Topic 14 o 64· Governmental economic appraisal
Valuation bi stated preference
Stated preference methods estimate the value o a non-merkit guid bi directly askin fowk whit they wad be willin tae pey for it, or willin tae accept in compensation tae gie it up, typically throu a structurt survey describin a hypothetical scenario. Contingent valuation is the best-kent technique in the faimily.
Why it maitters
Green Book Annex 2 (supplementary guidance on valuin non-merkit impacts) endorses stated preference methods for guids that hae nae observable merkit transaction tae infer value frae at aw — air quality, biodiversity, flood protection, the existence value o a landscape somebody micht never veesit (https://www.gov.uk/government/publications/the-green-book-appraisal-and-evaluation-in-central-government). Defra haes publisht its ain stated-preference guidance for environmental appraisal specifically acause sae muckle environmental value (habitat preservation, watter quality) haes nae proxy merkit ava, unlike, say, noise, that at least correlates wi observable hoose prices (see revealed preference valuation).
Stated preference's core appeal — it can value literally onythin, includin guids naebody haes ever transactit in — is an aw the source o its credibility problem. Acause respondents arenae actually spendin siller, contingent valuation surveys are vulnerable tae hypothetical bias (fowk owerstate willingness-tae-pey whan there is nae real budget constraint), embeddin effects (the same guid is valued differently dependin on whit else is in the survey), an starting-pynt bias in biddin-gemme designs. The 1993 NOAA panel on contingent valuation, convened efter the Exxon Valdez oil spill litigation, set oot design staundards — a binary "wad ye pey £X, aye/naw" referendum format raither nor open-endit biddin, an mandatory reminders o the respondent's actual budget constraint — that remain the reference staundard for defensible surveys.
The maths
Contingent valuation (referendum format):
Present a binary choice: "wad ye pey £X per year for ootcome Y? aye/naw"
Vary X randomly athort respondents.
Fit willingness-tae-pey as a function o the aye/naw response rate at ilka X.
Mean WTP = aurie unner the estimatit demand curve
Aggregate value = Mean WTP × affectit population
Choice experiment (discrete choice modellin) variant:
Present respondents wi repeatit choices atween bundles o attributes
(includin a cost attribute), estimate implicit prices for ilka
non-cost attribute frae the trade-affs respondents reveal.
The choice-experiment variant is generally preferred in current UK practice ower single-question contingent valuation acause forcin respondents tae trade aff several attributes agin cost repeatedly produces mair internally consistent, harder-tae-gemme estimates nor a single aye/naw question.
Wrocht example
National government: Defra commissions a contingent valuation survey tae value a river watter-quality improvement programme. A referendum-format survey o 2,000 hooseholds finds 62% wad pey £40/year via a hypothetical watter bill supplement, an the estimatit demand curve gies a mean willingness-tae-pey o £28/year per hoosehald.
Mean WTP = £28/hoosehald/year
Hooseholds in catchment = 340,000
Aggregate annual value = £28 × 340,000 = £9.52m/year
Ower a 20-year appraisal period at 3.5% discoont rate (annuity factor ≈ 14.2):
PV(benefit) ≈ £9.52m × 14.2 ≈ £135m
This aggregate figure is then compared agin the cost side o the programme's social cost-benefit analysis. The Green Book requires this kind o stated-preference evidence tae be reportit alangside its confidence interval an survey methodology, nae as a bare pynt estimate, jist acause the underlyin nummer is mair fragile nor a merkit price.
Charity: a heritage trust surveys visitors an non-visitors aboot willingness-tae-pey tae prevent the closure o a historic buildin neither group necessarily veesits (its existence value). Acause non-visitors wha will never see the buildin still report positive WTP, the survey captures existence an bequest value that a simple visitor-fee revenue coont (a revealed-preference proxy) wad completely miss — shawin stated preference's genuine advantage whaur nae merkit transaction o ony kind exists tae reveal value.
Link tae saftware ingineerin
Stated preference methods seldom apply directly tae saftware engineerin wark, but engineers biggin citizen consultation platforms, budget-participation tools, or public survey infrastructure are aften biggin the instrument the economics depends on. Gettin the survey design details richt — randomised bid amoonts, binary referendum framin ower open-endit questions, explicit budget-constraint reminders — isnae a UX nicety, it is whit mak the resultin valuation defensible unner scrutiny; a puirly designit in-app survey can invalidate months o subsequent economic analysis. See citizen satisfaction metrics for the mair general discipline o elicitin public opinion data that will bear analytical weicht.
Pitfaws
- Open-endit "hou muckle wad ye pey?" questions. Thir are faur mair prone tae strategic an anchorin bias nor binary referendum framin; the NOAA panel's recommendation tae uise a referendum format exists jist acause open-endit elicitation performs badly.
- Nae reminder o the respondent's actual budget constraint. Athoot it, stated WTP routinely exceeds whit the same fowk wad pey whan a real budget trade-aff is in play — hypothetical bias.
- Embeddin effects ignored. The same guid valued alane versus valued as pairt o a lairger bundle produces different WTP estimates; report whit else, if onythin, wis in the survey frame.
- Treatin a single survey's pynt estimate as settled. Green Book practice expects a range an a discussion o kent biases, nae a bare nummer cairrit forrit intae the cost-benefit table as if it wur a merkit price.
Sources
- HM Treasury. "The Green Book," Annex 2: valuing non-market impacts. https://www.gov.uk/government/publications/the-green-book-appraisal-and-evaluation-in-central-government
- Defra. "Valuing environmental impacts: practical guidelines" (contingent valuation and choice experiment guidance). https://www.gov.uk/government/collections/valuing-environmental-impacts
- Arrow K, et al. "Report of the NOAA Panel on Contingent Valuation." Federal Register, 1993.
- Mitchell RC, Carson RT. "Using Surveys to Value Public Goods: The Contingent Valuation Method." Resources for the Future, 1989.