1# RECORD 828367334a6cb48f8254ab88d9172664e3afd7a22# TITLE: WORLD HERITAGE DESIGNATION AND PROPERTY: THE IMPACT OF PRESERVATION ON THE PRICE OF HISTORIC ASSETS3# YEAR: 2022.0 VENUE: International Journal of Strategic Property Management DOI: 10.3846/ijspm.2022.181564# AUTHORS: Eyal Salinger; D. Shefer; Nir Mualam5# PAGES: 13 CANDIDATE TABLE PAGES (1-based): [6, 7, 8, 9]6# ABSTRACT: The restrictions and requirements imposed by historic preservation regulations bring about many changes to the rights of property owners. They might impose additional costs, most notably by prohibiting the demolition of designated buildings and thus decreasing property-development opportunities. The objective of this study is to examine what happens following world heritage designations; specifically, if and how such designations impact the property values of historic buildings. Using the hedonic price method, we measure the value of the option to demolish and rebuild that is denied to owners of designated buildings. We also measure the value of preservation regulations, expressed in the prices of apartments in designated buildings. The study area is the “White City” of Tel Aviv, which UNESCO designated as a world heritage site. The findings suggest that the market price of a designated building is on average 12.5% lower than its theoretical value if the building was not subject to historic preservation regulations. Moreover, a premium of approximately 14% was found in the price of apartment units in designated buildings if the building underwent preservation. These findings could have a direct impact on public policies designed to promote the preservation of historical buildings in world heritage sites.789===== PAGE 1 (score 8, numeric-paren 1, table-words 0, var-words 7) =====1011Copyright © 2022 The Author(s). Published by Vilnius Gediminas Technical University12This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unre-13stricted use, distribution, and reproduction in any medium, provided the original author and source are credited.14Introduction15When a site is designated as World Heritage by UNESCO 16(The United Nations Educational, Scientific and Cultur-17al Organization), the designation can result in multiple 18benefits such as attracting investment opportunities, in-19creasing tourism, and the regeneration of the surround-20ing urban environment. Alongside these benefits, this 21sought-after status also imposes certain responsibilities. 22Public officials at the local, regional and national levels of 23government must enact laws and adopt policies that make 24it possible to protect registered World Heritage properties 25according to the World Heritage Convention, also known 26as the Convention for the Protection of the World Cultural 27and Natural Heritage (World Heritage Committee, 1972).28In turn, these policy changes in local regulation may 29spur other changes in the city, resulting in conflicts and 30uncertainty. One such change that may be expected is the 31effect on the rights and privileges of property owners who 32suddenly find themselves bound by regulations and re-33strictions. This paper examines these issues by looking at 34the changes brought about by UNESCO’s designation of 35Tel Aviv’s “White City” as a World Heritage site.36To do this, we examine the market value of heritage 37properties in a district earmarked by UNESCO as World 38Heritage. Market value is defined as the price at which 39both buyers and sellers are willing to do business (Ap-40praisal Institute, 1992). This price is assumed to reflect 41the most profitable use of the property – its highest and 42best use. From the perspective of owners, there are several 43feasible routes to handle heritage properties. For exam-44ple, a non-designated building – i.e., a property not se-45lected for preservation – can be renovated, expanded, or 46demolished, and in the latter case, a new building can be 47constructed in its place. While preservationists and World 48Heritage enthusiasts look at a variety of social and com-49munity values in pursuing heritage protection, property 50owners may adopt a different perspective. In particular, 51they may look to maximize their property’s value in ac-52cordance with the principle