# RECORD e1b44941a0d59a0ba33dc85660a7305568feae7f # TITLE: Assessing the economic premium of additional elevator: PSM hedonic analysis in Beijing, China # YEAR: 2024.0 VENUE: International Journal of Strategic Property Management DOI: 10.3846/ijspm.2024.22606 # AUTHORS: Zhao-Wen Zhang; Yihua Mao; Yueyao Shui; Ru-Yun Deng # PAGES: 12 CANDIDATE TABLE PAGES (1-based): [4, 6, 7, 9] # ABSTRACT: Although many Chinese cities have made remarkable progress in promoting age-friendly housings, few empirical studies related to the premium price of housings with additional elevators installed have been reported. In this study, we constructed propensity score matching–hedonic price models as well as quantile–hedonic models to assess the impact of adding elevators on housing prices. This study concludes: (1) The impact of additional elevators on housing prices varies depending on the floor level. (2) Single-facing and older housings are more likely to be negatively affected by the addition of elevators, and this interaction increases with floor level. (3) Adding an elevator reduces the price of low-priced housings on low floors and raises the price of high-priced housings on low floors, while for middle-floor and high-floor housings, the premium increases with the price of the housing. (4) Compared with the commercial community housings, the elevator premium of the affordable community housings is higher. Our findings provide an important reference for local governments to formulate appropriate charging and compensation mechanisms for adding elevators to multistory housings to promote retrofitting of aging communities. ===== PAGE 1 (score 12, numeric-paren 0, table-words 0, var-words 12) ===== This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Copyright © 2024 The Author(s). Published by Vilnius Gediminas Technical University ISSN: 1648-715X / eISSN: 1648-9179 INTERNATIONAL JOURNAL OF STRATEGIC PROPERTY MANAGEMENT INTERNATIONAL JOURNAL OF STRATEGIC PROPERTY MANAGEMENT 2024 Volume 28 Issue 6 Pages 381–392 https://doi.org/10.3846/ijspm.2024.22606 ASSESSING THE ECONOMIC PREMIUM OF ADDITIONAL ELEVATOR: PSM HEDONIC ANALYSIS IN BEIJING, CHINA Zhao ZHANG, Yihua MAO*, #, Yueyao SHUI, Ruyu DENG College of Civil Engineering and Architecture, Zhejiang University, Hangzhou, China Article History: Abstract. Although many Chinese cities have made remarkable progress in promoting age-friendly housings, few empirical studies related to the premium price of housings with additional elevators installed have been reported. In this study, we constructed propensity score matching–hedonic price models as well as quantile– hedonic models to assess the impact of adding elevators on housing prices. This study concludes: (1) The im- pact of additional elevators on housing prices varies depending on the floor level. (2) Single-facing and older housings are more likely to be negatively affected by the addition of elevators, and this interaction increases with floor level. (3) Adding an elevator reduces the price of low-priced housings on low floors and raises the price of high-priced housings on low floors, while for middle-floor and high-floor housings, the premium in- creases with the price of the housing. (4) Compared with the commercial community housings, the elevator premium of the affordable community housings is higher. Our findings provide an important reference for local governments to formulate appropriate charging and compensation mechanisms for adding elevators to multistory housings to promote retrofitting of aging communities. ■received 11 March 2024 ■accepted 3 November 2024 Keywords: age-friendly retrofit, additional elevator, housing price, floor level, affordable community. * Corresponding author. E-mail: 12012142@zju.edu.cn # Center for Balance Architecture of Zhejiang University, Hangzhou, China # Binhai Industrial Technology Research Institute of Zhejiang University, Tianjin, China 1. Introduction Compared with young and middle-aged people, the mo- bility of the elderly is limited, and the lack of elevators in multistory buildings has become a major health obstacle for the elderly (Chen et al., 2023). Therefore, elevator in- stallation is an important part of age-friendly community retrofits (Yang & Li, 2023). The Chinese government has put enormous effort into promoting and installing elevators in multistory buildings, and “installing elevators” has been in- cluded in the State Council’s government work report for five consecutive years since 2018. Though various regional governments in China have made great efforts to develop the project of adding elevators to existing residential build- ings, the actual process remains far from the policy goal. The implementation of the work of installing elevators in old neighborhoods has been hindered by reasons related to fund-raising, cost-sharing, approval procedures, opera- tion, and supervision, among others (Ning, 2014). Howev- er, according to the