1# RECORD 6c0fa6707be0f8c5edd0d4088f7dfb356412099f2# TITLE: The demand for air quality: evidence from the housing market in Bogotá, Colombia3# YEAR: 2018.0 VENUE: Environment and Development Economics DOI: 10.1017/s1355770x180000504# AUTHORS: Fernando Carriazo; John Gomez-Mahecha5# PAGES: 18 CANDIDATE TABLE PAGES (1-based): [12, 14, 15]6# ABSTRACT: Abstract Identifying a demand function for air quality is a key input to calculate welfare measurements of pollution abatement policies. We defined intra-urban housing submarkets to apply a Second Stage hedonic pricing model that allowed us to identify an inverse demand function for PM10 reductions in Bogotá. The monthly benefits of compliance with the U.S. Environmental Protection Agency standard (50 μg/m3 – annual average), and the far more stringent World Health Organization standard (20 μg/m3 – annual average) are valued at US$12.16 and US$189.64 per household, respectively. These values, in turn, represent about 1.35 per cent and 21.04 per cent of the average household income. The hedonic model applied hereby shows that intra-urban housing submarkets are suitable for the identification of a demand function to be used by policy makers interested in evaluating non-marginal benefits (costs) from air quality improvements (deterioration).789===== PAGE 1 (score 13, numeric-paren 0, table-words 0, var-words 13) =====1011Environment and Development Economics (2018), 23, 121–13812doi:10.1017/S1355770X1800005013EDE14RESEARCH ARTICLE15The demand for air quality: evidence from the16housing market in Bogotá, Colombia17Fernando Carriazo1∗and John Alexander Gomez-Mahecha2181 Universidad del Rosario, Calle 12C No. 6-25 – Bogotá DC, Colombia and192 Universidad de Los Andes, Bogotá, Colombia20*Corresponding author. Email: fernando.carriazo@urosario.edu.co21(Submitted 30 June 2016; revised 1 January 2018; accepted 7 January 2018)22Abstract23Identifying a demand function for air quality is a key input to calculate welfare measurements24of pollution abatement policies. We defined intra-urban housing submarkets to apply a Sec-25ond Stage hedonic pricing model that allowed us to identify an inverse demand function26for PM10 reductions in Bogotá. The monthly benefits of compliance with the U.S. Environ-27mental Protection Agency standard (50 µg/m3 – annual average), and the far more stringent28World Health Organization standard (20 µg/m3 – annual average) are valued at US$12.1629and US$189.64 per household, respectively. These values, in turn, represent about 1.35 per30cent and 21.04 per cent of the average household income. The hedonic model applied hereby31shows that intra-urban housing submarkets are suitable for the identification of a demand32function to be used by policy makers interested in evaluating non-marginal benefits (costs)33from air quality improvements (deterioration).34Keywords: Air pollution; Bogotá; EPA; hedonic models; housing markets35JEL Classification: Q51; Q53; R31361. Introduction37This study estimates welfare benefits related to air quality improvements that are indi-38rectly revealed through property values in Bogotá. Most stations from Bogotá’s mon-39itoring network recorded admissible levels for carbon monoxide (CO), sulfur (SO2)40and nitrogen dioxides (NO2), and ozone (O3). However, as judged by the local pol-41lutants standards, readings for PM10 exceeded the annual local air quality standard42of 50 µg per cubic meter, especially in Puente Aranda, the city’s main industrial area43(Gaitán et al., 2007).44Most studies examining individual preferences for clean air are based on the hedo-45nic model framework (Rosen, 1974) using the so-called First Stage (FS) estimation,46where residential prices are a function of structural characteristics, and neighborhood47and environmental amenities surrounding the housing unit location. Previous hedonic48© Cambridge University Press 2018. This is an Open Access article, distributed under the terms of the Creative Com-49mons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution,50and reproduction in any medium, provided the original work is properly cited.51https://doi.org/10.1017/S1355770X18000050 Published online by Cambridge University Press525354===== PAGE 12 (score 16, numeric-paren 0, table-words 3, var-words 8) =====555613257Fernando Carriazo and John Alexander Gomez-Mahecha58Table 1. Hedonic Price Function estimation59160261362463All64Apart.65Condos66House67PM1068−0.015372∗∗∗69−0.0061217∗∗70−0.024382∗71−0.021776∗∗∗72073074−0.0175−0.0176Structural77Yes78Yes79Yes80Yes81Security82Yes83Yes84Yes85Yes86Cadastral87Yes88Yes89Yes90Yes91Owner bias92Yes93Yes94Yes95Yes96Other pollutants97Yes98Yes99Yes100Yes101Spatial fix effects102Yes103Yes104Yes105Yes106Nσ 2107μ form1086,061 PM10, Strata,1094,153 PM10, Strata,110319 PM10, Strata,1111,562 