1 Structural change and economic growth in selected emerging economies

Structural change and economic growth in selected emergingeconomies Muhamed Zulkhibri Islamic Research and Training Institute, Islamic Development Bank, Jeddah, Saudi Arabia Ismaeel Naiy

International Journal of Development Issues

Structural change and economic growth in selected emerging economies

Muhamed Zulkhibri Ismaeel Naiya Reza Ghazal

Article information:

To cite this document:

Muhamed Zulkhibri Ismaeel Naiya Reza Ghazal , (2015),"Structural change and economic growth inselected emerging economies", International Journal of Development Issues, Vol 14 Iss 2 pp 98 -116

Permanent link to this document:

http://dx.doi.org/10.1108/IJDI-09-2014-0064

Downloaded on: 05 March 2016, At: 00:38 (PT)

References: this document contains references to 61 other documents

To copy this document: permissions@emeraldinsight.com

The fulltext of this document has been downloaded 438 times since 2015*

Users who downloaded this article also downloaded:

Hakeem Ishola Mobolaji, Kamil Omoteso, (2009),"Corruption and economic growth in some

selected transitional economies", Social Responsibility Journal, Vol 5 Iss 1 pp 70-82 http://

dx.doi.org/10.1108/17471110910940014

Muhammad Tahir, Toseef Azid, (2015),"The relationship between international trade openness

and economic growth in the developing economies: Some new dimensions", Journal of ChineseEconomic and Foreign Trade Studies, Vol 8 Iss 2 pp 123-139 http://dx.doi.org/10.1108/

About Emerald www.emeraldinsight.com

Emerald is a global publisher linking research and practice to the benefit of society The companymanages a portfolio of more than 290 journals and over 2,350 books and book series volumes, aswell as providing an extensive range of online products and additional customer resources andservices

Emerald is both COUNTER 4 and TRANSFER compliant The organization is a partner of theCommittee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative fordigital archive preservation.

*Related content and download information correct at time of download

Structural change and economic growth in selected emerging

economies

Muhamed Zulkhibri

Islamic Research and Training Institute, Islamic Development Bank,

Jeddah, Saudi Arabia

Ismaeel Naiya

Economic Research and Policy Department, Islamic Development Bank,

Jeddah, Saudi Arabia, and

Reza Ghazal

Business and Management Sciences Department, University of Kurdistan-Hawler (UKH), Kurdistan, Iraq

Abstract Purpose – This paper aims to investigate the relationship between structural change and economic growth

for a panel of four developing countries, namely, Malaysia, Nigeria, Turkey and Indonesia over 1960-2010.

Design/methodology/approach – The study extent the growth equation by incorporating degree of

openness, labour and investment and construct structural change indices – modified Lilien index and the norm of absolute values It utilizes the recently developed panel cointegration techniques to test and estimate the long-run equilibrium of the growth equation.

Findings – The results confirm that structural change and economic growth are cointegrated at the

panel level, indicating the presence of long-run equilibrium relationship However, the impact of structural change on economic growth seems to be small and evolve slowly.

Originality/value – The findings indicate the need for policymakers to identify the binding

constraints that impede growth and the importance of institutionalize policy to encourage investment in productive sectors.

Keywords Economic growth, Panel cointegration, Structural change Paper type Research paper

1 Introduction

As the economic development involves shifting resources from low to high productivitysectors, it implies that economic development is a process of structural transformation.Structural transformation is also associated with other forms of changes such as social andpolitical transformations in form of changes in institutions, demography and labourmigration from rural to urban areas, etc It generally involves improvements in technologyand innovation, institution, human resource development and all changes that lead toincrease in levels of productivity in economic activities It has been observed that during thepast few decades, many countries were able to achieve sustained economic growth anddevelopment through structural transformation of their economies

Analysis of the relationships between growth and changes in economic structures

can be traced back to the classical era (Lucas, 1988) The modern analyses of structural

change started with Fisher (1935) and Clark (1940) who proposed the division of

economic activities into primary, secondary and tertiary sectors which served for

quantitative structural analyses Moreover, Kuznets (1971) proposed similar

classification of the economy into agriculture, industry and services sectors, arguing

that long-run economic development is accompanied by shifts in the allocation of

resources (especially labour) from primary sector (agriculture) to secondary sector

(industry) and subsequently to tertiary sector (services)

