1# RECORD 71a2f8a0af3e00fb77bb5a3c7e34234f9d6c71e72# TITLE: The Effect of War on Spatial Differentiation of Real Estate Values and Urban Disorder in the Damascus Metropolitan Area3# YEAR: 2024.0 VENUE: Urban Science DOI: 10.3390/urbansci80401834# AUTHORS: Mounir Azzam; Valerie Graw; A. Rienow5# PAGES: 27 CANDIDATE TABLE PAGES (1-based): [16, 17]6# ABSTRACT: The Syrian war, which commenced in 2011, transformed the Damascus real estate market due to heightened insecurity and tenure disputes. Using the hedonic price models, including 2411 housing transactions over the period 2010–2022, this study aims to understand the spatial dynamics of the real estate market in wartime. Our findings indicate that war variables have had a significant impact on the differentiation of property prices. Notably, property attributes have a more substantial impact on real estate values than district location, with severely damaged buildings in Damascus City resulting in an 89% decline in prices, while prices in Rural Damascus districts have decreased by 50%. Additionally, this study examines the urban texture of Damascus using correlation and homogeneity statistics derived from the gray-level co-occurrence matrix obtained from Google Earth Engine. Our findings show that correlations were highly differentiated, particularly among Rural Damascus districts, with a total decline of 87.2%. While homogeneity values decreased overall between 2015 and 2019, they improved slightly after 2019. This study guides decision makers in mitigating severe property value variations across war-affected urban areas by fostering spatial justice in property rights and promoting balanced investment and sustainable real estate development during the post-war recovery phase.789===== PAGE 1 (score 9, numeric-paren 0, table-words 1, var-words 7) =====1011Citation: Azzam, M.; Graw, V.;12Rienow, A. The Effect of War on13Spatial Differentiation of Real Estate14Values and Urban Disorder in the15Damascus Metropolitan Area. Urban16Sci. 2024, 8, 183. https://doi.org/1710.3390/urbansci804018318Academic Editors: Shivanand Balram,19Raja Sengupta and Jorge Rocha20Received: 12 September 202421Revised: 18 October 202422Accepted: 18 October 202423Published: 22 October 202424Copyright: © 2024 by the authors.25Licensee MDPI, Basel, Switzerland.26This article is an open access article27distributed under the terms and28conditions of the Creative Commons29Attribution (CC BY) license (https://30creativecommons.org/licenses/by/314.0/).32Article33The Effect of War on Spatial Differentiation of Real Estate Values34and Urban Disorder in the Damascus Metropolitan Area35Mounir Azzam *36, Valerie Graw and Andreas Rienow37Institute of Geography, Ruhr University Bochum, Universitätsstraße 150, 44801 Bochum, Germany;38valerie.graw@ruhr-uni-bochum.de (V.G.); andreas.rienow@ruhr-uni-bochum.de (A.R.)39* Correspondence: mounir.azzam@ruhr-uni-bochum.de40Abstract: The Syrian war, which commenced in 2011, transformed the Damascus real estate market41due to heightened insecurity and tenure disputes. Using the hedonic price models, including422411 housing transactions over the period 2010–2022, this study aims to understand the spatial43dynamics of the real estate market in wartime. Our findings indicate that war variables have had a44significant impact on the differentiation of property prices. Notably, property attributes have a more45substantial impact on real estate values than district location, with severely damaged buildings in46Damascus City resulting in an 89% decline in prices, while prices in Rural Damascus districts have47decreased by 50%. Additionally, this study examines the urban texture of Damascus using correlation48and homogeneity statistics derived from the gray-level co-occurrence matrix obtained from Google49Earth Engine. Our findings show that correlations were highly differentiated, particularly among50Rural Damascus districts, with a total decline of 87.2%. While homogeneity values decreased overall51between 2015 and 2019, they improved slightly after 2019. This study guides decision makers52in mitigating severe property value variations across war-affected urban areas by fostering spatial53justice in property rights and promoting balanced investment and sustainable real estate development54during the post-war recovery phase.55Keywords: real estate differentiation; spatial justice; hedonic price; urban texture analysis; reconstruction56strategy571. Introduction58The political economy of urban processes for Syrian cities has been dominated for59decades by the socialist system. Although the central ideology was egalitarian, there was60significant rural–urban variation [1,2]. The urban fabric of Damascus was distinguished by61clusters of cellular communities in which rural–urban interdependence triggers the orga-62nizing collective consumption, where real estate markets are among a set of items subjected63to this consumption in order to alleviate the bottleneck of urban inflation [3,4]. Within64each district, there was a high level of social mixture in the high-quality neighborhoods,65and there were significant contrasts with the adjacent informal housing quarters. Despite66many clear observations of urban disparity in the Damascus metropolitan area, especially67in the aftermath