interviews conducted by this research group in Beijing with residents, communities, streets, and staff of the government departments in charge, the great- est difficulty in installing elevators is the development of the charging and compensation mechanism for installing elevators, which cannot be separated from the discussion on the premium for installing elevators (Liu & Sun, 2019). The hedonic price model suggests that the price of a housing is determined by a set of features of the housing, which combine to form a package of features that affect the price of the housing (Tsai, 2022). Compared with mul- tistory housings without elevators, those with elevators are favored by most prospective buyers due to their conveni- ent travel characteristics, and these buyers show a higher willingness to pay (Dai et al., 2022). The economic benefits brought by this willingness to pay make the elevator con- figuration of multistory housings one of the features de- termining the price of housings, which pushes up the price of multistory elevator houses (Ma et al., 2022). In recent years, many scholars at home and abroad have explored the impact that the addition of elevators to multistory houses has brought to the lives of residents and hous- ing prices based on the perspectives of case studies (Liu & Sun, 2019) and qualitative discussions (Zhou & Tang, 2019). However, few studies have quantitatively analyzed the economic premium of the addition of elevators based on the hedonic price model. To that end, we sought to answer the following ques- tions: First, will the provision of additional elevators change the floor structure of housing prices? In other words, com- pared with high-floor flats, can middle-floor and low-floor flats without elevators be sold at higher prices due to their ===== PAGE 4 (score 26, numeric-paren 0, table-words 2, var-words 19) ===== 384 Z. Zhang et al. Assessing the economic premium of additional elevator: PSM hedonic analysis in Beijing, China the variables such as transportation facilities, education- al facilities, living facilities, landscape and parks, and the straight-line distance to Tiananmen Square. Finally, the data were screened according to the research purpose to remove the samples with crippled or unreliable informa- tion. Finally, 17,952 valid samples were obtained, of which the number of samples without elevator was 17,095, and the number of samples with additional elevator was 857. Table 1 summarizes the definitions of all the variables as well as the results of descriptive statistics. 3.3. Research design and method First, we constructed a hedonic price model to explore the effect of the floor on the price of housings without elevators versus housings with additional elevators. After a series of preliminary tests, we found that compared with other forms of models, the logarithmic hedonic price mod- el could better explain the relationship between housing prices and explanatory variables with the best model ac- curacy (Malpezzi, 2002). This model form has been devel- oped and tested over a long time, proven to be reasonable and reliable (Wen et al., 2022). The first model was de- signed as follows: 0 1 2 ln i i j k P LF HF X = α + β + β + γ + λ + δ + ε ∑ . (1) retrofitting of elevators in old neighborhoods, Beijing has the advantages of early retrofitting work, strong govern- ment support, and a deep public understanding of the policy, which lays a solid foundation for our study, which is why we chose Beijing as a case city for the study. 3.2. Data and variable description The research data described herein were acquired from Lianjia’s second-hand real estate transaction website. As an intermediary organization with over 50% market share of second-hand housing in Beijing, Lianjia’s second-hand housing transaction records are highly comprehensive and representative. The transaction time is limited to March 2017 to March 2021 after the first additional elevators started to be put into use. The database contains infor- mation such as housing price, area, age, decoration status, floor, and orientation. According to the policy related to the additional elevator and the original elevator in Beijing, the data of the total floor height of 4–7 floors were re- tained, and based on the related literature, 1–2 floors were defined as “low floor”, 3–4 floors were defined as “middle floor”, and 5–7 floors were defined as “high floor” (Ma et al., 2022). Subsequently, the POI geographic informa- tion data were used to mine the residential neighborhood characteristics and location characteristics to determine Table 1. Variable definition and statistic description Variable Description Min. Max. Mean S.D. Housing price (P) Total transaction price of a housing (CNY) 505,000 26,000,000 4,162,290 1,773,343 Area (AR) Building area of a housing (m2) 13.51 338.86 77.97 30.62 Age (AG) Age of building (years) 2 71 23.21 9.00 Floor (F) Floor level of a housing 1 7 3.12 2.43 Low