ABSTRACT
The Trump administration introduced 25% tariffs on steel imports in March 2018. To minimize the adverse effects of these tariffs on downstream US producers who import steel products, the administration simultaneously introduced exclusion requests for tariff exemptions. In this paper, we investigate whether companies from states where Trump won the 2016 presidential election were more likely to receive tariff exemptions than were firms that applied for exemptions in non-Trump states. Our estimation result suggests that firms located in Trump states were more likely to be granted exemptions. In addition, firms with lower sales-to-employment ratios, a signal of lower efficiency, were more likely to have exclusion requests approved if they were from states where Trump won the election.
Acknowledgments
This paper is a revised version of KIEP working paper 20-01. We thank Kyong Hyun Koo, Unjung Hwang, Chankwon Bae and seminar participants at KIEP for their helpful comments and suggestions. All remaining errors are our own.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1 It also introduced 10% tariffs on aluminium imports but our analysis focuses only on the steel tariffs.
2 There is an extensive list of literature on the political economy of trade policy. A recent review is provided by McLaren (Citation2016).
3 There is an issue concerning the location of firms. Some firms in our sample operate in multiple states. These firms usually have headquarters or head-offices on the address provided in the data but also have extra production facilities in other states. In this case the location (at the state level) of a firm is not clear. For a robustness check we use a sample only including firms that have more than half of their employees located on the address or that have major production facilities in the same state as their headquarters are located in. The result is almost identical to the baseline result using the whole sample. Detailed explanation and estimation result are available upon request.
4 The results from a simple mean difference test (T-test) also reveal that SER and firm’s age are not statistically different by the trump-win dummy at conventional significance levels (1%, 5%, and 10%).
5 For the baseline model, ln(age), obj, HTS code and country origin of imports are included.
6 We identify the 12 most requested HTS codes at the 4-digit level (covering about 90% of products) and create 12 HTS code dummies accordingly. Each of these dummies takes the value 1 if a firm requested a product that belongs to the corresponding HTS code and 0 otherwise. The country origin of imports dummy is defined similarly.
7 The result is also robust to adding the industry level SER constructed using BLS data. Additionally, age has a negative sign, which means that young firms are favoured. One way to interpret this is that the Trump administration favours more vulnerable firms (young firms) to prevent severe economic damages.
8 The Rust Belt includes Wisconsin, Ohio, Indiana, Michigan, Pennsylvania, West Virginia, and Illinois.
9 We also use the average unemployment rate from 2013 to 2017 (five years) instead of the unemployment rate in 2017. The implications are the same.
10 We use the 10-year average of the industry share from 2008 to 2017.