Mapping the Midweek Mountain: The New Geography of Hybrid Work

Battle Count: 77

This paper is fundamentally about labor economics, urban geography, and behavioral work patterns. It has no direct relevance to quantitative trading strategies, market microstructure, or financial modeling. Indirect relevance might exist for real estate investment trusts (REITs) analysis or labor market indicators, but the paper does not engage with financial markets or trading.

The analytical methodology itself (descriptive statistics, threshold-based classification, time allocation computation) is straightforward. However, the primary barrier is data access: obtaining and processing 41 billion geolocation pings from a commercial provider (Veraset), matching them to building polygons, cross-referencing SEC filings, and integrating SafeGraph POI data requires significant computational infrastructure and commercial data licenses. The geospatial matching and filtering pipeline is non-trivial.

The core dataset (Veraset geolocation data) is commercially licensed and not publicly available. The methodology for identifying employees, inferring home locations, and classifying WFO/WFH days is described in detail, but replication requires access to proprietary mobile data. SafeGraph POI data and SEC 10-K filings are publicly available. The final sample sizes (3,237 in 2019; 6,338 in 2022; 3,092 in 2023) are small relative to the raw data, and the filtering criteria are specific.