# RECORD a5c3a8b4f02ea9b02009cba2eba334012eb7b212 # TITLE: Economic Impact of a Heritage Policy on Residential Property Values in a Historic District Context: The Case of the Old City of Quebec # YEAR: 2018.0 VENUE: The Review of Regional Studies DOI: 10.52324/001c.7989 # AUTHORS: Nicolas Devaux; É. Berthold; Jean Dubé # PAGES: 19 CANDIDATE TABLE PAGES (1-based): [12, 13] # ABSTRACT: Previous studies have attempted to measure the willingness-to-pay to locate inside a historic district. Yet not enough attention has been paid to the implicit value attributed to the impact of heritage policies within the historic district. This paper fills that gap by investigating the impact of such policies on individual condominium properties for the case of the Old City of Quebec (Canada), designated by law as a protected area and included on the World Heritage List (UNESCO). More precisely, this research measures the impacts of the reorganization and landscaping of one particular street located south-west of the protection area in the old part of Quebec City. A difference-in-differences (DID) estimator based on a hedonic repeated sales approach is proposed to isolate such effects based on proximity to the site under study. Estimation results suggest that the reorganization of the street had no significant impact on the closest properties’ prices, but had negative effects for properties located within 150 to 450 meters of the street. These observations outline complex relations between heritage policies and urban landscaping projects. ===== PAGE 1 (score 9, numeric-paren 0, table-words 1, var-words 7) ===== (2018) 48, 279–297 © Southern Regional Science Association 2018. ISSN 1553-0892, 0048-749X (online) www.srsa.org/rrs The Review of Regional Studies The Official Journal of the Southern Regional Science Association Economic Impact of a Heritage Policy on Residential Property Values in a Historic District Context: The Case of the Old City of Quebec* Nicolas Devauxa, Etienne Bertholdb, and Jean Dubéc a Department of Societies, Territories, and Development, University of Quebec in Rimouski, Canada b Department of Geography, Laval University, Canada c School of Planning and Development, Laval University, Canada Abstract: Previous studies have attempted to measure the willingness-to-pay to locate inside a historic district. Yet not enough attention has been paid to the implicit value attributed to the impact of heritage policies within the historic district. This paper fills that gap by investigating the impact of such policies on individual condominium properties for the case of the Old City of Quebec (Canada), designated by law as a protected area and included on the World Heritage List (UNESCO). More precisely, this research measures the impacts of the reorganization and landscaping of one particular street located south-west of the protection area in the old part of Quebec City. A difference-in-differences (DID) estimator based on a hedonic repeated sales approach is proposed to isolate such effects based on proximity to the site under study. Estimation results suggest that the reorganization of the street had no significant impact on the closest properties’ prices, but had negative effects for properties located within 150 to 450 meters of the street. These observations outline complex relations between heritage policies and urban landscaping projects. Keywords: historic districts, landscaping policy, urban policy, hedonic pricing model, urban regeneration JEL Codes: O18, R31, R52, R58 1. INTRODUCTION Cultural heritage has gained more importance in the field of urban studies (Harrison, 2013). Public policies are an essential part of heritage conservation that protect the history of an area (Graham, Ashworth, and Tunbridge, 2000; Saint-Pierre, 2008). As with many cultural policies, heritage policies may produce cultural benefits for local communities and nations, among which collective identity flourishes (Hobsbawm and Ranger, 1983; Harrison, 