# RECORD fc9d39b019f0526d9688d6cea3c777de313aea1d # TITLE: Factors influencing the value of houses: Estimates for the city of Cagliari, Italy # YEAR: 2015.0 VENUE: nan DOI: 10.1016/j.landusepol.2014.08.012 # AUTHORS: C. Zoppi; M. Argiolas; S. Lai # PAGES: 45 CANDIDATE TABLE PAGES (1-based): [39, 40, 41, 42, 43, 44] # ABSTRACT: Abstract The aim of this paper is to analyze the relationship between housing values and a set of determinants, related both to the urban environment and to the structural characteristics of the housing market, in the metropolitan area of Cagliari. In order to achieve this aim, a sample of residential properties spread across the urban context was taken into account. For every single residential unit we study the value of houses, identified as their estimated value, cadastral value, rent value, value supplied by the National Observatory on Real Estate Market, and finally sale value as related to factors which are identified as relevant variables in several studies concerning the real estate market. The adopted approach implies data collection concerning value and characteristics of houses. The resulting dataset is geocoded and spatially analyzed, in order to identify spatial autocorrelation of the value of houses and its correlations with respect to the characteristics of houses through the hedonic approach. The methodological approach can be easily replicated and exported with reference to other Italian and European urban contexts and results could be straightforwardly comparable. Policy implications of the findings could be a point of reference for future Italian and European planning policies concerning housing markets and the improvement of the quality of urban life. ===== PAGE 1 (score 11, numeric-paren 0, table-words 0, var-words 11) ===== This is the accepted version of the following article: Zoppi C., Argiolas M., Lai S. (2015) Factors influencing the value of houses: Estimates for the city of Cagliari, Italy. Land Use Policy, 42, pp. 367-380 ISSN: 0264-8377. http://dx.doi.org/10.1016/j.landusepol.2014.08.012 1. Introduction Our interpretive point of view concerning the value of houses is that this value reflects the quality of urban life. The improvement or decline in the quality of urban life determines bene- fit or damage to homeowners, since they experience a change in the quality of life, and to landlords, who receive higher or lower rents. So, in our view the value of a house is essential- ly related to its character of a composite good, which is bought and sold in the housing market as a parcel of characteristics, which determine its market price (among many, Palmquist, 1984, and Cheshire and Sheppard, 1995). As a consequence, we propose to study the quality of life concerning an urban context through the analysis of the housing market where we observe equilibrium prices concerning purchases and sales of parcels of housing values’s determinants. Such determinants are grouped into four distinct categories as follows: i. structural characteristics of the residential unit (such as unit size, distance from the shoreline, qualitative indexes accounting, inter alia, for the building age, the apartment level and the maintenance level); ii. neighborhood demo- graphic characteristics (such as residential density both in the