented framing

What: We can frame sustainability around OUTCOMES (e.g. per-capita income/consumption should not decline for future generations) or around OPPORTUNITIES (future generations should have the same opportunities to create wealth or enjoy the environment — what they do with them is up to them).

Why: Any policy prescription implicitly embeds assumptions about WHAT we are preserving (income? consumption? wealth? utility? nature? ecosystems?) — be aware of these implicit choices.

Remember: Outcome vs. opportunity is a SEPARATE dimension from weak vs. strong. You can be strict or flexible about either.

The Sustainable Development Goals (SDGs)

The SDG architecture

What: In 2015 the UN adopted 17 SDGs (Sustainable Development Goals), subdivided into 169 targets, measured by 304 indicators (the lecture quoted both "304" and elsewhere "231" appears in other materials — know it's ~17 goals / ~169 targets / ~300 indicators).

Why: Connects environmental, economic, and human-development dimensions as ONE agenda ("sustainable development").

Remember: No need to memorize all numbers — the profs want order-of-magnitude / qualitative sense of "how are we doing."

SDG 1 — End poverty

What: Target 1.1: by 2030 eradicate extreme poverty. The threshold traces back to the 2005 Millennium Development Goals "$1.25/day"; today it is $2.15/day in 2017 PPP (Purchasing Power Parity) dollars (≈ $3/day today). Indicator: share of population below the international poverty line.

Why: Note the missing dimensions an economist spots immediately: "$1.25 WHERE and WHEN" — i.e. you need price/time indexing (PPP, base year).

Remember: Extreme poverty fell from >40% in the early 1990s (nearly half the planet) to ~10% today, mainly thanks to economic growth. Still ~1 in 10 in extreme poverty — glass half full AND half empty.

SDG 7 — Affordable, reliable, sustainable, modern energy

What: Target 7.1: universal access to affordable, reliable, modern energy by 2030. Indicator 7.1.1: share of population with electricity access.

Remember: ~90% now have electricity access (down from 75% access / 25% lacking in 1998; dipped slightly with the 2022 energy crisis). But by stricter definitions ~half the world lacks RELIABLE energy (power outages are a way of life in many emerging markets) and ~1/4 lack clean cooking fuels (still cook with biomass burned indoors). Reliability matters for industry (need costly backup, often polluting diesel generators — a barrier to entry), hospitals/refrigerated supply chains, electrified transit. Indoor biomass smoke is atrocious for women and children (particulates).

SDG 13 — Climate action

What: Targets are looser (integrate climate measures into policy/planning); indicator = total greenhouse-gas emissions per year.

Remember: We are NOT on track for 2°C. On the current policy path (business-as-usual, BAU), central estimate is ~3.5°C end-of-century warming. Good news: BAU projections have fallen from above 4°C (2014-era scenarios) to the latest IPCC's ~2.5–3.5°C — still far above 2°C and above the cost-benefit optimum.

Carbon capture caveat

What: Emissions inventories have uncertainty (e.g. how much oil burned in Afghanistan?), but rising atmospheric CO₂ concentrations are measured unambiguously; some gases (methane) are remotely sensed.

Remember: As of recently, only ~27 direct-air-capture facilities and ~47 commercial point-source carbon-capture-and-storage plants worldwide — mostly pilot/small. Net-zero scenarios assume a massive scale-up of a technology still in its infancy.

Microeconomics refresher

Utility functions

What: A function $u(\cdot)$ that assigns a number (a "score") to each consumption bundle such that more-preferred bundles get higher numbers. Utility is just that numerical score. A mathematical representation of preferences.

Why: Lets us rank choices and maximize welfare.

Remember: Worked example — coffee & croissants, $u = \sqrt{\text{coffees}} \cdot \sqrt{\text{croissants}}$. Bundle (9,9) → $u=9$; (25,4) → $u=10$; (4,16) → $u=8$. So this $u$ ranks (25,4) ≻ (9,9) ≻ (4,16) — which does NOT match "Funny," who preferred more croissants. A valid utility function must give higher scores to more-preferred bundles; otherwise it's the wrong representation.

Consumption — what it really is

What: Consumption is everything we use and value, not just "excess." Includes healthcare (~10% of global spending, ~16% in Switzerland), food (~half of consumption in low-income countries), housing, AND leisure/time with friends and family.

Why: "Consume" sounds ominous (doom-scrolling TikTok, plastic toys) but really captures what we care about.

Remember: Leisure is part of the consumption bundle.

Concave utility → decreasing marginal utility

What: We typically assume utility is concave ⇒ decreasing marginal utility of income/consumption: the first franc gives a bigger utility gain than the millionth franc.