floor (LF) Whether the housing is located on a low floor (1 = yes; 0 = no) 0 1 0.16 0.37 High floor (HF) Whether the housing is located on a high floor (1 = yes; 0 = no) 0 1 0.36 0.48 Elevator (E) Whether the housing with an installed elevator (1 = yes; 0 = no) 0 1 0.05 0.21 South (S) Whether the housing with a south-facing orientation (1 = yes; 0 = no) 0 1 0.70 0.46 Decoration (D) Whether the housing well decorated (1 = yes; 0 = no) 0 1 0.41 0.49 Bedroom number (BN) Number of bedrooms 0 6 2.12 0.71 Living room number (LN) Number of living rooms 0 4 1.14 0.39 Educational facilities (EF) Primary school, secondary school, and senior high school within 1 km of a housing, each one takes 1 score and total 3 scores 0 3 2.43 0.80 Living facilities (LF) Shopping malls, 3A hospitals, post offices and banks within 1 km of a housing, each one takes 1 score and total 4 scores 0 4 3.41 0.73 Traffic facilities (TF) Bus stops and subway stations within 1 km of a housing, each one takes 1 score and total 2 scores 0 2 1.62 0.50 Landscapes & parks (LP) Landscapes and parks within 1 km of a housing, each one takes 1 score and total 2 scores 0 2 1.38 0.76 Distance to distance (D_TAM) Straight–line distance from a housing to Tiananmen (m) 289.02 111,815.46 15,335.10 8,461.19 ===== PAGE 5 (score 15, numeric-paren 0, table-words 4, var-words 7) ===== International Journal of Strategic Property Management, 2024, 28(6), 381–392 385 In the model, P is the price of the housing, LF is wheth- er the housing is located on a low floor, HF is whether the housing is located on a high floor, Xi is the other control variables, where the discrete variables are in their original form, and the continuous variables were introduced into the model using logarithmic form, α0, β1, β2, and γi are the coefficients of the characteristic variables to be estimated, λj is the year fixed effect, δk is the district fixed effect, and ε is the error term. Unlike urban public goods with complete non-exclu- sivity, the impact of urban semi-public goods with par- tial non-exclusivity, such as elevators, and electric vehicle charging piles (Gao et al., 2022) on the price of housings is more complicated. The benefits of these urban semi-public goods can only be enjoyed by residents living within spe- cific boundaries and thus are usually defined as dichoto- mous variables (Tsai, 2022). If these dichotomous variables are directly added into the OLS hedonic price model, the systematic differences between dwellings equipped with urban semi-public goods and those not equipped with urban semi-public goods are often overlooked, result- ing in a selectivity bias in the sample. Consequently, a PSM model with whether or not to equip an additional elevator as the core explanatory variable was constructed to remove the effect on the model caused by endogeneity from sample self-selection (Wang & Li, 2022). To enhance the robustness of the empirical results, we chose three matching methods: K-nearest neighbor matching (KNM), radius matching (RM), and kernel matching (KM). Further- more, we retained the maximum loss data of the samples under the three different matching methods and con- structed the model based on the matched data. 0 1 ln i i j k P E X = α + β + γ + λ + δ + ε ∑ ; (2) 0 1 2 ln i i j k P E E F X = α + β + β × + γ + λ + δ + ε ∑ . (3) E is an additional elevator variable in the model, and the interaction term E × F for floor and additional elevator were added to examine the interaction effect of floor and additional elevator on housing prices, and the other vari- ables are the same as in Model 1. In addition, to examine the interaction effect of orien- tation and additional elevator on housing prices, the sin- gle orientation feature variable and the interaction term E × Single for single orientation and additional elevator were added to the model, and the final model form was obtained as follows: 0 1 2 3 ln . i i j k P E Single E Single X = α + β + β + β × + γ + λ + δ + ε ∑ (4) Single is the single orientation variable in the model, E × Single is the interaction term between the two, and the other variables are the same as in Model 1. Likewise, the interaction term E × lnAG for age and additional elevator were added to the model to examine the interaction effect of age and additional elevator on housing prices, and the final model form was obtained as follows: 0 1 2 3 ln ln ln . i i j k P E AG E AG X = α + β + β + β × + γ + λ + δ + ε ∑ (5) The ordinary least squares (OLS)-based regression method ignores the real estate market differentiation ef- fect of the additional elevator premium. To capture the change in the additional elevator premium in the distribu- tion of housing prices in the overall market, we introduced a quantile regression model to assess the heterogeneous impact of the additional elevator on the housing prices at different price points in the form of the following model: 0 ln ( ) ( ) ( ) j i i j k P E X = α τ + β τ + γ τ + λ + δ + ε ∑ ∑ , (6) where τ is the quantile point, and the other variables are consistent with the baseline model. 