JournalofQualityandTechnologyManagement

VolumeIX,IssueI,June2013,Page131154

EDUCATION, POVERTY AND ECONOMIC GROWTH

IN SOUTH ASIA: A PANEL DATA ANALYSIS

M. Afzal1, M. Shafiq2, N. Ahmad3, H.M. Qasim4, K. Sarwar5

1,4Department
of Economics, University of the Punjab, Lahore
2Institute
of Quality & Technology Management, University of the
Punjab, Lahore
3Department of Economics, University of the Sargodha, Sargodha
5Department of Economics, Lahore Leads University, Lahore

ABSTRACT

The study investigated the linkages between education, poverty and economic
growth in Bangladesh, India, Pakistan and Sri Lanka. Panel data for the period
of 1995-96 to 2012-13 was used. The findings confirmed positive relationship
between education and economic growth, while poverty is found to be inversely
related to economic growth for these countries of South Asia. The study
recommends that each sample country should adopt poverty reduction and
education enhancing policies to accelerate her economic growth. To compete with
other countries of the region, Pakistan has to raise her average level of education.
These countries should design such policies that engage their population in a
more productive uses so that the economic growth of these countries may further
be enhanced.

Keywords: Growth, Education, Poverty, and Panel Data

Jel Classification: C23, I21, I32 and O40

INTRODUCTION

Education (Edu.) is very important and one of the essential factors for
economic growth (econ. growth). Any country cant attain sustainable
econ. growth and development without intensive and extensive
investment in Edu.. Edu. broadens peoples thinking about themselves
and the rest of the world. It improves living standards of the people and
also increases their productivity and vision. Edu. role is vital to increase
economic proficiency and social uniformity by increasing the value and
productivity of their labor force (LF).

Electronic copy available at: http://ssrn.com/abstract=2814563

Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

Investment in Edu. and expenditures on job trainings were small in 19th

century. During the second half of the 20th century, acquiring Edu. and
skills have become important for an individual as well as for the nations
productivity. Economists have long been aware of the importance of Edu.
in the process of econ. growth and econ. development. Smith (1776)
stressed the importance of Edu. in his book entitled An Inquiry into the
Nature and Causes of the Wealth of Nations. He noted that the
acquisition of Edu. and knowledge becomes fixed capital of a person. A
nation cannot be well developed without investing in Edu.. Edu. is the
first step in the way and process of economic development (econ.
development). It is two-way process, at one side, it enhances the econ.
growth and on the other hand, it reduces poverty (Pov.) by increasing the
productivity. It plays a central role in the accumulation of human capital
abilities and increases econ. growth through better skills and knowledge.

The significance of Edu. and human capital in the way of econ. growth
has been brought out in many studies (Romer 1989, 1990; Denison, 1985;
Islam, Wadood & Tariq, 1995; Afzal, Rehman, Farooq & Sarwar, 2011 and
Afzal, Malik, Begum, Sarwar & Fatima, 2012). The discussion on human
capital was started in the early 1960s. Human capital got importance by
emergence of endogenous growth theories presented by Lucas (1988),
Romer (1989, 1990) and Mankiw, Romer & Weil (1992). In case of
Pakistan, Afzal, Rehman, Farooq & Sarwar (2011) and Afzal, Malik,
Begum, Sarwar & Fatima (2012) explained that the Edu. is one of the
major indicators of human capital. Lucas (1988) developed one of the first
endogenous growth models and said that Edu. is one of the significant
elements that generates the technological development in an economy.
Mankiw, Romer & Weil (1992) first time included human capital in their
production function. They found that the higher level of human capital
leads to the higher rate of econ. growth. Denison (1985) considered Edu.
as a foremost contributor to econ. growth. Economies such as Hong
Kong, Korea, Singapore and Taiwan have attained extraordinary rates of
econ. growth by making huge investment in Edu. (World Bank, 1993).
Krugman (1994) credited much of East Asian tigers achievement to their
swift enlargement of Edu.. Barro (1991) found that schooling was closely
associated with econ. growth.

