Indirect trade and direct trade: Evidence from Japanese firm transaction data

• DOI: http://doi.org/10.1111/twec.12997
• Date: 01 February 2021
• Author: Yukiko Umeno Saito,Tadashi Ito
• Published date: 01 February 2021

444

|

wileyonlinelibrary.com/journal/twec World Econ. 2021;44:444–461.

© 2020 John Wiley & Sons Ltd

Received: 11 August 2019

|

Revised: 2 May 2020

|

Accepted: 4 June 2020

DOI: 10.1111/twec.12997

ORIGINAL ARTICLE

Indirect trade and direct trade: Evidence from

Japanese firm transaction data

TadashiIto1

|

Yukiko UmenoSaito2

1Gakushuin University, Tokyo, Japan

2Waseda University, RIETI, Tokyo, Japan

Funding information

Japan Society for Promotion of Science KAKENHI, Grant/Award Number: 17K03707 and 18H03637; Japan Society for

Promotion of Science KAKENHI, Grant/Award Number: 18H00859

KEYWORDS

indirect trade, regional economies, wholesalers

1

|

INTRODUCTION

By exporting goods or services, firms not only expand their markets but also benefit from acquiring

new knowledge and techniques, thus raising their productivity.1 Imports, on the other hand, give firms

the chance to procure better inputs in terms of lower prices and/or higher quality, thus enabling more

efficient production.2 As the costs of transportation and transactions steadily decrease, interfirm trans-

action networks expand on a global scale, and worldwide production system and global value chains

(GVCs) emerge.

Since Ricardo (1817), the international trade literature has long investigated the merits of trade. A

recent development, the so-called “heterogeneous firms trade model” pioneered by Eaton and Kortum

(2002) and Melitz (2003), shows that for domestic firms, trade liberalisation favours only the most pro-

ductive firms because it allows them to enter foreign markets. In contrast, the least productive firms are

forced to exit the market because of tough competition with more foreign competitors. However, there are

firms in the middle of these two groups of firms, which benefit from indirectly participating in GVCs in

exporting their goods or importing inputs through wholesalers or other manufacturing firms.

Helpman, Melitz, and Yeaple (2004) argue that because foreign direct investment (FDI) or exports

incurs fixed costs relevant to each mode, firms need to be sufficiently productive to absorb the FDI fixed

cost or export fixed costs to engage in these modes. Usually, FDI fixed costs are larger than export fixed

1Bai, Krishna and Ma (2017), Crespi, Criscuolo and Haskel (2008), Harris and Li (2007), Shevtsova (2015), among others,

show productivity enhancing effects of exports.

2Amiti and Konings (2007) show the positive impacts of imported inputs for firms’ productivity. Antras and Helpman (2004)

and Antràs, Fort and Tintelnot (2017) provide theoretical models on outsourcing. Merits pertaining outsourcing in domestic

firm transactions are shown by Bernard, Moxnes and Saito (2019).

|

445

ITO and SaITO

costs. As a result, only the most productive firms, having lower marginal costs, can serve foreign markets

through FDI. The next most productive firms serve foreign markets by exporting goods produced in their

own countries and the least productive firms sell their products only in their domestic markets. Ahn,

Khandelwal, and Wei (2011), Akerman (2018), Bernard, Jensen, Redding, and Schott (2010) and Crozet,

Lalanne, and Poncet (2013) incorporate indirect exports into this framework. Namely, although some

firms are not sufficiently productive to serve foreign markets, they can export their products through

intermediaries, which reduce export-related fixed costs by spreading the fixed costs among many ex-

port clients. Thus, many small- and medium-sized firms can enjoy benefits of trade. Firms in regional

economies, especially, are smaller and less productive and thus less likely to directly export their goods.

However, there is a larger scope for firms in regional economies to benefit from indirect exporting.

This paper attempts to establish stylised facts on direct and indirect trade and its relationship with

firm performance using firm transaction data of Japanese firms.3 The main f indings are as follows:

1. The size of firms, in terms of the number of employees and sales values, is smaller in re-

gional areas than in metropolitan areas.4 Even controlling for firm size, firms in regional areas

are less likely to export, which indicates higher trade costs in regional economies. This is

probably due to regional economies having less information on overseas markets lacking

sufficient infrastructures for export activities.

2. Approximately 40% of firms in regional areas are engaged in either direct or indirect exports/im-

ports, among which indirect exports/imports represent more than 30%. This share rises to close to

70% in terms of the number of employees and sales values.

3Similar analyses using the same data set with this paper have been completed by Fujii, Ono and Saito (2016), which finds that the

distributions of sales and labour productivity are ordered for direct, indirect and non-exporters, and also by Okubo, Ono, Saito (2014),

which finds that wholesalers tend to locate closer to their manufacturing buyers and farther from their manufacturing sellers.

4The metropolitan area is called the ‘core’ region, whereas the regional area is called ‘periphery’ region.

FIGURE 1 Firm distribution by the number of employees, the TSR data set and the Census.

Source: Authors’ computation from the TSR and the Census [Colour figure can be viewed at wileyonlinelibrary.com]

0%

0–4

5–9

10–19

20–29

30–49

50–99

100–299

300–999

1000–1999

2000–4999

5000–

10%

20%

30%

40%

50%

60%

70%

Distribution of number of employees

census TSR