4. Results and discussion 4.1. Impact of the floor level on housing prices Table 2 presents the results of OLS regression analysis for the sample without an elevator and the sample with an additional elevator. We compared the results of the re- gressions for the two sets of samples and analyzed them. Based on our analysis, we conclude that for the sample without elevator, the impact coefficient of low floors on housing prices is 0.004 and non-significant, and the impact coefficient of high floors on housing prices is −0.063 and significant at the 1% level, indicating no significant differ- ences between housing prices of middle floors and those of low floors, and that the housing prices of high floors have decreased by 6.3% compared with middle floors. This suggests that for housings that are not equipped with el- evators, residents are more concerned about the positive externalities of the convenience of travel on the middle and low floors than the negative externalities due to their uncomfortable environmental factors. They show a higher willingness to pay for middle and low-floor housings com- pared with high-floor housings. While in the analysis of Model 2 on the sample of adding elevators, the coefficient of the impact of low floors on the price of housings is −0.055 and significant at the 5% significant level, the coef- ficient of the impact of high floors on the price of hous- ings is 0.014 and non-significant, the price of housings in the middle floors and high floors is not significantly differ- ent, and the price of housings in the low floors compared with those in the middle floors decreases by 5.5%. Most additional elevator housings were built in the 20th century in older communities, where drainage and community en- vironment are poorer, and residents living there are more likely to feel the noise, mosquitoes, humidity, and air pol- lution caused by low floor (Maloutas & Botton, 2023). In this case, they express a higher willingness to pay for the additional elevator high-floor housings that compensate ===== PAGE 6 (score 79, numeric-paren 41, table-words 12, var-words 6) ===== 386 Z. Zhang et al. Assessing the economic premium of additional elevator: PSM hedonic analysis in Beijing, China for the negative externalities of accessibility. Comparing the results of the two sets of regressions reveals a differ- ence in the effect of the floor on the price of dwellings in the no-elevator sample and the additional-elevator sam- ple. Therefore, H1 is partially verified. 4.2. Impact of additional elevators and floor level on housing prices Subsequently, we applied the PSM–hedonic model to analyze the differential premium of adding an elevator to housings on different floors. After using the PSM method to deal with endogeneity due to sample self-selection, the maximum loss data of the samples under the three differ- ent matching methods were retained, and based on the matched data, the economic premium generated by the addition of elevators on different floors of housings was analyzed. Based on our results (Table 3), we conclude that the effect of adding an elevator is significant at the 1% lev- el for the full sample as well as for the middle- and high- floor samples, while the effect on the low-floor sample is not significant. Specifically, the estimated coefficient value for adding an elevator in the full sample is 0.054, a result that suggests that the addition of an elevator housing is 5.4% more expensive than a housing without an elevator. Furthermore, residents living in housings on middle and high floors are willing to pay more to live in a housing with an elevator, and the magnitude of the effect is in the order Table 2. Empirical results of without elevator sample and installed elevator sample Variable Without elevator sample Installed elevator sample Coef. p-value Coef. p-value LF 0.004 0.277 –0.055** 0.048 HF –0.063*** 0.000 0.014 0.477 lnAR 0.776*** 0.000 0.827*** 0.000 lnAG –0.040*** 0.000 –0.034* 0.083 S 0.032*** 0.003 0.027 0.282 D 0.052*** 0.000 0.064*** 0.001 BN 0.034*** 0.000 0.058*** 0.001 LN 0.014** 0.015 0.020 0.371 EF 0.007*** 0.000 0.006*** 0.000 LF 0.018*** 0.000 0.033*** 0.007 TF 0.071*** 0.000 0.055** 0.011 LP 0.023*** 0.000 0.003 0.826 lnD_TAM –0.159*** 0.000 –0.158*** 0.000 Constant 13.617*** 0.000 13.714*** 0.000 Year FE Yes Yes District FE Yes Yes Adj. R2 0.820 0.713 Observations 17095 857 Notes: ***, **, and * indicate that the estimates are significant at the 1%, 5%, and 10% levels. Table 3. Results of basic regression and interaction analysis of the effect of floor level and additional elevator on housing prices Variable Whole sample LF sample MF sample HF sample Model 2 Model 3 Model 2 Model 3 Model 2 Model 3 Model 2 Model 3 E 0.054*** –0.023*** 0.024 0.011 0.070*** –0.007*** 0.119*** –0.061*** E × F 0.025*** 0.008*** 