Educational institutions, investments in Edu., quality Edu. and equal

access to Edu. play the imperative role in the alleviation of Pov. and in
enhancing the econ. growth in the country (Chaudhary & Rehman, 2009;

132|

Electronic copy available at: http://ssrn.com/abstract=2814563

 JournalofQualityandTechnologyManagement

Santos, 2009; Moaz & Neeman, 2008). Edu. also plays a significant role in
the reduction of the income discrepancies (Danacica, Belascu & Llie,
2010). Pov. is also thought to be the basic root cause of terrorism and child
labor. People commit crimes because they are not capable to fulfill their
basic needs of life. Edu. helps to lower the crime rate, terrorism and child
labor through reducing the Pov. (Kruger & Maleckovca, 2003; Veron &
Fabre, 2004).

South Asian countries like Bangladesh, India, Pakistan and Sri Lanka had
failed to sustain their econ. growth due to two main reasons. Firstly, Edu.
is always neglected by the higher authorities and secondly, Pov. is
increased with the passage of time in South Asia. Edu. is triggering econ.
growth through many factors like enhancing the employment
opportunities, improving health facilities, reducing fertility rate,
development of technology and source of political stability. Education
reduces the Pov. rate by generating more employment opportunities.
Edu. is strongly connected with Pov., as parents seem to be unwilling to
send their children for Edu. due to chronic and extensive Pov.

Edu. is considered very important factor for econ. growth but the
relationship between Edu. and econ. growth is not so simple and always
direct (Moroto, 2000: and Afzal, Malik, Begum, Sarwar & Fatima, 2012).
Many other factors may affect the relationship between Edu. and econ.
growth. Pov. has become very crucial variable that directly or indirectly
affects the linkages between Edu. and econ. growth.

Most of the studies have examined the linkage between Edu. and econ.
growth in Pakistan and in other countries based on time series data. There
exists hardly any study in empirical literature that covers the Edu. and
econ. growth relationship in the presence of other variables like Pov. in
panel data analysis environment. Keeping in the view the importance of
Edu. for econ. growth and advantages of using panel data, this study was
design to explore the relationship between Edu. and econ. growth in case
of selected South Asian countries including Pakistan. Pov. is included in
the model as it seems to have its impact on the relationship between Edu.
and econ. growth. Physical capital (PK) and labour force (LF) are also
included in the model because both variables are the very basic
components of econ. growth models.

|133
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

To analyze the nature of relationship among Edu., Pov., and econ. growth
with the inclusion of LF and PK in case of selected South Asian
developing countries is the first objectives of this research work. The
second main objective of this paper is to recommend some policy options
that will lead to promote Edu., accelerate econ. growth and reduce Pov. in
selected countries of South Asian.

The connection among Edu., Pov. and econ. growth is affected by many
other factors like terrorism, inflation, debt accumulation, political
instability, rule of law, trade openness, foreign aid, fertility rate, and
institutional and sociological factors. Due to limited time and data
constraint, the present research work is in the context to only examine the
relationship among three variables i.e., econ. growth, Edu., and Pov. by
including only PK and LF as supporting variables.

REVIEW OF LITERATURE

Edu. and continuous process of econ. growth are considered the most
important sources of high living standards of a nation. Review of studies
that explores the linkages among Edu., Pov. and econ. growth is being
presented:

Afzal, Rehman, Farooq & Sarwar (2011) examined the relationship

between Edu. and econ. growth of Pakistan utilizing time series data for
the period of 1971-72 to 2008-09. Edu. was measured through ten different
indicators in their study. To examine the short-run (SR) and long-run (LR)
relationships, they have applied ARDL Approach to Co-integration. To
check causality, the Toda-Yamamoto (TY) causality technique was used.
The findings of their study confirmed the long run relationship between
Edu., PK, LF and econ. growth in Pakistan. The results of causality
confirmed the bi-directional causality between Edu. and econ. growth.
They recommend more investment in university Edu. that, in turn, leads
to more econ. growth in Pakistan.

