marcin jozwiak Warehouse in Sady, Poznań County, Poland
Warehouse in Sady, Poznań County, Poland. (Image: Marcin Jozwiak via Unsplash)

Trade insurance matters more in times of uncertainty

Evidence from the COVID-19 pandemic and the global financial crisis.

By Matthieu Crozet, Banu Demir and Beata Javorcik

Global trade fell sharply during the first few months of the COVID-19 pandemic. We argue that uncertainty played an important role in driving the trade collapse since trade flows insured by letters of credit (LCs) were more resilient than other flows. By contrast, during the global financial crisis, when the supply of LCs decreased, the opposite trend was observed: trade flows insured by LCs experienced a more severe decline than other flows.  

Cross-border trade is risky. Trade partners have to decide who will bear the risks associated with each transaction, be it a financial, commercial, political or adverse business risk. But risks can be mitigated. One option is using a trade-financed instrument called ‘letter of credit’ (LC). In an LC-financed transaction, the importer’s bank pledges to pay for the goods on the importer’s behalf, provided the exporter meets all requirements specified in the contract. The risk of non-payment for goods already shipped or of non-delivery of pre-paid goods is thereby eliminated. Yet LCs are costly as they involve bank fees and charges. About 12.5 per cent of world trade is covered by LCs1.

In our paper2, we introduce a new index that captures product-level LC-intensity and use it to compare the performance of trade in products that typically rely on LCs relative to other products during the COVID-19 crisis and the global financial crisis.

A new product-specific measure of the intensity of LC use

Based on unique data from Turkey detailing financial terms of international trade transactions, we construct a measure that captures product-specific reliance on trade insurance. Our measure (LC-Int) captures the incidence of the use of LCs across products, controlling for the characteristics of countries that export or import them. LC-Int is available for 1,196 products (defined as distinct 4-digit Harmonized System (HS) codes), of which 188 are agricultural and agri-food products.3

The figure below presents the median value of LC-Int by broad product category. The products with the highest LC-Int values include metals and minerals (such as ferrous products, tar, crude petroleum oils, pitch coke, etc.), as well as machinery and transport vehicles (such as rail locomotives). The new index also shows considerable variation across products even within the same product category.  For instance, “Silk-worm cocoons suitable for reeling" (HS5001) are among the products with the highest LC-Int value, while another product belonging to the same 2-digit HS heading “Silk waste (including cocoons unsuitable for reeling, yarn waste and gametted stock)" (HS5003) is among the products with the lowest LC-Int value. Similarly, the index value for “Live bovine animals” (HS0102) is in the top decile, while that for “Meat of bovine animals; fresh or chilled” (HS0201) is only in the 3rd decile.

Reliance on trade insurance: letters of credit intensity index

Note: The figure shows the median value of LC-Int for each sector relative to the overall median over all products.

LC-Int exhibits intuitive correlations with other product characteristics. Capital goods and consumer durables use LCs more intensively than other types of goods. LC-Int also increases with average shipment size and fraction of shipments by sea. These patterns are intuitive as trading partners have a greater incentive to insure larger shipments or shipments with longer transit times due to higher risks. By contrast, trading firms have less incentive to insure transactions with more trusted parties, as captured by the degree of relationship stickiness4.

LC-backed trade has been more resilient during the COVID-19 pandemic

Global trade fell sharply during the first few months of the COVID-19 pandemic: the value of global trade declined by 16 per cent and 18 per cent year-on-year in April and May of 2020, respectively. U.S. exports witnessed a decline of 30 per cent and 35 per cent, respectively, during the same months.

In its early months, the pandemic created heightened uncertainty, with governments around the world introducing lockdowns and imposing restrictions, thereby causing severe disruptions to economic activity. Uncertainty about which firms would survive the downturn increased the risks associated with international trade transactions. Exporters worried about not being paid for the goods they had shipped and importers feared that pre-paid imports would never be delivered. 

We examined whether products that traditionally rely more heavily on LCs exhibited greater resilience relative to other products during that time. The evolution of monthly U.S. exports, illustrated in the figure below, is consistent with our assumptions. The April/May drop in monthly exports is clearly visible, but the dip in U.S. exports of products that traditionally rely more heavily on LCs was less pronounced and rebound was faster.

The collapse of U.S. exports during the Covid-19 pandemic

Notes: The grey bars (left axis) show the monthly U.S. exports during 2019 and 2020. The blue and orange lines (right axis) show the year-to-year growth rates of U.S. monthly ex- ports of products with higher than median LC-Int and lower than median LC-Int respectively.

To formally test our hypothesis, we use monthly data on U.S. and EU-15 exports disaggregated by destination country and 4-digit HS product codes for the period April 2017 to December 2020. We test whether the trend in the export of products typically traded using LCs differed relative to that of other products exported during the pandemic. We include an extensive set of fixed effects in the estimation to absorb (i) any variation in the year-on-year growth of exports for a given country pair in a given time period (year-month), such as a slowdown in the national economy and lockdown; (ii) pre-existing trends in trade flows for a given product and country pair; (iii) change in export supply of a given product from a given origin; and (iv) product-specific seasonality. We also allow for differential monthly trends with respect to other product characteristics, such as contract intensity, average shipment size, relationship stickiness and income elasticity. 

Our estimation results confirm our assumption: exports of products that traditionally rely more heavily on LCs were more resilient during the pandemic relative to other products. Specifically, a one-standard deviation increase in LC-Int was associated with a 1.3 log-points larger increase in monthly U.S. exports during the pandemic crisis (i.e. from February to August 2020). The estimated effect is significant as it corresponds to about one-fourth of the average annual change in monthly trade flows in our data. The effect estimated for EU-15 exports is comparable to that obtained for the U.S. sample. For exports from both the U.S. and EU-15, we find that the results are more significant for exports destined for countries with a high number of COVID-19 cases.

LC-backed trade took a big hit during the global financial crisis

Between the third quarter of 2008 and the second quarter of 2009, world trade plunged – an episode referred to as the Great Trade Collapse5. The Great Trade Collapse was triggered by the global financial crisis, a shock that caused economic downturn, accompanied by reduced availability of financial instruments, including LCs. Using annual data on U.S. and EU-15 exports for the 2003-2009 period, we demonstrate that the export of products that rely more heavily on LCs experienced a more severe decline in exports to countries affected by the global financial crisis.


Times of crises, be they economic or financial, are often associated with a decrease in international trade. Product reliance on LCs has a direct impact on the resilience of trade flows. The export of products insured through LCs is more resilient during periods of increased uncertainty, in particular. By contrast, financial crises that negatively affect the supply of LCs are associated with a greater decline in trade of LC-intensive goods. These patterns highlight the importance of distinguishing between different causes of crisis-related drops in international trade to forecast future recoveries.  

  • Matthieu Crozet is Professor of Economics at the Université Paris-Saclay.  
  • Banu Demir is Assistant Professor of Economics at the Bilkent University.
  • Beata Javorcik is Chief Economist at the European Bank for Reconstruction and Development (EBRD) and Professor of Economics at Oxford.

Disclaimer: The views expressed in this article are those of the authors based on their experience and on prior research and do not necessarily reflect the views of UNIDO (read more).

Read next