Connected for Better or Worse? The Role of Production Networks in Financial Crises 
This paper argues that a country’s production network can play a key role in exacerbating or mitigating a Sudden Stop—an abrupt halt of capital inflows to that country. It also contests that differences in the structure of production networks help explain why crises are systematically more severe in emerging economies than in advanced economies. The authors develop a simple two-sector small open economy model in which a collateral constraint, Fisherian deflation, and input–output linkages jointly determine the response of tradables consumption and relative prices to tradables productivity shocks. They then use the results from that model and input–output data from the Organisation for Economic Co-operation and Development in a three-sector quantitative model to assess the degree to which differences in the structure of production networks explain the difference in Sudden Stop severity between emerging and advanced economies.
Key Findings
- The structure of production networks in emerging economies is often commodity intensive and characterized by weak intersectoral linkages. By contrast, advanced economy production networks tend to feature nontradables that supply inputs to all other sectors of the economy.
- Economies with weak intersectoral linkages and commodity-intensive production networks are more sensitive to Sudden Stops than economies in which nontradables supply inputs to and use inputs from the tradable sector.
Implications
The paper’s results highlight that role that the structure of an economy’s production network plays in affecting the likelihood of Sudden Stops occurring and the severity of those that do occur. The findings also show that the effectiveness of macroprudential and industrial policies may depend on an economy’s underlying production architecture.
Abstract
We study how production networks shape the severity of Sudden Stops. We build a small open economy model with collateral constraints and input–output linkages, derive a sufficient statistic that maps network structure onto the amplification of tradable shocks, and show that a planner optimally introduces sectoral wedges to reduce amplification. Using OECD input–output data and Sudden Stop episodes, we document systematic network differences between emerging and advanced economies and show they predict crisis severity. A calibrated three-sector DSGE model disciplined by these differences reveals that endowing an advanced economy with an emerging-market production network moves most of the way toward the observed emerging–advanced Sudden Stop gap.