PM10, Strata,112Strata∗PM10,113Strata∗PM10,114Strata∗PM10,115Strata∗PM10,116Spatial fix eff.117Spatial fix eff.118Spatial fix eff.119Spatial fix eff.120Standard errors in parentheses.121∗p < 0.1, ∗∗p < 0.05, ∗∗∗p < 0.01.122polling all types of properties. Columns (2), (3), and (4) show the submarket estimation123results for apartments, condominiums, and houses, respectively. All four models include124variables (shown in the online appendix table A5) related to structural dwelling charac-125teristics (Structural), neighborhood characteristic variables related to security (Security),126distances to principal amenities (Distance), and structural characteristics of the nearby127houses (Cadastral), locality fixed effects, CO, SO2 and O3 among others. The dependent128variable includes both the reported value of rent for renters and hypothetical (imputed)129rent value for owners. We intended to capture differences between these two prices by130adding a dummy equal to one in case the dwelling is rented as well as its interactions131with all the variables. We called this specification an ‘owner bias specification’. Finally,132the structure for the asymmetrical error variance is specified in the last row, following133the specification in equation (4).134Regarding PM10, our variable of interest, results suggest a negative relationship135between PM10 concentration and rental prices. The model predicts that an increase of1361 µg/m3 is accompanied by a monthly average rent reduction of 0.61 per cent for apart-137ments, of 2.43 per cent for condominiums, and of 2.17 per cent for houses. An increase138of 7.71 µg/m3 (the standard deviation of the PM10 concentration by household) reduces139rent prices by 4.7 per cent in the first submarket, 18.73 per cent in the second submarket,140and 16.73 per cent in the third one (1.62, 6.93, and 8.63 per cent of HH monthly income,141respectively).142Coefficient results for other variables (shown in the online appendix) had the143expected signs. The number of rooms (numCuartos) has a positive effect on the rents,144increasing them from 2.97 to 6.98 per cent for each additional room. An increase in one145bathroom (banios) increased rent from 12.29 to 16.23 per cent. The model also shows146that an increase in one case of homicide in the block is associated with a decrease in the147rent from 0.67 to 9.95p per cent. Neighborhood quality, measured by the cadastral score148for the census track where the residence is located, also has a positive effect on price, as149was expected.150https://doi.org/10.1017/S1355770X18000050 Published online by Cambridge University Press151152153===== PAGE 13 (score 19, numeric-paren 0, table-words 4, var-words 11) =====154155Environment and Development Economics1561331575.2 Second stage estimation – demand for reductions of air pollution158Based on the predicted effect of the PM10 concentration on property rents, the implicit159price for the reduction of air pollution (W(PM10ij)) is calculated. Because defining160apartment, condominium, and house submarkets already solves the SS identification161issue by estimating information for multiple markets, the only problem that is still162unresolved is the endogeneity issue.163A 2SLS method is used to obtain consistent estimations of the demand parameters164under a linear specification. Optimum instrumental variables could be defined based165on those variables that affect PM10 concentrations, but not house prices. The selected166instruments are rainfall, wind speed and a PM10 spatial lag. At a large geographical scale,167rainfall would likely have a direct effect on property values, as Albouy et al. (2016) found168in the U.S for climate valuation. However, at an intra-urban scale, we argue that rainfall is169a suitable instrument since variation in rain patterns among neighborhoods is not likely170to influence property values within a city. Nonetheless, rain directly influences pollution171levels by washing the atmosphere, even at a small spatial scale. Also, there is no reason172to expect any direct effect on wind speed on dwellings’ prices in Bogotá. Still, previous173works have identified a strong relationship between wind speed and the concentration174of several pollutants (Verma and Desai, 2008), including PM10 (Tian et al., 2008). Tian175et al. (2008) identify this relationship as not linear so our instruments also include the176quadratic and cubic form of this variable. Finally, following Anselin and Lozano-Gracia177(2008) design, we included a PM10 spatial lag as an instrument. Table 2 shows a summary178of the first stage results of the 2SLS in column (1) and the second stage in column (2),179correcting again by clusters in the estimation.8 As was expected, more rain is associated180with less concentration of PM10, wind speed