of the highest present value 53yield. In such cases, demolition would be considered when 54the expected value gained from future streams of revenues 55would equal or be greater than the net revenue derived 56from simply renovating or expanding (Amin & Capozza, 571993; Brueckner, 1980).58International Journal of Strategic Property Management59ISSN: 1648-715X / eISSN: 1648-9179602022 Volume 26 Issue 5: 397–40961https://doi.org/10.3846/ijspm.2022.1815662*Corresponding author. E-mail: nirm@technion.ac.il63WORLD HERITAGE DESIGNATION AND PROPERTY: 64THE IMPACT OF PRESERVATION ON THE PRICE OF HISTORIC ASSETS65Eyal SALINGER 66 , Daniel SHEFER, Nir MUALAM 67 *68Faculty of Architecture and Town Planning, Technion – Israel Institute of Technology, Haifa, Israel69Received 12 April 2022; accepted 21 November 202270Abstract. The restrictions and requirements imposed by historic preservation regulations bring about many changes to the 71rights of property owners. They might impose additional costs, most notably by prohibiting the demolition of designated 72buildings and thus decreasing property-development opportunities. The objective of this study is to examine what happens 73following world heritage designations; specifically, if and how such designations impact the property values of historic 74buildings. Using the hedonic price method, we measure the value of the option to demolish and rebuild that is denied to 75owners of designated buildings. We also measure the value of preservation regulations, expressed in the prices of apart-76ments in designated buildings. The study area is the “White City” of Tel Aviv, which UNESCO designated as a world herit-77age site. The findings suggest that the market price of a designated building is on average 12.5% lower than its theoretical 78value if the building was not subject to historic preservation regulations. Moreover, a premium of approximately 14% was 79found in the price of apartment units in designated buildings if the building underwent preservation. These findings could 80have a direct impact on public policies designed to promote the preservation of historical buildings in world heritage sites.81Keywords: heritage, cultural property, valuation, property value, market price, historic preservation, UNESCO.828384===== PAGE 6 (score 17, numeric-paren 0, table-words 3, var-words 8) =====858640287E. Salinger et al. World heritage designation and property: the impact of preservation on the price...88were designated prior to their sale. All observations for 89non-designated/designated buildings are for buildings that 90were not renovated/not preserved prior to the transaction. 91In addition, we created a database for apartment transac-92tions which contains information about 237 sales, 91 of 93which are in designated buildings that were earmarked for 94protection prior to the sale.95Our sample is unique, as to the best of our knowledge 96there have been few if any studies investigating the price 97impact of WH and ensuing preservation policies on des-98ignated buildings still endowed with development rights. 99Moreover, based on data from a real estate-appraiser prac-100tice, adjustments were made for the transaction price of 101buildings in cases involving a building sold with statutory 102tenants’ rights and in cases in which a building permit 103was already issued for (additional) construction prior to 104the transaction. In the first instance, the expected cost of 105evacuating the statutory tenants, based on comparable 106transactions, was added to the transaction price. In the 107second scenario, the cost of obtaining a permit, in terms 108of both money and time saved, was taken into considera-109tion when determining transaction price. Therefore, the 110sale price represents a building free of statutory tenants 111that was granted permission for additional construction.112Tables 1 and 2 below define the main variables that 113were tested in the hedonic price method.114Table 1. Main variables included in the hedonic price method – building transactions115Variables116Description117Mean118Std. Dev.119Dependent variable120Ln_BUILD_PRICE121Ln nominal selling price of the building, in New Israeli Shekels (NIS)12216.171230.59124Independent