Structural change is a complex, intertwined phenomenon It should be emphasized

that any disruption of the process of structural transformation can have far-reaching

consequences on sustained economic growth and development, and poverty reduction

This theory has been supported by series of empirical studies on developed and the

newly industrializing economies, which revealed a steady decline of the share of labour

in agriculture sector, a passing increase and peak in the proportion of labour in

manufacturing sector and a consistent rise in the share of labour in services, reflecting

the transition from agrarian to post-industrial stage

Although the relationship between structural change and economic growth has been

empirically established in developed economies, it does not hold in most developing

countries with different technological, demographic and political set-ups that constitute

a different environment for structural transformation Many developing countries are

having high population and labour supply growth that exceeds the absorptive capacity

of their manufacturing sector Consequently, surplus labour released from the

agriculture sector may not be directly absorbed in the manufacturing, which may

compound problems of unemployment, inequality and poverty

This paper contributes to the literature by re-examining the relationship between

structural change and economic growth and providing comparative analysis from four

developing countries, namely, Indonesia, Malaysia, Nigeria and Turkey using panel

cointegration approach over the period 1960-2010 The empirical evidence on the

proximate determinants of structural transformation, particularly in emerging markets,

remains scant Hence, the study specifically addresses to what extent structural change

influences the behaviour of aggregate economic performance over the past five decades

or vice versa Given the heterogeneity of structural transformation processes in

developing countries, the choice of these countries stem from relatively similar

structural economic characteristics:

• these countries are oil exporting countries except for Turkey;

• these countries originally depend on exports of agricultural commodities; and

• the growth rate was relatively similar before it began to embark on economic

transformation

This paper is structured as follows Section 2 provides the overview of structural change

and growth literature Section 3 provides briefly overview of structural change and

growth in Turkey, Malaysia, Nigeria and Indonesia Section 4 describes the method and

data used for empirical analysis Section 5 discusses the results of the empirical models,

and Section 6 finishes with the main conclusions

99

Structural change and economic growth

Despite its high relevance for growth theory, business cycle theory and labourmarket theory as well as for economic policy, the topic of structural change is frequentlyneglected in economic research in the earlier period Structural change analysis received

a major boost in the 1990s, to find explanations for the process of technological changeand its effects on the economy (da Silva and Teixiera, 2006) To date, there exist varieties

of theoretical approaches that are concerned with the explanation of structural shiftsbetween the three broad sectors of the private economy and among the industries withinthese sectors

The most popular literature are the two schools of growth theory, which contradicteach other: “balanced growth school” and “structural change school” The balancedgrowth school stands for the mainstream neo-classical growth theory The models of thebalanced growth school are micro-founded (i.e they use utility functions) On the otherhand, the structural change school stands for the structural change models where nobalanced growth paths exist as long as structural change takes place (Baumol, 1967;

Laitner, 2000;Acemoglu and Guerrieri, 2008)

Kuznets (1973)points to two of the central causes of structural change that are stillrelevant in the more recent theoretical literature on the topic (i.e varying incomeelasticities of demand and the differential impact of technological progress) However, in

a more recent contribution, Baumol et al (1989)observe that the great diversity ofproductivity developments across industries and sectors and emphasize not only thefact that structural change is a long-term phenomenon, but also that productivitygrowth is particularly relevant in the long run

The structural change school is not necessarily consistent with Kaldor’s (1961)

stylized facts Kaldor’s stylized facts require that capital, consumption and output grow

at a constant rate that the growth path must be balanced, while the structural changeschool features unbalanced growth paths Furthermore, as the structural change schooldoes not rely on balanced growth paths, the analysis is relatively complicated.Therefore, the models from the structural change school make very simple assumptions

In a more recent literature, most studies aimed at reconciling structural changedefined, for instance, in terms of employment shifts towards the “stagnant sector” – withbalanced (or constant) aggregate growth The bulk of this literature, which includes

Meckl (2002),Foellmi and Zweimüller (2008)andBonatti and Felice (2008)has relied on

a demand-side explanation for structural change, namely, on non-homothetic (hierarchical)utility functions consistent with Engel’s law – to derive this result.Ngai and Pissarides(2007), andAcemoglu and Guerrieri (2008), however, focus on the supply side

Timmer and de Vries (2009)used part of the Groningen Growth and Development Center(GGDC) data set to investigate structural transformation in Asia and Latin America,concluding that growth accelerations are mostly explained by within-sector productivity

rather than labour reallocation (i.e structural change) They also argue that productivity

growth in market services is more important than productivity improvements in

manufacturing In addition,McMillan and Rodrik (2011)complemented the full GGDC data

set with national sources for China, Turkey and nine African countries (including Ethiopia)

Their main conclusion is that Africa and Latin America have experienced growth-reducing

structural change since 1990 – which largely explains the difference between these regions’

productivity performance and that of Asia This means that labour has moved to

lower-productivity sectors, therefore dampening economic and productivity growth

A sizeable body of literature has attempted to account for the broad set of empirical

regularities that characterize structural transformation, emphasizing shifts across

sectors in output and employment both across countries and over time At the empirical

level, studies on the impact of structural change on aggregate growth are provided by

