Once again “smile curve”: Is chain upgrading possible?

• Published date: 01 February 2023
• Author: Halit Yanikkaya,Abdullah Altun,Pınar Tat
• Date: 01 February 2023
• DOI: http://doi.org/10.1111/rode.12927

REGULAR ARTICLE

Once again “smile curve”: Is chain upgrading

possible?

Halit Yanikkaya | Abdullah Altun |Pınar Tat

Department of Economics, Gebze Technical University, Kocaeli, Turkey

Correspondence

Pınar Tat, Department of Economics,

Gebze Technical University, Kocaeli,

Turkey.

Email: pinartat@gtu.edu.tr

Funding information

Turkish Academy of Sciences

Abstract

The geographic fragmentation of the production pro-

cess inevitably leads to the question of whether or not

there is a relationship between the level of value-added

and the position of production stages along with a

global production/value chain, known as the “smile

curve”as the theoretical literature suggests. This study

investigates the relationship between value-added and

production stages across more than 34 sectors over

40 countries for two different periods, 1995–2011 and

2000–2014, by taking into account a variety of different

measures of production stage and country-sector het-

erogeneity. In the first step, we utilize the decomposi-

tion methodology of Wang, Wei, Yu, and Zhu (2017) to

track the production stages across country sector. In

the second step, we test our hypotheses by employing

the fixed effects (FE) estimation technique. The results

show that the relationship between value-added con-

tent in output and backward length exhibits a U-shaped

distributional pattern. The significant results for both

developed and developing economies indicate the

potential of economies to benefit from the functional

upgrading along with production stages. Our general

conclusion regarding production chains is also valid for

stages mainly related to the global value chain (GVC),

especially for developed countries. Furthermore, capital

intensity and total factor productivity appear to be cru-

cial factors for improvements in process upgrading.

Received: 11 October 2021 Revised: 7 July 2022 Accepted: 1 August 2022

DOI: 10.1111/rode.12927

62 © 2022 John Wiley & Sons Ltd. Rev Dev Econ. 2023;27:62–88.wileyonlinelibrary.com/journal/rode

Given the robust and positive impacts of chain

upgrading in all countries considering both the total

production process and GVC part, industries should

exploit the opportunities for higher value-added in the

global production system.

KEYWORDS

backward production lengths, GVCs, sectoral value-added

content, smile curve

JEL CLASSIFICATION

F14, F41, F68, O24

1|INTRODUCTION

In the past three decades, rapidly enhancing technology and decreasing transportation costs

tightly connect modern economies through value-added and production chains. To describe this

supply chain phenomenon principally depending on the global division of labor and specializa-

tion, Baldwin (2011) uses the term “unbundling of production.”The production story of the

iPhone can be taken as one of the striking examples of this new production/trade model. Even

though an iPhone is assembled in China, this process generates a value-added, which is at most

5% of the iPhone's total value-added (West, 2018). Moreover, increasing trends of a variety of

length measures for two different periods in Figure 1exhibit a higher scale of production stages

in the global supply chains in the contemporary world.

1

The geographic fragmentation of the

production process inevitably leads to the following question: Is there any relationship between

the level of value-added and the position of production stages along with a global production/

value chain in a way that sectors located in upstream or downstream parts of value chains pro-

duce more value-added than sectors with intermediate levels of up(down)streamness?

1.95

2.00

2.05

2.10

2.15

2.20

2.25

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014

Backward Length_1st Set Length_1st Set Backward Length_2nd Set Length_2nd Set

FIGURE 1 Length over years. Notes: Backward length is calculated by following Wang et al. (2017), and

length is calculated by following Fally (2012). The first set includes the years 1995–2011, and the second set is

over the period 2000–2014. The first set includes 40 countries, and the second set includes 43 countries. [Colour

figure can be viewed at wileyonlinelibrary.com]

YANIKKAYA ET AL.63