variables125Variables controlling for location characteristics126LOCATION_PEDESTRIAN_STREET A dummy variable that equals to 1 if the building is exposed to 127positive externalities: e.g., located on a pedestrian street1280.241290.43130LOCATION_MAIN_ROAD131A dummy variable that equals to 1 if the building is exposed to 132negative externalities: e.g., located on a main transportation artery1330.211340.41135BUFFER_DESIGNATED136A dummy variable that equals to 1 if there are more than 25 137designated buildings located within 100 meters of the sold property1380.201390.40140Variables controlling for structural characteristics141FACADE142A dummy variable that equals to 1 if the building has a wide façade 143or if it is a corner building1440.511450.50146Ln_PARCEL_AREA147Ln parcel area in sq. m.1486.301490.43150YEAR_BUILT151Year of construction152194915323.5154UNITS155Number of apartments in the building1569.211575.67158BUILD_ATTACHED159A dummy variable that equals to 1 if the building has a joint wall 160with an adjacent building1610.081620.27163BUILD_PARKING164A dummy variable that equals to 1 if parking, or the possibility of 165parking, is currently available within the building’s plot1660.441670.50168Variables controlling for planning regulations that apply to the plot169Ln_BUILD_%_USED170Ln development rights used, calculated as the total gross built-up area 171divided by the plot area, multiplied by 100 (equals 0 if the highest 172and best use of a non-designated building is to demolish and rebuild)1733.971741.97175NOT_FOR_DEMOLITION176A dummy variable that equals to 1 for all designated buildings and 177for non-designated buildings if the highest and best use of non-178designated buildings is not to demolish the building1790.811800.39181DESIGNATED182A dummy variable that equals to 1 if the building is designated for 183preservation (all types of designations)1840.561850.50186DESIGNATED_DETAILED187A dummy variable that equals to 1 if the building is designated for 188preservation according to a detailed statutory plan which sets very 189specific provisions and limitations1900.1751910.38192DESIGNATED_LOCAL_SEVERE193A dummy variable that equals to 1 if the regulation pertaining to 194a designated historic building sets “severe” (grade 1) management 195restrictions- thus curtailing the ability of owners to use, alter, or add 196to an existing building1970.1751980.38199DESIGNATED_LOCAL_REGULAR200A dummy variable that equals to 1 if the historic building is designated 201for “ordinary” more flexible (grade 2) preservation according to existing 202regulation. That is, it is not subject to severe restrictions on property2030.212040.41205DESIGNATED_INTERNATIONAL206A dummy variable that equals to 1 if the building is designated as 207historic and was built in the early International Style2080.272090.44210DESIGNATED_ECLECTIC211A dummy variable that equals to 1 if the building is designated as 212historic and was built in the eclectic Style2130.272140.44215216217===== PAGE 7 (score 19, numeric-paren 0, table-words 3, var-words 9) =====218219International Journal of Strategic Property Management, 2022, 26(5): 397–409220403221Variables222Description223Mean224Std. Dev.225Ln_PARCEL_AREA_DESIGNATED226An interaction variable of the variable Ln_PARCEL_AREA with the 227variable DESIGNATED2283.582293.24230Ln_BUILD_%_USED_DESIGNATED An interaction variable of the variable Ln_BUILD_%_USED with the 231variable DESIGNATED2322.662332.42234Variables controlling for time235TRANSACTION_YEAR236A dummy variable that equals to 1 for the year the building was sold 237(except for the year 1998)238UNEMPLOYMENT239Unemployment rate in Tel Aviv, sorted by year2406.672411.37242End of Table 1243Table 2. Main variables included in the hedonic price method – apartment transactions244Variables245Description246Mean247Std. Dev.248Dependent variable249Ln_APT_PRICE250Ln nominal selling price of the apartment, in New Israeli Shekels (NIS)25113.952520.38253Independent variables254Variables controlling for location characteristics255LOCATION_PEDESTRIAN_256STREET257A dummy variable that equals to 1 if the property sold enjoys positive 258externalities: e.g., located on a pedestrian street2590.222600.42261LOCATION_MAIN_ROAD262A