Sample Report: Global Cross-Border B2C E-Commerce 2015
562FinalPaper
1. The Effect of Special 301 on US Trade Volume
Elise Hauser
Joy Zhou
Chang Liu
Introduction
In 1989 the United States Department of State, through the Office of the US Trade
Representative, began publishing the Special 301 Report. This report includes a list of countries
that have been reported by private companies as violating international intellectual property
rights laws. Countries on the Special 301 list are further classified into three separate
designations: Watch List, Priority Watch List, and Priority Foreign Country (2013 Special 301
Report).
Once a country is listed on the Special 301 list the State Dept. begins efforts to correct
issues in trade relationships with the listed countries. This should increase trade activity with
these countries, which in turn should increase total US trade volume (2013 Special 301 Report).
In this study we find a positive relationship between our Special 301 variables and the
absolute value of net exports, indicating that the Special 301 Report does increase the volume of
US international trade.
Literature Review
A recent study concerning the effect of Special 301 that we based our initial research on
is “Has Special 301 Promoted US Manufacturing Exports?” by David Riker (2012). This study
measures the impact of Special 301 designations on US export revenues and measures the
consequent impact on employment and research and development (R&D) expenditures in the US
manufacturing sector.
2. Riker found in his econometric analysis that Special 301 priority designations had a
significant, positive impact on US manufacturing export revenues: In the studied sample
manufacturing export revenues increased between $6 billion and $11 billion per year. This
impact on export revenues translated into an impact on US manufacturing employment, as well
as increasing the research and development (R&D) expenditures of the US manufacturing sector.
The study found that manufacturing employment increased by 17,000 – 35,000 each year and
R&D spending increased $172 million - $384 million per year.
The econometric model in this study is based on a gravity model of international trade,
which relates the value of trade between two countries to the product of their aggregate incomes
and to the distance between the countries. This model uses country-level panel data regarding
each export destination country as a panel. This model also uses dummy variables to measure the
effect of Special 301.
There are several differences between our study and this study. First, we focus on the
effect of Special 301 on the absolute value of US net exports, not just manufacturing exports.
Second, we use time series data over a period of 40 years, from 1973 to 2013. Additionally, we
do not use one dummy variable to indicate the presence of Special 301, but a variable indicating
the number of countries listed on Special 301, as well as variables indicating the number of
countries on the Watch List, Priority Watch List and Priority Foreign Country List each year.
Finally, we do not use panel data, but time series simple linear regression after taking the first
difference.
Hypothesis
Based on our understanding of the Special 301 process, economic theory, and the
literature, we hypothesize that the variables representing Special 301 should have a significant
3. effect on trade. The former Special 301 reports may also have an effect on trade, so we also look
at lags of Special 301. We hypothesize that Special 301 variables will have positive, statistically
significant effects on the volume of US international trade.
Further, we expect that FDI will have a strong effect on the absolute value of net
exports. Based on a study of international technology spillovers to U.S. manufacturing firms via
FDI between the years of 1987 and 1996, the size of FDI spillovers is economically important,
accounting for about 14% of productivity growth in U.S. firms (Keller, Yeaple, 2003). The
growth in U.S. firms would lead to more trade between America and foreign countries.
The dollar index, which calculates the strength of the American dollar relative to
other countries’ currencies might play a role in international trade. Additionally, we expected to
see a positive relationship between the volume of foreign trade and GDP in U.S and GDP in the
rest of the world.
Data
The explanatory variables that we use in our model are: dollar index, FDI level into the
US, FDI flow into the US, US GDP per capita, GDP per capita of the rest of the world, and
Special 301 variables.
Our dependent variable is the absolute value of US net exports. We chose to use the
absolute value of US net exports as our dependent because we are interested in the relationship
between Special 301 and the volume of US international trade. We were unable to find reliable
data for total US international trade volume, so we took a variable we had already collected, net
exports, and took the absolute value. It is important to note that this variable is a substitute for
what we are really interested in measuring, the volume of all US international trade.
4. Where available we collect quarterly data from year of 1973 to year of 2013 for each
variable. Variables that were reported annually instead of quarterly were interpolated to be
measured quarterly.
