Real Estate Pictures: The Role of Furniture Preferences in Subjective Valuation

Abstract

In real estate pictures, home staging, which is the temporary display of high-end furniture, is meant to positively impact the sales process for the involved parties: the broker, the seller and also the buyer. In this article we focus on the impact of home staging on the price aspect in this process. In our experimental design to isolate the effect of furniture on valuation, we show pictures with furniture to one group and those without furniture to the other group. The main task for both groups of students is to estimate the market value of an empty apartment. Additionally, they must state their preferences for furniture in an apartment. On aggregate, we do not find significant differences in market value estimates between the two conditions, which is in line with the recent literature. At subsets based on individually indicated preferences for furniture, we find two main differences: those in the group with lower indicated preferences for furniture do not adjust their value estimates in a second estimation task, where the presence of furniture is changed, while those in the second group adjust their value estimates in the second estimate, where the adjustments are significant with a magnitude of approximately 11% based on their preferred condition. We conclude that if a client indicates a high preference for furniture, then the presented furniture affects his or her perception of the property. Additionally, we monitor the dwell time, which is significantly longer for the furnished condition and is an indication of information reception.

Keywords:

• Real estate valuation
• furniture preference
• home staging

Notes

1 For a detailed review of the role of behavioural factors in real estate economics, see (Salzman & Zwinkels, 2017).

2 This floor plan includes or excludes furniture in the respective treatment.

3 A two-sided t-test results in a p-value of 0.9222 evidencing to no difference in means for the first MV estimation task. The p-value for the second task is 0.0018 and thus shows significantly different mean MV estimates. For the mean values see Table D1 in Appendix D.

4 The preference for furniture has a mean of 2.8, which is evidence of the negligible importance of furniture regarding home preferences. See Figure E1 in Appendix E for a histogram of the six means in the objective Likert scale rating and Appendix B for the summary statistics of all the preferences). The other 11 of the 12 variables have mean values between 6.8 and 8.5 on the 10-point Likert scale. A regression analysis of the effect of the 12 Likert-scale variables on the MV estimates in the first and second task shows no significant influence of any of the variables.

5 A power analysis using the "powerMediation” package with its “SSizeLogisticBin” function is used to calculate the sample size. The minimum sample size is 147 when assuming no effect and expecting a shift of 10% of participants. The sample proportion is set to 0.5 with an alpha of 0.05 and power of 0.8.

6 Additionally, the subjective “light” variable shows a very significant impact in the logit model.

7 We assume more interest in the real estate topic due to the fact that the students have made a conscious decision to study real estate and facility management. We also tested the difference in mean estimates for the group of real estate and facility management students and find no significant difference (results not reported due to small sample sizes: N = 19 for group U and N = 37 for group F).

Additional information

Funding

This work was supported by Tiroler Wissenschaftsförderung (grant number: 18668) and OeNB Jubiläumsfonds (grant number: 18530).