Dowrick and Gemmel (1991),Bernard and Jones (1996),Broadberry (1997,1998),Foster

Broadberry and Irwin (2006),Nordhaus (2008),Restuccia et al (2008),Dietrich (2009),

Duarte and Restuccia (2010),Jiang (2011),Singh and Cortuk (2011)andHartwig (2012)

Nonetheless, these empirical studies have left many questions unanswered and at best

provided mixed results

3 Overview of structural change and economic growth

The economic performance of these countries can be measured by gross domestic

product (GDP) per capita In 1960, GDP per capita stood at US$200 for Indonesia,

US$812 for Malaysia, US$279 for Nigeria and US$1,582 for Turkey Indonesia had the

least GDP per capita among the four countries with the GDP per capita stood at US$200,

just about 25 per cent of Malaysia’s GDP per capita, 72 per cent of Nigeria’s and just 13

per cent of Turkish’s in 1960 In 1990, Indonesia’s GDP per capita was 23 per cent of

Malaysia’s, 165 per cent of Nigeria’s and 17 per cent of Turkish, and by 2010, it was 22,

212 and 21 per cent of Malaysia, Nigeria and Turkey, respectively

In terms of value-added growth, it shows that manufacturing value added has risen

between 1960 and 2010 in Indonesia, Malaysia and Turkey, while a declining trend can

be examined in the case of Nigeria This means that manufacturing has not been

successful in Nigeria compared to the other three countries It is implied that the

Nigerian economy has not been effectively transformed This is also because

manufacturing sector, which is supposed to be the major driving force of the economy,

is not taking over from agriculture sector both in terms of contribution to GDP

However, as development persists, manufacturing sector generates more employment

before the tertiary sector (service) finally takes over

Figure 1shows the trends of economic growth and sectoral transformation as well as

how structural transformation takes place in these countries At the beginning, the

agricultural sector dominated most of the economies providing the highest contribution

to both output and employment As predicted, as economic growth increases,

agricultural contribution to the economy declines, while that of manufacturing and

services increases with the latter higher than the former in the case of Indonesia,

Malaysia and Turkey Recent research by Maddison (1991)andBuera and Kaboski

(2009) also shows for many countries that structural change involves three distinct

patterns: a decline in agriculture, a rise in services and a hump-shaped pattern in

manufacturing labour

101

Structural change and economic growth

In the case of Nigeria, the trend was distorted with the manufacturing sector unable topick up, while the agricultural sector has started regaining its relevance since 1980s Therising contribution of the agricultural sector in the face of rising agriculturalemployment in Nigeria suggests that there was no increase in productivity, and thechange was due to increase in labour in the sector which the increase in the industrialsector could be attributed to the activities of the oil subsector and not manufacturing.

It has been observed as inFigure 1for these four economies that achieving successfulstructural transformation from traditional sector to modern are those countries with higherlevel of economic growth It can also be observed that structural transformation andeconomic growth are twin sisters However, these observations do not resolve the issues ofcausation, i.e structural transformation causes economic growth or vice versa This isbecause if economic growth is found to be the cause of structural transformation, then therecommendation would be to focus on policies that enhance growth and vice versa

4 Data and methodology

The real GDP per capita (LGDP) and real GDP data are taken from the World Bank’sWorld Development Indicators (WDI) via the World Bank database LOPEN is the sum

of export and import values to GDP ratio and is used as a proxy of openness, andLINVEST is the share of investment per GDP, a proxy of the capital stock in absence ofdata for gross fixed capital formation Data for LOPEN and LINVEST are obtained fromPenn World Table version 7.1 Data for LLABOR are proxy by persons employed (in

0 200 400 600 800 1,000 1,200 1,400

0 10 20

0 5 10 15 20 25 30 35 40 45 50

0 10 20 30 40 50 60 70

0 10 20 30 40 50 60 70

Agriculture Manufacturing Services GDPPC

Note: GDPPC stands for GDP per capita income (right axis) Source: WDI (2011) and Central Bank of Nigeria (2011)

thousands of persons)[1] All data are transformed into logarithmic form except for

structural change indices The detailed description of the data used in the empirical

analysis is shown inTable I

Two different indices for structural change measurements are calculated following

Dietrich (2009)andSingh and Cortuk (2011) The first index is the simplest measure of

structural change, the norm of absolute values (NAV):

