1# RECORD aa81dc840f8ab6c12b272e830117706e34b5a89a2# TITLE: Fresh air in the city: the impact of air pollution on the pricing of real estate3# YEAR: 2024.0 VENUE: Environmental science and pollution research international DOI: 10.1007/s11356-023-31668-14# AUTHORS: Oktawia Miłuch; K. Kopczewska5# PAGES: 24 CANDIDATE TABLE PAGES (1-based): [14]6# ABSTRACT: This study investigates the impact of ambient air pollution on housing prices in Warsaw, Poland, by examining spatial dependencies. The high concentration of particulate matter (PM_10, PM_2.5 and PM_1) is expected to reduce real estate values. Using a hedonic model with approximately 15,000 observations and a spatial error model, we did not find evidence of this impact. Standard and premium housing submarkets differ in price determinants, but both are insensitive to environmental issues. This could be explained by the lack of comprehensive intra-urban historical information on air pollution, which limits investors’ rationality and their ability to properly value real estate based on environmental issues. Additionally, measurement and aggregation issues, along with low pollution variability within the city, may contribute to the insignificance of this information in real estate sales prices. Our empirical research confirms a strong link between air pollution and weather conditions within the city, where low temperatures and low-speed southern winds worsen contamination levels, while high temperatures and westerly winds improve air quality. Furthermore, we find that incorporating pollution data using PM yearly mean concentration works better in modelling than the PCA-reduced air pollution index.789===== PAGE 1 (score 20, numeric-paren 0, table-words 1, var-words 18) =====1011Vol:.(1234567890)12Environmental Science and Pollution Research (2024) 31:7604–762713https://doi.org/10.1007/s11356-023-31668-1141 315RESEARCH ARTICLE16Fresh air in the city: the impact of air pollution on the pricing of real 17estate18Oktawia Miłuch1 · Katarzyna Kopczewska1 19Received: 8 September 2023 / Accepted: 18 December 2023 / Published online: 2 January 2024 20© The Author(s) 202421Abstract22This study investigates the impact of ambient air pollution on housing prices in Warsaw, Poland, by examining spatial 23dependencies. The high concentration of particulate matter (PM10, PM2.5 and PM1) is expected to reduce real estate values. 24Using a hedonic model with approximately 15,000 observations and a spatial error model, we did not find evidence of this 25impact. Standard and premium housing submarkets differ in price determinants, but both are insensitive to environmental 26issues. This could be explained by the lack of comprehensive intra-urban historical information on air pollution, which limits 27investors’ rationality and their ability to properly value real estate based on environmental issues. Additionally, measure-28ment and aggregation issues, along with low pollution variability within the city, may contribute to the insignificance of 29this information in real estate sales prices. Our empirical research confirms a strong link between air pollution and weather 30conditions within the city, where low temperatures and low-speed southern winds worsen contamination levels, while high 31temperatures and westerly winds improve air quality. Furthermore, we find that incorporating pollution data using PM yearly 32mean concentration works better in modelling than the PCA-reduced air pollution index.33Keywords Air pollution · Spatial econometrics · Real estate valuation34JEL Classification Q53 · R31 · C21 · O1835Introduction36Despite the extensive research on air quality, the real estate 37value created by clean air in an urban environment is still 38unknown as the empirical studies are far from conclusive. 39There are three reasons for this.40The first one, theoretical, lies in rationality assump-41tion and the interaction of many factors impacting housing 42prices. Environmental characteristics are not market goods; 43thus to estimate the value of air quality, the most common 44approach in economic studies is to link it with real estate 45value from a given area. In a pioneering work, Ridker and 46Henning (1967) assumed that air quality was one of the 47variables explaining a potential home buyer’s utility func-48tion. With the hedonic price theory development, the air 49pollution assessment was mainly based on housing prices. 