dummy variable equals to 1 if the property sold is exposed to 263negative externalities: e.g., located on a main transportation artery2640.042650.19266BUFFER_DESIGNATED267A dummy variable that equals to 1 if there are more than 25 268designated buildings located within 100 meters of the sold property2690.142700.34271GARDEN272A dummy variable that equals to 1 if the property sold is within 150 273meters of a public garden2740.462750.50276Variables controlling for structural characteristics277Ln_TOTAL_APT_AREA278Ln total apartment area, in sq. m.2794.272800.32281FLOOR282Floor on which the apartment is located2833.022841.19285PARKING286A dummy variable that equals to 1 if there is a parking spot within 287the building’s plot2880.192890.39290ELEVATOR1291A dummy variable that equals to 1 if there is an elevator in the 292building when the apartment sold is on the second floor or higher2930.442940.50295YEAR_BUILT296Year of construction297195629822.74299Ln_UNITS_PER_DUNAM300Ln number of apartments per dunam3013.133022.67303APT_RENOVATED304A dummy variable that equals to 1 if the apartment sold had been 305renovated3060.203070.40308BUILD_PRESERVED309For designated buildings, a dummy variable that equals to 1 if the 310building was preserved before the transaction3110.0633120.24313BUILD_RENOVATED314For non-designated buildings, a dummy variable that equals to 1 if 315the building was renovated before the transaction3160.173170.38318Variables controlling for planning regulations that apply to the plot319DESIGNATED320A dummy variable that equals 1 if the building is designated for 321preservation (all types of designations)3220.383230.49324DESIGNATED_DETAILED325A dummy variable that equals 1 if the building is designated for 326preservation according to a detailed statutory plan which sets very 327specific provisions and limitations3280.163290.37330DESIGNATED_LOCAL_SEVERE331A dummy variable that equals to 1 if the regulation pertaining to a 332designated historic building sets “severe” management restrictions 333(grade 1 buildings)- thus curtailing the ability of owners to use, alter, 334or add to an existing building3350.053360.22337DESIGNATED_LOCAL_REGULAR338A dummy variable that equals to 1 if the historic building is 339designated for “regular” preservation according to existing regulation. 340That is, it is not subject to severe restrictions on property3410.173420.37343Variables controlling for time344TRANSACTION_YEAR345A dummy variable for each of the years sampled (excluding 1998)346UNEMPLOYMENT347Rate of unemployment in Tel Aviv3487.243491.39350351352===== PAGE 8 (score 86, numeric-paren 46, table-words 13, var-words 4) =====353354404355E. Salinger et al. World heritage designation and property: the impact of preservation on the price...3565. Findings357The results of our analysis are presented in Tables 3 and 4. 358Table 3 shows that the independent variables that were 359expected to have an impact on transactions have indeed 360had the expected negative or positive impact on pricing. 361Model 1 depicts a simple comparison of the price of desig-362nated and non-designated buildings in UNESCO’s White 363City. As can be seen from the results, DESIGNATED vari-364able is not significant at the 10% level. A plausible reason 365for this is that designated buildings vary immensely in 366regard to the level of instituted preservation. While some 367are granted a range of developer incentives by the Preser-368vation Plan, others are left with fewer incentives in order 369to ensure their preservation. Put differently, incentives for 370designated buildings differ according to the main charac-371teristics of the building, which determine the value of the 372building and the highest and best use as if the building 373was not designated.374Thus, in Model 2 and Model 3 we replace the 375DESIGNATED variables with two main characteristics: 376parcel area and built-up area (i.e., development rights 377already used by the building’s owners). In Model 2 all 378variables are significant at the 10% level. The signs of the 379interaction variables, positive for ln_BUILD_%_USED_380DESIGNATED and negative for ln_PARCEL_AREA_381DESIGNATED, imply a stronger impact of built-up area 382for designated