Order versus disorder – the impact on value estimates of durable consumption goods

ABSTRACT

The purpose of this article is to investigate if easily removable disorder (i.e. a negative externality displayed in pictures) influences value estimates of two durable goods (car and apartment). We examine the effect with a 2 × 2 experimental setup consisting of 137 estimates from students. While we assume an influence of the disorder in the case of the car, there must not be any value influence in the apartment task. The first estimate of the car decreases by 11.9% in the case of disorder, but the effect does not decrease in the second estimation. For the apartment, the disorder effect reduces the estimates by 13.3%. In a second estimate, the effect on value weakens but still is present. The study shows that (i) images lead to an incorrect assessment and (ii) disorder is perceived differently for the two goods.

KEYWORDS:

• Durable goods
• good appearance
• disorder value effect

JEL CLASSIFICATION:

• D12
• D46

Highlights

• We investigate the impact of disorder on value estimates of a car and an apartment.

• In a 2 × 2 experimental design, university students estimate values based on pictures.

• Apartment task: in the first estimate, the effect of disorder is overestimated by 13.3%.

• In the second apartment estimate, the effect weakens, but still is higher than removal costs.

• Car task: similar magnitude of deduction for disorder between and within treatments.

Data availability statement

The authors confirm that the data supporting the findings of this study are available within the article or its supplementary materials.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.

Notes

¹ for more details see: LimeSurveyhttp://www.limesurvey.org

² You do not get monetary compensation when viewing a property and posting your private valuation/price.

³ We are well aware that the sequence of our experiment (car – apartment – car – apartment) might have an influence on the posted values, as both goods were not estimated in isolation. This investigation is left open for upcoming research.

⁴ for an introduction to the anchoring bias see (Tversky and Kahneman 1974)

⁵ Is used for all statistical tests in this article.

⁶ Comparing the barplots with the statistics it seems that the percentage change is incorrect and does not match the depicted deviation in Figure 1. This is due to the calculation of the percentage values: The basis of the 11.9% and the 21.9% difference is the mean of the results of treatment B stage 3 and stage 4, respectively.