Export diversification, margins and economic growth at industrial level: Evidence from Thailand

• DOI: http://doi.org/10.1111/twec.12921
• Published date: 01 October 2020
• Author: Juthathip Jongwanich
• Date: 01 October 2020

2674

|

wileyonlinelibrary.com/journal/twec World Econ. 2020;43:2674–2722.

© 2020 John Wiley & Sons Ltd

Received: 7 March 2019

|

Revised: 24 October 2019

|

Accepted: 7 January 2020

DOI: 10.1111/twec.12921

ORIGINAL ARTICLE

Export diversification, margins and economic

growth at industrial level: Evidence from Thailand

JuthathipJongwanich

Faculty of Economics, Thammasat University, Bangkok, Thailand

KEYWORDS

economic growth, export diversification, export margins

1

|

INTRODUCTION

After the global financial crisis, concerns about the impact of external shocks have been widely dis-

cussed and domestic demand-led growth has been proposed to be a viable policy option for Asian

countries (ADB, 2009; Prasad, 2009). However, for many developing Asian economies including

Thailand whose domestic market is relatively small and who have long been engaged with multi-

national enterprises, turning to domestic demand-led growth strategies is considerably challenging.

In addition, a country potentially gains potent benefits from exporting, including realising scale

economies, foreign exchange earnings and productivity improvements as a result of global exposure

(Bustos, 2011; Fernandes, 2007). For those countries, export-led growth models have continued to

be implemented after the global financial crisis. Export diversification and upgrading have been pro-

posed in policy circles as a compromise solution to relying on the export sector. In terms of export

diversification, it is argued that diversification helps reduce export instability as it provides a hedge

against price variations and shocks in specific product markets (Bertinelli, Fudanoki, Komori, Peiro,

& Rossi, 2006; Harding & Javorcik, 2007). In addition, it is argued that countries/firms that can pro-

duce many products with their comparative advantages have a high capability of absorbing or adapting

to foreign technologies, the accumulation of skills and the ability to conduct learning-by-doing. Thus,

export diversification has the potential to have a positive impact on both productivity and growth.

However, in theory, role of export diversification on productivity and growth is unclear. The idea

of export diversification tends to contradict traditional trade theory, particularly the Ricardian and/or

Hecksher–Ohlin models, wherein countries should specialise and be actively concerned with factor

accumulation, not diversification. However, new trade theory emphasising firm heterogeneity tends

to suggest a complex relationship between trade diversification and productivity (Cadot, Carrère,

The author would like to thank referees for useful comments and suggestions. Initial funding from the Thailand Research

Fund (TRF) for this project is highly appreciated. The previous version was presented at the 40th Symposium, Re-designing

Thailand: How to Grow Sustainably, organised by the Faculty of Economics, Thammasat University at the Stock Exchange

of Thailand on 17 January 2017, and Asia Economic Outlook and Challenges to Growth and Stability, 23 November 2018,

Korea University, Seoul, Republic of Korea. Last, but not least, the author would like to thank Sornsawan Sailamai for being

an excellent research assistant.

|

2675

JONGWANICH

& Strauss-Kahn, 2011). In addition, recent empirical studies (Cadot et al., 2011; Imbs & Wacziarg,

2003; Mohan, 2016) show the non-monotone pattern of export diversification and per capita income

with initial diversification and subsequent respecialisation when income reaches a certain level. This

could to some certain extent imply that the role of export diversification would become less relevant

for growth when countries become richer and produce more complex products.

In addition, while export diversification/growth can emerge from both intensive and extensive

margins, how these two margins contribute to economic growth is debatable. Intensive margins refer

to an increase in exports through expanding existing products (traditional products), while extensive

margins refer to expanding exports through creating new products and/or developing new trading

partners. On the one hand, Evenett and Venables (2002); Brenton and Newfarmer (2007); Cadot et

al. (2011) find that export diversification was mostly explained by intensive margins. However, ex-

panding exports through such margins could create downside risks since a country/firm may overly

rely on a fixed basket of export products, which may lead to a decline in export prices, along with

an increase in volatility arising from exogenous shocks. On the other hand, Hummels and Klenow

(2005), as well as Pham and Martin (2007), find that extensive margins (new products) are cru-

cial in contributing to export growth/diversification. Hidalgo and Hausmann (2009), Hausmann,

Hwang, and Rodrik (2007), Hausmann and Klinger (2007) point out that for ensuring improvement

in economic development, exports should be expanded into more complex of production. However,

Brenton and Newfarmer (2007) show that extensive margins, in terms of expanding existing products

to new geographical markets, are more crucial in explaining export growth than the discovery of new

products.

