Do international flights promote FDI? The role of face‐to‐face communication

• Published date: 01 November 2019
• Date: 01 November 2019
• Author: Kiyoyasu Tanaka
• DOI: http://doi.org/10.1111/roie.12437

Rev Int Econ. 2019;27:1609–1632. wileyonlinelibrary.com/journal/roie

|

1609

© 2019 John Wiley & Sons Ltd

1

|

INTRODUCTION

Information collection is crucial for managerial decision‐making processes in foreign direct invest-

ment (FDI) to ensure the success of risky investments in an unfamiliar environment. Advances in

information and communication technologies (ICTs) have reduced barriers to acquiring codified busi-

ness information via telephone calls, video conferences, and internet access. Declining communica-

tion costs help to coordinate complex production tasks from a distance (Baldwin & Evenett, 2015).

Meanwhile, there remain coordination costs regarding face‐to‐face contacts because in‐person meet-

ings are still crucial input for building business relationships, negotiating business contracts, and mon-

itoring offshore production (Storper & Venables, 2004). International business travel is key to acquire

uncodified and relationship‐specific information (Harvard Business Review, 2009). However, there

is limited evidence on the role of face‐to‐face contacts in FDI activity, making it a crucial empirical

question.

Received: 9 January 2019

|

Revised: 28 May 2019

|

Accepted: 28 June 2019

DOI: 10.1111/roie.12437

ORIGINAL ARTICLE

Do international flights promote FDI? The role of

face‐to‐face communication

KiyoyasuTanaka

Institute of Developing

Economies,JETRO, Chiba, Japan

Correspondence

Kiyoyasu Tanaka, Institute of Developing

Economies, JETRO, 3‐2‐2 Wakaba,

Mihama‐ku, Chiba‐shi, Chiba 261‐8545,

Japan.

Email: kiyoyasu_tanaka@ide.go.jp

Funding Information

I acknowledge the financial support of

JSPS Grant‐in‐Aid for Young Scientists (B)

Grant Number 16K17129 and for Scientific

Research (A) 18H03637

Abstract

International passenger flights facilitate business travel for

in‐person meetings abroad. However, the significance of

face‐to‐face communication (FFC) is not clear. To identify

the FFC channel, this paper examines whether flights pro-

mote FDI more strongly for multinational firms that face

relatively intensive FFC in foreign production. Expatriate

employees are used as a proxy for the FFC intensity. Using

firm‐level data on Japanese multinational firms for the pe-

riod 1989 to 2006, I show that more frequent flights increase

new FDI entry, with the larger positive effects for multi-

nationals in high FFC sectors. The results support the FFC

channel to connect flights and FDI.

JEL CLASSIFICATION

F21; F23; L9

1610

|

TANAKA

In this paper, I examine whether international passenger flights promote FDI activity through

a channel of face‐to‐face communication (FFC). Air transportation services facilitate international

business trips through a reduction in travel time and costs for in‐person meetings in foreign countries.

Such a decline in FFC costs encourages multinational firms to establish a foreign production plant.

Moreover, flight frequencies should have varying impacts on multinational firms that may incur vary-

ing intensities of in‐person contacts in foreign production. To assess the hypothesis, I use firm‐level

data on 3,030 Japanese multinational firms investing across 79 foreign markets for the period 1989

to 2006. Estimating a linear probability model for new FDI entry at the firm level, I examine whether

a weekly frequency of air passenger flights increases the probability of observing FDI entry and the

extent to which such an impact depends on a proxy variable for FFC intensities in foreign production.

There are two major challenges in my empirical strategy. First, air services are endogenously de-

termined through a supply–demand relationship of passengers. To address an endogeneity issue, I es-

timate the impact of a country‐level flight frequency on firm‐level FDI decisions. Because individual

firms do not have a strong influence on aggregate demand for passenger flights at the bilateral level,

my specification reduces a reverse causality bias. Additionally, I follow the identification strategy in

Rajan and Zingales (1998) by estimating the differential effect of a country‐level flight variable across

the firms that may respond differently to this variable. Finally, I exploit world heritage sites as a plau-

sible instrument for passenger flights to address an endogeneity bias in flights.

A second challenge is a lack of systematic data on in‐person meetings held by multinational firms

for starting up a new foreign plant. To fill this gap, I construct a new proxy for an intensity of in‐per-

son interactions at the sector level by using data on foreign affiliates’ employees who are sent on

international assignment from a home to a host country. Expatriate employees help to coordinate local

management with the headquarters and to facilitate face‐to‐face contacts with local workers, business

partners, and government officials in foreign countries (Collings, Scullion, & Morley, 2007; Edström

& Galbraith, 1977; Harzing, 2001; Moreira & Ogasavara, 2018). They play a crucial role in knowl-

edge flows through face‐to‐face interactions. Thus, multinational firms in a greater need for FFC

should employ a higher number of expatriate employees in foreign production. By estimating the de-

terminants of expatriate employees at the affiliate level, I quantify the relative importance of expatri-

ates across manufacturing sectors, which can serve as a proxy for FFC intensity in foreign production.

The main results are summarized as follows. Passenger flight frequencies significantly increase the

probability that Japanese firms invest in a foreign market. The positive flight effects are significantly

higher for Japanese firms in manufacturing sectors that are relatively FFC intensive. To gauge the eco-

nomic magnitude of flight effects, I show that an increase in flight frequencies by one more flight per

day is predicted to increase the likelihood of new FDI in high FFC sectors by 0.032 percentage points.

As the mean of new FDI entry is 0.4%, a marginal effect of flights on FDI entry is large. Additionally,

I show the robustness of the positive flight effects to an endogeneity bias by using world heritage sites

as an instrument for flight frequencies. The results remain similar after accounting for alternative

sector‐level interactions with flights, including a share of technicians or managers, task routineness

index (Costinot et al., 2011), importance of relationship‐specific investments (Nunn, 2007), and time

sensitiveness (Blyde & Molina, 2015). Taken together, the results support the hypothesis that more

frequent passenger flights can lower the FFC costs for FDI decisions.

This paper makes an empirical contribution to the literature on FDI determinants. As information

collection and communications are crucial in FDI decisions, Blonigen and Piger (2014) show that

FDI activity depends on host‐country communications infrastructure as measured by telephone traffic

and internet users.1

Blonigen, Ellis, and Fausten (2005) argue that information sharing among firm

networks reduces information costs, thereby encouraging FDI activity.2

Furthermore, Stein and Daude

(2007) argue that differences in time zones discourage frequent real‐time communications, which