Methodology to Determine the Number of Jobs in Each Industry Expected to be Harmed by the Korea FTA

Table 2.3 of the U.S. International Trade Commission’s study on the Korea FTA predicted that the U.S. trade balance would worsen in electronics, metal products, motor vehicles and parts, other transportation equipment, iron-containing metals, textiles, and apparel under the FTA, among other sectors.

Determining the number of jobs in these at-risk industries in each congressional district was accomplished through a four-step process. First, a correspondence between the sectors in the Global Trade Analysis Project (GTAP) model and sectors defined in the North American Industrial Classification System (NAICS) was determined. Second, records of companies matching the appropriate NAICS codes were obtained from the Hoover's Inc. company establishment database, including their address and, usually, the latitude and longitude of the location of the establishment. Third, for the few records where the latitude and longitude was not available, the University of Southern California's WebGIS Services was used to determine the latitude and longitude from the address information. Fourth, each establishment, represented by its latitude and longitude, was tested for inclusion in each polygon defined by the borders of each U.S. congressional district with the aid of geographic shapefiles obtained from the U.S. Census Bureau.

To create the GTAP-NAICS concordance table, first a concordance table of GTAP and Harmonized Tariff Schedule (HS) classifications was obtained from the GTAP website. Then, a HS-NAICS concordance table was obtained from the Foreign Trade Division of the U.S. Census Bureau. The 10-digit HS codes of the Census Bureau concordance table were truncated to six digits to make them compatible with the GTAP concordance table, but duplicated HS codes in the resultant table were not removed. The two concordance tables were then merged.

In many cases, one six-digit NAICS code was associated with more than one GTAP sector in the resultant concordance table. To resolve this issue and obtain a nearly one-to-one NAICS-GTAP concordance table, each NAICS codes associated with more than one GTAP sector was reassigned to the GTAP sector that was associated with the greatest number of 10-digit HS codes that were, in turn, linked to the specified NAICS code. In this way the method aimed to assure that the GTAP sector most closely related to the NAICS code was assigned to the NAICS code. In cases where there were “ties" between two GTAP sectors, the NAICS code was assigned to both GTAP sectors.

The NAICS and GTAP industrial classifications do not match perfectly. That is, products produced by a given NAICS sector may fall into more than one GTAP sector. There is more than one way to construct a NAICS-GTAP concordance table. Each method would give slightly different results for the relationship between the NAICS and GTAP classifications systems. Besides determining the relationships based on the number of tariff lines, there is also a method based on the volume of U.S. imports and exports in those tariff lines. The concordance table constructed based on trade volume is very similar to the concordance table constructed based on the number of tariff lines, with only a few NAICS codes assigned to different GTAP sectors.