of the war [5–7], the understanding of spatial inequality of property values68remains unsystematic.69Living in proximity to areas affected by conflict poses significant risks, which should70be taken into account when valuing real estate [8,9]. Syria has endured more than a decade71of war, resulting in extensive property damage and reducing many cities to rubble [10].72Given this, the impact of war on property values is particularly pronounced in affected73districts, where it can be expected that prices for damaged or conflict-adjacent properties74will be lower than for those unaffected by conflict. In this study, the pertinent question is to75what extent can war negatively impact the spatial differentiation of real estate values, and76how can they be monitored?77Urban Sci. 2024, 8, 183. https://doi.org/10.3390/urbansci804018378https://www.mdpi.com/journal/urbansci798081===== PAGE 16 (score 60, numeric-paren 32, table-words 6, var-words 9) =====8283Urban Sci. 2024, 8, 1838416 of 2785poral dependencies within the hedonic model. Nonetheless, these temporal classifications86have been instrumental in our analysis of mean housing prices before, during, and after87the war, as detailed in Sections 3.2 and 3.3.88Models 1 and 2 introduce property attributes for 12 districts in the Damascus Gov-89ernorate and 37 surrounding districts in the Rural Damascus Governorate, respectively,90to determine the relationship between the property values and the effects of war. Models913 and 4 introduce district attributes and monitor the spatial differentiation of property92values by comparing transactions in different areas. The hedonic models report coefficients,93standard errors, and p-values to indicate the mathematical change, accuracy, and statistical94significance of estimates. The coefficients of each variable have been interpreted as percent-95age changes in property prices relative to a baseline reference. Given that the dependent96variable is transformed to the natural logarithm of real estate transaction prices, we apply97the exponential function to these coefficient estimates to facilitate their interpretation as per-98centage changes associated with a one-unit change in each respective independent variable.99Specifically, we compute the percentage change using the formula (exp (β) −1) × 100.1003.4.1. Property Characteristics101Looking at the coefficients and standard errors in models 1 and 2 show that all war102variables are statistically significant at the one percent level and represent large implied103economic impacts (Table 3). There is a strong statistical association between the decline104in property values and the extent of building destruction. This suggests that, as buildings105become more destroyed, prices will drop more. Despite the high negative signs, the106magnitudes indicate a potentially considerable willingness to pay for homes with poor107structural conditions. However, this trend should be viewed with suspicion, as it may108be motivated by investment rather than living, as suggested by the saying “buy on the109cannons, sell on the trumpets” [80,81].110Table 3. Hedonic regression results with property attributes.111Variable112Model 1113Damascus Districts114Model 2115Rural Damascus Districts116Intercept11710.8546 (0.3768) ***1189.0971 (0.2714) ***119Area1200.0032 (0.0008) ***1210.0104 (0.0011) ***122Age123−0.0018 (0.0025)124−0.0040 (0.0020) *125State_p1260.3427 (0.0368) ***1270.2380 (0.0246) ***128View1290.1568 (0.0379) ***1300.1489 (0.0247) ***131F_right132−0.6878 (0.3477) *133−0.3506 (0.2494)134P_right135−0.6146 (0.3457)136−0.3471 (0.2478)137N_bed1380.1361 (0.0337) ***139−0.0201 (0.0268)140N_bath1410.0600 (0.0530)142−0.0436 (0.0370)143N_balcony1440.0491 (0.0514)1450.0229 (0.0345)146Parking147−0.0572 (0.0692)1480.2003 (0.0734) **149Elevator150−0.0411 (0.0799)1510.0432 (0.0420)152SDB153−2.2511 (0.4437) ***154−0.6999 (0.0974) ***155MDB156−1.3450 (0.1674) ***157−0.6056 (0.0596) ***158LDB159−1.3786 (0.1184) ***160−0.5678 (0.0552) ***161UB162−1.2019 (0.0826) ***163−0.4687 (0.0539) ***164R21650.6031660.474167Adjusted R21680.5961690.469170Number of observations1719001721511173Legend: * p < 0.05; ** p < 0.01; *** p < 0.001.174Model 1 indicates that SDB in Damascus City reduces prices by about 89%, while MDB175and LDB decrease prices by about 74%. Although UB properties should have significantly176higher values, the coefficients show similar large decreases of around 70%. However, in177model 2, which depicts the situation in Rural Damascus districts, the figures indicate that178179180===== PAGE 17 (score 50, numeric-paren 24, table-words 6, var-words 9) =====181182Urban Sci. 2024, 8, 18318317 of 27184property values decrease proportionally less compared to Damascus City. Prices drop by18550% for SDB and gradually decrease based on the level of damage, to reach 37% for UB.186The results of the regression analysis indicate that the war variables had a dominant187effect on real estate values compared to other independent variables. The coefficients of age,188ownership, full and partial right ownership, and number of rooms were not statistically189significant. In model 1, a one-unit increase in the share of the area led to a 0.3% increase190in the sales price, while a one-year increase in the age of the property