# RECORD 66b74975968370c75371207845d963de2413a23e # TITLE: Does urban centrality influence residential prices? An analysis for the Barcelona Metropolitan Area # YEAR: 2017.0 VENUE: nan DOI: 10.7764/rdlc.16.1.57 # AUTHORS: Carlos Marmolejo-Duarte # PAGES: 9 CANDIDATE TABLE PAGES (1-based): [7] # ABSTRACT: DOI: 10.7764/RDLC.16.1.57 The bid rent theory (BRT), originally conceived for a monocentric city suggests a trade-off between land value and transport costs. Thus, in most of the practical applications, the simply distance/time/cost to the CBD is used as a proxy of accessibility. Nonetheless, in contemporary metropolises, employment and services do not cluster in one CBD but in many centers, furthermore the centrality quality does not follow a smooth gradient as distance to centers increases. Consequently, taking the distance/time/cost to centers in the context or hedonic models is problematic for collinearity issues and too simplistic since it directly assumes a smoothed gradient function.  In this paper we test in Barcelona Metropolitan Area, a very well recognized polycentric city, whether some continuous indicators of centrality are key determinants of housing prices. Using listing prices, a hedonic model is built, and the asking price is regressed over two continuous indicators of centrality, one of them calculated departing of the spatial-temporal behavior of people, which itself is a novelty in this kind of studies.  The results suggest that continuous centrality indicators do exert a moderate influence on housing prices after controlling for other structural and locative attributes.  Nevertheless, the main determinants of prices are related to the socioeconomic stratification not accessibility as suggested by BRT. Energy class appears also as a factor influencing dwellings’ price. ===== PAGE 1 (score 8, numeric-paren 0, table-words 0, var-words 8) ===== 57 Does urban centrality influence residential prices? An analysis for the Barcelona Metropolitan Area ¿Incide la centralidad urbana en la formación de los precios residenciales? Un análisis para la Barcelona Metropolitana Carlos Marmolejo-Duarte (Main and Correspondant Author) Departamento de Tecnología en la Arquitectura, Centro de Política de Suelo, Escuela Técnica Superior de Arquitectura de Barcelona, Universidad Politécnica de Cataluña Diagonal, 649, 4a, Arq Legal, 08028, Barcelona, Spain carlos.marmolejo@upc.edu Manuscript Code: 638 Date of Acceptance/Reception: 27.03.2017/21.05.2015. DOI: 10.7764/RDLC.16.1.57 Abstract The bid rent theory (BRT), originally conceived for a monocentric city suggests a trade-off between land value and transport costs. Thus, in most of the practical applications, the simply distance/time/cost to the CBD is used as a proxy of accessibility. Nonetheless, in contemporary metropolises, employment and services do not cluster in one CBD but in many centers, furthermore the centrality quality does not follow a smooth gradient as distance to centers increases. Consequently, taking the distance/time/cost to centers in the context or hedonic models is problematic for collinearity issues and too simplistic since it directly assumes a smoothed gradient function. In this paper we test in Barcelona Metropolitan Area, a very well recognized polycentric city, whether some continuous indicators of centrality are key determinants of housing prices. Using listing prices, a hedonic model is built, and the asking price is regressed over two continuous indicators of centrality, one of them calculated departing of the spatial-temporal behavior of people, which itself is a novelty in this kind of studies. The results suggest that continuous centrality indicators do exert a moderate influence on housing prices after controlling for other structural and locative attributes. Nevertheless, the main determinants of prices are related to the socioeconomic stratification not accessibility as suggested by BRT. Energy class appears also as a factor influencing dwellings’ price. Key words: Hedonic price models, real estate valuation, polycentrism, time-geography, Barcelona. Resumen La teoría de la renta ofertada (TRO) originalmente concebida en un contexto monocéntrico sugiere la existencia de un efecto compensatorio entre el valor del suelo y los costes de transporte. Por ende, en la mayor parte de las aplicaciones empíricas, la simple distancia/tiempo/coste al centro es usada como un indicador de la accesibilidad. Sin embargo, en la mayor parte de las metrópolis contemporáneas tanto el empleo como los servicios se localizan en varios centros, mientras que la cualidad de centralidad no se desvanece de forma uniforme a medida que la distancia a los centros incrementa. Por tanto, utilizar la