Restaurant Prices: Measuring Service Inflation from Scanner Data

Measuring service prices remains particularly challenging due to their inherent heterogeneity and lack of standardization, unlike goods whose prices are easily captured through scanner data. This study exploits newly available scanner data from approximately 1000 Japanese restaurants covering the period 2018–2025, which record detailed information on every order. We construct indices for prices, quantities, and sales. Although restaurant items are highly heterogeneous and less standardized than goods, price indices remain stable across different formulas. The indices exhibit only minimal biases from intratemporal (cross‐sectional) and intertemporal substitution. Moreover, scanner‐based restaurant prices align closely with CPI measures.