50They assumed that when controlling structural attributes, 51the observed differences in real estate prices exhibit the site 52characteristics value, such as environmental amenities (see 53meta-analysis by Smith and Huang 1995). However, such 54a perspective implicitly assumes that investors1 can con-55sciously choose (and valuate) at a given moment in time all 56of the features of real estate. This is mostly true in relation 57to the size, location, or interior features such as balcony, etc., 58but difficult in the case of air pollution. This is because the 59current measurements of air quality are usually well known, 60but the long-term historical and prospective information 61about the environmental conditions of a given location is 62Responsible Editor: Eyup Dogan63 * Katarzyna Kopczewska 64 65kkopczewska@wne.uw.edu.pl66 67Oktawia Miłuch 68 69oktawiamiluch28@gmail.com701 71Faculty of Economic Sciences, University of Warsaw, Ul. 72Dluga 44/50, 00‑241 Warsaw, Poland731 In our research, we use the word investors to refer to people will-74ing to buy a property for their own use as well as developers building 75houses for sale. The two groups share the same goal — to maximise 76their utility/profits; therefore, clean air in the neighbourhood is advan-77tageous for both. In the case of individuals, the theory explaining this 78approach is presented in the literature review.798081===== PAGE 14 (score 81, numeric-paren 52, table-words 4, var-words 21) =====8283761784Environmental Science and Pollution Research (2024) 31:7604–7627 851 386line. As one could expect, properties situated closer to the 87city centre have a higher sqm price than those further apart, 88regardless of their total value. This relation is reflected by 89the significant and positive value of the coefficient specify-90ing distance to the city centre; nevertheless, the effect is 91more visible for exclusive apartments, which is probably due 92to their concentration in central areas.93Lastly, the impact of transit time by public transport 94varies between standard and luxury flats. The former ben-95efits from a shorter commuting time, while the effect is the 96Table 3 Spatial error models for 97submarkets98*p < 0.1; **p < 0.05; ***p < 0.0199Dependent variable: price of a square meter of a flat100Standard flats101Exclusive flats102(1)103(2)104Internal features of the flat105Surface area (sqm)106 − 36.59***10710.36*** 108Year of construction10929.02***11036.24***111Number of rooms112 − 270.41***113 − 858.58***114Floor11521.19**116 − 99.00117Floors in building118 − 49.85***119 − 78.01120Market type (secondary)1211120.63***1221533.71***123Air conditioning124709.32***125949.89**126Garage127477.42*** 128 − 181.82129Elevator130310.65***131638.83*132Balcony133 − 221.80***134 − 363.79135Garden13645.04137755.41138Terrace13999.67140777.06**141Districts of Warsaw142Bialoleka143 − 943.59*** 144861.48145Bielany146735.38*** 1475973.90**148Mokotow1491193.84***1502934.15151Ochota152515.90**153 − 123.84154Praga Polnoc1557.53156939.74157Praga Poludnie158 − 289.421593864.10160Rembertow161 − 1547.48***1622496.67163Srodmiescie1643224.36***1656497.90***166Targowek167 − 817.08***168 − 1879.58169Ursus170 − 914.99***1713565.19172Ursynow1731651.40***1745955.07**175Wawer17695.251777658.26**178Wesola1791205.13***1803483.14181Wilanow1822876.65***1836459.78***184Wlochy185 − 455.91**1861701.08187Wola1881162.64***1892233.34190Zoliborz1912216.74***1923782.91*193Relative location194Distance to city centre195 − 366.55***196 − 1641.29***197Transit time198 − 14.63**19999.83** 200Air quality201Mean PM10202 − 60.07***203 − 89.05204Mean PM2.520592.58**206 − 421.30207Mean PM1208 − 12.31209811.13**210Constant211 − 41,112.88***212 − 49,346.83***213λ2140.485***2150.382***216Observations21713,5392181399219Log Likelihood220 − 122,031.70221 − 13,918.14222σ22233,815,348.0022424,991,628.00225Akaike Inf. Crit226244,137.5022727,906.28228Wald test (df = 1)2291939.54***230111.83***231LR test (df = 1)2321545.56***23392.71***234235236===== PAGE 15 (score 27, numeric-paren 0, table-words 3, var-words 21) =====2372387618239 240Environmental Science and Pollution Research (2024) 31:7604–76272411 3242opposite considering the latter. Little is known about the 243impact of real transit time using public transport. Most of the 244papers examine the influence of transportation infrastructure 245in the neighbourhood on real estate value, which may not 246fully reflect the real possibilities of using public transport. 247However, in our case, the opposite signs in the two models 248might be explained by homebuyers’ needs. While individu-249als with lower budgets seeking to purchase a flat may use 250public transport on an everyday basis, people who can afford 251luxury apartments may ignore this factor.252The environmental variables include the yearly mean 253concentration of particular matters: PM10, PM2.5 and PM1. 