census ward and in the city dis- trict in which the property is located, or the number of permanent foreign residents living in the district); iii. plan-related characteristics (such as the presence of residential zones within a given distance from the property, proximity to parks or other green areas, and to common public services), and iv. land cover types. In order to analyze the relationship between housing prices and the aforementioned potential constituent characteristics, we pursue an approach based on a hedonic model in order to figure out the general willingness to pay for a specific commodity among the municipal area of Cagliari (Sardinia). ===== PAGE 39 (score 37, numeric-paren 0, table-words 11, var-words 11) ===== Variable Definition Mean St.dev. Characteristics of housing units, vector HUNIT in (5) AREA Finished interior area (m2) (source: 2012 direct survey) 109.43 34.89 Q_POS Position quality (presence and quality of panoramic views, distance from other buildings and structures / daylighting quality, apartment level). 4.52 1.84 Q_TYP Typological quality (building and apartment maintenance level, quality of construction, equip- ment and mechanical system conditions, building age). 4.19 1.41 DISCOAST Distance from the coastline (m) (source: Spatial Dataset of the Regional Geographic Infor- mation System of Sardinia7) 1788.15 877.80 Demographic characteristics of the neighborhood where a house is located, vector DEMOG in (5) DENSITY Population density in the Census tract (residents/km2) (source: 2001 National Survey of the Italian National Institute of Statistics concerning population and houses) 21704.12 10632.79 FOR_2012 Permanent foreign residents in the neighborhood (foreign residents) (source: 2012 Survey of the Municipality of Cagliari) 354.17 203.23 RES_2012 Residents in the neighborhood (residents) (source: 2012 Survey of the Municipality of Cagliari) 7645.28 2978.05 Plan-related characteristics of the neighborhood where a house is located, vector PLANREL in (5) PL_ZONE Dummy, location in a residential completion area (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: Au- gust 1, 2014]) 0.12 0.33 A_ZONE Area of the “A” zone in a buffer of 150 m around the location of a house (m2) (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: August 1, 2014]) 4753.14 11935.82 B_ZONE Area of the “B” zone in a buffer of 150 m around the location of a house (m2) (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: August 1, 2014]) 33033.85 14514.09 C_ZONE Area of the “C” zone in a buffer of 150 m around the location of a house (m2) (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: August 1, 2014]) 400.78 2262.48 EZ_ZONE Area of the “EZ” zone in a buffer of 150 m around the location of a house (m2) (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: August 1, 2014]) 678.98 3287.24 MIXUSE Percent area of the “IS” zone in a buffer of 150 m around the location of a house (percent) (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: August 1, 2014]) 12.66 11.78 PARKS Area of the “S3”and recreational “G” zones in a buffer of 800 m around the location of a house (m2) (source: Masterplan of the City of Cagliari, available on the Internet at: http://www.comune.cagliari.it/portale/it/puc.wp [last accessed: August 1, 2014]) 24.17 13.68 