Looking At The Effects Of Steel Anti-Dumping Duty Orders Using Advanced Analytics And Large Data Sets (Part 1)

NU Borders is a Washington DC and Boston-based tech company that focuses on analyzing large amounts of trade data for commercial and Federal Government customers to identify trends, anomalies, investigative targets and more. Our Business Intelligence for Trade Enforcement (BITE) platform is used by multiple customers to analyze data, match names on screening lists, perform deep-dive data science and perform compliance-related checks on import/export shipments.

This blog post is written in two parts, with Part 1 focusing on U.S. trade data and a statistical analysis that reflects trade trends and patterns after an anti-dumping duty order for steel flanges was enacted in March of 2018, particularly for trade from China and India. Part 2, to be published soon, will focus on the specifics around potential transhipment schemes that our team found in the data. This is follow-up work that the BITE team initially performed in 2019, and is now updated with new U.S. import data.

The anti-dumping order was designed due to certain countries using unfair trade practices to support their industrial and commercial base. These unfair trade practices include government subsidies for energy, labor compensation, and/or tax breaks, which artificially lower the costs for manufacturing commodities. When these commodities are imported into the United States at Less Than Fair Value (LTFV), U.S. competitors suffer due to the dumping of the commodities into the U.S. market and are often forced out of business. In order to level the playing field, many governments – including the U.S. – impose Antidumping and Countervailing Duties (AD/CVD) orders which impose higher rates of duty on violating countries and companies. Exporters from those countries frequently transship commodities through countries without AD/CVD orders and claim those transshipment points as country of origin.

China has a history of flooding the market with products and forcing out competition due to their ability for low overhead. During 2018, the U.S. published multiple sanctions packages against China due to the steel flange product. Using the BITE platform and almost 500 million trade records (primarily manifest data), the team identified multiple exports from various countries to see if they had any relation to Chinese companies after the anti-dumping order was enacted.

The manifest data used in this analysis includes core data elements for each transaction, such as the method of transport, shipper, consignee (importer), country of origin and receipt, brand, product description, harmonized system (HS) code and shipment dollar value. HS codes are a numerical system developed and managed by the World Customs Organization and used to classify trade products and goods on a global level. The HS codes are consistent internationally up to the sixth digit.

The HS codes listed in the antidumping act were 7307.21.1000 and 7307.21.5000. These HS codes were queried in the Chinese, Indonesian, Indian, Malaysian, and U.S. manifest datasets. With guidance from NU Borders federal border security and law enforcement subject matter experts, the team of data scientists started to focus on the nexus of trade between Vietnam, Indonesia, India and Malaysia, and the connection to Chinese companies that would use ports in these countries to circumvent the ADD orders in the U.S. By looking into the manifest data, the team hoped to identify a connection between companies that appeared in U.S. data as well as other Asian countries' data sets. The BITE-ingested U.S. import manifest data consists of over seventy-six million records (or unique trade transactions).

After filtering for HS code 730721 for all data in the system, approximately twenty-three thousand records remain, with the top country importers shown below in Figure 1:

Fig. 1 - Top Countries US Imports 730721

Fig 2. : Total shipments with HS code 730721

Further filtering to show shipments prior to the ADD order are shown below in Figure 3 (Notice Vietnam is low on the list):

Fig 3: Shipments prior to ADD order enacted in 2018

In Figure 4 (below), that reflects top countries importing steel flanges into the U.S. after the AD/CVD order, notice that Vietnam has increased its exports into the U.S. as well as Korea, Romania and Singapore

Fig. 4: Shipments by country into the U.S. after the ADD order was enacted.

Part 1 of our analysis provides us with enough information to conclude that further investigation of the data is needed. We now know that there was a spike in steel flange shipments from certain countries, after the enactment of the anti-dumping duty order, with a decrease in imports from China and India. We will begin to explore these spikes in more detail, including going down to the shipment / entity level to identify trends and reasons for the spikes and troughs.

This analysis was led by Kathryn Williams, a DC-based data scientist with NU Borders. For further information on any of the content, please reach out to bite@nuborders.com

‍