0.019*** 0.033*** Constant 13.533*** 13.462*** 13.120*** 13.099*** 13.608*** 13.573*** 13.813*** 13.721*** Controls Yes Yes Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Yes Yes Yes District FE Yes Yes Yes Yes Yes Yes Yes Yes Adj. R2 0.814 0.814 0.794 0.794 0.818 0.818 0.824 0.825 Observations 17750 2617 8643 6124 Notes: ***, **, and * indicate that the estimates are significant at the 1%, 5%, and 10% levels. ===== PAGE 7 (score 72, numeric-paren 36, table-words 13, var-words 6) ===== International Journal of Strategic Property Management, 2024, 28(6), 381–392 387 of high floors (0.119) > middle floors (0.070). This suggests that the positive externality of convenience of travel from additional elevators gradually increases as the floor rises, and H1 is verified. However, residents living on the lower floors are unwilling to pay more for living in a housing equipped with an original elevator. This is ascribable to the fact that the additional elevator does not reduce their travel costs and even negatively affects their lives due to the impact on the ventilation and lighting of the housing. In addition, in order to further analyse the change in the economic premium of adding elevators with the floor, we added an interaction term E × F to Model 2 and plotted Figure  1. The positive coefficient of E × F shows that as the floor increases, the coefficient of adding elevators will increase, further illustrating the floor differences in the premium of adding elevators. Besides, we further confirmed the nonlinear relationship between floor and the coefficient of adding elevators through a regression with low, medium and high floor sub-samples. Specifically, the coefficient of adding elevators does not increase uniformly with the floor level. It increases slightly at low floor level, while increases dramatically at medium and high floor level. 4.3. The moderating effect of orientation and age on the premium for additional elevators The addition of an elevator, while convenient for residents to get around, also negatively impacts the safety, silence, lighting and ventilation in the housing. This negative impact is more pronounced for single-facing and older housings than for newer and multi-facing housings. On the other hand, the floor level is an important factor influ- encing the silence and lighting of a house, and the effect of adding an elevator on the silence and lighting of houses on different floors may also be different (Jia et al., 2014). To explore the impact of orientation and housing age on the premium of adding elevators for housings on differ- ent floors, we introduced the single orientation variable into the model as well as the interaction term E × Single and E × lnAG for hedonic price analysis. The results of the analysis are summarized in Table 4. The direction and sig- nificance of the effect of single orientation on the price of dwellings is not stable in the regressions for the full sam- ple as well as for the low-, middle- and high-floor samples. However, the coefficient on the interaction term E × Single and E × lnAG are significantly negative in the full sample as well as in the regressions for the low-, middle- and high-floor samples. This suggests a moderating effect of orientation and housing age on the premium for adding an elevator. In contrast, the benchmark model only cap- tures the average premium for adding an elevator. With this, H2, H3 was tested. To compare the difference between the premium for adding an elevator to a single-facing housing and a multi-facing housing on different floors, we plotted a line graph of the premium for adding an elevator (Figure 2). The coefficients of the model’s add-elevator variable are the coefficients of the premium for adding an elevator to a multi-facing housing. The coefficient on the interaction term is the coefficient on the housing price loss of add- ing an elevator to a single-facing housing compared with a multi-facing housing. The coefficient on the premium Figure 1. Relationship between additional elevator economic premium coefficients and floor level Table 4. Interaction analysis of the effect of orientation, age and additional elevator on housing prices Variable Whole sample LF sample MF sample HF sample Model 4 Model 5 Model 4 Model 5 Model 4 Model 5 Model 4 Model 5 E 0.090*** 0.133*** 0.045** 0.040* 0.094*** 0.103*** 0.156*** 0.217*** lnAG –0.065*** –0.062*** –0.055*** –0.059*** –0.083*** –0.080*** –0.083*** –0.071*** Single –0.000 –0.016 0.029*** –0.025*** E × Single –0.123*** –0.091*** –0.125*** –0.151*** E × lnAG –0.015*** –0.005** –0.013*** –0.039*** Constant 13.594*** 13.535*** 13.175*** 13.140*** 13.590*** 13.551*** 13.928*** 13.782*** Controls Yes Yes Yes Yes Yes Yes Yes Yes Year FE Yes Yes Yes Yes Yes Yes Yes Yes District FE Yes Yes Yes Yes Yes Yes Yes Yes Adj. R2 0.815 0.815 0.794 0.794 0.820 0.818 0.826 0.824 Observations 17750 2617 8643 6124 Notes: ***, **, and * indicate that the estimates