distancia/tiempo/coste a los centros en el contexto de los modelos hedónicos es problemático debido a la multicolinealidad y demasiado simple, puesto que se asume una función de un gradiente suavizado. En este artículo probamos hasta qué punto dos indicadores continuos de centralidad son elementos clave en la formación de precios residenciales en la policéntrica área metropolitana de Barcelona. A partir de precios de oferta se construye un modelo hedónico en el cual los precios son regresados sobre dos indicadores de centralidad, uno de ellos construido a partir del análisis del comportamiento espacio-temporal de la población, lo cual en sí mismo es una novedad en este tipo de estudios. Los resultados sugieren que los indicadores de centralidad continuos ejercen una influencia moderada sobre los precios una vez que se controlan los atributos arquitectónicos y locativos de la vivienda. Sin embargo, los principales determinantes de los precios están relacionados con la estratificación socioeconómica y no con la accesibilidad como sugiere la TRO. La clase energética aparece también como un factor con influencia en el precio de la vivienda. Palabras clave: Modelos de precios hedónicos, valoración inmobiliaria, policentrismo, geografía del tiempo, Barcelona. Introduction The “standard urban model” based in the bid rent theory, with roots in the von Thünen’s (1826) and Hurd’s (1924) works, as it was depicted by Alonso (1965), explains that both land rent and urban density declines as the distance to center increases. In such model, it is assumed that most of the employment is concentrated in the center, which foster a daily commuting between peripheral residential zones and it. By having the employment clustered in the center it also is assumed that services follow such spatial trend. All in all, result in increasing transport cost and time (Haig, 1926) as distance to center increases, and consequently a trade off with rent allocated to residential land. It is the existence of the land rent gradient that underlies the formation of density in a competitive market scenario. If it is considered that developers invest capital in land and building when developing a site, and constant returns per unit of land are relaxed (i.e. once substitution between land and building costs is allowed), real estate developers economize on the use of expensive land in more central locations where prices peak. In optimizing developments, they add more building capital per unit of capital invested on land (i.e. they build multi-story structures instead of low rise ones), despite the fact that ===== PAGE 7 (score 21, numeric-paren 0, table-words 2, var-words 10) ===== 63 Table 3. Model coefficients. Source: Own elaboration. Model Unstandardized Coefficients Standardized Coefficients t Sig. B Std. Error Beta 14 (Constant) 10.297 0.055 186.731 0.000 Area (Sq. M.) 0.018 0.001 1.023 23.435 0.000 Residents with university degree (%) 0.005 0.001 0.112 4.452 0.000 AC 0.101 0.013 0.095 7.503 0.000 Area (Sq. M.)^2 -4.104E-05 0.000 -0.505 -11.880 0.000 Built density 0.038 0.006 0.097 6.264 0.000 Interaction lift*story level 0.013 0.002 0.070 5.846 0.000 Swiming pool 0.136 0.026 0.055 5.204 0.000 Number of baths 0.062 0.012 0.065 5.189 0.000 Heating 0.046 0.013 0.046 3.453 0.001 PCA factor for high professional profile 0.061 0.014 0.096 4.466 0.000 Residential buildings with doorkeeper (%) 0.005 0.001 0.063 4.128 0.000 Centrality synthetic index 0.010 0.003 0.049 3.481 0.001 Renewal 0.043 0.017 0.027 2.512 0.012 Energy qualification 0.009 0.004 0.022 2.076 0.038 Figure 2 Analysis of residues. Source: Own elaboration. Conclusions Our findings suggest that, apart from structural variables concerning the quality and fittings of studied flats, centrality does exert an intermediate impact on real estate values. By using continuous centrality indexes such as that related to built density or the actual use of the city (synthetizing the spatial-temporal behavior of population) some of the habitual problems in multicentric hedonic price models have been solved with basic econometric approaches. As it has been found previously in Barcelona (Roca, 1986, 1988, Roca et al., 2003, Marmolejo, 2008), most of the locative variables with the highest incidence on residential values are associate to the spatial segregation of population. So, locations characterized by very well educated population employed in managerial and professional positions do imply a significant and important positive impact on asking prices in our model. According to the b coefficient, which measures semi-elasticity in our log-lin model, by increasing 1% the neighbors with university studies, asking prices rise 0.5%, that ===== PAGE 8 (score 13, numeric-paren 0, table-words 0, var-words 12) ===== 64 implies that the value gap between the zone with the lowest presence of high educated residents and the zone with the highest level of such population foster residential values 3.2 times (320%!), by solely modifying this factor. Whether such finding is masking a higher acquisitive power and not or not only the existence of a market premium linked to the social perception (prestige/prejudice) remains in the impossibility to isolate the income effect since in Spain such data is not asked in the Census. It is necessary, thus, to review the enthroned importance of trade-off theory by relativizing it with other important factors such as socioresidential segregation. Especially in cities where real estate markets arbitrate both the formation of land values and land use allocation. Acknowledgments This paper has been produced in the framework of two research projects LugarES (MINECO CSO-2012-33441) and EnerVALOR (MINECO FEDER BIA2015-63606-R) leaded by the author. References Aguirre, C., & Marmolejo, C. (2011). El impacto del policentrismo sobre la distribución espacial de los valores inmobiliarios: un análisis para la Región Metropolitana de Barcelona. Revista de la construcción, 10(1), 78-90. https://dx.doi.org/10.4067/S0718-915X2011000100008 Anas, A., Arnott, R. & Small, K. (1998). Urban Spatial Structure. Journal of Economic Literature, 36(3), 1426-1464. Alonso, W. (1965). Location and land Use: toward a general theory of land rent. Cambridge: Harvard University Press. Batty, M. (2001). Polynucleated Urban Landscapes. Urban Studies, 38(4), 635-655. https://doi.org/10.1080/00420980120035268 Cerda Troncoso, Jorge, & Marmolejo Duarte, Carlos. (2010). De la accesibilidad a la funcionalidad del territorio: una nueva dimensión para entender la estructura urbano-residencial de las áreas metropolitanas de Santiago (Chile) y Barcelona (España). Revista de geografía Norte Grande, (46), 5-27. https://dx.doi.org/10.4067/S0718-34022010000200001 García-López, M.A. & Muñiz, I. (2010). The Polycentric Knowledge Economy in Barcelona. Urban Geography, 31(6), 774-799. http://dx.doi.org/10.2747/0272-3638.31.6.774 Secchi, B. (1968). Hacia la Comprensión de la Metropoli. In Análisis de las Estructuras Territoriales (First Edition). Barcelona: Gustavo Gili. Hurd, R. (1924). Principles of City Land Values. (First Edition). New York, USA: Arno Press. Kau, J. B. & Lee, C. F. (1976). Capital-Land Substitution and Urban Land Use. Journal of Regional Science, 16(1), 83-92. http://dx.doi.org/10.1111/j.1467-9787.1976.tb00950.x McMillen, D.P. (1996). One Hundred Fifty Years of Land Values in Chicago: A Nonparametric Approach. Journal of Urban Economics, 40, 100-124. https://dx.doi.org/10.1006/juec.1996.0025 McMillen, D.P. & McDonald, J.F. (2002). Land Values in a Newly Zoned City. The Review of Economics and Statistics, 84(1), 62-72. https://dx.doi.org/10.1162/003465302317331928 Marmolejo, C. (2008). La incidencia de la percepción del ruido ambiental sobre la formación de los valores residenciales: un análisis para Barcelona. Revista de la Construcción, 7(1), 4-19. Marmolejo, C. & González, C. (2009). Does noise have a stationary impact on residential values? Journal of European Real Estate Research, 2(3), 259- 279. http://dx.doi.org/10.1108/17539260910999992 Marmolejo, C. & Cerda, J. (2012). La densidad-tiempo, otra perspectiva de análisis de la estructura metropolitan. Scripta Nova, 16, 15. Marmolejo, C. & Cerda, J. (2014). En búsqueda de los lugares estructurales: un análisis de comportamiento espacio temporal de la población para la identificación de centralidades urbanas. In 10º Congreso Ciudad y Territorio Virtual. Monterrey, México. McDonald, J. F. (1987). The Identification of Urban Employment Subcenters. Journal of Urban Economics, 21, 242-258. https://dx.doi.org/10.1016/0094-1190(87)90017-9 McDonald, J. F. & McMillen, D.P. (1990). Employment Subcenters and Land Values in a Polycentric Urban Area: the Case of Chicago. Environment and Planning A, 22(12), 1561-1574. https://dx.doi.org/10.1068/a221561 Mills E.S. & Hamilton B.W. (1984). Urban Economics. Glenview, IL., USA: Scott Foresman. O’Sullivan, A. (2007). Urban Economics (Sixth Edition). McGrawHill. Pena Trapero, J. B. & Pena Trapero, X. (1977). Problemas de la medición del bienestar y conceptos afines. Una aplicación al caso español. Madrid, España: Instituto Nacional de Estadística. Roca, Josep (1986). Los precios del suelo en el ámbito metropolitano. Barcelona, España: Corporación Metropolitana de Barcelona. Roca, Josep (1988). La estructura de los valores urbanos: un análisis teórico-empírico. Madrid, España: Instituto de Estudios de Administración Local.