Chaudhary, Iqbal & Gillani (2009) investigated the causality between

higher Edu. and econ. Growth for the time period of 1972-73 to 2005-06
for Pakistan. The results of their study confirm the LR relationship
between Edu., PK, LF and real GDP. They find unidirectional causality
that runs from real GDP to higher Edu.. Chaudhary, Iqbal & Gillani (2009)
used university enrollment as a measure of higher Edu.. But the definition

134|
 JournalofQualityandTechnologyManagement

of higher Edu. in Pakistan is little bit different. In Pakistan, higher Edu.

includes college Edu. and university Edu.. If proper proxy of higher Edu.
was used, then it may possible that higher Edu. may cause also to econ.
growth.

Afzal, Farooq, Ahmad, Begum, & Quddus (2010) analyzed the SR and LR
relationship between school Edu. and econ. growth in Pakistan. Time
series data for the period of 1970 to 2008 was used. The SR results of
ARDL approach confirmed the inverse relationships between school Edu.
and econ. growth. Direct relationship was found between school Edu. and
econ. growth in the LR. Inflation was found to hurt the econ. growth both
in the SR as well as in the LR. Their study recommended that the
reduction in inflation and Pov. was necessary to enhance the school Edu.
and econ. growth for Pakistan.

Permaani (2008) explored the determinant of econ. growth in East Asian

countries. He used panel data set for the period of 1965-2000. Schooling
years as a measure of human capital was used. He estimated the Labour
Augmented Solow Model. Human capital proved itself a significant
determinant of econ. growth according to the results of Permaanis study.

Francis & Iyare (2006) checked the causality between Edu. and econ.
development in the Caribbean. They used Vector Error Correction Model
(VECM) by using time series data from 1964-65 to 1998-99. The results
confirmed bidirectional causality between Edu. and income in the SR, but
no causality was found between Edu. and income in the SR and LR in
Barbados, Trinidad and Tobago. Their study suggested more income
spending on Edu. for more econ. development.

Islam, Wadud & Islam (2007) checked causality between Edu. and econ.
growth in Bangladesh by using data from 1976 to 2003. They have used
multivariate approach rather than bivariate approach in their analysis.
They found bidirectional causality between econ. growth and Edu. and
confirmed the LR relationship between Edu. and econ. growth.

Podreca & Carmeci (2004) explored the linkage between Edu. and econ.
growth. They used panel data set of five year moving average for the time
span of 1960-1990 by taking 81 developed and least developed sample
countries. The average year of schooling was used as proxy of Edu.
Generalized Method of Movement (GMM) was applied for empirical

|135
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

analysis. They found positive impact of average year of schooling on

econ. growth in case of all sample countries.

Danacica, Belascu & Llie (2010) used time series data for the span of 1980-
2008 to explore the causal nexus between higher Edu. and econ. growth in
case of Romania. The results of their study confirmed that there was a LR
relationship between higher Edu. and econ. growth. One way causality
running from econ. growth to higher Edu. was also found. However, this
study faced a serious drawback. They used Johansen & Juelius (1990,
1995) co-integration technique on just 28 observations and the optimal
lag-length was four. The estimated results may mislead and data may face
the loss of degree of freedom due to short data span.

Kakar, Khilji & Khan (2011) have examined the LR relationship between
Edu. and econ. Growth for Pakistan. They used educational expenditure
as a proxy of human capital and time series data for the period of 1980-
2009. The results of their study confirmed the LR relationship between
Edu. and econ. growth. They also found that the PK and LF helped to
enhance the econ. growth with Edu.. Their study recommends that the
government should make such policies that help to improve the quality of
Edu. in order to accumulate human capital for econ. development.

Barro & Lee (2010) investigated the linkage between econ. growth and
Edu. in case of 146 countries by utilizing large panel data set. GMM
system estimator proposed by Areccano & Bovver (1995), and Blundell &
Bond (1998) was applied for analysis. Average year of schooling of
working age population and educational attainment of worker at
different level of Edu. were used as measures of human capital. The result
of their study showed that the schooling had a positive effect on econ.
growth in all sample countries.

Veron & Fabre (2004) explored the effect of Pov. and educational policies
on child labour, econ. growth and school attendance. Tradeoff between
human capital accumulation and child labour has been found. Pov. trap
may occur when parents are not choosing quality of Edu.. Further, they
said that public Edu. system more easily generated Pov., while the private
Edu. system improved the growth through quality of Edu. and lesser
chances of Pov. trap.

