Environmental Valuation - Hedonic Pricing

Environmental Valuation

Hedonic Pricing Method

18/02/09

Carlos Ferreira

1

Revealed Preference ● ●

●

●

Goods are composed by a bundle of attributes Consumer utility results from the combination of all attributes of the good...

Consumer's WTP is a function of those attributes We should be able to model WTP for a particular good as a function of the levels of each attribute

18/02/09

Carlos Ferreira

2

Determining the implicit price of the environmental good 1.Choose a market-priced good whose price is influenced, among other things, by the environmental good/liability. 2.Measure, as accurately as possible, the quantity of the all characteristics that determine the price of the good (including the environmental good) 3.Regress the Utility function 18/02/09

NOTE: Carlos ΔU ≈ Ferreira WTP

3

Markets used ●

Housing (property)

●

Housing (renting)

●

Wages

●

Rationale: there's an implicit value to environmental amenities, which impacts market prices - Individuals are WTP more for a house in a location where env. amenities are better, ceteris paribus. - Individuals are WTA working in environmentally risky/unpleasant conditions for higher wages, ceteris paribus.

18/02/09

Carlos Ferreira

4

Utility Function: Housing Property Market P = f (H1, …, Hm; N1, …, Nn; A1, …, Ap) H = housing characteristics N = neighbourhood characteristics A = environmental characteristics

U = f (H, N, A), subject to I = PG.G + PH.H G = Other Goods the subject consumes

Considering the Housing Market is in equilibrium, PWTP = PWTA, and utility constant (at W*), we obtain the house price acceptance function:

PWTP = f (U*, I, A, PG) Differentiating with respect to the specific environmental amenity (A) and maintaining utility constant at U, we obtain the willingness to pay for improvement in A. 18/02/09

Carlos Ferreira

5

Problems with the model ●

Assumes equilibrium in the relevant market

●

Omitted variable bias

●

Error of measurement bias

●

Multicollinearity

●

Heteroscedasticity

●

No specified Functional Form

●

Market segmentation

●

Expected versus actual characteristic levels

●

Attitudes to risk

18/02/09

Carlos Ferreira

6

Example

Hedonic Valuation of Marginal Willingness to Pay for Air Quality in Metropolitan Damascus

18/02/09

Carlos Ferreira

7

Objectives 1. Estimate the MWTP for air quality improvement in Damascus; - high levels of Total Suspended Particulate (TSP) 2. Test the validity of applying the HP method in Damascus - high variation in pollution in different areas - only two types of common housing - developing country 18/02/09

Carlos Ferreira

8

Market Scope

●

●

●

Traditional housing (rarer, more expensive, seldom transacted) excluded from the analysis Two small districts (elite, exceptionally expensive) excluded from the analysis Used two kinds of property value: price of the house (stock variable) and rent of the house (flow variable)

18/02/09

Carlos Ferreira

9

Data Issues ● ●

● ●

●

Data on property values not readily available Need to interview residents to obtain property values, incomes and structural quality “Expert advice” by local real estate agencies TSP values are the average of the minimum and maximum concentration read in late Summer 2000. Data on neighbourhood location and accessibility directly observed

18/02/09

Carlos Ferreira

10

Functional Form and Variables ln(PRICE) = ln(AREA) + ln(ROOM) + ln(FLRLEV) + HQSTR + LQSTR + LIGHT + DNBR + + LHMG + ln(HEALTH) + ln(CENTER) + ln(INCOME) + ln(TSP)

Functional form chosen for ●

Simplicity

●

Highest adj. R2 (0.85)

18/02/09 ●

Carlos Ferreira

Signs and magnitudes of coefficients conform to expectations

11

Results ●

● ●

The Marginal WTP for a unit-decrease in TSP is approximately $60.00 The author considers such value “reasonable” The study has demonstrated the feasibility of using the HP method in the context of developing countries.

However, for all its strengths we believe the study deserves a further critical appreciation 18/02/09

Carlos Ferreira

12

Criticisms ●

Variables Used - Price: mixing property and rental markets - TSP: usage of min-max average values; data pre-dates study by some years; other types of pollution ignored – valuable proxy? - Structural quality: obtained through homeowners and realtor's descriptions

●

Criteria for choosing Functional Form is somewhat loose

18/02/09

Carlos Ferreira

13

In conclusion ●

HP method is deceptively simple

●

A sturdy theoretical approach

●

Various practical considerations - Variables used - Functional Forms

●

●

●

Results very sensitive to inefficiencies in the market considered Better at evaluating the effect of policies ex-post than at predicting them Presently, the scope for usage seems somewhat limited

18/02/09

Carlos Ferreira

14

References ●

●

●

Alsherfawi A., M. (2005). Hedonic Valuation of Marginal Willingness to Pay for Air Quality in Metropolitan Damascus. Forum of International Development Studies, no. 30: 23 – 34. Garrod, G & Willis, K. G. (1999). Economic Valuation of the Environment. Cheltenham: Edward Elgar. Hanley, N., Shogren, J. F. & White, B. (2007). Environmental Economics in Theory and Practice. Hempshire: Palgrave Macmillan.

18/02/09

Carlos Ferreira

15