buildings and a weaker impact of parcel 383area for designated buildings compared to non-designated 384properties. The variable controlling for development rights 385equals zero for non-designated buildings that can be de-386molished. The same variable has a value attached to it in 387cases of designated buildings that cannot be demolished, 388or in cases of non-historic properties whose best use is 389to keep them intact (the average ln value being approxi-390mately 5). Thus, we added a dummy variable in Model 1 391and Model 2 that equals to 1 for (a) designated buildings 392Table 3. Estimated coefficients for building transactions393Building type394Model 1395Model 2396Model 3397Model 4398Model 5399All400All401All402Non-designated Non-designated403Coefficient 404(t-value)405Coefficient406(t-value)407Coefficient 408(t-value)409Coefficient410(t-value)411Coefficient 412(t-value)413Independent variables414CONSTANT41512.703416(27.09)***41711.850418(18.74)***41911.023 420(20.69)***42111.738422(17.99)***42311.366 424(19.30)***425Ln_PARCEL_AREA4260.761427(9.70)***4280.913429(8.74)***4301.014 431(11.49)***4320.881433(8.22)***4340.901 435(9.13)***436Ln_BUILD_%_USED4370.572438(5.82)***4390.351440(2.13)**4410.273442(1.28)443UNEMPLOYMENT444–0.203445(–8.80)***446–0.211447(–9.88)***448–0.196 449(–9.06)***450–0.176451(–5.32)***452–0.152 453(–5.25)***454LOCATION_PEDESTRIAN_STREET4550.350456(4.59)***4570.367458(5.11)***4590.383 460(5.15)***4610.417462(2.46)**4630.359 464(2.25)**465LOCATION_MAIN_ROAD466–0.111467(–1.55)468–0.120469(–1.71)*470–0.079471(–0.87)472FAÇADE4730.147474(2.40)**4750.164476(2.77)***4770.195 478(3.24)***4790.252480(2.79)**4810.283 482(3.39)***483YEAR_20094840.551485(3.94)***4860.512487(3.86)***4880.524 489(3.72)***4900.561491(3.49)***4920.566 493(3.52)***494Ln_PARCEL_AREA_DESIGNATED495–0.230496(–1.81)*497–0.440 498(–6.34)***499Ln_BUILD_%_USED_DESIGNATED5000.292501(1.87)*5020.537 503(5.98)***504NOT_FOR_DEMOLITION505–3.00506(–5.79)***507–1.904508(–2.28)**509–1.489510(–1.35)511DESIGNATED5120.074513(0.88)514DESIGNATED_LOCAL_REGULAR515–0.093516(–1.05)517BUFFER_DESIGNATED5180.029519(0.39)520Number of observations5216352263523635242852528526F=52735.9352841.9652951.5253024.8453137.71532Adjusted R Square5330.8615340.8695350.8515360.8765370.872538Note: * ρ < 0.1; ** ρ < 0.05; *** ρ < 0.01.539540541===== PAGE 9 (score 99, numeric-paren 60, table-words 10, var-words 6) =====542543International Journal of Strategic Property Management, 2022, 26(5): 397–409544405545and for (b) cases in which the highest and best use of a 546non-designated building is to keep the existing building 547(NOT_FOR_DEMOLITION). The variable has a negative 548sign, correcting for the non-continuous distribution of the 549variable controlling for development rights.550Model 3 uses the stepwise method and refers to the 551variables ln_PARCEL_AREA and ln_PARCEL_AREA_552DESIGNATED. The results suggest that the impact of the 553parcel area (in ln) is more pronounced for non-designated 554buildings than for designated buildings. A change of 1% in 555the parcel area (Ln_PARCEL_AREA) affects the price for 556non-designated buildings by 1.014%, implying a high de-557mand for a limited supply of large plots in the White City 558of Tel Aviv. As for designated buildings, where demolition 559is prohibited, and hence property development opportuni-560ties are restricted, an interaction variable of the parcel area 561with designated buildings (ln_PARCEL_AREA_DESIG-562NATED) is included with a negative sign, mitigating the 563impact of the parcel area. According to Model 3, for des-564ignated buildings, a change of 1% in the parcel area affects 565the price by 0.574% (1.014 – 0.440 = 0.574). Moreover, the 566variable controlling for development rights (BUILD_%_567USED_DESIGNATED) has a significant positive sign only 568for designated buildings. A change of 1% in the utilized 569development rights affects the price of designated build-570ings by 0.537%.571Table 4. Estimated coefficients for apartment transactions572Building type573Model 6574Model 7575Model 8576Model 9577All578All579Non-designated580Non-designated581Coefficient582(t-value)583Coefficient 584(t-value)585Coefficient586(t-value)587Coefficient 588(t-value)589Independent