Using Input-Output Networks to Study Industry Linkages and Stock Lead-Lag
Summary
The document uses national input-output tables to map intermediate flows between industries and represent those relationships as networks. It compares China across several historical snapshots with the United States in a later snapshot, using network centrality to examine industries’ positions in the production chain and how the structure changes. The account reports that manufacturing industries are prominent in China’s network, while several large service and property-related sectors are also central. It distinguishes centrality from economic influence: real estate ranks highly in centrality, but construction, chemicals, and metal processing show stronger demand-pushing effects. In the US comparison, central industries are mostly in services.
The document then tests whether shocks transmitted through supply chains could create lead-lag opportunities in related stocks. It reports significant in-sample lead-lag behavior among Chinese industry indices, but finds little connection to actual supply-chain links and unstable behavior over time. Out-of-sample strategy results are described as poor. These findings limit the investment case: network structure can describe economic relationships, but the reported evidence does not support a reliable trading strategy based on those links.
Key ideas
- Input-output tables can be turned into networks that describe intermediate flows and industry positions.
- Network centrality and an industry’s ability to stimulate demand elsewhere are distinct measures.
- The reported industry network structures differ between China and the United States.
- Industry stock indices showed in-sample lead-lag effects, but these effects were not closely tied to supply-chain links.
- The reported lead-lag relationships were unstable, and strategies based on them performed poorly out of sample.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.