LL2-23 digest — Part D front matter and Ch23 Understanding and accounting for the costs of inaction#

Late lessons from early warnings II (EEA 2013), pp. 561–580 (PDF 563–582).

Authors and form. Mikael Skou Andersen and David Owain Clubb, affiliations not stated. Several headline numbers come from Andersen’s own work. There are no panels. Part D has no introductory text (pp. 561–563 are a title page, contents and a list of panels). The chapter is a methods review framed by the Stern Review, with five illustrative cases: lead, mercury, nitrate, air pollution (SO2) and ozone-depleting substances (ODS).

Core argument. - Policymakers often respond to early warnings only once the “costs of inaction” have been estimated (p. 564). Those costs are simply the benefits of action by another name (p. 565). - The authors accept this hurdle and argue it can be cleared in precaution’s favour. Estimates should rest on evidence rather than proof, use low social discount rates, report ranges and be built with other disciplines, not by “economists alone” (pp. 564, 566, 574, 577). - The thesis: “Precautionary action can be justified by using credible estimates of the costs of inaction” (p. 577).

Key evidence. - Lead. - IQ losses were converted into lost lifetime income, 2–3 % per IQ point. This helped persuade the Reagan administration (pp. 566–567). - Across different studies, years and currencies, the value per gram of lead ranges from 8 US cents at a 10 % discount rate, through EUR 1.50 at a 1.4 % social rate, to EUR 5.90 undiscounted. The chapter attributes the spread to the discount rate (p. 568). - Cardiovascular harm is omitted, so these are lower bounds (p. 568). - The EU phased lead out in stages and met “little protest” when it banned it in 2000 (p. 567). - Mercury. It took 12 years to move from recognising Minamata disease (1956) to the government’s 1968 statement on its cause. Damage is estimated at USD 1,500–6,000+ per kg. Greenland bears USD 59m a year with almost no local emissions (p. 569). - Nitrate. - The EU dropped its 25 mg/l guide value in 1998 for “absence of scientific proof”. The Iowa study, published after the value was dropped (2001), suggested, “along with other types of studies”, effects at 15–25 mg/l. - The evidence is ambiguous. The chapter says no European cost-benefit analysis exists, despite citing a “tentative cost-benefit assessment” (Van Grinsven et al. 2010). The authors put UK risk costs at EUR 2.6bn a year (calculation not shown). They say the figures “would suggest” precaution is warranted and that the guide value “deserves reconsideration” (pp. 570–571). - Air pollution. - EU-25 costs of particles from SO2, NOx and primary PM combined are EUR 276–427bn a year (3–5 % of GDP), depending on whether deaths are valued per life (VSL) or per life-year (VOLY). They reach 15–22 % of GDP in Poland and other new Member States. - Damage runs at EUR 5–9 per kg of SO2, against abatement costs starting below EUR 1 (pp. 571, 573). - The EPA required a reanalysis of the Pope cohort, which held. Tall chimneys exported acidification to Scandinavia (p. 572). - Japan’s 1975 abatement push had “practically negligible” macroeconomic cost and yielded patents (p. 573). - ODS. The chapter says the Montreal Protocol also avoided about 215 GtCO2-eq, more than Kyoto’s required cuts. That figure treats the whole counterfactual stream as avoided and probably overstates the saving (about 135 Gt is commonly cited; verify). The authors infer that early warners deserve credit equal to the Protocol’s. A 60-year latency locks in further skin cancers. The ozone layer is expected to recover in 2050–2075 (pp. 575–576). - Discounting. In Stern’s figures, the costs of inaction fall from 14.7 % to 4.2 % of GDP as the discount rate rises from 1.3 % to 2.8 % (p. 574).

The authors’ lessons. From the evidence: estimates are often lower bounds, valuation choices dominate, conservative bounds already justify air pollution abatement, willingness-to-pay surveys are unreliable, and proof lags evidence. Advocacy: low social discount rates, reconsidering the abandoned nitrate guide value, resurfacing playgrounds, interdisciplinary estimation, and crediting early warners equally with Montreal (pp. 568, 571, 576–577).

Main mechanisms. In practice, a money figure for harm serves as the burden of proof. The costs of acting are tangible and the costs of not acting diffuse. Proof is slow because studies are costly and harms appear late, so harm goes undercounted. Valuation choices carry ethical judgements. Harm is displaced across borders, and persistent stocks lock it in. Staged transitions reduce resistance, patent incentives reportedly steer which technology is chosen (one source), and linked systems produce co-benefits.

Transferable insights (the full list of 19 is in the notes). 1. Requiring a money figure for harm shifts the burden onto warners (pp. 564, 570–571). Suggestive. 2. Concentrated, near-term costs of acting outweigh diffuse, delayed costs of inaction (pp. 565, 570). Moderate. 3. Demanding proof rather than evidence delays action. Standards can be dropped just before the evidence arrives (pp. 566, 570). Moderate. 4. Estimates count only what is already quantified, so they tend to be lower bounds (pp. 564, 568, 576). Moderate. “Grossly underestimated” is asserted. 5. Technical valuation choices shift results several-fold, from about 1.5× to over 10× (the largest spread comes from comparing different studies), and embed ethics (pp. 568, 569, 571, 574). Strong for sensitivity. 6. Report ranges and test whether the conservative bound still justifies action (pp. 571, 573). Moderate. 7. Dispersal fixes export harm to others (pp. 569, 572). Strong for these cases. 8. Persistent hazards lock in future harm (pp. 568, 575–576). Strong. 9. Staged, incentive-aligned transitions reduce resistance (p. 567). Suggestive. 10. Patent incentives can favour the more harmful option (p. 567). Suggestive. 11. Early abatement can be cheap and innovative (p. 573). Suggestive. 12. Linked systems produce co-benefits that single-issue appraisals miss (p. 576). Moderate. 13. A tool’s political legitimacy decides whether its results matter (pp. 571, 577). Suggestive.

Main caveats. - One-sided. The chapter weighs the costs of action only where they are small (Stern’s 1 %, SO2 below EUR 1 per kg, Japan). It mentions PCB’s “billions” of abatement cost without weighing them against damage. It gives no false alarms, and its cases are chosen with hindsight. - Nitrate overclaimed. The conclusion says estimates “help prompt immediate action” (p. 577), yet no action followed in the chapter’s own account. The UK figure rests on one US cohort and the lead author’s model. IARC’s 2010 Group 2A classification (ingested nitrate or nitrite under conditions leading to endogenous nitrosation, a probable carcinogen) goes unmentioned (verify). - Contestable claims: - unreferenced claims that IQ drives national affluence, with Figure 23.1 taken from an advocacy website (p. 568); - extrapolated figures, such as lead costs of 4–6 % of GDP (p. 568) and Japanese abatement at 6.5 % of GDP (p. 573); - minor errors: an air pollution episode in Wallonia dated “1929”, probably the 1930 Meuse Valley fog (verify); Table 23.1 described as a social cost of carbon; the US ozone withdrawal uncited. - Glossed debates. It calls Stern “broadly accepted” and may read Heinzerling et al. selectively (verify). - Justice raised but not analysed. - In its favour. It is candid about contested evidence, reports ranges and admits political limits. It makes economics an ally of precaution.