1# RECORD 6663c2076db87ffe9baf2162abc65f7ffa450ab02# TITLE: Evaluating the effects of supervised consumption sites on housing prices in Montreal, Canada using interrupted time series and hedonic price models3# YEAR: 2024.0 VENUE: Drug and Alcohol Dependence Reports DOI: 10.1016/j.dadr.2024.1002424# AUTHORS: M. Schaefer; D. Panagiotoglou5# PAGES: 16 CANDIDATE TABLE PAGES (1-based): [13, 14, 15]6# ABSTRACT: Background In 2017, three brick and mortar supervised consumption sites (SCS) opened in Montreal, Canada. Opponents argued the sites would attract people who use drugs and reduce local real estate prices. Methods We used interrupted time series and hedonic price models to evaluate the effects of Montreal’s SCS on local real estate prices. We linked the Quebec Professional Association of Real Estate Brokers’ housing sales data provided by Centris Inc. with census tract data and gentrification scores. Homes sold within 200 m of the SCS locations between 1 January 2014 and 31 December 2021 were included. We adjusted for internal (e.g., number of bed/bathrooms, unit size) and external attributes (e.g., neighbourhood demographics), and included a spatio-temporal lag to account for correlation between sales. For sensitivity analysis we used site-specific dummy variables to better account for unmeasured neighbourhood differences, and repeated analyses using 500 m and 1000 m radii. Results We observed a price shock after the opening of the first two SCS in June 2017 (level effect: −10.5%, 95% CI: −19.1%, −1.1%) but prices rose faster month-to-month (trend effect: 1.1%, 95% CI: 0.7%, 1.6%) after implementation. Following the implementation of the third site in November 2017 there was no immediate impact (level effect: 2.4%, 95% CI: −10.4%, 17.0%) but once more prices roses faster (0.9%, 95% CI: 0.4%, 1.5%) thereafter. When we replaced neighbourhood attributes with a site-specific dummy variable, we observed the same pattern. Sales’ prices dropped (level effect: −9.6%, 95% CI: −15.0%, −3.8%) but rose faster month-to-month (trend effect: 0.9%, 95% CI: 0.6%, 1.2%) following June 2017’s SCS implementations, with no level effect (4.9%, 95% CI: −7.3%, 18.6%) and a positive trend (0.9%, 95% CI: 0.5%, 1.3%) after November 2017’s SCS opening. In most 500 m and 1000 m radii models, there were no immediate shocks following SCS opening, however, positive trend effects persisted in all models. Conclusion Our models suggest homes sold near SCS may experience a price shock immediately post-implementation, with evidence of market recovery in the months that follow.789===== PAGE 1 (score 6, numeric-paren 0, table-words 0, var-words 6) =====1011Title: Evaluating the effects of supervised consumption sites on housing prices in Montreal, Canada 12using controlled interrupted time series and hedonic price models 13Authors: Maximilian Schaefer MScPH1, Dimitra Panagiotoglou PhD1 141Department of Epidemiology, Biostatistics, and Occupational Health, McGill University, Montreal, 15Canada 16 17Corresponding author: 18Dimitra Panagiotoglou 19Department of Epidemiology, Biostatistics and Occupational Health 20McGill University 211200-2001 McGill College Avenue 22Montreal, QC H3A 1G1 23Canada 24 25Contributors 26DP conceptualized the study, acquired funding, and obtained the data. All authors accessed and 27verified the underlying data. MS conducted the analyses. All authors interpreted the results. DP drafted 28the original version of the manuscript. MS compiled tables and figures. All authors reviewed and edited 29the manuscript. All authors approved the final version and accept responsibility for the decision to 30submit the manuscript for publication. 31 32Declaration of interests 33We declare no competing interests. 34 35Data sharing 36Due to commercial restrictions, supporting data is not available. 