Workshop on Well-Being II

FEBRUARY 25, 2008

Following the first workshop a month ago, today I attended the second of a series of “Workshops on Well-being” at the LSE. Below are some (very) impressionistic notes.

Presentation by Paul Dolan and Robert Metcalfe: Valuing non-market Goods: Preference based and experience based methods

  1. How do Value non-market goods?
  • Preferences
    • revealed (observed market behaviour)
    • stated (contingent hypothetical market)
  • Experiences: subjective well-being
  • Traditionally (implicitly) assume all of these are equivalent
  1. urban renewal in swansea (from 2001)
  • 2 areas: hafod and landore
  • hafod has had renewal (500/950 homes)
  • landore (675) has not
  • compare the two (omitting those who have not yet had renewal and in-movers)
  1. Improvements:
  • front boundary walls
  • street lighting
  • etc
  • landore: house cost 95k, income 16k
  • hafod

  1. Estimate revealed preferences using land-registry data and dummy for renewal.

  2. WTP: card they fill in saying what they are willing to pay for various improvements (per month for 3 years).

  3. “Thinking about your life and personal circumstances, how satisfied are you with your life as a whole?”

  4. 364 out of 1625 (22.4%) response rate (low but representative)

  • slightly biased to renewal area

  1. Revealed prefs: no effect on prices of renewal (and apparently this had also been found in a much larger

  2. WTP: £750 over 3 years (£250 a year)

  3. 6.5% increase in life satisfaction (7.1 to 7.7 on a 10 point scale)

  • controlling for marital status, age, gender, income etc
  • approx 19k in monetary terms
  • but problems of endogeneity of income
  • instrument using partner in employment and rented accommodation well-being hit now is £6350
  • cost of renewal was ~14k
  1. 2nd experiment: renewal in Port Talbot but not Neath
  • 8k for improvements + £250 for home safety per household
  • WTP: £500 (over 3 years)
  • SWB: 12.5k
    • but 2.5k for repair and 10k for safety!
    • income instrumented by WTP
  1. What explains the discrepancy?
  • revealed preference may be wrong because of unobserved effects (e.g. improvements in Hafod).
  • WTP: evidence that per month/year figures would just go on forever (even though told just for 3 years).
  • SWB Income Compensation: might be discounted PV of long-term benefits.
  • WTP accumulate over 12-20 years (average time people are in the house) or discounted SWB IC over 25 years results in WTP value = SWB IC value.