presents a non-linear relation with PM10181and the spatial lag (WPM10) is highly significant. The Hausman test, an instruments182relevance test, and Hansen’s J statistic for over-identification are also reported in table 2.183All three tests suggest that the model is well specified.184The second column also shows expected results in the estimation of the linear demand185for reduction in air pollution. The results show that air pollution reduction is a normal186good, meaning that it is affected positively by income. Other socio-demographic vari-187ables are reported in table A6 in the online appendix. Head of Household’s (HH) age188and education have a positive effect on the willingness to pay for pollution reductions.189We also found that some characteristics of the dwellings are substitutes of air quality,190including number of bedrooms (numCuartos) and neighborhood quality (Puntaje).191Finally, the SS hedonic estimation shows that the demand for air quality satisfies the192law of demand. In a low-middle income city like Bogotá, we expect a high response of air193quality to implicit price changes. At the mean, the demand for air quality will decrease194by 1.39 per cent with an increase of 1 per cent in price. Although this elasticity seems195relatively high (−1.39), our estimated elasticity lies in the range of previously estimated196elasticities using SS of the hedonic models for air quality. Brasington and Hite (2005)197and Zabel and Kiel (2000) find more inelastic demand functions, −0.12 and −0.219198respectively, while Chattopadhyay (1999) found a more elastic demand (−2.04)1995.3 Welfare measures200Welfare measures for non-localized amenities like air quality or any housing charac-201teristic whose changes affect all agents in the market should be calculated based on the2028Complete results are available in the online appendix. Please see table A6.203https://doi.org/10.1017/S1355770X18000050 Published online by Cambridge University Press204205206===== PAGE 14 (score 16, numeric-paren 0, table-words 4, var-words 6) =====207208134209Fernando Carriazo and John Alexander Gomez-Mahecha210Table 2. Demand for air quality estimation211(1)212(2)213First stage214Second stage215PM10216−0.057522∗∗∗217−0.01218Rainfall219−0.0057113∗∗2200221Wind speed22219.996∗∗∗223−5.92224Wind speed2225−9.3602∗∗∗226−3.17227Wind speed32281.4206∗∗∗229−0.54230WPM102311.0011∗∗∗232−0.02233Ingresos234−5.03 × 10−062350.0001853∗∗∗23602370238R22390.28240N2415,6942425,694243Other socio-demographical variables244Educations, HH sex, HH age, HH number of kids,245partipating in env. organizations, other dwelling246characteristics247Hausman pValor2480.01249Relevance pValor2500251Hansen pValor2520.73253Var. Est.254Robust cluster255Robust cluster256Standard errors in parentheses.257∗∗p < 0.05, ∗∗∗p < 0.01.258demand from the SS and not from the FS hedonic estimation (Taylor, 2003). The area259under the derived demands will show the total willingness to pay for a certain reduction260of PM10.261Figure 2 shows the demand curves estimated in table 2. The shaded areas of the graph262show the total monthly willingness to pay per household for air quality that satisfies263both the WHO standard (left) and the national standard (right) given an initial level264of pollution of 51.78 µg/m3. A reduction in air pollution of PM10 up to 50 µg/m3 will265equate with a total willingness to pay of US$12.16 per month per household, or 1.35266per cent of monthly income. The total willingness to pay for a PM10 reduction up to26720 µg/m3 amounts to US$189.64 per month per household, which represents 21.04 per268cent of monthly income. Table 3 presents these results and disaggregates the effects by269low, medium, and high socioeconomic strata.270https://doi.org/10.1017/S1355770X18000050 Published online by Cambridge University Press271272273===== PAGE 15 (score 15, numeric-paren 8, table-words 1, var-words 3) =====274275Environment and Development Economics276135277Figure 2. Demand for Air Quality278Table 3. Welfare measures279Elasticity (%)280TWP for 20 µm/m3281TWP for 50 µm/m3282General283−1.31284189.64285(21.04)28612.16287(1.35)288Low Strata289−0.82290152.42291(37.47)29210.07293(2.48)294Med Strata295−1.52296211.41297(19.88)29813.38299(1.26)300High Strata301−2.88302263.69303(8.12)30416.31305(0.5)306Percentage of monthly income in square brackets.307We carried out several robustness checks: first, we tested the suitability of the pro-308posed market segmentation by applying a Chow Test and a Two Step Clustering Method.309Second, we tested different functional forms of the hedonic function as well as different310air quality measures. Third, we compared OLS estimations of the FS hedonic model with311the Frontier Model Specification. Finally, we checked the robustness of the demand func-312tion