2013). Urban heritage policies should be distinguished from a standard landscaping policy because they include intangible and symbolic goals. Such policies can also have important economic effects. Such impacts can be expressed through economic benefits related to increased tourism, but also through externalities for residents and commercial businesses. For the former, the classical input-output analysis is used to evaluate the direct, indirect, and spinoff effects. For the latter, hedonic modeling has proved to be an elegant way of measuring the implicit price (or willingness-to-pay) of being close to such Nicolas Devaux (Department of Societies, Territories, and Development) is Professor at Universtiy of Quebec in Rimouski, Rimouski, Qc, Canada. Etienne Berthold (Department of Geography) and Jean Dubé (School of Planning and Development) are Professors at Laval University, Quebec City, Qc, Canada. Corresponding Author: N. Devaux E-mail: Nicolas.devaux@uqar.ca ===== PAGE 12 (score 54, numeric-paren 42, table-words 1, var-words 1) ===== 290 The Review of Regional Studies 48(3) © Southern Regional Science Association 2018. significant effect is noted. In contrast to the network distance variables, the significant effect is limited to the second buffer when Euclidean distance is used and corresponds to a CAD$11,268 price decrease, a comparable impact on the previous specification. Table 3: Estimation Results Model (1) Model (2) Model (3) Independent Variables Coef. Sig. Coef. Sig. Coef. Sig. 1997-Q2 reference reference reference 1997-Q3 -0.2288 -0.2385 -0.2299 1997-Q4 0.5699 *** 0.5566 *** 0.5608 *** 1998-Q1 -0.5747 ** -0.5879 ** -0.5826 ** 1998-Q2 -0.0553 -0.0752 -0.0713 1998-Q3 -0.0183 -0.0479 -0.0453 1998-Q4 0.0503 0.0254 0.0340 1999-Q1 0.0686 0.0196 0.0192 1999-Q2 0.1000 0.0627 0.0690 1999-Q3 0.0303 0.0210 0.0208 1999-Q4 0.1623 0.1242 0.1351 2000-Q1 0.1597 0.1397 0.1463 2000-Q2 0.0783 0.0675 0.0660 2000-Q3 0.1142 0.1130 0.1129 2000-Q4 0.1507 0.1292 0.1434 2001-Q1 0.1242 0.0999 0.1142 2001-Q2 0.1082 0.0884 0.0909 2001-Q3 0.1863 0.1611 0.1826 2001-Q4 0.1813 0.1626 0.1709 2002-Q1 0.2496 0.2309 0.2354 2002-Q2 0.3019 * 0.2969 * 0.3000 * 2002-Q3 0.2132 0.1955 0.2072 2002-Q4 0.3549 ** 0.2879 * 0.3142 * 2003-Q1 0.2500 0.2431 0.2505 2003-Q2 0.5753 *** 0.5585 *** 0.5674 *** 2003-Q3 0.3297 * 0.2783 * 0.2904 * 2003-Q4 0.6495 *** 0.6340 *** 0.6331 *** 2004-Q1 0.4444 *** 0.4320 *** 0.4328 *** 2004-Q2 0.5782 *** 0.5693 *** 0.5823 *** 2004-Q3 0.5748 *** 0.5434 *** 0.5683 *** 2004-Q4 0.6266 *** 0.6094 *** 0.6181 *** 2005-Q1 0.6875 *** 0.6722 *** 0.6814 *** 2005-Q2 0.6989 *** 0.6855 *** 0.6922 *** 2005-Q3 0.7318 *** 0.7406 *** 0.7152 *** ===== PAGE 13 (score 86, numeric-paren 75, table-words 1, var-words 1) ===== DEVAUX, BERTHOLD, & DUBÉ: HERITAGE POLICY AND RESIDENTIAL PROPERTY VALUES 291 © Southern Regional Science Association 2018. Table 3 (Continued) Model (1) Model (2) Model (3) Independent Variables Coef. Sig. Coef. Sig. Coef. Sig. 2005-Q4 0.6736 *** 0.6771 *** 0.6647 *** 2006-Q1 0.6978 *** 0.7290 *** 0.7047 *** 2006-Q2 0.6741 *** 0.6681 *** 0.6591 *** 2006-Q3 0.7554 *** 0.7540 *** 0.7331 *** 2006-Q4 0.7701 *** 0.7803 *** 0.7594 *** 2007-Q1 0.8387 *** 0.8398 *** 0.8070 *** 2007-Q2 0.8567 *** 0.8792 *** 0.8712 *** 2007-Q3 0.8403 *** 0.8573 *** 0.8271 *** 2007-Q4 0.9031 *** 0.9078 *** 0.9005 *** 2008-Q1 0.8385 *** 0.8540 *** 0.8377 *** 2008-Q2 0.9492 *** 0.9573 *** 0.9276 *** 2008-Q3 1.0158 *** 1.0422 *** 1.0145 *** 2008-Q4 0.9313 *** 0.9123 *** 0.8963 *** 2009-Q1 0.7928 *** 0.7868 *** 0.7669 *** 2009-Q2 0.8729 *** 0.8946 *** 0.8788 *** 2009-Q3 0.8376 *** 0.8351 *** 0.8054 *** 2009-Q4 1.0667 *** 1.0831 *** 1.0760 *** 2010-Q1 0.9712 *** 0.9812 *** 0.9726 *** 2010-Q2 1.1202 *** 1.1241 *** 1.1060 *** 2010-Q3 0.9001 *** 0.9329 *** 0.9170 *** 2010-Q4 1.1378 *** 1.1314 *** 1.1238 *** 2011-Q1 1.1329 *** 1.1350 *** 1.1015 *** 2011-Q2 1.0868 *** 1.0993 *** 1.0848 *** 2011-Q3 1.3397 *** 1.5400 *** 1.5564 *** Grant for renovations (dummy) 