Firm-Driven Unilateralism
What is National Trade Estimate (NTE)?
Pursuant to the Trade Act of 1974 and the amendments made in 1984, the United States Office of the Trade Representative (USTR) has been vested with the authority to investigate and identify countries that pose significant threats to the fair trade practice with the United States. National Trade Estimate (NTE) reports are the proceedings of these investigations carried out by USTR. While the reports have been published since 1986, the Wayback Machine only has access to the old versions of USTR's website starting from 1995. That being said, this project involves 28 years of data spanning from 1995 to 2022.
NTE report features roughly 60 countries every year, with each chapter dedicated to each country, including major areas of trade-relevant concerns, e.g., import tariffs, export subsidies, investment barriers, technical barriers to trade, and so forth. These categories are not uniform across entire set of reports but differ from administration to administration or even from year to year, although some of them consistently appear throughout the reports. I recategorize them into 15 issue areas that appear most frequently throughout the reports: Import Policies; Export Subsidies; Standards, Labeling and Certification; Government Procurement; Intellectual Property Rights; Services Barriers; Investment Barriers; Anti-competitive Pratices; Technical Barriers to Trade; Sanitary and Phytosanitary Barriers; E-commerce; Barriers to Digital Trade; Agriculture; Trade Remedies; and Other Barriers. For more information on the data, refer to this hyperlinked page.
Introducing an Original Proxy, "the USTR Demands"
Wordfish method
Overall consistency
The means of the USTR demands estimate by administration does not experience much fluctuation over time, suggesting US trade politics agenda is relatively stable across administrations.
Stratification in sentiments?
It is interesting to see that the trend of the USTR demands coincides with the rise of China, and the fall of Japanese economy, while EU has been subject to ever-increasing complaints from US pharmaceuticals about drug pricing and reimbursement policies of its member states. Moreover, the long-standing battle between Boeing and Airbus only settled around 2020.
Excavating the underlying dimension with latent semantic analysis
Negative terms associated with LSS_IPR
Negative terms associated with LSS_Investment
The highlighted words in the two plots above are negative seed words that have been used to determine the sentiment score of each country-year chapter within an NTE report.
The sentiment scores under the contexts of intellectual property rights (IPR) and investment barriers show strong and negative correlation with the USTR demands estimate. Considering that IPR and investment regulations are the major concerns of US superstar firms investing in their assets abroad, this further justifies that the USTR demands are reflections of US elite firms' demands to host countries they invest in.
Main Results
PTA and USTR demands
This figure shows difference-in-differences estimates computed with PanelMatch (Imai , Wang, and Kim, 2023). The dependent variable is the USTR demands. Treatment is signing a PTA with the United States, or inclusion of the four provisions as listed in the figure. The sample is limited to WTO developing countries.
Signing a PTA with the United States reduces the USTR demands for a short run. The four provisions included in the US PTAs have similar effect of reducing the USTR demands, as they are relevant to regulatory reforms that US elite firms seek for in host countries. The effect only applies to WTO developing countries; PTAs do not make any difference in reducing USTR demands for non-developing countries.
USTR Demands and Aid
This figure plots the marginal effect of a one standard deviation increase in the USTR demands on US aid obligations, conditional on the UNGA voting distance. For democratic developing countries, at the lower levels of UNGA distance at t-1, increase in the USTR demands at t-2 leads to the increase of US aid. This can be interpreted as "policy-buying" behavior of the United States with aid commitments to its ideological friends in UNGA.
USTR Demands and Special 301
This figure plots the marginal effect of a one standard deviation increase in the USTR demands on the probability of Special 301 targeting, conditional on the UNGA voting distance. For democratic developing countries, the more distant a country gets from the United States in the UNGA voting space at t-1, the more likely it is subject to the Special 301 targeting, if the USTR demands increases at t-2. Ideologically distant states to the United States in the UNGA voting space are threatened with Special 301 if they get faced with the US demands regarding policy changes to accommodate the business interests of US elite firms.
PTA and Special 301
This figure shows the diff-in-diff estimates computed with PanelMatch.The dependent variable is the binary variable indicating whether a country is listed in any three categories of Special 301 watch list in a given year. Treatment is signing a PTA with the United States. The matched sets have been refined by balancing on the USTR demands estimate. The analysis includes a moderator variable, which indicates a country's WTO developing country status.
Signing a PTA with the United States, for WTO developing countries, reduces the likelihood of Special 301 targeting for a short period of time compared to the developing countries that do not join any US PTA. I find no effect of PTA with countries that are not defined as WTO developing countries.