for air quality to different instruments. Complete results of robustness checks are313discussed in the online appendix (please see tables RC1 and RC2 in the online appendix).3146. Discussion315Results from the FS hedonic model confirm that air quality is capitalized into property316values in Bogotá. However, such capitalization varies from one housing submarket to317another. A Chow test concludes in favor of separate housing submarkets and, therefore,318this test suggests a non-unique HPF for Bogotá.319The proposed market segmentation within the city helped us estimate a SS hedonic320model to identify a demand for air quality while relying on information obtained from321https://doi.org/10.1017/S1355770X18000050 Published online by Cambridge University Press322323324===== PAGE 16 (score 11, numeric-paren 0, table-words 0, var-words 11) =====325326136327Fernando Carriazo and John Alexander Gomez-Mahecha328the market process. This market segmentation could be defined by non-market valu-329ation practitioners interested in applying the SS hedonic model within any urban area330where apartments, houses and condominiums are the most prevalent types of residential331housing.332We address identification and endogeneity issues by using housing submarkets and333an IV econometric strategy. Results of the SS confirm the demand law and the hypothesis334that air quality is a normal good. An identification of a demand function for air quality is335a suitable and very flexible tool for policy makers interested in identifying non-marginal336benefits (costs) from air quality improvements (deterioration). We illustrated the flex-337ibility with which the demand function could be used to calculate benefits for various338ambient quality scenarios: benefits from compliance to the EPA standard amount to339US$12.16/month/household, whereas compliance with the more stringent WHO stan-340dard brings benefits close to US$189.64/month/household. The latter may seem high341in magnitude for a city like Bogotá. Nonetheless, these estimates correspond to non-342marginal changes of pollution levels calculated from the area below a demand curve343(Total Willingness to Pay.) These estimates do not entail the usually reported marginal344implicit price from FS hedonic applications. Furthermore, the non-marginal changes345correspond to a reduction of more than 50 per cent of the average PM10 pollution levels.346Nonetheless, our estimated values from revealed preferences could be incorporated in a347cost/benefit analysis for regulatory policies to control urban air pollution.348Supplementary material349The supplementary material for this article can be found at https://doi.org/10.1017/350S1355770X18000050.351References352Albouy D, Graf W, Kellog R and Wolff H (2016) Climate amenities, climate change, and American quality353of life. Journal of the Association of Environmental and Resource Economists 3(1), 205–45.354Anselin L and Le Gallo J (2006) Interpolation of air quality measures in hedonic house price models: spatial355aspects. Spatial Economic Analysis 1(1), 31–52.356Anselin L and Lozano-Gracia N (2008) Errors in variables and spatial effects in hedonic house price models357of ambient air quality. Empirical Economics 34(1), 5–34.358Bartik TJ (1987) The estimation of demand parameters in hedonic price models. The Journal of Political359Economy 95(1), 81–88.360Bayer P, Keohane N and Timmins C (2006) Migration and hedonic valuation: the case of air quality. Journal361of Environmental Economics and Management 58(1), 1–14.362Boyle MA and Kiel KA (2001) A survey of house price hedonic studies of the impact of environmental363externalities. Journal of Real Estate Literature 9(2), 117–144.364Boyle KJ, Poor PJ and Taylor LO (1999) Estimating the demand for protecting freshwater lakes eutrophi-365cation. American Journal of Agricultural Economics 81(5), 1118–1122.366Brasington DM and Hite D (2005) Demand for environmental quality: a spatial hedonic analysis. Regional367Science and Urban Economics 35(1), 57–82.368Carriazo F, Ready R and Shortle J (2013) Using frontier stochastic models to mitigate omitted variable bias in369hedonic pricing model: a case study for air quality in Bogotá, Colombia. Ecological Economics 91, 80–88.370Chattopadhyay S (1999) Estimating the demand for air quality: new evidence based on the Chicago housing371market. Land Economics 75(1), 22–38.372Chay KY and Greenstone M (2005) Does air quality matter? Evidence form the housing market. Journal of373Political Economy 113(2), 376–424.374Cheshire P and Sheppard S (1998) Estimating the demand for housing, land, and neighbourhood charac-375teristics. Oxford Bulletin of Economics and Statistics 60(3), 357–382.376https://doi.org/10.1017/S1355770X18000050 Published online by Cambridge University Press377