136|
 JournalofQualityandTechnologyManagement

Chaudhary & Rehman (2009) explored the role of Edu. in reducing the
Pov. in Pakistan by using time series data from 1972 to 2007. The results
of this study confirmed that the primary and middle Edu. were positively
but insignificantly related to Pov., while university Edu. was negatively
and significantly related to Pov.. Their results also confirm the negative
but insignificant relationship of econ. growth with Pov.. This study is
based on time series data but does not explore the problem of stationarity.
On the other hand, the results of Autoregressive regression and
correlation matrix do not support each other and multicolinearity is also
seen in their model.

Rahman & Peng (2012) investigated the LR relationship between different

level of Edu. and econ. growth in case of Bangladesh and China by using
time series data from the period of 1980-2009. Co-integration, ECM and
Granger Causality were applied for empirical analysis. Real GDP, gross
enrollment at primary, secondary and tertiary levels of Edu. were the
variables of their study. The findings of Rahman & Peng (2012) showed
that the primary and tertiary level of Edu. have positive effect on econ.
growth of Bangladesh. In case of China, secondary enrollment and
tertiary enrollment had significant positive effect on econ. growth.
Unidirectional causality existed from econ. growth to higher Edu. and
from secondary to primary Edu. in case of Bangladesh. In China,
unidirectional causality existed from primary and secondary Edu. to econ.
growth and bidirectional causality between primary and secondary Edu..

Shahbaz, Iqbal & Butt (2011) explored the causality between human
development and econ. growth by using large panel data set of ten Asian
developing countries i.e., Pakistan, Bangladesh, India, Indonesia,
Malaysia, Korea, Philippines, Singapore, Sri Lanka and Thailand. The
results showed homogeneous causality running from human
development to econ. growth. A heterogeneous causality has found from
econ. growth to human development. Only Korea and Singapore have
bidirectional causality between human development and econ. growth.
The causality was not found between human development and econ.
growth in Bangladesh and Philippine,.

Afzal, Malik, Begum, Sarwar & Fatima (2012) for Pakistan utilized the
time series data on Edu., Pov., real GDP and PK for the span of 1971-72 to
2009-10. The findings of their ARDL model confirmed positive and
significant effect of PK on econ. growth both in the SR and LR. Edu.

|137
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

affected the econ. growth positively and significantly only in the LR. Pov.
and econ. growth were inversely related to each other only in the LR. The
results of Toda-Yamamoto Augmented Granger Causality (TYAGC) Test
confirmed the bidirectional causality between Edu. and econ. growth,
between econ. growth and Pov. and between Pov. and Edu.. They
recommended that Pov. reduction and Edu. enhancing strategies must be
adopted to accelerate the process of econ. growth of the country.

Dauda (2009) examined the LR relationship between Edu. and econ.

growth with inclusion of PK and LF. Johansen Cointegration and Error
Correction Model (ECM) were applied for the period of 1997-2007 in case
of Nigeria. Government expenditure on Edu. was used as proxy of Edu..
Empirical results confirmed the LR relationship between Edu., econ.
growth and gross capital formation. The study also found that the Edu.
and PK have positive and significant impact on econ. growth, but LF has
insignificant impact on econ. growth. Dauda (2009) claimed that
educational investment positively correlated with the econ. growth with
inclusion of a PK and LF.

Babalola (2011) explored the SR and LR relationship and causality

between Edu. and econ. growth using time series data from 1977-2008 in
case of Nigeria. Investment in Edu. was used as a proxy variable for
measuring Edu.. Johansen co-integration formulated by Johansen &
Javelins (1992) for LR relationship and Grange causality test (1996) for
causality were employed. He found co-integration and SR relationship
between Edu. and econ. growth in Nigeria. The results of the study also
showed the unidirectional causality running from econ. growth to Edu..

Solaki (2006) investigated the LR and SR relationship between human

capital and econ. growth for Greece by utilizing time series data from
1961-2006. Primary, secondary and tertiary Edu. was used as a proxy of
human capital. Johansen co-integration, Error Correction Model and
Grange causality tests were used. He found the LR and SR relationship
between Edu. and econ. growth. Higher Edu. and government
expenditure on Edu. cause the econ. growth. He found that in case of
primary and secondary Edu. causality running from econ. growth to
Edu.. He suggested that the government should increase investment in
Edu., particularly in higher Edu. for econ. growth.