Why: This is the engine behind almost every Drupp-half topic: equity weighting, progressive cost-sharing for climate (the rich lose less welfare per franc given up), and intertemporal discounting via income growth.

Remember: Implications: (1) richer individuals/countries are best placed to give up consumption with least pain for sustainability goals; (2) in very poor settings (<$2/day) it can be reasonable to prioritize human-development goals over climate. Economics characterizes trade-offs; society chooses the values/weights — that's above the economist's pay grade.

Capital and the weak-vs-strong distinction

Three types of capital + comprehensive wealth

What: Capital = a STOCK that delivers a FLOW of goods and services over time. Three types:

• Human capital ($K_h$): education, knowledge, skills, the workforce, and health (health determines our ability to produce the service flow).
• Physical capital ($K_p$): machines, buildings, roads, computers. (Money is financial capital — classified a bit differently but also economic capital.)
• Natural capital ($K_n$): forests, atmosphere, oceans, soil, water, AND non-renewable mineral/oil stocks in the ground. (Once oil is extracted it becomes a commodity; the in-ground stock is the natural capital.)

Remember: Comprehensive wealth $K = K(K_h, K_p, K_n)$ = total value of all capital stocks. HOW to aggregate is a hard, open research question — do we sum, take the minimum, the geometric mean? The lecture deliberately stays "hand-wavy" on the functional form.

Weak vs. strong sustainability — the substitutability question

What:

• Weak (flexible) sustainability: use and invest in capital stocks such that comprehensive wealth $K$ does NOT decline over time. Capitals can SUBSTITUTE for one another.
• Strong (strict) sustainability: use and invest such that NO single stock falls below a critical threshold — substitution NOT permitted below that threshold, even if you could get a lot of other capital in exchange at that margin.

Why: The terminology is value-loaded ("weak" sounds bad — nobody advocates "weak"). Think "flexible vs. strict."

Remember: It all boils down to: how SUBSTITUTABLE are man-made and natural capital? Examples of the question: man-made vs. natural water filtration; aquaculture vs. wild fishing; artificial pollination vs. bees; can we shield ourselves from UV if we destroy the ozone layer? If a natural service has NO technological substitute, strong sustainability is the sensible framing. Example substitution under weak: cut trees → build a school → train children (traded natural for human capital).

Weak vs. strong applied to climate

Remember: Weak/flexible approach to climate = price emissions (carbon tax or emissions-trading scheme) so every actor trades off costs and benefits at the margin → a balanced economic/human/environmental outcome. Strong/strict approach = hold $T < 2$°C no matter the cost. (Caveat raised in class: a pure "2°C no matter what" statement is incomplete because it ignores the possibility that the cost itself could push other capital below critical levels — an implicit assumption.)

Tax vs. ETS aside: Under certain conditions a carbon tax and an emissions-trading scheme (ETS) are formally equivalent at the margin for firms. They differ in WHAT is certain: with an ETS you fix the QUANTITY of emissions (price adjusts with booms/recessions); with a tax you fix the PRICE (quantity adjusts). Neither is inherently "stricter."

The Global Energy Challenge

Energy and prosperity

What: Global energy consumption ≈ 650 exajoules (2023–24) — "several million nuclear bombs." Top consumers: China, the US, India (China rising further). Most energy still comes from fossil fuels: ~80% globally, ~50% in Switzerland.

Why: Decarbonizing the GRID ≠ decarbonizing ENERGY: renewable electricity has surged, but electricity is only ~20% of total energy consumption.

Remember: No prosperous country uses little energy — log GDP per capita vs. log energy per capita is tightly positive. There is variation (Switzerland is richer than the US and uses far less energy → efficiency gains are possible) but no rich country is also low-energy. Even accounting for outsourced manufacturing, the statement holds.

Emerging markets are the protagonists

What: Emerging markets overtook rich countries as the world's primary energy consumers around 2010. Going forward, essentially ALL the growth in energy demand comes from emerging markets; developed-country demand is roughly flat.

Why: The key future question is: which technologies will emerging markets use to meet existing-unmet and growing demand?

Remember: We should EXPECT and HOPE for rising energy demand in emerging markets as people escape poverty. Rich-country actions still matter (innovation, tech leadership), but always bear emerging markets in mind.

The "Energy Pentagon" — five trade-off dimensions

What: The energy system trades off (at least) five dimensions vs. the SDGs: (1) affordability, (2) reliability, (3) climate, (4) particulate pollution / air quality, and (5) fiscal sustainability (a dimension that's often overlooked — subsidies require fiscal space, which a fiscally-constrained world often lacks).

