321 bottleneck products threaten Japanese manufacturing output
A Bank of Japan study analyzing 105,481 transaction relationships across 15,022 firms reveals that 321 bottleneck products pose severe macroeconomic risks. Inventory buffers delay major manufacturing output declines by four months during 50 percent supply disruptions.
Inside the two-thousand-supplier pyramid
Researchers mapped 1,105,563 firm-product transactions among 13,013 manufacturing firms by integrating Teikoku Databank share data with official business surveys.
Applying authority centrality identified 197 downstream focal firms at the top of production hierarchies.
The automotive sector features exceptional scale, averaging nearly 2,000 tier suppliers per focal firm across up to six transactional tiers.
Demand shifts from car manufacturers exert a far stronger pull on individual suppliers than focal firms in other industries.
Construction machinery also exhibits extended chains of 3,000 to 4,000 suppliers, whereas electrical and IT networks remain smaller domestically due to international horizontal integration.
Four months on the inventory buffer
The study identifies 321 bottleneck products across 286 firms based on high betweenness centrality, market shares exceeding 25 percent, and high price heterogeneity.
Concentrated in electronics, machinery, and chemicals, these inputs form critical chokepoints.
Simulating a 50 percent supply reduction reveals that inventory buffers absorb shocks for four months before aggregate output drops sharply.
Transportation equipment suffers the most severe indirect losses due to its broad component reliance.
Post-2010 inventory buildups noticeably improve resilience relative to mid-2000s lows.
Granular precision with glaring blind spots
By mapping firm-product networks, the study provides a groundbreaking look at Japanese supply chain vulnerabilities.
However, omitting wholesale intermediaries and international trade flows creates major blind spots that underestimate cross-border risks.
Central banks must integrate global trade data to make these models operationally useful.