Nowcasting Gentrification Using Airbnb Data

There is a rumbling debate over the impact of gentrification: presumed gentrifiers have been target protests and attacks in some cities, while they welcome as generators new jobs taxes others. Census data fails to measure neighborhood change real-time since it usually updated every ten years. This work shows that Airbnb can be used quantify track changes. Specifically, we consider both structured (e.g., number listings, reviews, listing information) unstructured user-generated reviews processed with natural language processing machine learning algorithms) for three major New York City (US), Los Angeles Greater London (UK). We find (especially its part) appears nowcast gentrification, measured changes housing affordability demographics. Overall, our results suggest from online platforms create socioeconomic indices complement traditional measures are less granular, not real-time, more costly obtain.