led to only a 0.18%191decrease. Similarly, in model 2, an increase of one unit in the share of the area and192age of the property was associated with only a 1% increase and 0.4% decrease in sales193price, respectively.1943.4.2. District Characteristics195Introducing a set of war variables as control variables in district attribute models 3196and 4, Table 4 shows a significant impact on the pattern of hedonic results in terms of197coefficient magnitude and statistical significance. This leads to a better monitoring of198spatial differentiation of housing prices and more credible estimates of war’s impact on199property values based on district damage classification. Model 3 shows that State_N, QI,200and accessibility are statistically significant categorical variables, attracting a positive price201premium of 17.5%, 47.3%, and 20% in the sales price for each category, respectively.202Table 4. Hedonic regression results with district attributes.203Variable204Model 3205Damascus Districts206Model 4207Rural Damascus Districts208Intercept2099.6587 (0.6956) ***2108.8499 (0.6902) ***211State_n2120.1616 (0.0516) **2130.3971 (0.0294) ***214HD215−0.0184 (0.0505)2160.1048 (0.0217) ***217QI2180.3872 (0.0496) ***2190.1436 (0.0262) ***220Regular221−0.3343 (0.5304)222−0.1010 (0.6822)223Slums224−0.2968 (0.5293) ***225−0.0509 (0.6800)226Accessibility2270.18214 (0.06331) **228−0.0496 (0.0214) *229D_city_center2300.00006 (0.00004) ***2310.000040 (0.000009) **232SDD233−0.4491 (0.1621) **234−0.6483 (0.0526) ***235MDD236−1.2068 (0.2198) ***237−0.4209 (0.0709) ***238UD239−0.7933 (0.0931) ***240−0.5484 (0.0733) ***241D_conflict_point2420.00005 (0.00002) ***2430.000002 (0.000016)244R22450.4272460.432247Adjusted R22480.4202490.427250Number of observations2519002521511253Legend: * p < 0.05; ** p < 0.01; *** p < 0.001.254Model 4 reveals that State_N has the greatest impact on property values in Rural255Damascus districts, with a 48.7% increase in selling price for each score, followed by QI,256which drives a 15% increase. HD also has a significant association, resulting in an 11%257increase in selling price. However, HD in model 3 is associated with a negative sign258(−1.8% decrease in selling price), indicating that high-density housing during wartime is259relatively safe and can increase the property value, while conflict-prone areas that have260been converted to low-density residences because of the destruction. In contrast, low-261density neighborhoods in Damascus City are commonly perceived as exclusive housing262areas with high market values. Conversely, informal settlements are frequently associated263with overcrowding, substandard housing conditions, and decreasing property prices.264The war dummy variables in district attributes resulted in negative coefficients similar265to those observed in property attributes. However, model 3 shows that the impact of war is266more pronounced in the Damascus Governate. For instance, at Tadamon (MDD), property267prices plummeted by 70%, while Al Kadam (SDD) decreased by 36% and UD by 54.7%. In268contrast, model 4 indicates that MDD areas had the smallest decline of approximately 34%,269270271===== PAGE 18 (score 3, numeric-paren 0, table-words 0, var-words 3) =====272273Urban Sci. 2024, 8, 18327418 of 27275although there were no significant spatial differences. On the other hand, SDD and UD276resulted in a decrease of 48% and 42%, respectively.277As might be expected, real estate values become more important when looking at278property features, irrespective of the district’s condition. In the post-war period, the results279indicate a higher willingness to pay for well-maintained housing in SDD, which offers280affordable property prices and attractive future prospects.281The analysis reveals that proximity to conflict points is statistically significant only282in model 3 and has a positive sign in both models, indicating that the distance to conflict283points has a significant impact on property values. In other words, residents are paying less284to reside close to conflict points. Specifically, a one-kilometer decrease in distance to conflict285points lowers property values by 5% in Damascus districts and by only 0.2% in Rural286Damascus districts. The distance to the city center, measured from the Great Umayyad287Mosque, is also statistically significant in both models and has a positive sign. However,288this does not necessarily lead to improved property values. This finding suggests that the289real estate market in Damascus is less centralized.2903.5. Texture Analysis291This section systematically explores the impact of spatial changes on real estate differ-292entiation by analyzing the changes in GLCM homogeneity and correlation values in relation293to property prices. The study compares results between two phases: the most heavily294damaged phase of the war in Damascus (2015–2019) and the post-war phase (2019–2022)295following local reconciliation agreements between conflicting parties. The GLCM homo-296geneity values ranged from 0.0057 to 0.0296, while the correlation values ranged from 0.38297to 0.9, categorized into low and high classes based on the midpoint (Figure 7).298Figure 7. Reference image, homogeneity plan, and correlation plan for 2015, 2019, and 2022 in the299study area.300