Real Estate Agent Response to Hispanic/Latino Clients: Homophily, Local Demographic, and Regional Differences

How to cite

Hanson, Andrew; Hawley, Zackary. 2024. "Real Estate Agent Response to Hispanic/Latino Clients: Homophily, Local Demographic, and Regional Differences." AEA RCT Registry. AEARCTR-0014969. https://doi.org/10.1257/rct.14969-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Andrew Hanson* , Zackary Hawley

* Primary investigator

Registry ID
AEARCTR-0014969
DOI
https://doi.org/10.1257/rct.14969-1.0
First registered
Last updated
Status
Completed
Trial dates
2022-01-01 – 2024-06-15
Countries
United States of America ()
JEL codes
J15, C93, L85
Registry record
http://www.socialscienceregistry.org/trials/14969

Abstract

We conduct a large-scale correspondence experiment across 11 U.S. states to assess differential treatment of Hispanic/Latino clients by real estate agents. The experiment covers an area with a wide range of underlying Hispanic/Latino population and a subject pool of agents that is ethnically diverse.

Intervention

We test for differential treatment by real estate agents between Hispanic/Latino and white clients using a correspondence experiment. We contact agents via e-mail and monitor response to an initial inquiry about help with a housing market transaction. We then use response/non-response to test for differential treatment by a set of randomly assigned client characteristics- client race/ethnicity, gender, and side of the market.

Experimental design

There are three primary client characteristics in the experiment: ethnicity (white or Hispanic/Latino), gender (male or female), and side of market (buyer or seller). We contact real estate agents in the subject pool using e-mail correspondence from clients with one of the ethnicity/gender/side of market combinations. The outcome of interest is if a real estate agent sends a response to the initial correspondence.

The name of each client signals ethnicity and gender to the agent, with the text of correspondence and e-mail subject signaling the side of the market for an intended transaction. We create a standardized structure of correspondence to send messages to agents, and then randomly assign content of the message within that structure.

Design details

Randomization method
Randomization done in standard software program.
Randomization unit
Randomization at the individual level (real estate agent).
Clusters
0
Observations
37,862
Arms
19,020 white clients, 18,842 Hispanic clients. 18,803 male clients, 19,059 female clients. 18,891 buyer clients, 18,971 seller clients.
Primary outcomes
Real estate agent response or non-response to clients with different ethnicity, gender, and operating on different (buy/sell) side of the housing market.

Cited by

No citation data available for this registration yet.

Keywords