138|
 JournalofQualityandTechnologyManagement

Chandara (2010) examined causality between government expenditure on

Edu. and econ. growth for India for the period of 1951-2009. Linear and
nonlinear Granger causality tests (1969) were employed. Statistical
technique constructed by Back & Brock (1992) and modified by Hoekstra
& Jones (1994) was used to check causality between Edu. and econ.
growth. The empirical results confirmed bi-directional causality between
econ. growth and Edu..

Brempong, Paddison & Mitiku (2006) analyzed the relationship between

higher Edu. and econ. growth in African countries. Panel data set for the
time span of 1960-2000 was used. Enrollment at higher Edu. and
educational attainment were used to measuring human capital.
Augmented Neoclassical Growth model of Mankiw, Romer & Weil
(1992) was employed for estimation. The empirical results of their study
confirmed positive significant impact of higher Edu. on econ. growth.

Hassan & Ahmed (2006) examined the relationship between Edu. and
econ. growth in 39 Sub Saharan African countries. Panel data for the
period of 1975-2005 was used. Primary and secondary school enrollments
ratios were used as a measure of Edu.. The augmented Solow model
proposed by Mankiw, Romer & Weil (1992) was applied for analysis. The
result of their study reveals that the primary and secondary school
enrollments ratios have positive impact on econ. growth.

Keeping in view the above discussion, the primary objective of the

present research work was to examine the role of Edu. for econ. growth
and vice versa with the inclusion of PK, LF and Pov. for Bangladesh,
India, Pakistan and Sri Lanka. The present study differs from most of the
above mentioned previous studies in the following respects: (a) this study
applies more suitable econometric technique i.e., Panel data analysis to
check the robustness of the results. (b) the present study also used more
comprehensive measures in its analysis i.e., Gross Enrollment Index (GEI)
for measuring Edu., Human Pov. Index (HPI) for measuring Pov. and
GDP index for measuring econ. growth, while most of the previous
studies on the relationship among Edu., Pov. and econ. growth have used
naive measures of Edu., Pov. and econ. growth in their analysis. There
exists hardly any study in literature that examined Edu. econ. growth
relationship in the presence of other variables like PK, LF and Pov. using
panel data approach in case of South Asian region.

|139
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

DATA SOURCES AND METHODOLOGY

The reliability of empirical results always depends upon reliable data,

data frequency, data span, data sources and the methodology used in the
analysis. This section includes data sources and methodology that have
been used in the analysis.

Data Sources

The present study used annual time series data from 1995-96 to 2012-13 in
case of Bangladesh, India, Pakistan and Sri Lanka. The data on the
variables gross domestic product per capita (GDPpc), Edu., PK, LF and
Pov. have been drawn from World Development Indicators (WDI) and
United Nations Development Program (UNDP).

Methodology

To examine the relationship among econ. growth, Edu., and Pov., the
Fixed Effects Model(FEM) was applied to estimate the following
equation:

GDPIi,j = 0 + 1PKij + 2LFij + 3 Eduij +4 HPIij + ij (Model 1)

Where:

GDPIij = GDP index

PKij = Physical capital
LFij = Labour force
Edu.ij = Education. Education was measured by two indicators i.e., Gross
Enrolment Index (GEI) and Education Index (Edu.Index)
HPIij = Human Poverty Index
ij = Error term

GDP Index (GDPI)

The gross domestic product (GDP) represents the market value of all final
goods and services that are produced within the boundaries of a country
in a year. GDP captures the overall wellbeing of an economy. The present
research work used a more comprehensive measure of econ. growth i.e.,
GDP index (GDPI), rather than using GDP or real GDP to measure econ.

140|
 JournalofQualityandTechnologyManagement

growth. The present study has developed GDPI by using the 2000 UNDP
methodology as:

Where:

GDPpc = The GDP per capita.

max = The highest value of the GDPpc
min = The minimum of the GDPpc

In the above case, the maximum and minimum values of GDPpc are
40000 and 100, respectively.