37 38Funding 39This study was supported by an Insights Development Grant from the Social Sciences and 40Humanities Research Council of Canada (Grant # 430-2022-00061) to DP. DP is also supported by a 41Junior 1 Scientist Award from the fonds de recherche du Québec – Santé (Award #297427). The funders 42of the study had no role in the study design, data collection, data analysis, data interpretation, or writing 43of the report. 44 45 46 47 48All rights reserved. No reuse allowed without permission. 49(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 50The copyright holder for this preprint51this version posted June 12, 2023. 52; 53https://doi.org/10.1101/2023.06.09.2329120954doi: 55medRxiv preprint 56NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.575859===== PAGE 13 (score 72, numeric-paren 40, table-words 3, var-words 11) =====6061Table 1: Comparison of house and neighborhood features of treated and control units pre-intervention (1 January 2014 – 31 July 2017) 62 63Primary analysis 64Subset analysis 65 66Control units 67N=1,537 68Treated units 69N=765 70p-value 71Control units 72N=830 73Treated units 74N=714 75p-value 76Closing price, mean (SD) 77$419,001.00 78($339,096.23) 79$271,473.73 80($123,347.53) 81<0.001 82462,517.94 83(405,532.79) 84271,662.04 85(125,255.45) 86<0.001 87Housing features, mean (SD): 88 89 No. of bathrooms 901.26 (0.47) 911.09 (0.30) 92<0.001 931.32 (0.51) 941.09 (0.30) 95<0.001 96 No. of bedrooms 971.60 (0.70) 981.87 (0.71) 99<0.001 1001.67 (0.72) 1011.86 (0.71) 102<0.001 103 No. of extra rooms 1041.96 (1.19) 1052.27 (1.14) 106<0.001 1071.92 (1.26) 1082.26 (1.14) 109<0.001 110 Floor size (in m2) 11180.85 (40.57) 11283.87 (27.15) 1130.063 11486.81 (44.49) 11583.71 (27.10) 1160.105 117 Distance to closest 118shelter/SCS 119684.38 (239.35) 120607.41 (276.54) 121<0.001 122632.41 (259.01) 123605.04 (282.36) 1240.047 125Neighbourhood demographics, mean (SD): 126 127 Age of the population 12832.91 (4.53) 12935.3 (2.12) 130<0.001 13131.53 (3.96) 13235.80 (2.05) 133<0.001 134 Household income 135$36,615.76 136($17,124.33) 137$28,663.55 138($4,515.64) 139<0.001 140$26,304.72 141($15,998.87) 142$28,520.07 143($3,833.68) 144<0.001 145 Household size 1461.68 (0.18) 1471.77 (0.10) 148<0.001 1491.69 (0.18) 1501.77 (0.10) 151<0.001 152Proportion of population, %: 153 154 155 156 157 158 159 Indigenousb 1600.41 1611.08 162<0.001 1630.25 1641.03 165<0.001 166 Visible minoritiesc 16741.87 16815.65 169<0.001 17049.76 17115.57 172<0.001 173 Without secondary school 174completedd 1752.23 17612.04 177<0.001 1782.70 17912.07 180<0.001 181 With postsecondary 182education 18389.20 18473.75 185<0.001 18687.83 18773.75 188<0.001 189 Unemployed (rate) 1908.49 1918.40 1920.555 19310.39 1948.33 195<0.001 196Proximity on scale from 0 to 1, mean (SD): 197 198 Employmente 199 200 201 2020.37 (0.08) 2030.18 (0.05) 204<0.001 205 Pharmacyf 206 207 208 2090.26 (0.17) 2100.14 (0.09) 211<0.001 212 Childcare facilityg 213 214 215 2160.35 (0.08) 2170.34 (0.11) 2180.017 219 Healthcare facilityh 220 221 222 2230.11 (0.05) 2240.04 (0.05) 225<0.001 226 Grocery storef 227 228 229 2300.27 (0.11) 2310.19 (0.11) 232<0.001 233 Primary schoolg 234 235 236 2370.05 (0.05) 2380.27 (0.10) 239<0.001 240All rights reserved. No reuse allowed without permission. 241(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 242The copyright holder for this preprint243this version posted June 12, 2023. 