Research

Exchange Rate Pass-through and Expenditure-Switching Revisited (Job Market Paper) under Census Bureau Project #2874.

Expenditure-switching is the main channel through which exchange rates transmit to the real economy. Conventional wisdom holds that this channel's strength is increasing in exchange rate pass-through into prices: assuming the import demand elasticity is independent of pass-through, larger effects of exchange rates on prices yield larger substitution of spending between domestic and foreign goods. In this paper, I show that this conventional wisdom does not hold. Using confidential US micro-data and a panel-data local projection technique, I show that quantity-exchange rate elasticities are similar across high and low pass-through environments. In essence, low pass-through is subject to a larger import demand elasticity than is high pass-through. I then propose an extension of a standard small open economy New Keynesian model by adding a layer of import buying (retail) firms, in which both exporting and importing firms are subject to price rigidities. I show empirically and theoretically that the ``import buyer rigidity" dampens overall adjustment, but less so under low pass-through because in this case the pass-through is more persistent. The model thus accounts for why the quantity-exchange rate elasticities are similar across pricing regimes. I conclude by exploring the implications of this framework for monetary and exchange rate policy, actually finding a stronger expenditure-switching channel under low pass-through.

Media: Marginal Revolution

The Macrofinancial Link Between Tariffs, Exchange Rates, and Trade (with Victor Orestes) (Preliminary- Updated Frequently)

We document how currency markets and trade flows respond to tariffs imposed by and on the US as related to other countries' macrofinancial position. We show that countries which maintain higher interest rates than the US depreciate much more strongly -- to the point of offsetting the tariffs on impact -- than their low-interest counterparts. However, these effects are not as persistent as the tariff shocks. Our results highlight a US hegemonic asymmetry: tariffs imposed on the US have little effect on currency markets, US demand for high-interest countries' goods is relatively elastic, but the latter's demand for US exports is not. Monetary policy can be an effective tool to target the exchange rate fluctuation as it has a similar incidence as tariffs. Finally, we present evidence that the interest rate analysis could draw from trade-network fundamentals. To rationalize our findings, we modify a baseline model of exchange rate determination using the interest rate as a "sufficient statistic" wedge in fundamentals. Our model indicates that the financial market imperfections we observe in data distort the global response to tariff escalation.

Research In Progress

Broken Links: The Disruptive Impact of Import Competition on Local Supply Chains and Employment (with Daron Acemoglu, David Autor, David Dorn, and Gordon Hanson) under Census Bureau Project #1684 and #3223.

Although the substantial job loss that followed from the surge of imports from China is well documented, why import flows created such large adverse effects on local labor markets is poorly understood. This paper documents the overlooked role of supply chain disruptions. We build empirical measures of local and national supply linkages by exploiting commodity-level input-output tables and the gravity-like structure of supply relationships. Consistent with standard input-output models, we find that establishments whose customers are adversely affected by Chinese import shocks see a drop in their own output and employment. The standard model further suggests that establishments whose suppliers are exposed to rising import competition stand to benefit from the availability of less expensive Chinese imports. Contrary to this prediction, we document that establishments whose suppliers compete with cheaper imported substitutes actually experience falling sales and employment effects. These “downstream” impacts appear to reflect costly disruptions to US supply chains, whose operation depends on local long-term relationships. Supporting this interpretation, we show that it is local, rather than national, downstream effects that are most consequential, and that these downstream impacts are driven by customer-supplier linkages involving significant relationship specificities rather than arms-length transactions. We conclude that domestic firms are challenged in exploiting potential gains from cheaper imported inputs when long-term supply chain relationships are threatened.

Import Buyer Price Rigidities and the Transmission of Open-Economy Shocks under Census Bureau Project #2874.

I show that average shock-specific demand elasticities are not sufficient statistics for the aggregate import response to policy shocks. Import buyer rigidities, i.e. the price stickiness of importing firms, shape firms' import response differently for each of tariff and monetary policy shocks as well as the resulting exchange rate shocks. That these rigidities are heterogeneous implies a nonlinearity in the firm distribution that changes the aggregate import response to these ``joint shocks." This paper presents a series of empirical facts regarding differences between shocks, develops a small open economy New Keynesian model with heterogeneous import buyer rigidities to theoretically establish the wedge in import responses for a range of joint shocks, and then quantifies these wedges. 

Import Buyer Price Rigidities and Firm Adjustment to Input Cost Shocks under Census Bureau Project #2874.

How do import buyer rigidities -- i.e. price stickiness of the importing firms -- change firm adjustment to imported input cost shocks? Import buyer rigidities slow imported input cost adjustment, but how do they change overall firm adjustment? In a standard model, when an imported input becomes more costly, firms substitute towards labor. Do import buyer rigidities also dampen this process, or do they generate new cost pressures that require adjustment on other margins? I answer this question using confidential Census Bureau micro-data.

Older Research