Product life‐cycle, knowledge capital, and comparative advantage

• Author: Xiaoping Chen,Yuchen Shao
• Published date: 01 February 2020
• Date: 01 February 2020
• DOI: http://doi.org/10.1111/roie.12449

252

|

wileyonlinelibrary.com/journal/roie Rev Int Econ. 2020;28:252–278.

© 2019 John Wiley & Sons Ltd

1

|

INTRODUCTION

Research and development (R&D) is crucial for a country's long‐term economic growth. Thus, un-

derstanding the comparative advantage in doing R&D for different countries is important. Recent

literature has highlighted the role of knowledge capital endowment in shaping a country's comparative

advantage and trade patterns. The knowledge capital endowment measures the available resources

or capacity a country has to carry out new R&D activities.1

Similar to physical capital, the relative

endowment of knowledge capital also generates comparative advantage across industries. However,

different from physical capital, innovation output may be imitated. As a result, institutions of intellec-

tual property that restrict imitation may shape the trade patterns.2

As highlighted by Bilir (2014), the

product life‐cycle (PC) length is associated with the risk of imitation. In this paper, we investigate the

role of PC in shaping trade patterns between countries.

We follow Bilir (2014) and measure the PC length by the average time lag between new patents

and their cited existing patents in a given industry.3

PC can be interpreted as the life‐cycle of tech-

nologies or patents in each industry. Technologies or patents emerge at different frequencies across

Received: 31 January 2018

|

Revised: 10 September 2019

|

Accepted: 12 September 2019

DOI: 10.1111/roie.12449

ORIGINAL ARTICLE

Product life‐cycle, knowledge capital, and

comparative advantage

XiaopingChen1

|

YuchenShao2

1School of Social Sciences,Nanyang

Technological University, Singapore,

Singapore

2School of Economics,Nanjing University,

Nanjing, China

Correspondence

Yuchen Shao, School of Economics,

Nanjing University, Nanjing, China.

Email: shaoych04@gmail.com

Funding information

National Natural Science Funds (China),

Grant/Award Number: 71903087

Abstract

This paper empirically investigates the effect of product

life‐cycle on trade patterns. We measure the product life‐

cycle length with patent citation data at the industry level.

Using bilateral trade data from 2002 to 2006, we find that

countries with more knowledge capital endowment export

more in industries with shorter product life‐cycles. We

show that this pattern is largely driven by the risk of imita-

tion. The pattern is reversed when imitation is limited by

stronger intellectual property rights.

JEL CLASSIFICATION

F10; F14

|

253

CHEN aNd SHaO

industries. Over time these technologies or patents will mature and then become obsolete and never be

cited again by newly developed technologies or patents. It is not surprising to observe that PC differs

across industries. Some industries, such as steel, paper, or cement manufacturing, have long PC with

little variation in technology incorporated over time. In these industries, new technologies emerge less

frequently. Existing technologies will be used and cited for a longer period of time. Whereas in some

other industries, such as electronic or pharmaceutical products, the PC may be quite short. In such

industries, new technologies and patents arrive more frequently.4

Heterogeneous PC lengths across

industries can have important effects on R&D investments. Bilir (2014) highlighted the role of PC

in multinational firms’ outsourcing decisions and showed that the risk of imitation is higher when

the PC length is longer. On the other hand, the data seem to support that innovation is more costly

in industries with longer PC length.5

If this is true, standard trade theory predicts that countries with

more knowledge capital endowment should have the comparative advantage in industries with long

PC length. However, the imitation risk tends to push countries’ R&D investment away from these

long‐PC industries. In the end the trade pattern is determined by the comparative advantage effect and

the imitation risk.

We look into the international trade data to see how PC length affects the trade patterns. We mea-

sure the PC length using patent citation data and the knowledge capital stock is calculated with coun-

try‐level time‐series R&D expenditure data using the perpetual inventory method. We then bring these

measures together with other country‐level control variables to the bilateral trade data. Our empirical

analysis shows that countries with more knowledge capital endowment tend to export more in those

short‐PC industries, indicating that imitation risk drives the allocation of R&D resources across in-

dustries. When imitation risk is limited by good intellectual property rights (IPRs), this pattern will be

reversed, suggesting a comparative advantage for countries with more knowledge capital endowment

in industries with longer PCs. These results are robust to various alternative measures, controlling for

other sources of comparative advantage and using alternative methods and samples.

This paper contributes to one of the fundamental questions in the field of international trade, the

determinants of trade patterns. We highlight the roles of PC as well as IPRs institution in shaping

international trade flows. There also seems to be a new source of comparative advantage originating

from PC length and knowledge capital endowment. There is a large number of literature on compar-

ative advantage: country factor endowment and industry factor intensity such as capital endowment

and capital intensity, skilled labor endowment and skill intensity in Romalis (2004), Chor (2010),

Cai and Stoyanov (2016) among others; country institutions and industry‐level institutional depen-

dence as in Nunn (2007), Levchenko (2007), Essaji (2008), Nunn and Trefler (2014), Maskus and

Yang (2018); finance development and financial dependence as in Beck (2002), Do and Levchenko

(2007), Manova (2013) for example. Our paper finds some evidence for another source of compara-

tive advantage. Countries with more knowledge capital endowment have comparative advantage in

long‐product life‐cycle (long‐PC) industries where innovation seems to be harder. This type of com-

parative advantage may be more important in the modern economy as R&D becomes more important

and global.

Our paper also adds to the literature on PC. Our measure of PC follows Bilir (2014), where she

highlights the effect of PC length and IPR on the organization of multinational production. We share

similar insight with Bilir (2014) as we both emphasize the effect of PC on imitation risk. Our measure

of PC, using the citation lag between patents, is somewhat different from the original PC proposed by

Vernon (1966), where it describes the maturation process of a product. In a sense, one can interpret our

PC as the patent cycle. It measures the average life cycle of patents in a given industry. These cycles

will affect the allocation of R&D investments across industries for different countries, thereby shaping

their comparative advantage and trade patterns. There is another stand of literature also connecting PC