254The statistically significant coefficients of PM10 and PM2.5 255in the model related to standard apartments indicate that 256their prices are associated with air pollution levels. However, 257while PM10 negatively impacts the house value (by 0.5%), 258according to our estimates, PM2.5 increases the sqm price 259(by 0.7%). Similar conclusions can be drawn for PM1 in the 260case of exclusive apartments (by 3.9%). The insignificant 261coefficients of PM10 and PM2.5 for luxury properties, as well 262as the estimates of PM1 for standard properties, evidence 263a lack of association between air pollution and housing 264prices. The results may be unexpected; however, there is 265some speculation about why this occurs. Fernández-Avilés 266et al. 2012 and Tang and Niemeier (2021), using differ-267ent pollutants and techniques to measure air quality, have 268drawn similar conclusions on the influence of air pollution 269on real estate value. Among studies which have published 270insignificant results relating to air pollution impacting house 271prices, some authors have acknowledged that the irrelevant 272effect of environmental coefficients is because the given pol-273lutant does not tend to exceed air quality standards (Kim 274et al. 2003). Examining the literature, the results of other 275research suggest that when, in general, ambient air pollu-276tion is relatively low, the pollutants’ influence on property 277value is more likely to be insignificant. However, compared 278to other metropolitan cities in Europe, air quality in Warsaw 279is quite poor. Moreover, as explained in the section con-280cerning air quality data analysis, during winter in the worst 281scenario, the number of days with exceeded norms of PM10 282and PM2.5 exceeded the level of 20. Altogether, neither the 283insignificant coefficient of the PM1 in the model describ-284ing standard properties nor estimates of PM10 and PM2.5 285for luxury flats are, as suggested by other researchers, are 286related to superior air quality. The significant and negative 287impact of PM10 on the prices of standard flats, as suggested 288by Lu and Lee (2022), may be explained by the fact that 289particulate matter with the largest diameters is more percep-290tible than other pollutants included in the study below. The 291positive sign displayed by the PM2.5 coefficient in the stand-292ard real estate model, as well as PM1 increasing the value 293of exclusive properties, constitute counter-intuitive results. 294According to Fernández-Avilés et al. 2012, these unexpected 295conclusions may arise from the difference between the sub-296jective environmental perception of homebuyers and objec-297tive conditions. The rational consumer makes the decision 298which maximises utility based on all available information. 299However, the environmental condition of a given property 300may be out of the scope of a potential homebuyer. Unlike 301long-term air pollution data, information on the accessibil-302ity of public services or transportation is readily available. 303Therefore, in the Polish context of high demand for real 304estate, purchasing a house might not depend on the air qual-305ity in the area but on affordability and facilities available in 306the vicinity. Based on the analysis of the spatial distribu-307tion of PM10, PM2.5 and PM1, it happens to be that, in most 308cases, the most polluted districts have the most developed 309commercial services, communication and leisure facilities, 310etc. For instance, Bielany and Śródmiescie in both models 311positively impact the sqm price of a dwelling. The former 312is the central district of Warsaw, while the latter is a former 313industrial area with a steel mill, which is today an attractive 314place to live. As the conclusions from the literature about 315the impact of air quality on housing prices are ambiguous, it 316seems that in general, nations around the globe differ in their 317attitude toward clean air value. In a significant number of the 318papers based on data from Asia, air pollution decreases real 319estate prices, while outcomes of studies referring to other 320parts of the world are not as straightforward, which is also 321the case in Warsaw.322As one can observe, according to p-values, not all vari-323ables in both models are significant. The main purpose of 324the research was to question whether air quality impacts real 325estate prices; thus, its foremost role is explanatory. As writ-326ten by Shmueli (2010), “trimming potentially theoretically 327meaningful variables is not advisable unless one is quite 328certain that the coefficient for the variable is near zero, that 329the variable is inconsequential, and that trimming will not 330introduce misspecification error “. Hence, some insignificant 331variables, important from the theoretical point of view, are 332included in the final models.333Policy implications334The analysis of air quality data has highlighted the need for 335improvements in Warsaw’s air pollution policy as the ambi-336ent air quality over the analysed time was unsatisfactory. 337Compared to other European cities, the capital of Poland 338experiences some of the most severe air pollution, mostly 339from fuel combustion from the residential sector and trans-340port, which leads to comparatively high social costs per cap-341ita. Therefore, municipal actions towards those two sources 342of pollution to improve air quality should be reconsidered.343In the case of pollution from transportation, one of the most 344common approaches widely applied worldwide is the restric-345tion of access to the city centre for the most polluting cars. 346