Our Special 301 variables include: a variable for the number of countries listed in Special
301, and variables for each the number of countries on each the Watch List, Priority Watch List
and Priority Foreign Country List. This data is generated annually, so we interpolate it to be
quarterly (2013 Special 301 Report).
To control for effects of other variables, we gather quarterly data from 1973 to 2013 for
dollar index from the Federal Reserve Economic Data compiled by the St. Louis Fed (FRED),
which measure the relative strength of the dollar, weighted against a broad assortment of trade
partners; FDI level and flow in US (FRED); US GDP per capita (FRED); GDP per capita of rest
of the world (World Bank). We use the world total of GDP and subtract US GDP to find the
GDP of the rest of the world.
The US government began to implement Special 301 in 1989. So, we can separate data
set into two parts: before and after 1989.
From the data, we can see that the absolute value of US exports increased greatly in the
second period with the number of Special 301 variables increasing, which means that the volume
of US international transactions increased.
The table below provides summary statistics of the data set.
5. Table 1:
Methodology
After we collected our variables we began the process of determining our model and
calculating our coefficients. The first step was to specify our model. We decided that in addition
to the variables we had collected we were also interested in their lags. We include two lags in the
model to account for effects that took a year or two years to manifest. Next, we tested each of
our variables for unit roots. Since we were dealing with time series data we were concerned that
the variables might not be stationary. To test for unit root we used the Dicky Fuller test, where
we regressed the difference of the variable of interest on the lag of itself and compared the
6. resulting t-statistic with the Dicky Fuller distribution critical values. We did find some evidence
of unit roots in several of our variables, including both measures of FDI and the variable
measuring the number of Priority Foreign Countries. To correct for unit root we used a
difference model where we modeled the difference of all of our variables.
We were also concerned about spurious relationships between some of our independent
variables and the dependent variable. Specifically, we were concerned that the volume of US
international trade was increasing over time due to other factors, some un-measurable. Likewise,
as the Special 301 list became a more institutionalized policy tool, the number of countries on
the list could grow each year for no other reason than the list becoming a more standard policy
feature. To control for the risk of a spurious regression we added a time trend. Upon further tests
of our model we found a second time trend (t2) to be highly significant, so we include two time
trends in the model as well.
Once we had created a model that included all of our differenced dependent variables
regressed on all the independent variables differenced, two lags of the differenced variables, and
two time trends, we were able to run our simple linear regression model.
Results
Below, in Table 2 we see the results from our regression.
8. l2dpfc 1 1.11255 6.75337 0.16 0.8694
l2dgdp 1 -0.04251 0.04649 -0.91 0.3622
l2dfgdp 1 0.03555 0.04722 0.75 0.4529
t 1 0.30590 0.19958 1.53 0.1278
t2 1 -0.00235** 0.00118 -1.98 0.0495
Note: *and **denote statistical significance at the 10% and 5% levels, respectively.
According to Table 2, the change in the level of current FDI has a significantly positive
relationship with the absolute value of net exports. Since the level of current FDI calculates the
cumulative quantity, it has a more significant effect on trade than the flow of current FDI. The
increasing current FDI would lead to productivity gains for domestic firms and boost the
interaction in trade. It is assumed that companies would generate various types of positive
externalities, or spillovers, to domestic firms (Keller, Yeaple, 2003). Governments all over the
world often spend large amounts of resources in order to attract multinational companies to their
country because of this positive effect of current FDI on their trade.
The current change in the number of the countries in the Special 301 Report also has a
significant association with the absolute value of net exports. The Special 301 Report is
published every year, so the firms probably refer to the latest results of the report. The former
reports have less effect on the decisions in collaborating with other foreign companies, which
leads to less significant results of the lags of Special 301 variables. Only the change in first lag of
the countries on watch list has the negative effect on the absolute value of net exports. The
current total number of the countries in Special 301 Report has a positive effect on the absolute
value of net exports. However, the number of the countries on the Watch List, Priority Watch
List and Priority Foreign Country list each have negative effects on the absolute value of net
exports. Since what we are really interested in studying is the total dynamics of trade, negative
9. coefficients do not necessarily mean that the Special 301 decreases trade. Likely we are
observing an increase in imports, which will decrease the absolute value of net exports, but could
still imply increased trade volumes. Since there are so many factors which can influence imports
and exports, the system through which Special 301 list interacts with trade is complicated. This
paper is only focused on the effect that Special 301 has on aggregated trade. Therefore, no matter
the positive or negative effect on net exports (which can be decomposed into different effects on
imports and exports), we can clearly see that Special 301 variables increase the dynamics of
international trade.
