How Do the Trans‐Pacific Economies Affect the USA? An Industrial Sector Approach

• Author: David D. Selover,Takeshi Yagihashi
• DOI: http://doi.org/10.1111/twec.12471
• Date: 01 October 2017
• Published date: 01 October 2017

How Do the Trans-Pacific Economies

Affect the USA? An Industrial Sector

Approach

Takeshi Yagihashi and David D. Selover

Department of Economics, Old Dominion University, Norfolk, VA, USA

1. INTRODUCTION

OVER the past three decades, the Trans-Pacific region has grown to become the main

trading partner of the USA. This region, made up of over 40 countries spread across

Asia, Oceania and the Americas, accounts for roughly two-thirds of the US trade volume. The

fast growth in trade is due to various factors such as improved air transportation, increased

use of computers in manufacturing, industrial agglomeration and reduction in tariffs through

free trade agreements in the Trans-Pacific region (see Helpman, 2006; Hummels, 2007;

Williams, 2013). The US administration has repeatedly made remarks about the economic

importance of the region (e.g. ‘the pivot to Asia’ and ‘America’s Pacific Century’) and is cur-

rently negotiating a regional trade agreement known as the Trans-Pacific Partnership (TPP).

1

While being economically tied to the region is beneficial for the US economy in the long run

(Petri and Plummer, 2012), there are also concerns among the public that this would subject

the US economy to external shocks, such as the Asian financial crisis of the late 1990s and

the disruptions of the Japanese Tohoku earthquake and tsunami of 2011. Thus, it is of prime

interest to US policymakers to study how shocks in the Trans-Pacific region affect the US

economy.

Our main purpose is to quantify how the US economy reacts at the aggregate and the sec-

tor levels to shocks from the Trans-Pacific region. Specifically, we are interested in answer-

ing the following series of questions: How does an overall Trans-Pacific shock affect the

USA? Which sectors in the USA are particularly vulnerable? Which sectors in the Trans-

Pacific region most strongly affect the US economy? Past studies have emphasised the posi-

tive role that bilateral trade plays in business cycle transmission (see Frankel and Rose,

1998; Baxte r and Kouparitsas, 2005). However, as argued by Bems et al. (2010), bilateral

trade does not determine exposure to a specific foreign country if there are vertical produc-

tion linkages that involve several countries. Furthermore, Crosby (2003) finds that within

the Asia–Pacific region, the standard measure of bilateral trade does not explain output

co-movement.

There are several characteristics of the Trans-Pacific region that complicate our analysis.

First, countries in this region differ not only in size and stage of development, but also in

economic structure and how much they depend on foreign trade. Thus, selecting a few aggre-

gate variables to represent this region is difficult. The second characteristic is that vertical

production linkages are prominent in the region (Kimura et al., 2007). A leading example is

seen in the auto industry, where car parts manufactured in Canada and Japan serve as

1

For more information on the trade policy under current Obama administration, see Barfield (2009).

©2016 John Wiley & Sons Ltd 2097

The World Economy (2017)

doi: 10.1111/twec.12471

The World Economy

essential inputs for automobile assembly in the USA and Mexico. Similarly, in the electronics

industry, semiconductors manufactured in Japan and Taiwan are used as components for

machinery manufacturing in Malaysia and Thailand. Such vertical production linkages are

expected to strengthen the transmission of business cycles across countries (see Burstein

et al., 2008; Ng, 2010).

To incorporate these characteristics in our analysis, we use a large panel of time-series data

consisting of disaggregated industrial production indexes. Our data cover 16 major industrial

sectors from nine countries in the Trans-Pacific region. We treat the Trans-Pacific region

(excluding the USA) as an independent regional unit by aggregating output across countries

and/or sectors to study its impact on the USA. We employ a factor-augmented vector autore-

gression (FAVAR) approach (Bernanke et al., 2005) to reduce the dimensions of our data and

to simulate the transmission of shocks from the Trans-Pacific region to the USA. FAVAR has

two advantages compared with the traditional VAR. First, we can incorporate a broader set of

information related to the unknown transmission mechanism by utilising detailed sectoral

data. Vasishtha and Maier (2013) claim that this aspect of FAVAR is particularly relevant

when the business cycle transmission involves more than two countries. Second, the FAVAR

analysis allows us to study the transmission of shocks in several dimensions, that is the effect

of overall and sectoral shocks in the Trans-Pacific on the US economy for both overall and

sectors.

There are several findings in this paper. First, while a positive shock in the overall

Trans-Pacific output generates positive responses in the US output for almost all sectors, the

magnitude of the response differs substantially by sector. For example, the primary metals,

machinery, electrical equipment and computer/electronics sectors show large responses fol-

lowing Trans-Pacific shocks, whereas the food, beverage/tobacco and petroleum/coal sectors

show almost no response. We find that the openness to trade and the durability of goods

can explain part of the results. Second, we find that shocks in five Trans-Pacific sectors, that

is, plastics/rubber, primary metals, machinery, computer/electronics and transportation equip-

ment, result in large responses in the overall US output, indicating that these sectors are of

particular importance to the US economy. The large responses of US output, particularly

from shocks in relatively small sectors, such as plastics/rubber and primary metals, are sur-

prising given that the past literature (e.g. Helg et al., 1995) has found that sectoral shocks

have little overall effect. Third, to explore the possible reasons for the large effect, we

examine the same-sector responses for the above five sectors. We find that for primary met-

als, computer/electronics and transportation equipment, the same-sector response is larger

than the overall response, indicating that vertical production linkages within these sectors

are relatively strong. Finally, our main results are robust to alternative specifications such as

the inclusion of Chinese industrial production and oil prices, different numbers of factors

and lags, different restrictions for the identification of shocks and alternative method of

aggregation.

This paper contributes to the literature in several ways. First, using a disaggregated sectoral

data set, we provide a coherent explanation on how exogenous shocks in the Trans-Pacific

region affect the USA in terms of both the national economy and the individual sectors.

While the international economics literature has long noted the importance of sectoral distinc-

tion in explaining the behaviour of trade and output, the question of how the transmission of

shocks from sector to sector compares to the transmission from sector to overall output

remains relatively unexplored. Our results show that sectors that are not particularly large in

©2016 John Wiley & Sons Ltd

2098 T. YAGIHASHI AND D. D. SELOVER