variables590CONSTANT59111.026592(56.41)***59311.014594(56.37)***59511.145596(42.88)***59711.062598(43.19)***599Ln_TOTAL_APT_AREA6000.702601(22.84)***6020.700 603(22.79)***6040.708605(16.37)***6060.721607(17.07)***608PARKING6090.116610(4.43)***6110.126612(5.01)***6130.123614(4.43)***6150.128616(4.67)***617ELEVATOR16180.07619(2.93)***6200.085621(3.98)***6220.044623(1.48)6240.056625(2.01)**626FLOOR6270.023628(2.93)***6290.023630(2.92)***6310.021632(2.26)**6330.019634(2.04)**635Ln_UNITS_PER_DUNAM636–0.112637(–3.16)***638–0.109639(–3.08)***640–0.154641(–3.46)***642–0.140643(–3.16)***644DESIGNATED645–0.025646(–1.15)647BUILD_PRESERVED6480.129649(3.14)***6500.105651(2.80)***652BUILD_RENOVATED6530.022654(0.80)6550.033656(1.15)657LOCATION_PEDESTRIAN_STREET6580.08659(3.65)***6600.075661(3.46)***6620.074663(2.53)**6640.066665(2.27)**666LOCATION_MAIN_ROAD667–0.182668(–4.01)***669–0.190670(–4.23)***671–0.213672(–2.25)**673–0.212674(–2.24)**675BUFFER_DESIGNATED6760.061677(2.27)**6780.057679(2.12)**6800.073681(1.62)682YEAR_2001683–0.192684(–5.18)***685–0.199686(–5.41)***687–0.204688(–4.21)***689–0.196690(–4.07)***691YEAR_20066920.164693(5.52)***6940.164695(5.54)***6960.195697(4.84)***6980.191699(4.78)***700YEAR_20077010.277702(9.67)***7030.273704(9.69)***7050.302706(7.55)***7070.297708(7.52)***709YEAR_20087100.389711(14.09)***7120.393713(14.29)***7140.371715(11.03)***7160.362717(10.94)***718YEAR_20097190.474720(13.20)***7210.490722(17.88)***7230.478724(13.10)***7250.474726(13.20)***727Number of observations728237729237730146731146732F=733121.43734138.5273589.43736102.85737Adjusted R Square7380.8917390.8917400.8957410.894742Note: * ρ < 0.1; ** ρ < 0.05; *** ρ < 0.01.743744745===== PAGE 10 (score 16, numeric-paren 1, table-words 2, var-words 11) =====746747406748E. Salinger et al. World heritage designation and property: the impact of preservation on the price...749the sale of the apartment (BUILD_PRESERVED). Thus, 750there was a major investment in preserving the building 751which is much higher compared to a lower investment for 752the renovation of non-designated buildings prior to the 753sale of the apartment (BUILD_RENOVATED).3 This re-754sult can be explained by the fact that buyers of apartments 755usually do not have the resources to preserve the building 756in its entirety; they often will appreciate preservation if the 757building had already been preserved.758An interesting result from Model 6 is that the variable 759controlling for the physical condition of non-designated 760buildings (BUILD_RENOVATED) is not significant at the 76110% level. This variable was omitted in Model 7 (the step-762wise method), implying that the value of an apartment 763unit in a non-designated building is not explained by the 764physical condition of the building (whether renovated or 765not). Two factors may explain this result: first, 39 of 40 766non-designated buildings in the sample that were reno-767vated have an elevator; and second, 23 out of 40 build-768ings offer parking space. Both variables were included in 769the model with a significant positive impact. These results 770suggest that the increase in value expressed by these two 771variables together (23.5% in Model 7) may also reflect the 772contribution of the renovation to the price of the building.773In addition, Model 7 illustrates that the positive ex-774ternalities that preservation generates increases the prices 775of apartments in the urban environment by 6% (BUFF-776ER_DESIGNATED). This implies that apartment buyers 777in the White City generally appreciate locations within 778a given conservation area, where preservation gener-779ates positive externalities (spillover effect). Thus, World 780Heritage sites may generate a premium on apartments in 781already preserved buildings as well as on apartments lo-782cated near an ensemble of heritage assets. World Heritage 783could then have a cumulative effect which depends on a 784concentration of heritage properties in one location. As 785more buildings are preserved, property values in the ur-786ban environment are expected to increase by more than 7876%. However, it should be noted that the BUFFER_DES-788IGNATED variable was not significant at the 10% level in 789Model 1 for