The change in dollar index, GDP in U.S. and GDP in the rest of the world do not have
statistically significant effects on the absolute value of net exports in the model. Probably
because the data for the dollar index only includes some countries’ currencies, it cannot reflect
the total influence of currency on trade. Additionally, GDP per capita sometimes does not
contribute to the trade between countries. For instance, China has low GDP per capita relative to
the rest of the world, but has a large trade volume with America due to the large population in
China.
The time trend variable also has significant effect on the absolute value of net
exports. The trade has an increasing trend with the development of the world. The time trend
variable captures the unknown factors in world development.
Conclusion
Studying the effects of the Special 301 Report is a complex process, as it involves many
countries of various size economies, with various trade policies and with different levels of
economic development. Based on the available data we were able to build a model that used first
10. differenced time series to measure US international trade volume. We collected data for the
absolute value of US net exports, as well as several variables to measure trade volume, including
FDI, relative dollar value, and US and world GDP per capita. We also measured Special 301
with several different variables, one that measured the total number of countries on the list, as
well as three variables that measured the number of countries on the designations Watch List,
Priority Watch List, and Priority Foreign Country.
In our study we found that the contemporary variables for Special 301 had a significant
effect on the absolute value of US net exports. The lags (1 or 2) of any of the Special 301
variables had a less significant effect on the absolute value of US net exports. We found that the
first lag of the watch list countries had slightly significantly negative effect on trade, but
generally the lagged Special 301 variables had little significance. Additionally, while we found
that variable for the total number of countries on the Special 301 list had a statistically
significant, positive effect on the absolute value of US net exports, the designations had a
negative effect. Since our dependent variable is a measure of exports, negative values don’t
necessarily indicate less total trade, but likely an increase in imports.
Our findings here support the findings of other studies, that the Special 301 Report does
have a positive impact on the volume of US international trade. From our study it remains
unclear specifically how the Special 301 Report affects trade in terms of imports versus exports,
but it is clear that this program has increased total US international trade volumes.
11. Bibliography
Keller, W., & Yeaple, S. R. (2003). Multinational enterprises, international trade, and
productivity growth: firm-level evidence from the United States (No. w9504). National
Bureau of Economic Research.
Riker, D. (2012). Has special 301 promoted us manufacturing exports? Review of International
Economics, 20(1), 288-298. Retrieved from
http://onlinelibrary.wiley.com.prox.lib.ncsu.edu/doi/10.1111/j.1467-9396.2012.01022.x/p
df
(2013). 2013 special 301 report. Retrieved from
http://www.ustr.gov/sites/default/files/05012013%202013%20Special%20301%20Report
.pdf
(2014, November 20). Real trade weighted us dollar index: Broad. Retrieved from Federal
Reserve Economic Data website: http://research.stlouisfed.org/fred2/series/TWEXBPA
(2014, November 20). Rest of the world; foreign direct investment in us; asset level. Retrieved
from Federal Reserve Economic Data website:
http://research.stlouisfed.org/fred2/series/ROWFDNQ027S
(2014, November 20). Rest of the world; foreign direct investment in us; asset flow. Retrieved
from Federal Reserve Economic Data website:
http://research.stlouisfed.org/fred2/series/ROWFDIQ027S
(2014, November 20). Real gross domestic product per capita. Retrieved from Federal Reserve
Economic Data website: http://research.stlouisfed.org/fred2/series/A939RX0Q048SBEA
(2014, November 20). Gdp per capita. Retrieved from Data website:
http://data.worldbank.org/indicator/NY.GDP.PCAP.CD
