A Data-Driven Framework for Tourism Accommodation Policy: Managing Neighbourhood Level Visitor Pressure through Short-Term Rental Analytics

Abstract

Short-term rentals shape where visitors stay, how accommodation demand is distributed across neighbourhoods, and how tourism pressure is dispersed within major cities. Yet policy responses are still often grounded in platform snapshots, citywide totals, or ex post housing arguments, leaving destination managers with limited neighbourhood-level tools for accommo-dation governance. This paper develops a data-driven framework for tourism accommodation policy based on administrative registry data and illustrates it with the New York City short-term rental registration-and-listing dataset reported in the city’s FY25 registration materials. The framework proceeds in four stages. First, it distinguishes legal market entry from observed market activation by linking active registrations to advertised listings. Second, it constructs neighbourhood-level measures of accommodation scale, intensity, platform dependence, and renewal vulnerability. Third, it combines these dimensions into a tourism accommodation pressure score designed for destination management rather than generic compliance monitoring. Fourth, it evaluates alternative areaprioritisation rules under fixed policy capacity. In the case study, the legal market contains 3,194 registered active listings linked to 3,073 visible listings, indicating that 22.6% of active registrations were not connectedto a visible listing in the published data. Spatial concentration is substantial: the Gini coefficient for advertised listings across ZIP codes is 0.560, and the ten largest ZIP-code markets account for 30.0% of attached listings. Platform dependence is also pronounced, with 95.2% of attached listings attributable to Airbnb. Policy simulation further shows that prioritisation by the composite pressure score captures 49.7% of registrations expiring within 180 days when 20 ZIP codes can be targeted, compared with 43.4% under volume-only targeting and 18.6% under random selection. The findings show how admin-istrative short-term rental records can be translated into a practical tourism policy instrument for balancing accommodation supply, neighbourhood stability, and selective intervention.