Remember: With current technology there's no "silver bullet." Example for heating: natural gas (good particulates/reliability, bad climate), biomass/wood pellets (climate-ish good but bad particulates), electric (depends entirely), heat pumps (excellent climate & particulates, but not yet affordable). Each option scores differently across the five.

Why fossil fuels persist — affordability

Remember: Median levelized cost of electricity (LCOE) for coal and gas is still below wind/solar in many places (though there's huge spread — solar is far cheaper in some places, and wind/solar costs are falling fast). LCOE is NOT a true apples-to-apples comparison: a MWh of solar isn't worth a MWh of gas because solar is INTERMITTENT and not DISPATCHABLE (grids must meet demand instantly). Making intermittent sources comparable requires STORAGE, which adds cost (Switzerland is well-placed for storage; many places need batteries). "Solar with backup" often still lacks a clear cost advantage.

Costs of decarbonization policy on affordability

Remember: The EU Emissions Trading System (ETS) (started 2005, world's first multi-country greenhouse-gas ETS; permits ~€60–80/ton recently; firms increasingly buy permits at auction rather than free allocation) reduced emissions AND spurred low-carbon patenting — but tightening shocks raised consumer energy prices. US state Renewable Portfolio Standards (RPS) (electricity operators must source X% from renewables; gas plants buy renewable-energy certificates) boosted renewables but also raised consumer energy prices after adjusting for confounders.

Climate impacts on the economy

Three takeaways about climate damages

What/Remember: (i) Wide range of outcomes — agriculture, health/mortality, productivity, migration, etc. (ii) Heterogeneous — creates winners and losers across regions/sectors (it's not all bad). (iii) Net large and negative — central estimate ≈ 5% GDP-equivalent annual loss from end-of-century warming.

Why "5% is a lot": COVID-2020 was a ~3% global GDP loss — one of the biggest shocks in living memory. Climate's ~5% is an ANNUAL loss, year after year, far larger cumulatively.

Channels (illustrating the breadth)

Remember: Agricultural yields (heat/rain/frost) AND indirect ag effects (crop-insurance premiums rise with weather variability; out-migration from degrading areas — seen in US and sub-Saharan data; new drought/heat-tolerant seed varieties developed for crops facing more extreme weather). Health: excess mortality from heat (esp. elderly; some offsetting benefit from fewer extreme-cold days, but net upward); healthcare utilization rises across the WHOLE age distribution (more ER visits, accidents, psychotic episodes — usually ignored in standard estimates); workplace injuries (documented for Switzerland via Suva claims — e.g. more people fall off ladders because poor sleep from heat with no AC). Air conditioning strongly mitigates many heat damages.

Heterogeneity example — energy demand

Remember: Warming generally RAISES electricity demand (cooling) and LOWERS heating-fuel demand everywhere. For some (e.g. Switzerland) the heating saving outweighs the cooling rise → net benefit on that dimension. But for hot emerging economies the cooling-driven electricity demand increase is enormous relative to today's generation: Nigeria ~+2000%, Indonesia ~+99%, Pakistan ~+171% of current electricity output (driven partly by expected economic growth). Estimates use Monte-Carlo confidence intervals over climate/economic/population uncertainty.

Particulate pollution

Why particulates matter

What: Arguably the biggest environmental public-health issue in the world. Ambient (outdoor) fine particulate matter (PM2.5 — particles smaller than 2.5 micrometers) alone kills ~4.7 million people/year (up from 4.1M five years ago) — not even counting indoor pollution.

Why context: Cancer ~10M/yr, diarrheal disease ~1.5M, road injuries ~1.2M, extreme-weather/climatic-disaster deaths ~50,000/yr (very different order of magnitude). So particulates are vastly under-discussed.

Remember: PM10 = particles <10 micrometers (bigger than a comparison they draw to a human hair); PM2.5 = the tiny nasty ones that penetrate deep into the lungs, into the blood, and across the blood-brain barrier (causally linked, via wind-shift studies, to dementia). Effects reach far beyond deaths: in-utero exposure (compared across siblings) lowers 4th-grade school performance; a ~$700 classroom air filter improved US students' test scores as much as cutting class size by a third; reducing PM raises worker productivity and cognition.

Particulates and the energy-policy trade-off

Remember: Particulates are tied to the energy system (coal/biomass combustion emit them directly; some sources emit precursors). Since ~1/4 of the world lacks clean cooking fuels, making modern fuels (LPG, gas, electricity) MORE expensive risks pushing marginal households BACK to dirty biomass → more indoor PM exposure. A classic SDG trade-off.