Gross Enrollment Index (GEI)

Edu. is one of the most important sources of econ. Growth and

development. This study used a relatively better measure of Edu. i.e.,
Gross Enrollment Index (GEI). GEI was constructed by combining
primary, secondary, and tertiary gross enrollment ratios (GER). GER was
an indication of different levels of Edu.. The GEI was calculated by using
2000 UNDP methodology as follows:

Where:

GER = The gross enrollment ratio

max = The maximum value of the GER
min = The minimum of the GER.

In this case, the maximum and the minimum values of GER are 100and
0, respectively.

|141
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

Education Index (Edu. Index)

Edu. Index, a key indicator that is utilized as a proxy to measure

educational development of a country. Edu.Index was constructed by
using 2000 UNDP methodology. In Edu.Index, adult literacy rate (ALR)
with two-thirds weighting and the combined primary, secondary, and
tertiary gross enrollment ratio (GER) with one-third weighting were
added together. This measure for education in case of Pakistan was
already used by Afzal, Rehman, Farooq and Sarwar (2011) and Afzal,
Malik, Begum, Sarwar & Fatima (2012).

Afzal, Rehman, Farooq and Sarwar (2011) constructed education index as

Education index is constructed by adding together adult literacy rate
index (ALI) with two-third weightage and the combined primary,
secondary, and tertiary gross enrollment ratio index (GEI) with one-third
weightage. The adult literacy rate (ALR) gives an indication of the ability
to read and write, while the gross enrolment ratio (GER) gives an
indication of the level of education.

where 0 and 100 are the minimum and maximum values of the
variables adult literacy rate and gross enrollment ratio (p. 323).

Physical Capital (PK)

More savings leads to more capital that, in turn, was used for investment
purposes (Solow, 1956). A significant role of PK is recognized in the
process of econ. growth and hence development. PK was measured by the
rate of growth of gross fixed capital formation.

142|
 JournalofQualityandTechnologyManagement

Labour Force (LF)

A countrys skilled LF helps in her eco. growth as well as development

process. LF was measured by growth rate of population in the present
research work. It is one of the important components of production
function (Hall, 2000).

Human Poverty Index (HPI)

Pov. being an important variable directly or indirectly affects each of Edu.

and econ. growth. This research work used a better comprehensive
measure of Pov. known as Human Poverty Index (HPI). HPI is an
indication of the standards of living in an economy. The data for the
variable HPI has been taken from UNDP to complement the Human
Development Index (HDI). HPI is considered as a better reflect of the
extent of deprivation as compared to the HDI.

Advantages of using Panel Data

Studies on cross sectional and time series data have faced certain
problems and limitations. To avoid such limitations of each cross section
and time series data, the present study utilizes panel data to analyze the
trends of different cross section over time. The advantages of using
panel data can be summarized as follows: (a) they provide more efficient
estimations of parameters by considering broader sources of variations,
(b) they outsource more information to the analyst, and (c) they allow the
study of the dynamics behavior of the parameters (d) Pooling increases
the number of observations. (e) Panel data permits controlling for some
types of unit heterogeneity. (f) Panel data allows testing theories that
make forecasts in space and time. (g) Panel data allows to explicitly
modeling dynamics. Accordingly, we use panel data to

a) get rid of omitted variable bias

b) make the best of the available information
c) test theories that predict changes
d) Test theories that predict parameter heterogeneity. (Asteriou,
2006, p. 268).

|143
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

Fixed Effects Model (FEM)

FEM was used to explore the relationship between explained and

explanatory variables within an entity. FEM assumes that some factors
within the entity may have impact on explained variables. FEM controls
these factors by introducing dummy variables for time invariant
characteristics i.e., colonial origin, religion and race etc. Time invariant
features are unique and exclusive for each cross section. It is not
correlated with other individual characteristics. So, each entity is
different, therefore, error and intercept terms of each entity is not
correlated with others. FEM has constant slopes and different intercept
term for each cross section unit. It can also be said that Fixed Effects
Estimator treated the entity specific or group specific. This means that it
allows the different constant for each entity. The equation of FEM is
presented as follow:

Where:

Y = Dependant variable
= Intercept for each cross section unit
X = Explanatory variable
i = Each cross section entity or unit
j = Time period
u = Error term or disturbance term