are significant at the 1%, 5%, and 10% levels. ===== PAGE 8 (score 28, numeric-paren 0, table-words 9, var-words 10) ===== 388 Z. Zhang et al. Assessing the economic premium of additional elevator: PSM hedonic analysis in Beijing, China for adding an elevator to a single-facing housing is thus the sum of the coefficients on the add elevator variable and the coefficients on the interaction term in the model. For the full sample, the premium coefficient of adding an elevator for multi-directional housings is 0.090, whereas the premium coefficient of adding an elevator for single- directional housings is −0.033. That is to say, adding an elevator for single-directional housings not only does not increase the price of housings but also decreases the price of housings by 3.3%, which reflects a strong nega- tive externality of adding an elevator for single-directional housings. Regarding floor heterogeneity, the coefficient of housing price loss for adding an elevator to a single- facing house increases as the floor rises. This is related to the fact that higher floors have better light and views. Compared with the middle and lower floors with average light and views, the shading of an additional elevator has the greatest impact on the lives of residents of the higher floors who would otherwise have a wider view. For multi- facing housings, adding an elevator will increase the price of housings on the middle and upper floors. For single- facing housings, adding elevators will only increase the price of housings on the higher floors, reducing the price of housings on the middle and lower floors. This is the result of the trade-off between the negative externality of adding elevators affecting lighting and ventilation and the positive externality of reducing travel costs. Similarly, in order to compare the difference in the premium for adding an elevator to newer and older housings on different floors, we plot Figure 3. The negative coefficient on E × lnAG indicates that the premium coefficient for adding an elevator decreases as the age of the housing increases. Specifically, the coefficient of premium for adding an elevator is 0.093 for newer housings (with lnAG one standard deviation below the mean) and 0.081 for older housings (with lnAG one standard deviation above the mean) in the full sample, suggesting that negative externalities such as safety hazards and noise associated with the addition of elevators have a stronger impact on older housings with a lower level of construction. In terms of floor heterogeneity, the coefficient of E × lnAG increases with the rise of floors, which is due to in the additional elevator buildings, due to the initial design did not reserve the space for the winch, and thus the winch is often installed on the top floor, causing great noise nuisance to the residents of the upper floors. For older housings with thinner walls, the intolerable noise caused by the winch has a stronger impact on the lives of residents of the upper floors. 4.4. The socially heterogeneous impact of additional elevators on housing prices From the quantile regression results in Table 5 and Figure 4, we deduce that the impact of the variable of additional el- evator on the price of housings at different quantile points is different. We only report the coefficients of adding an elevator due to space considerations. In the first column, the impact of additional elevator on the price of housings at Q0.1 is not significant, and the semi-elasticity coeffi- cient of additional elevator on the price of housings at Q0.9 reaches the maximum value of 0.102. Furthermore, the coefficient shows a gradual increase from the low- to the high-quantile points, which indicates that buyers of high-priced housings will pay more attention to the el- evator configuration of the housing, while purchasers of low-priced housings do not pay much attention to the ad- ditional elevator, H4 was tested. As stated earlier, buyers of high-priced housings generally have higher incomes; they have higher demand for life-enhancing amenities, such as additional elevators as well as for the environment of the community in which their housings are located. The plot ratio of the high-end community is generally lower, and the negative externalities of the additional elevator affect- ing ventilation and lighting are limited, which tend to pro- duce a relatively high premium for the additional elevator. For low-priced housings, the relevant facilities and the sur- rounding environment are poor, and the community floor area ratio is high. At this time, the addition of elevators will largely affect the ventilation and lighting, thus affecting buyers’ willingness to pay. Therefore, the quantile regres- sion results show the above trend. Next, to analyze whether there is a difference in the premium for adding an elevator for low-, middle- and Figure 2. Additional elevator economic premium coefficients in single and multiple