Fiscal sustainability

Fiscal space and the energy transition

What: Subsidies feel like "free money / manna from heaven" to voters, but governments need fiscal SPACE to provide them.

Remember: Many economies carry debt-to-GDP ratios of ~100–150% (Switzerland is a low-debt outlier). Demographic change (more retirees, falling worker-to-retiree ratio) will strain public finances further. COVID fiscal response: rich countries spent >10% of GDP; many emerging markets couldn't. About 600 million people live in countries where the government gets >30% of revenue from fossil-fuel production.

Why it's a real dilemma: Nigeria — ~300M people, government has only ~$100/person/year for everything, and ~50% of that comes from oil. "Phase out oil / go net-zero" leaves such countries without a good answer — a genuine sustainable-development tension.

SDG trade-offs and how to escape them

SDG trade-off examples

Remember: Phasing out oil ⇔ lost government revenue in fossil-dependent countries (Nigeria). Carbon pricing cuts emissions BUT raises consumer electricity prices (regressive). Pricing clean cooking fuels too high pushes poor households back to dirty biomass → worse indoor PM. Subsidizing clean tech competes with healthcare/education spending (fiscal). Always look for the SDG vs. SDG tension.

Innovation as the escape hatch

What: The first thing that should come to mind at ETH: innovation.

Why: Cheap scalable direct air capture, plummeting renewables and storage costs, lab-grown meat, cheaper heat pumps — these MITIGATE or dissolve the trade-offs.

Remember: With innovation, win-wins exist. (Degrowth was raised as an alternative angle — deferred to the growth lecture.)

Critical thinking about Generative AI

Evaluating Gen-AI answers — three dimensions

What: Evaluate any AI answer on (1) correctness (is it factually right?), (2) what's missing (which dimensions are ignored?), (3) uncertainty (does it reflect uncertainty appropriately?).

Why: Saul Perlmutter's concern: Gen-AI's confident style is often ANTITHETICAL to scientific thinking, which is inherently PROBABILISTIC — we know almost nothing for certain, only with confidence up to a level. Confident simple answers are "catnip" for our brains (which is why politicians and AIs give them) but bad for understanding reality.

Remember: When asked "how should the world make its energy system more sustainable," ChatGPT exhibited a Western/rich-country bias, a climate bias (ignoring water use, biodiversity, broader particulate pollution), overconfidence / no acknowledgment of uncertainty, and ignored economic trade-offs. Take-home: even with AI, you must know what QUESTIONS to ask and bring uncertainty/scientific thinking. (Examples of past scientific over-confidence: babies-on-belly → SIDS; "ulcers caused by stress" until Barry Marshall ingested H. pylori and won a Nobel.)

Key formulas & one-line takeaways

Key formulas

$$\text{Comprehensive wealth: } K = K(K_h,\, K_p,\, K_n)$$

$$u = \sqrt{\text{coffees}} \cdot \sqrt{\text{croissants}} \quad\text{(example utility; concave } \Rightarrow \text{ decreasing marginal utility)}$$

One-line takeaways

• Brundtland (1987): meet present needs without compromising future generations; sustainability is intertemporal choice under scarcity.
• Two core challenges: how to MEASURE sustainability, and how to ACHIEVE it.
• Extreme poverty: >40% (early 1990s) → ~10% today; electricity access ~90% (10% lack it), ~half lack RELIABLE energy, ~1/4 lack clean cooking fuels; on track for ~3.5°C.
• Concave utility ⇒ decreasing marginal utility — the engine of equity weighting and discounting.
• Three capitals ($K_h, K_p, K_n$) → comprehensive wealth $K$; aggregation is hard (sum? min? geometric mean?).
• Weak (flexible) sustainability: $K$ non-declining, capitals substitutable. Strong (strict): no stock below a critical threshold. The whole question is SUBSTITUTABILITY of man-made vs. natural capital.
• ~80% of global energy is still fossil; electricity is only ~20% of energy; all demand growth is in emerging markets.
• The "Energy Pentagon": affordability, reliability, climate, particulates, fiscal sustainability.
• Climate damages: wide-ranging, heterogeneous (winners & losers), net large-negative ≈ 5% GDP-equivalent/yr — vs. COVID's 3%.
• PM2.5 kills ~4.7M/yr — the world's biggest environmental health issue; crosses the blood-brain barrier.
• ~600M people live in countries getting >30% of government revenue from fossil fuels (e.g. Nigeria) — fiscal sustainability is a real trade-off.
• Innovation can shrink the trade-offs (win-wins). Gen-AI answers carry Western/climate bias and false confidence — always ask what's missing and how uncertain.