Artificial land cover of the neighborhood where a house is located, variable LANDCOV in (5) LC_URB Artificial surfaces, urban fabric in 2008 (m2) (source: CORINE Land Cover Map of Sardinia – 2008 Edition, level 2, code 1.1) 64577.89 9560.18 Spatially-lagged dependent variables (see paragraph 4.1.1) CAD_LAG Spatially-lagged dependent variable, spatial lags of variables reported in Table 2, CAD_VAL 706.25 132.22 EST_LAG Spatially-lagged dependent variable, spatial lags of variables reported in Table 2, EST_VAL 2261.10 327.11 OMI_LAG Spatially-lagged dependent variable, spatial lags of variables reported in Table 2, OMI_VAL 2313.08 264.05 RENT_LAG Spatially-lagged dependent variable, spatial lags of variables reported in Table 2, RENT_VAL 7.79 0.78 SUPP_LAG Spatially-lagged dependent variable, spatial lags of variables reported in Table 2, SUPP_VAL 2497.37 346.92 Table 3. Definition of characteristics of houses and of neighborhoods where houses area located, and descriptive statistics. 7 Available on the Internet from the Regional Geoportal, at: http://www.sardegnageoportale.it/index.html [last accessed: August 1, 2014]. ===== PAGE 40 (score 23, numeric-paren 0, table-words 6, var-words 5) ===== Variable Coefficient Stand.error t-statistic Hypothesis test: coefficient=0 Constant 542.2240 203.9922 2.658 0.0083 AREA 0.4507 0.4957 0.909 0.3640 Q_POS 7.8002 9.0757 0.859 0.3908 Q_TYP 13.5892 11.9822 1.134 0.2577 DISCOAST -0.0371 0.0297 -1.248 0.2130 DENSITY -0.0021 0.0017 -1.270 0.2052 FOR_2012 0.1165 0.1406 0.829 0.4077 RES_2012 -0.0037 0.0086 -0.433 0.6651 PL_ZONE -200.8341 147.2615 -1.364 0.1737 A_ZONE -0.0035 0.0045 -0.784 0.4337 B_ZONE -0.0050 0.0030 -1.687 0.0927 C_ZONE -0.0018 0.0089 -0.206 0.8370 EZ_ZONE 0.0076 0.0057 1.325 0.1863 MIXUSE -2.2312 1.9821 -1.126 0.2612 PARKS 8.80E-05 0.0001 0.588 0.5572 LC_URB 0.0016 0.0022 0.713 0.4763 CAD_LAG 0.3395 0.1458 2.329 0.0205 Adjusted R-squared= 0.0900 Table 4. OLS results, dependent variable CAD_VAL: the regression model includes the covariates of Table 3. ===== PAGE 41 (score 23, numeric-paren 0, table-words 6, var-words 5) ===== Variable Coefficient Stand.error t-statistic Hypothesis test: coefficient=0 Constant 1549.8007 224.0705 6.917 0.0000 AREA -2.8768 0.4852 -5.929 0.0000 Q_POS 71.3490 8.9995 7.928 0.0000 Q_TYP 43.0423 11.8714 3.626 0.0003 DISCOAST -0.0822 0.0305 -2.696 0.0074 DENSITY -0.0048 0.0017 -2.903 0.0040 FOR_2012 0.3796 0.1425 2.663 0.0082 RES_2012 -0.0164 0.0086 -1.912 0.0569 PL_ZONE -237.8951 146.9027 -1.619 0.1065 A_ZONE -0.0060 0.0045 -1.348 0.1787 B_ZONE -0.0059 0.0030 -1.990 0.0475 C_ZONE -0.0179 0.0088 -2.039 0.0423 EZ_ZONE -0.0088 0.0056 -1.557 0.1205 MIXUSE -5.7069 1.9711 -2.895 0.0041 PARKS 0.0002 0.0001 1.524 0.1287 LC_URB -0.0002 0.0022 -0.081 0.9357 EST_LAG 0.4775 0.0630 7.577 0.0000 Adjusted R-squared= 0.5208 Table 5. OLS results, dependent variable EST_VAL: the regression model includes the covariates of Table 3. ===== PAGE 42 (score 23, numeric-paren 0, table-words 6, var-words 5) ===== Variable Coefficient Stand.error t-statistic Hypothesis test: coefficient=0 Constant 1941.9215 110.8157 17.524 0.0000 AREA -0.4112 0.2190 -1.878 0.0614 Q_POS 21.5585 4.0668 5.301 0.0000 Q_TYP 1.9950 5.3698 0.372 0.7105 DISCOAST -0.0838 0.0140 -5.986 0.0000 DENSITY -0.0009 0.0008 -1.168 0.2436 