Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
To explore the relationship between spatial location and quality differentiation, we build a dataset of over 30,000 restaurants rated by TripAdvisor, across large UK cities. We document several new stylized facts. Top-rated restaurants tend to be spatially more concentrated around the city center than bottom-rated ones. Whereas top-rated restaurants tend to locate close to other top restaurants, bottom-rated restaurants tend to locate away from each other and closer to top-rated ones. Our theoretical model can explain the main features of observed spatial patterns. Consistently with the predictions of our model, our regression analysis finds that an increase in the population density in the city center is associated with a decrease in the spatial dispersion of both top and bottom restaurants, this reduction being larger in magnitude for top ones. Also, a larger quality difference between top and bottom restaurants increases both the absolute and relative dispersion of top restaurants.