The validity of the variables used in FEM can be checked through

performing the F-test. The hypothesis of F-test can be written as:

Baltagi (2005) said that it was simply the Chow test with Restricted
Residual Sum of Square (RRSS) that is obtained from OLS regression. The
Unrestricted Residual Sum of Square (URSS) is obtained by Least Squares
Dummy variables of Fixed Effects Regressions. F-test follows the Chi
square distribution. The rejection of the null hypothesis (H0) tells that the
estimation of FEM is consistent and efficient. F-test is calculated by using
the following formula:

144|
 JournalofQualityandTechnologyManagement

Where:

RRSS = Restricted Residual Sum of Squares

URSS = Unrestricted Residual Sum of Squares
N = Number of Cross Sections
K = Number of parameters to be estimated
T = Time period

EMPIRICAL RESULTS AND THEIR INTERPRETATION

This section presents empirical results and their interpretation.

Descriptive statistics

Descriptive statistics involved transformation of raw data in to a form

that would provide information to describe a set of factors in a situation.
This is done through ordering and manipulation of collected raw data. At
preliminary level of a research, it is of great interest to know frequently of
certain phenomenon occurs, the mean, the median, the maximum and
minimum values, as well as the extent of variability in the data set i.e.,
standard deviation, and histograms of the dependent and independent
variables. Descriptive statistics of the variables used in the analysis are
given in Table 1.

Table 1: Descriptive statistics

Mean Median Maximum Minimum

Overall (for all selected countries)
GDPI 52.85 53.00 67.00 35.00
PK 11.09 8.54 28.04 -11.96
LF 1.48 1.46 2.66 0.44
GEI 53.16 54.00 71.00 34.95
Edu.Index 59.34 54.00 87.00 38.00
HPI 31.30 33.10 46.00 17.70
Bangladesh
GDPI 43.81 44.50 50.00 35.00

|145
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

Mean Median Maximum Minimum

PK 8.40 8.38 12.06 1.80
LF 1.60 1.70 2.09 1.05
GEI 46.46 48.70 57.00 35.00
Edu.Index 45.56 45.00 54.00 38.00
HPI 41.21 42.50 46.00 34.90
India
GDPI 54.84 56.00 62.00 45.00
PK 9.34 8.00 18.85 -0.01
LF 1.50 1.43 1.78 1.30
GEI 58.91 61.00 63.80 52.12
Edu.Index 60.78 61.00 67.00 53.00
HPI 31.05 31.30 36.90 26.50
Pakistan
GDPI 51.26 52.00 56.00 46.00
PK 2.34 2.34 19.90 -11.96
LF 2.00 1.81 2.66 1.50
GEI 39.49 40.00 43.00 34.95
Edu.Index 46.15 46.00 53.00 38.00
HPI 36.77 36.30 43.00 30.00
Sri Lanka
GDPI 61.21 63.00 67.00 52.00
PK 24.18 24.18 28.04 20.04
LF 89.32 92.00 1.24 44.00
GEI 67.34 68.70 71.00 62.70
Edu.Index 83.43 83.00 87.00 81.00
HPI 17.28 17.60 21.00 17.70

The descriptive statistics given in Table 1 reveals that the overall average
score of GDPI is 52.85. In terms of GDPI, Bangladesh and Pakistan
perform below the overall averages score, while Sri lank and India
remains above the overall average. Regarding GEI, Sri Lanka is on the top
with score of 67.34. The overall average score of GEI is 53.16. Pakistan is
at the lowest among the selected countries in terms of GEI score. Pakistan
is in dire need to raise her average score of GEI. The mean value of HPI is
31.30. Sri Lanka performs well and got the mean HPI score of 17.28, while
Bangladesh is at the top with 41.21 mean score of HPI.