transactions (sales) of entire buildings. This 790can imply the existence of separate markets for those who 791buy entire buildings (i.e., real estate developers) and those 792purchasing a single apartment.793The interpretation of other variables included in Mod-794el 7 suggest that a change of 1% in the total apartment 795area (TOTAL_APT_AREA) affects the price by 0.7%. For 796example, for an apartment sold in the White City for NIS 7972,000,000, if the total area of the apartment increases by 7981% (say from 100 sq. m. to 101 sq. m.), then the price of 799the apartment will amount to NIS 2,014,000. This is in line 8003 In Model 6, the premium is calculated as the difference between 801the coefficient of the two variables (0.129 – 0.022 = 0.107). It is 802similar to the coefficient of BUILD_PRESERVED in Model 7 803(0.105). See also Halvorsen and Palmquist (1980) for the inter-804pretation of dummy variables in semi logarithmic equations.805In order to calculate the value of the option to demol-806ish and rebuild we applied Model 4 and Model 5 only for 807non-designated buildings. In Model 4 the variable control-808ling for development rights (Ln_BUILD_%_USED) was 809not significant at the 10% level. This variable was omitted 810in Model 5 (the stepwise method), implying that the value 811of a building is not explained by its physical attributes. 812One possible explanation of this result is that all buildings 813in the sample were in poor/deteriorating physical condi-814tion when sold. Moreover, in 40% of cases the highest and 815best use for the property was to demolish the building 816and build a new structure, or to remain indifferent about 817the option of demolition. Thus, the physical component 818attribute in these cases equals to zero2.819We made an out-of-sample prediction where the 820impact of preservation regulations on property value is 821calculated as the difference between the actual sale price 822of the building and its value if it were not earmarked for 823preservation. This calculation is performed by placing the 824attributes of the historic building (the independent vari-825able values) in Model 5 (non-designated buildings). When 826the actual sale price is lower than the estimated value of 827the building if it were not designated for preservation, 828then there is a decrease in the value of the property owing 829to the lack of the option to demolish and rebuild. Results 830suggest that in about 70% of the buildings sampled, the 831property value calculated in Model 5 is higher than the 832actual sale price. This implies a decrease in value owing to 833the lack of the option to demolish and rebuild. The aver-834age difference in natural log (Ln) for all 35 buildings sam-835pled is 0.21. However, when extreme results of differences 836(between the actual sale price of the building and the es-837timated value if it were not designated) are omitted, the 838average difference is 0.12, implying a decrease of 12.5% 839from the actual price of designated buildings because of 840the lack of the option to demolish and rebuild.841For buildings for which no decrease in value was 842found, the average plot area is 605 sq. m., and the average 843development rights used (in %) is 164%. For buildings for 844which a decrease in value was found owing to the lack of 845option to demolish and rebuild, the average parcel area 846was higher, 701 square meters, and the development rights 847used were lower, only 114%. This implies that the lack of 848the option to demolish and rebuild is most pronounced 849for designated buildings with a large plot area and limited 850development rights.851As for the results for apartment transactions (see Ta-852ble 4), the independent variables that were expected to 853have an impact on transactions have indeed had the ex-854pected negative or positive impact on pricing. Results of 855Model 6 suggest that, by itself, designation of heritage 856properties (expressed by the DESIGNATED variable) was 857not significant at the 10% level. Nevertheless, a premium 858of 11% on the price of apartments was found in designated 859buildings, but only if the building was preserved prior to 8602 The evaluation was based on real estate appraisers’ practice.861