0.0386 0.0231 0.0202 Impact - distance to d'Auteuil Street Network dist., 0 to 150 m (Buffer) -0.0405 Network dist., 150 to 300 m (Buffer) -0.2952 ** Network dist., 300 to 450 m (Buffer) -0.2376 * Network dist., 450 to 600 m (Buffer) 0.1353 Network dist., over 600 m reference Euclidean dist., 0 to 150 m (Buffer) 0.0102 Euclidean dist., 150 to 300 m (Buffer) -0.2476 ** Euclidean dist., 300 to 450 m (Buffer) 0.0883 Euclidean dist., 450 to 600 m (Buffer) 0.0795 Euclidean dist., over 600 m reference ===== PAGE 14 (score 21, numeric-paren 0, table-words 2, var-words 17) ===== 292 The Review of Regional Studies 48(3) © Southern Regional Science Association 2018. Table 3 (Continued) Model (1) Model (2) Model (3) Independent Variables Coef. Sig. Coef. Sig. Coef. Sig. N 744 744 744 R2 0.575 0.586 0.585 Akaike Information Criteria (AIC) 692.977 682.129 683.867 Baysian Information Criteria (BIC) 960.475 968.075 969.813 Legend: * p<0,05; ** p<0,01; *** p<0,001 Overall, the results point to three main conclusions. First, the assistance programs for renovations of properties located in Old Quebec have no impact on the selling price variations. This does not mean that condominium renovations are not valued. It is conceivable that renovated condominiums are sold at higher prices, meaning that public funds used for private renovations would be transformed in a greater profit margin for the seller and to tax revenues for the municipality. However, this is not the case here. One assumption would be that such renovations remove some “cachet” from the original apartments, even when offering some additional and more contemporaneous commodities. Another assumption would be that the effect is minimized considering that most of the condominiums in the area had been restored a few years before the analysis. In fact, as previously mentioned, two-thirds of the condominiums had been restored between 1980 and 1999. A final assumption, that could not be tested here, is that renovation may influence time-on-the-market, but not prices, with renovated condominiums being sold more rapidly than the others. Second, the results suggest a localized and negative effect of the landscaping renovation of d’Auteuil Street for condominiums located within 150 to 450 meters network distance and within 150 to 300 meters Euclidian distance. As a matter of fact, the reorganization and landscaping of d’Auteuil Street, turning from a four-lane street into a two-lane street, has brought more road traffic to the adjacent streets of the district. The reorganization of the street into a one- way road leading south also added high traffic volume in the heart of the historic district (more than 400 vehicles every hour from 7−9 am and 3−5 pm on work days). Noise can also be generated from the change in the material used to build the road, a result of the pavement of the road itself (made of stones, not of asphalt) on which vehicles travel up to 50 km/h (31 m/h).5 Unlike asphalted roads, a paved road generates an uneven friction between the car wheels and the ground, generating more noise. Consequently, the additional crowding effect and noise pollution can negatively influence house prices. Several studies have already shown how noise can lead to a depreciation of residential values, as such might be the case with airports and housing values (Baranzini and Ramirez, 2005; Cohen and Coughlin, 2008). Another negative externality could be related to the redesigned location of parking spaces. Surrounding streets may experience the drawback of the new street design since parking spaces are limited for residents inside the walled section of the Old City. In fact, the new one- way street may also change parking habits for vehicles heading north. This assumption seems plausible considering the scarcity of parking spaces in such dense and winding areas. 5 Sandberg (1987) has investigated the issue of road traffic noise in regard to different surface components, namely stones and asphalt.