FOR_2012 0.1709 0.0633 2.698 0.0074 RES_2012 -0.0190 0.0039 -4.934 0.0000 PL_ZONE -171.1874 66.2100 -2.586 0.0102 A_ZONE -0.0034 0.0020 -1.679 0.0943 B_ZONE -0.0025 0.0013 -1.861 0.0637 C_ZONE -0.0077 0.0040 -1.936 0.0538 EZ_ZONE -0.0066 0.0026 -2.603 0.0097 MIXUSE -3.7663 0.8908 -4.228 0.0000 PARKS 0.0003 .67E-04 4.638 0.0000 LC_URB -0.0006 0.0010 -0.637 0.5245 OMI_LAG 0.3096 0.0353 8.780 0.0000 Adjusted R-squared= 0.6725 Table 6. OLS results, dependent variable OMI_VAL: the regression model includes the covariates of Table 3. ===== PAGE 43 (score 23, numeric-paren 0, table-words 6, var-words 5) ===== Variable Coefficient Stand.error t-statistic Hypothesis test: coefficient=0 Constant 7.2509 0.3599 20.148 0.0000 AREA -0.0017 0.0007 -2.492 0.0133 Q_POS 0.0852 0.0129 6.618 0.0000 Q_TYP 0.0138 0.0170 0.814 0.4163 DISCOAST -0.0002 4.31E-05 -4.187 0.0000 DENSITY -2.91E-07 2.38E-06 -0.122 0.9029 FOR_2012 0.0005 0.0002 2.680 0.0078 RES_2012 -8.04E-05 1.22E-05 -6.591 0.0000 PL_ZONE -0.5894 0.2092 -2.817 0.0052 A_ZONE -4.33E-06 6.36E-06 -0.681 0.4965 B_ZONE -1.01E-05 4.24E-06 -2.385 0.0177 C_ZONE -3.22E-05 1.26E-05 -2.558 0.0111 EZ_ZONE -2.52E-05 8.06E-06 -3.131 0.0019 MIXUSE -0.0089 0.0028 -3.149 0.0018 PARKS 7.25E-07 2.09E-07 3.479 0.0006 LC_URB -4.73E-06 3.10E-06 -1.523 0.1289 RENT_LAG 0.2300 0.0347 6.629 0.0000 Adjusted R-squared= 0.5810 Table 7. OLS results, dependent variable RENT_VAL: the regression model includes the covariates of Table 3. ===== PAGE 44 (score 23, numeric-paren 0, table-words 6, var-words 5) ===== Variable Coefficient Stand.error t-statistic Hypothesis test: coefficient=0 Constant 1851.8704 131.5397 14.078 0.0000 AREA 0.3880 0.2667 1.454 0.1469 Q_POS -2.3448 4.9547 -0.473 0.6364 Q_TYP -2.0876 6.5567 -0.318 0.7504 DISCOAST -0.0849 0.0179 -4.738 0.0000 DENSITY -0.0019 0.0009 -2.052 0.0411 FOR_2012 0.2017 0.0775 2.602 0.0098 RES_2012 -0.0191 0.0047 -4.078 0.0001 PL_ZONE -128.3240 80.6387 -1.591 0.1126 A_ZONE -0.0026 0.0025 -1.072 0.2844 B_ZONE -0.0046 0.0016 -2.805 0.0054 C_ZONE -0.0080 0.0048 -1.652 0.0997 EZ_ZONE -0.0092 0.0031 -2.966 0.0033 MIXUSE -5.0426 1.0845 -4.650 0.0000 PARKS 0.0006 .88E-04 6.936 0.0000 LC_URB -0.0011 0.0012 -0.891 0.3736 SUPP_LAG 0.4332 0.0383 11.319 0.0000 Adjusted R-squared= 0.7514 Table 8. OLS results, dependent variable SUPP_VAL: the regression model includes the covariates of Table 3. ===== PAGE 45 (score 11, numeric-paren 0, table-words 3, var-words 5) ===== Covariate Dependent variable EST_VAL OMI_VAL RENT_VAL SUPP_VAL Sign (+/-) Significance (5%-10%-NO) Sign (+/-) Significance (5%-10%-NO) Sign (+/-) Significance (5%-10%-NO) Sign (+/-) Significance (5%-10%-NO) AREA - 5% - 10% - 5% + NO Q_POS + 5% + 5% + 5% - NO Q_TYP + 5% + NO + NO + NO DISCOAST - 5% - 5% - 5% - 5% DENSITY - 5% - NO - NO - 5% FOR_2012 + 5% + 5% + 5% + 5% RES_2012 - 10% - 5% - 5% - 5% PL_ZONE - 10% - 5% - 5% - NO A_ZONE - NO - 10% - NO - NO B_ZONE - 5% - 10% - 5% - 5% C_ZONE - 5% - 10% - 5% - 10% EZ_ZONE - NO - 5% - 5% - 5% MIXUSE - 5% - 5% - 5% - 5% PARKS + NO + 5% + 5% + 5% LC_URB - NO - NO - NO - NO EST_LAG + 5% OMI_LAG + 5% RENT_LAG + 5% SUPP_LAG + 5% Table 9. Synthesis of regression models’ estimates: sign and significance. If a coefficient’s estimate is not significant either at 5% or at 10%, then we put a “NO” in the significance column.