orientation samples Figure 3. Additional elevator economic premium coefficients in older and newer samples ===== PAGE 9 (score 77, numeric-paren 50, table-words 6, var-words 9) ===== International Journal of Strategic Property Management, 2024, 28(6), 381–392 389 high-floor housings at different quartiles, we re-estab- lished the quartile regression model based on the market segments of low, middle, and high floors. For low-floor housings, the premium coefficient of adding an elevator is significantly negative between Q0.1 and Q0.2, signifi- cantly positive between Q0.8 and Q0.9, and non-signifi- cant between Q0.3 and Q0.7, suggesting that adding an elevator reduces the price of low-priced housings on low floors and raises the price of high-priced housings on low floors. This is attributable to the weaker positive exter- nality of the additional elevator on low floors to facilitate travel. As the price of the housing rises, the grade of the community improves, and the floor area ratio increases, the negative externality of the additional elevator on low floors that affects the ventilation and lighting is weakened, its net externality shifts from negative to positive, and the premium coefficient also gradually increases. As for the middle- and high-floor housings, except for the premium coefficient of adding elevators on middle floors, which is not significant at Q0.1 and Q0.2, all other coefficients are significant, and the coefficients exhibit a trend of gradual increase. The elevator premium increases with the rise of the price of housings, similar to the results of the whole sample regression. Column (5) and (6) encapsulates the estimation results based on OLS regression and quantile regression using the sample data of commercial and protected housing, which show that the coefficients of additional elevator premi- ums for commercial and protected housing are 0.047 and 0.052, respectively, and both of them are significant at the 1% level. That is to say, under other conditions, the prices of commercial housing and affordable housing with elevators are respectively 4.7% and 5.2% higher than that of housing without elevators, and the premium for affordable hous- ing with elevators is slightly higher than that of commer- cial housing. The strong demand for additional elevators from mobility-impaired residents in the affordable housing community outweighs the depreciation of the investment value of additional elevators due to poor liquidity in the affordable housing market. Figure 5 show the changes in the coefficients of the premiums for adding elevators at different quartiles, H5 was tested. The results demonstrate that the premium coefficients of adding elevators for com- mercial housing housings show a gradual increase from the lower to the higher quartile points, which is not significantly different from the results of the previous analysis for the full sample. However, for affordable housing, the premium coefficient of adding elevators shows a trend of increasing Table 5. Results of quantile regression Quantile (1) (2) (3) (4) (5) (6) Whole sample LF sample MF sample HF sample AH sample CH sample OLS 0.054*** 0.024 0.070*** 0.119*** 0.052*** 0.047*** Q.0.1 –0.010 –0.099*** –0.009 0.064*** 0.019*** –0.015 Q.0.2 0.031*** –0.027* 0.014 0.083*** 0.050*** 0.027*** Q.0.3 0.038*** –0.009 0.024** 0.091*** 0.054*** 0.032*** Q.0.4 0.053*** –0.003 0.031*** 0.125*** 0.058*** 0.056*** Q.0.5 0.087*** 0.013 0.058*** 0.148*** 0.059*** 0.080*** Q.0.6 0.091*** 0.016 0.087*** 0.155*** 0.080*** 0.095*** Q.0.7 0.098*** 0.042 0.090*** 0.156*** 0.082*** 0.099*** Q.0.8 0.102*** 0.050* 0.091*** 0.158*** 0.073*** 0.103*** Q.0.9 0.102*** 0.055*** 0.093*** 0.159*** 0.062*** 0.108*** Observations 17750 2617 8643 6124 5996 11754 Notes: ***, **, and * indicate that the estimates are significant at the 1%, 5%. Figure 4. Quantile regression results of the whole sample and floor-level sub-samples Figure 5. Quantile regression results of commercial housing and affordable housing samples ===== PAGE 10 (score 12, numeric-paren 0, table-words 0, var-words 12) ===== 390 Z. Zhang et al. Assessing the economic premium of additional elevator: PSM hedonic analysis in Beijing, China and then decreasing from the low to the high tertile, which can be analyzed from the two aspects of the use value and investment value of residential housing. For low- and middle-priced housings, buyers are more concerned about the use value of housings, and the strong demand for addi- tional elevators from mobility-impaired residents in the af- fordable housing community triggers a higher willingness to pay. For middle- and high-priced housings, buyers are more concerned about the investment value of housings, and the poor liquidity in the housing market leads to the loss of the investment value of the additional elevator, which has never been noticed by scholars before (Ma et al., 2018). 