How does import processing time impact export patterns?

• Author: Kazunobu Hayakawa,Taiyo Yoshimi,Nuttawut Laksanapanyakul
• Date: 01 July 2019
• DOI: http://doi.org/10.1111/twec.12789
• Published date: 01 July 2019

ORIGINAL ARTICLE

How does import processing time impact export

patterns?

Kazunobu Hayakawa

1

|

Nuttawut Laksanapanyakul

2

|

Taiyo Yoshimi

3

1

Inter-disciplinary Studies Center, Institute of Developing Economies, Chiba, Japan

2

Science and Technology Development Program, Thailand Development Research Institute, Bangkok, Thailand

3

Faculty of Economics, Chuo University, Hachioji-shi, Tokyo, Japan

KEYWORDS

customs, import processing time, Thailand

1

|

INTRODUCTION

In today's interconnected global economy, efforts to streamline, speed up, and coordi -

nate trade procedures, as much as efforts to further liberalize trade policies, will drive

the expansion of world trade and help countries to integrate into an increasingly glob-

alized production system, rather than being left on the margins of world trade.

World Trade Organization (2015)

In the modern era, which features sophisticated international supply chains, the efficiency of

operations at one stage of the supply chain may have significant effects on production outcomes in

subsequent stages. The long import process (e.g., customs clearance) has hitherto been identified

as one of the major obstacles in international trade. The time required for import processing

depends on various factors, such as the efficiency of cargo handling at ports. The import process

also takes longer if customs personnel physically inspect cargos. Such sluggishness in the import

process has significant effects on firms' activities. For one, if the import process is sluggish and

the arrival of imported intermediate inputs takes a long time, producers' ability to ensure timely

production can be reduced. Furthermore, importers have to pay extra storage costs if the import

process takes longer. Such tendencies are more likely to be observed in economies in which inter-

national supply chains are developed. Consequently, smoothness or sluggishness at one stage may

affect all stages of an international production network.

This study theoretically and empirically investigates this time cost in the import process. Specifi-

cally, we examine how import processing times affect export activities at the establishment level. In

the theoretical section, we introduce two sources of the time cost: the benefit of quick delivery and

the storage cost at the port. Following Hummels and Schaur (2013), the former is modelled as a direct

effect of import processing time on the productivity parameter. However, the latter source is

approached in a similar way to Carballo, Graziano, Schaur, and Martincus (2014, 2016) where the

time cost of imports is introduced assuming that ad valorem trade costs are positively associated with

the time spent on importing. By introducing these two sources into the model developed by Kropf

Received: 15 April 2017

|

Revised: 21 November 2017

|

Accepted: 12 February 2019

DOI: 10.1111/twec.12789

2070

|

© 2019 John Wiley & Sons Ltd wileyonlinelibrary.com/journal/twec World Econ. 2019;42:2070–2088.

and Sauré (2014), we investigate the effects of import processing time not only on exports but also on

export shipment frequency and exports per shipment. As a result, we demonstrate that an elongated

import processing time reduces establishments' export shipment frequencies as well as exports per

shipment and thus total exports. These two sources of the time cost are basically associated with the

same qualitative consequences although the first source, the benefit of quick delivery, can be related

to the time sensitivity of imported products more closely than the second source.

In empirical terms, we employ highly detailed customs data for Thailand from 2007 to 2011 to

investigate these theoretical predictions. Import processing time is measured at the establishment

level using the difference between the dates on which import shipments arrived at ports and the

dates on which they were later released from the container yard. Specifically, it measures the

length of time for cargo handling and customs clearance. Results indicate that longer import pro-

cessing times reduce establishments' total exports, particularly by decreasing the number of export

shipments. Significantly negative effects on exports per shipment appear in some specific

instances, such as in the case of sea transportation. These findings are important in Thailand. For

example, the classification of Harmonized System (HS) codes is an important issue in customs

clearance. The classifications used by importers sometimes differ from those given by customs

officers. Since importers need to contact customs officers when such inconsistencies in HS codes

occur, it takes much longer to clear customs. It is, thus, important to clarify how import processing

times affect the business activities of firms in Thailand.

This paper relates to the extant literature in three important ways. First, there is a growing body

of research into trade frequency. Eaton, Eslava, Kugler, and Tybout (2008) conducted a microlevel

study on this literature and provided various basic statistics regarding the number and frequency of

export transactions in Colombia. Alessandria, Kaboski, and Midrigan (2010) demonstrated that the

existence of fixed costs per import shipment leads to lumpiness in import transactions. Kropf and

Sauré (2014) computed fixed costs per export shipment using Swiss export data. Hornok and

Koren (2015b) and Békés, Fontagné, Murakozy, and Vicard (2014) shed more light on the correla-

tion between shipment frequency and several variables. The former study used export data for the

United States and Spain, showing that export shipments are less frequent and larger when export-

ing to countries with larger per‐shipment costs. Using French export data, the latter study showed

that firms adjust to increased uncertainty by reducing the number of shipments and increasing their

shipment sizes. Our study is related more to the former case, as we focus on potential relationships

between imports and exports without considering uncertainty. Furthermore, our paper sheds light

on the correlation of export shipment frequency and size with a novel consideration, that is, time

needed for the import process to be finalised.

Second, our paper is also closely related to the literature on the effects of trade facilitation (see

Feenstra & Ma, 2014; Hornok & Koren, 2015a; Persson, 2013; World Trade Organisation, 2015).

For example, Feenstra and Ma (2014) found that port efficiency significantly affects extensive trade

margins. Country‐level studies on the effects of customs clearance time include Djankov, Freund, and

Pham (2010), Freund and Rocha (2011), and Portugal‐Perez and Wilson (2012). By estimating grav-

ity equations, these studies found a significant effect of customs clearance time on trade values, using

data obtained from the Doing Business of the World Bank. However, other studies have examined

the relationship between customs clearance time and firm‐level exports (Dollar, Hallward‐Driemeier,

& Mengistae, 2006; Li & Wilson, 2009; Shepherd, 2013). Also, Hillberry and Zhang (2015) empiri-

cally show that such customs clearance time depends particularly on customs' governance.

Notably, some firm‐level studies use actual shipment dates to measure the length of time in

customs clearance, as in our paper. Martincus, Carballo, and Graziano (2015) investigate the

effects of export customs time on firm‐level exports, whereas the effects of import customs time

HAYAKAWA ET AL.

|

2071