Author(s) :   Komal Sharma, Dr. Naveen Kumar

Abstract : Common belief proposes that individuals purchasing homes from outside the local area typically incur a premium. According to conventional contract theory, this premium may result from the elevated expenses associated with gathering information. On the other hand, behavioural economists contend that the premium is attributed to anchoring biases in the buyers' perceptions of information. While both theories endorse the notion of a price premium, conflicting empirical evidence exists. This study reexamines this puzzle and conducts a rigorous examination of the two alternative hypotheses. We utilize a substantial housing transaction dataset from India available online to shed light on this issue. A cutting-edge machine-learning algorithm, incorporating the latest advancements in natural language processing for multiple languages, has been devised to identify non-local Mainland Chinese buyers and sellers. Employing the repeat-sales method to mitigate omitted variable biases, it is observed that non-local buyers tend to purchase at higher prices, while non-local sellers engage in transactions at lower prices compared to their local counterparts. Leveraging a policy change in transaction tax exclusively targeting non-local buyers as a quasi-experiment, and using local buyers as counterfactuals, our findings reveal a shift in the non-local price premium to a discount post-policy intervention. This outcome suggests that the dominance of anchoring biases hypothesis is evident.

Keywords: Anchoring bias, Machine learning, Housing transaction.

DOI : 10.61161/ijarcsms.v11i8.10

Pages : 34-40

*Authors are invited to submit papers through E-mail at editor.ijarcsms@gmail.com