5. Conclusions and policy implications With the gradual aging of the population, age-friendly multistory houses equipped with elevators are increas- ingly favored by home-buyers. In this study, we used the PSM–hedonic and quantile–hedonic models to delve into people’s willingness to pay for an additional elevator. In contrast to existing studies, while analysing the change in the premium for additional elevators with the floor, we also discussed in depth the possible negative impact of additional elevators on residents’ lives through the mod- erating effect of orientation and age of the housing (Liu & Sun, 2019). Additionally, we comprehensively examined the differences in the premiums for adding an elevator in two housing markets, commercial community housings and affordable community housings, which has never ap- peared in previous studies (Ning, 2014). The above findings reveal an important reference for promoting the project of adding elevators to exist- ing housings in old neighborhoods. First, in carrying out renovation projects to add elevators to old residential ar- eas, when faced with the demands and conflicts of owners on different floors regarding damage and benefits, com- pensation and contributions, etc., grassroots staff such as street offices and neighborhood committees can refer to the quantitative results in this article, use quantitative data as the basis for communication work, and alleviate communication conflicts. This provides a basis for ap- propriately charging and subsidising residents on differ- ent floors. Second, when setting up a funding model for elevator retrofitting, the differences in price due to floor level, orientation, age, housing price and community type should be taken into account to determine a reasonable model for additional elevator installation. Specifically, for single-oriented buildings and buildings that are too old, the relevant departments need to carefully assess the lighting, sound insulation and safety of the building before adding an elevator to avoid negative impacts. For low-value communities and affordable housing communi- ties, the relevant departments should do a good job of subsidising and publicising the work to ensure that the benefits of adding elevators are enjoyed by residents in need as much as possible. Finally, the measurement of the benefits of retrofitting elevators in old communities and the construction of a funding mechanism provide a reference for the implementation of other improvement and upgrading urban renewal projects. For other types of urban renewal projects, local authorities can also first measure their economic benefits, guide the beneficiaries of the project to make reasonable contributions, solve the financing difficulties of urban renewal projects, alleviate the financial pressure on local governments, and promote the stable and long-term development of urban renewal. However, our study still has some limitations. First, as the same housing is not usually traded repeatedly for a short period of time, we could not get the transaction prices before and after the addition of lifts to the same housing, and we did not use methods such as DID to explore the changes in the price of housings before and after the ad- dition of elevators, which is an important direction for our future research. Second, due to the limitation of technical means, we could inaccurately obtain the lighting time of each housing and only used the orientation variable to ex- plore the negative externality of adding elevators, which concluded that the study was not precise and comprehen- sive enough. Therefore, using urban 3D modeling and other means to fully understand the impact of additional elevators on residential lighting is a research topic worthy of attention by future scholars. Finally, due to privacy protection reasons, we do not have access to personal data of home-buyers. If data on the personal characteristics or demand of home- buyers is available, it may be possible to further enrich understanding of the impact of additional elevators on housing prices from the perspective of demand differences. Acknowledgements The authors would like to thank the editor and the anon- ymous reviewers for the positive and constructive com- ments and suggestions. Funding This work was supported by the Center for Balance Architecture, Zhejiang University under Grant No. 20203512-28C; The National Natural Science Founda- tion of China under Grant No. 72174178; and National Key Research and Development Program of China under Grant No. 2022YFC3601604. Author contributions Zhao Zhang was responsible for formal analysis, writing original draft and writing review & editing. Yihua Mao was responsible for conceptualization, methodology and funding acquisition. Yueyao Shui was responsible for investigation. Ruyu Deng was responsible for data curation. Disclosure statement The authors declare that they do not have any competing fi- nancial, professional, or personal interests from other parties.