从港口到价格：全球供应链中断的通货膨胀效应

From Ports to Prices:The Inflationary Effects of Yang Jiao, Ting Lan, Yang Liu, and Xinrui Zhou WP/26/26 IMF Working Papersdescribe research inprogress by the author(s) and are published toelicit comments and to encourage debate.The views expressed in IMF Working Papers are 2026FEB IMF Working PaperResearch Department Authorized for distribution by Rudolfs Bems IMF Working Papersdescribe research in progress by the author(s) and are published to elicitcomments and to encourage debate.The views expressed in IMF Working Papers are those of the ABSTRACT:This paper examines the inflationary effects of shipping delays.We construct a novel measure ofport-to-port shipping time using real-time AIS maritime data and link itwith granular port-level trade and item-level price data. We document substantial heterogeneity in goods imports across ports and regions, variation inexposure to delays, and aggregate price responses to congestion shocks. Exploiting cross-product variations in RECOMMENDED CITATION:Jiao, Yang, Ting Lan, Yang Liu, and Xinrui Zhou, “From Ports to Prices: TheInflationary Effects of Global Supply Chain Disruptions,” IMF Working Paper 26/26. * The authors would like to thank Yuning Xu for excellent research assistance, Camelia Minoiu, and 2026 AEAAnnual Meeting participants for their valuable comments. Yang Jiao acknowledges financial support bySingapore Ministry of Education Tier 1 Academic Research Fund (24-SOE-SMU-017). Yang Jiao: Singapore WORKING PAPERS From Ports to Prices:The Inflationary Effects of Global Prepared by Yang Jiao, Ting Lan, Yang Liu, and Xinrui Zhou 1Introduction Global supply chains have made the world more interconnected than ever before, providinghouseholds and firms unprecedented access to goods, resources, and markets. However, thisinterconnectedness also creates vulnerabilities, as disruptions in one part of the value chain canquickly ripple across countries and sectors. Events such as natural disasters, trade wars, militaryconflicts, labor strikes, and pandemics can severely hinder the smooth flow of goods, resulting Shipping delays, a critical disruption in global supply chains, can have widespread economicimpacts, as ocean shipping is the primary mode of transportation for about 80% of global trade.Using a novel measure of shipping delays linked to granular port-level trade and item-level price 2Data Our analysis draws on three main data sources: granular port-level trade data from the USATrade Online database, maritime traffic data derived from the Automatic Identification System 2.1Granular Level Import Data The USA Trade Online database provides monthly import values at the HS 6-digit level for all USports disaggregated by transportation mode. For our analysis, we focus on 93 US mainland sea- 2.2Maritime Traffic Data The real-time maritime traffic data are drawn from the AIS. Transponders—mandatory for nearlyall cargo vessels—broadcast high-frequency information on vessel type, location, speed, and draught, which is collected by terrestrial stations and satellites to enable global tracking of shipmovements. Building on the machine-learning–based methodology of Cerdeiro et al. (2020),which identifies port boundaries and reconstructs port-to-port voyages, we construct port-to- port travel time measures for all US ports over the period January 2017 to July 2023.2 For each vessel arriving at a US port, we measure shipping time from its previous foreignport of call through its departure from the port. We then construct monthly port-level indicatorsof shipping delays by aggregating vessel-level shipping times to the port–month level using a 2.3Domestic Price Data Domestic price data come from the BLS Consumer Price Index at the item level. Our analysisincludes 99 items at monthly frequency,3covering household consumption categories such as food and beverages, apparel and footwear, recreational goods, household equipment andfurnishings, motor vehicles, medical care goods, education and communication products, and 3Stylized Facts By combining granular port-level import data with real-time maritime traffic data and detaileddomestic price data, we document three stylized facts. First, import shares vary substantially across regions and major ports, both within and acrossproduct categories, providing the basis for our empirical strategy to identify the causal effectof shipping delays on US CPI. For each product category, we calculate how import shares aredistributed across ports. Within the same product category, import shares differ substantiallyacross major ports—including those located in the same region. For example, in the case of a much more dispersed import pattern, with 36% entering through Los Angeles, 18% throughNewark, 11% through Savannah, 5% through Long Beach, and 5% through Charleston. Second, shipping delays vary widely in both intensity and timing across ports. Figure 1 plotsthe average shipping time for vessels arriving at the ma