With the unsolved debate, this study aims to contribute to the literature by examining the impact

of export diversification and margins on economic growth. In contrast to most previous studies,

which analyse such relationships using cross-country analysis, this study employs an in-depth anal-

ysis of Thai industries as a case study. The role of firm heterogeneity is, to a certain extent, better

reflected through employing industry/firm-level analysis than cross-country analysis.1

Diversifications are calculated by using export data from UNCOMTRADE at the 6-digit Harmonized

System (HS) classification of 2002. Then, we use product concordance, obtained from the World

Integrated Trade Solution (WITS) to match the HS 2002 code with 4-digit International Standard

Industrial Classification (ISIC) Rev 3 in examining the relationship between diversification and

growth. Our analysis focuses on total industries and five key subsectors, namely the processed food;

chemicals; plastics and rubber; textiles and apparel; and electronics and automotive sectors. Three

alternatives are used to measure diversification, that is the Herfindahl index (HHI), the Gini coeffi-

cient index and Theil's entropy index.

In addition, there are two alternative measures of export margins, both intensive and extensive,

used in this study in order to perform a robustness check. The first method involves using the within-

and between-group components of the Theil index to measure intensive and extensive margins, as

proposed by Cadot et al. (2011). However, such measures using a count of export lines to calculate the

extensive margin, which may have the inherent disadvantage of treating low- and high-value products

equally in calculating margins. In fact, the implications of margins arising from low- and high-value

products on growth could be different (Hummels & Klenow, 2005). Thus, we use an alternative mea-

sure to represent the margins by calculating them as reflecting their own share of the world market.2

1 Note that with data limitations in terms of firm-level data, our analysis is carried through industry-level data.

2 Note that we can define extensive margins in terms of world market share since, as mentioned in Cadot et al. (2011), the

opening of new export lines in developing countries tends to involve copying products from developed countries, not making

genuine innovations (inside-the-frontier innovations).

2676

|

JONGWANICH

Moreover, as extensive margins can refer to both exporting new products3 and expanding new mar-

kets, which could have different implications on economic growth (Brenton & Newfarmer, 2007;

Haddad, Lim, Pancaro, & Saborowski, 2013), this study employs both prospects in referring to the

extensive margins.

The rest of the paper is organised as follows. The following section provides an analytical frame-

work to lay down the groundwork for analysing the impact of export diversification and margins on

economic growth. Section 3 provides the methodology applied to calculate diversification and export

margins. Trends and patterns within diversification and the margins of Thai industries are presented

and compared with other Asian countries. Section 4 presents our empirical model, while the ensuing

results are outlined in Section 5. The final section provides our conclusions and points out potential

policy inferences.

2

|

ANALYTICAL FRAMEWORK

Gains from trade in terms of productivity and output growth have been studied for several decades.

Such research began with the standard neoclassical trade model, the Ricardian comparative advan-

tage model and/or Hecksher–Ohlin-based comparative advantage. In the 1980s, the introduction of

the monopolistic competition theory of international trade (Helpman & Krugman, 1985; Krugman,

1979) highlighted that the origins of gains from trade had shifted more to intra-industry trade.

Since then, the assumption of representative firms has been relaxed and the literature of firm het-

erogeneity has been growing (Bernard & Jensen, 2004; Melitz, 2003). International trade allows

better-performing firms to expand their product lines into larger markets, while resources are re-

allocated from less productive to more productive firms. This, therefore, leads to improvements in

both industry efficiency and overall productivity. The effect of international trade on productivity

and then economic growth has also been highlighted in the new/endogenous growth theory.4 Amiti

and Konings (2017) show that due to learning-by-exporting, firms who participate in foreign mar-

kets are more likely to experience productivity gains as opposed to non-exporters. The former re-

ceives new information about technological progress, product designs and quality of goods from

their foreign exposure, leading to productivity/growth promotion. Nonetheless, it is of note that

doubts upon the positive impact of trade liberalisation on productivity and growth are still voiced

in the literature (Lee & Kim, 2009; Rodriguez & Rodrik, 2001; Vamvakidis, 2002). Some studies

show that export/trade alone does not cause growth, depending on certain structural characteristics,

such as financial depth, inflation stabilisation, public infrastructure, governance, labour market

flexibility, ease of firm entry and ease of firm exit (Calderón & Poggioa, 2010; Chang, Kaltani, &

Loayza, 2009).

3 It is noteworthy that extensive margin in terms of new products according to Cadot et al. (2011) and Hummels and Klenow

(2005) refers to expanding exports through creating new products, which could be either high value-added products or low

value-added ones. Recent empirical studies, for example Hidalgo and Hausmann (2009), Filipe et al. (2012) and Krishna and

Levchenko (2013) show that more developed and wealthier countries are likely to produce and export higher value-added/

complex products. While income per capita in Thailand has increased over the past decades, exports in higher value-added

products have become evident. However, Filipe et al. (2012) show that among 124 countries, Thailand was still ranked at

59th, lower than Singapore, Malaysia and China. Thus, this would imply that the extensive margin referred to in this study

could include both high and low value-added exports, though it tends to move towards the former overtime. Its impact on

economic growth is, therefore, still unclear and worth examining.

4 See for example Aw and Hwang (1995); Bernard and Wagner (1997); Fernandes (2007); Bustos (2011).