244; 245https://doi.org/10.1101/2023.06.09.23291209246doi: 247medRxiv preprint 248249250===== PAGE 14 (score 18, numeric-paren 12, table-words 0, var-words 3) =====251252 Neighbourhood parkf 253 254 255 2560.25 (0.10) 2570.23 (0.15) 2580.015 259 Public transitf 260 261 262 2630.09 (0.03) 2640.04 (0.02) 265<0.001 266Gentrification measure 267 268 Gentrifiable in 2006i 2690.00 (0.00) 2700.62 (0.49) 271<0.001 2720.00 (0.00) 2730.60 (0.49) 274<0.001 275 Gentrified in 2016i 2760.00 (0.00) 2770.18 (0.39) 278<0.001 2790.00 (0.00) 2800.19 (0.40) 281<0.001 282aNaturalized Canadian citizens, permanent residents, temporary residents 283bFirst Nations, Métis, Inuk and/or Registered or Treaty Indians and/or membership in a First Nation or Indian band 284cPersons, other than Indigenous persons, who are non-Caucasian 285dNo certificate, diploma, or degree 286eWithin driving distance of 10 km 287fWithin walking distance of 1 km 288gWithin walking distance of 1.5 km 289hWithin driving distance of 3 km 290iUsing the Grube-Cavers indicator 291 292All rights reserved. No reuse allowed without permission. 293(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 294The copyright holder for this preprint295this version posted June 12, 2023. 296; 297https://doi.org/10.1101/2023.06.09.23291209298doi: 299medRxiv preprint 300301302===== PAGE 15 (score 15, numeric-paren 0, table-words 2, var-words 2) =====303304Table 2: Results of variations of controlled interrupted time series adjusting for housing and 305neighborhood attributes, without and with proximity measures (models 1 and 2 vs. 3 and 4, 306respectively), and restricted to pre-March 2020 sales (models 2 and 4). 307 308Model 1a 309(95% CI) 310Model 2b 311(95% CI) 312Model 3c 313(95% CI) 314Model 4d 315(95% CI) 316Intercept ( 317$44,602 318($34,419, $57,797) 319$44,238 320($32,288, $60,611) 321$93,330 322($64,537, $134,968) 323$70,231 324($44,636, $110,504) 325Time ( 3261.002 327(1.001, 1.002) 3281.001 329(1.000, 1.002) 3301.003 331(1.002, 1.004) 3321.002 333(1.001, 1.004) 334Group ( 3350.732 336(0.701, 0.765) 3370.761 338(0.725, 0.798) 3390.722 340(0.680, 0.766) 3410.716 342(0.668, 0.766) 343Group*Time ( 3440.998 345(0.997, 1.000) 3460.999 347(0.998, 1.000) 3480.997 349(0.995, 0.998) 3500.997 351(0.996, 0.999) 352Level ( 3531.077 354(1.054, 1.101) 3551.001 356(0.976, 1.027) 3571.045 358(1.015, 1.076) 3590.986 360(0.953, 1.021) 361Trend ( 3621.003 363(1.002, 1.004) 3641.007 365(1.005, 1.008) 3661.000 367(0.999, 1.001) 3681.003 369(1.001, 1.004) 370SCS level ( 3710.948 372(0.912, 0.986) 3731.000 374(0.955, 1.047) 3750.967 376(0.925, 1.011) 3770.997 378(0.945, 1.051) 379SCS trend ( 3801.006 381(1.004, 1.007) 3821.003 383(1.000, 1.005) 3841.009 385(1.007, 1.010) 3861.008 387(1.005, 1.010) 388Pseudo-R2 3890.837 3900.842 3910.851 3920.857 393n 3946,903 3954,801 3964,815 3973,313 398aControlling for housing and neighbourhood attributes with spatio-temporal price lag, excluding proximity scores 399bControlling for housing and neighbourhood attributes with spatio-temporal price lag; restricted to sales pre-400March 2020, excluding proximity scores 401cControlling for housing and neighbourhood attributes with spatio-temporal price lag, including proximity scores 402dControlling for housing and neighbourhood attributes with spatio-temporal price lag; restricted to sales pre-403March 2020, including proximity scores 404bold indicates statistical significance 405 406 407 408All rights reserved. No reuse allowed without permission. 409(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 410The copyright holder for this preprint411this version posted June 12, 2023. 412; 413https://doi.org/10.1101/2023.06.09.23291209414doi: 415medRxiv preprint 416417418===== PAGE 16 (score 2, numeric-paren 0, table-words 0, var-words 2) =====419420 421 422 423 424Figure 2: Comparison of price in treated vs. control sales pre- vs. post-intervention implementation 425controlling for house and neighborhood amenities, with and without proximity scores in models 426 427All rights reserved. No reuse allowed without permission. 428(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 429The copyright holder for this preprint430this version posted June 12, 2023. 431; 432https://doi.org/10.1101/2023.06.09.23291209433doi: 434medRxiv preprint 435