NAV ⫽ 0.5兺

i⫽1

n

The computation of the index takes the differences of the sector shares x ibetween two

points in time s and t are calculated Then the absolute amounts of these differences are

summed up and divided by two (as each change is counted twice) The index indicates

the degree of economies resources were reallocated between sectors However, the index

can be influenced by the level of industry aggregation chose for the construction of the

index[2

The second index is the modified Lilien index (MLI) The Lilien (1982) index

originally measured the standard deviation of the sectoral growth rates of employment

from period s to period t that account for the speed of structural change.Stamer (1998)

modified this index to fulfil the characteristics of a metric High values for this indicator

represents fast structural changes and significant reallocation of resources between

sectors The MLI is constructed as follows:

The data used in the calculation of the NAV and MLI for Indonesia, Malaysia and

Turkey are the value added (constant 2000 US$) for the three main sectors, agriculture,

industry and services These data are obtained from the WDI via the World Bank

database, while data for Nigeria are obtained from the annual sector shares of GDP

published in the Statistical Bulletin 2011 of the Central Bank of Nigeria The use of two

indices allows us to check the robustness of our analysis with respect to the structural

change measure

As shown in Figure 2, both indices are highly correlated for all economies The

indices identify periods with high rate of structural change and periods of relative

stability over the past 60 years The rate of structural change across the countries has

been high since at least in late 1960s through to the late 1970s, except for Malaysia onthese two measures partly due to driven by higher investment and growth Nonetheless,the structure of the economy was relatively stable for Indonesia and Nigeria, whereasMalaysia and Turkey experienced continuous structural changes throughout theperiods except for few years due to economic recession.

Most economic time series are non-stationary, and the first step in the estimation is todetermine the presence of a unit root Various tests are currently available to test for unitroots in panel data Therefore, we adopt the recently developed panel unit root testssuggested byLevin and Lin (1992),Im et al (2003),Maddala and Wu (1999),Breitung(2000)andHadri (2000) For the first three tests, the null hypothesis is that the individualtime series in the panel are jointly non-stationary For the Hadri test, the null hypothesis

is that the time series for each cross-section member is stationary around a deterministictrend Before proceeding to the econometric work, we apply the unit root tests to a set ofpanel data for Malaysia, Indonesia, Turkey and Nigeria to examine whether thestructural change indices and other macroeconomic series contains a unit root[3

The IPS panel unit test is essentially a test for a unit root in series, say, y, and has the

where ⌬ is the first difference operator, y it is a white noise disturbance term with

variance␴2 The lagged-dependent variable is included to allow for serial correlation

The null hypothesis of a unit root in the panel is defined as␤i ⫽ 0, for all i To test the

hypothesis,Im et al (2003) propose a standardized ˉt statistic given by:

Following the tests for unit roots is the test for co-integration In an economic sense,

two or more series are co-integrated when there exists a long-term equilibrium to which

they converge over time, and the white noise error can be interpreted as the

disequilibrium error Thus, if series x t and y ty are both stationary in the first differences

(I (1)), and the error term␮tis stationary in the levels (I(0)), then the two series would be

integrated of order CI(1,1)

To examine the possible long-run relationship between the structural change and the

GDP growth, we first employ panel unit root tests which can be arranged in groups by

cross-section dependence and independence, heterogeneous and homogenous unit roots

The seven of Pedroni’s tests are based on the estimated residuals from the following

long run model:

y it⫽ ␣it⫹␦i t⫹␤i x it⫹ ␧it (5)where␧it ⫽ ␳i␧i(t⫺1)⫹ w it are the estimated residuals from the panel regression, i⫽ 1,

… … N represents the panel member, t ⫽ 1 … … T, refers to the time period, y

represents the GDP, x represents the structural change and other explanatory

variables (openness, labour and investment) and␤irepresents the slope coefficient The

parameters␣itand␦iallow for possibility of country-specific effects and deterministic trend

effects, respectively ␧it represents the estimated residual deviations from the long-run

relationship

To test the null hypothesis of no cointegration,␳i⫽ 1,Pedroni (1999,2004) proposed

two types of cointegration tests: panel tests and group tests First, the panel tests based

on the within dimension method (panel cointegration statistics test) which includes four

statistics, namely, panel v-statistic, panel rho-statistic, panel PP-statistic and panel

ADF-statistic Second, the group tests based on the between dimension method (group

mean panel cointegration statistics test) which includes three statistics, namely, group

rho-statistic, group PP-statistic and group ADF-statistic These seven statistics are

asymptotically distributed as standard normal and panel cointegration test statistics

can be found inPedroni (1999, 2004) The statistics can be compared to appropriate

critical values, and if critical values are exceeded, then the null hypothesis of

no-cointegration is rejected implying that a long-run relationship between the variables

does exist

After establishing the panel cointegration, the long-run cointegration vector could

be tested by using many methods, for instance, the panel fully modified ordinary

105

Structural change and economic growth