A Big Data Approach to Housing Benefit
Tue, 10 Jan 2017 14:02:00 GMT
Quantitative Analysis of Business, Economics and Finance (QABEF) Research Group Seminar
Speaker: Dr Charles Rahal – University of Oxford
Wed, 18 Jan 2017, 13:15, BSG/20
All staff and students welcome
A Big Data Approach to Housing Benefit
We detail a Big Data exercise that scrapes and parses 658,612 listings on a leading real estate portal across the end of 2016, and develop tools to geocode and spatially join each property to its corresponding Broader Rental Market Area. We pursue two avenues of analysis. First, we calculate how affordable the housing stock is to housing benefit recipients. The main observation here is strong regional variation driven by heterogeneous rental market growth rates, exacerbated by the benefit cap pricing mechanism. We conclude this section with simulations which project the distributions of possible time paths until the benefit caps are due to be revised again (2020). In the second part of the work, we show that 6.93% of listings included a statement such as “No DSS” which refused housing benefit recipients as applicants at the first stage of the rental process. Rates of refusal vary strongly between (from 2.47% in London to 16.76% in Wales) and within statistical regions. Refusal to rent is significantly more prevalent among less expensive properties and has a strong positive correlation with other types of landlord refusal (e.g. against pet-owners). We also find significant variation across letting agencies: refusal rates reach over 93% where agencies allow property owners to write their own advertisements.