The results of the extent of relationship between variables are presented

in Table 2. The correlation results given in Table 2 depict the positive
correlation between GEI and GDPI, between Edu.Index and GDPI with
the value of 0.72 and 0.81, respectively. The correlation between PK and

146|
 JournalofQualityandTechnologyManagement

GDPI is found to be 0.53, while LF is correlated with GDPI with the value
of 0.54. HPI and GDPI are correlated with a value of 0.86.
F-test is the choice between Fixed Effects Estimation Method (FEM) and
Pooled Ordinary Least Squares Method (Pooled OLS). If the value of F-
test is significant then Fixed Effects gives the better results than that of
the Pooled OLS. The result of F-test, when education was measured by
GEI is presented in Table 3(a) and the result of F-test, when education
was measured by Edu.Index is presented in Table 3(b).

Table 2: Cross Correlation

GDPI GEI Edu.Index HPI PK LF

GDPI 1.00
GEI 0.72 1.00
Edu.Index 0.81 0.91 1.00
HPI 0.86 0.82 0.97 1.00
PK 0.53 0.71 0.77 0.71 1.00
LF 0.54 0.79 0.84 0.78 0.69 1.00

Table 3(a): F-Test estimates, when education is measured by GEI for specification 6

Effects Test Statistic d.f. p-value

Cross-section F 6.666 (3,56) 0.000
Cross-section Chi-square 19.542 3 0.000

Table 3(b): F-Test, when education is measured by Edu.Index for specification 12

Effects Test Statistic d.f. p-value

Cross-section F 7.989 (3,68) 0.001
Cross-section Chi-square 22.946 3 0.000

As the value of F-test is significant in Table 3(a) and 3(b), the model 1 is
estimated by applying Fixed Effects Method (FEM). The results of FEM of
model 1 are present in Table 4.

The results of twelve different specifications of model 1 are presented in

Table 4. Edu. is measured by its two indicators i.e., GEI and Edu.Index.
Edu. was measured by GEI in first six specifications, while Edu.Index was
used to measure Edu. in next six specifications. FEM is used to control the
country specific differences. The positive and significant effect of PK and

|147
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

Edu. on GDPI is found. PK and Edu. level of these countries may further
improved to further accelerate GDPI of these countries. HPI is found to be
inversely related to GDPI. This may be due to the fact that it is very hard
to produce more in the presence of Pov.. Pov. seems to be very extensive
in the countries being studied that, in turn, creates hindrance in the way
of development of these countries. The effect of LF on GDPI is found
positive but insignificant in most of the specifications including our
preferred specification i.e., specification 6. It may be the outcome that the
population of these countries may not seem to be much more productive.
Pakistan, India and Bangladesh are the top five populous countries of the
world.

The goodness of fit is well established in all twelve specifications as the

value of F statistic in all specifications in Table 4 is highly significant. A
large proportion of variation in GDPI is being explained by GEI or
Edu.Index, PK, LF, and HPI as the value of R2 and adj.R2 ranges from 78
to 86 percent and 77 to 85 percent, respectively. The F-value increases
remarkably in specifications 6 as compared to that found in the case of
other specifications, so specification 6 is our preferred specification
among all. The F-value consistently rejects the null hypothesis that all the
added explanatory variables in different specifications are jointly equal to
zero.

148|
 |149
JournalofQualityandTechnologyManagement

Dependent variable: Economic Growth measured by GDPI

1 2 3 4 5 6 7 8 9 10 11 12
Table 4: Fixed Effects Estimates of Model 1

0.05 0.005 0.005 0.003 0.005 0.004

GEI
(0.000) (0.000) (0.0167) (0.000) (0.012) (0.001)
0.562 0.567 0.830 0.229 0.872 0.558
Edu.Index
(0.001) (0.000) (0.012) (0.011) (0.009) (0.081)
0.007 0.005 0.001 0.001 0.001 0.001
PK
(0.100) (0.034) (0.007) (0.007) (0.017) (0.035)
0.015 0.014 0.039 0.052 0.058 0.054
LF
(0.727) (0.744) (0.213) (0.346) (0.290) (0.240)
0.006 0.007 0.005 0.004
HPI
(0.006) (0.000) (0.023) (0.014)
R2 0.78 0.79 0.79 0.84 0.79 0.86 0.81 0.81 0.82 0.81 0.83 0.84
Adj.R2 0.77 0.77 0.77 0.83 0.77 0.85 0.79 0.79 0.81 0.81 0.81 0.82
65.83 52.46 52.88 78.12 43.76 164.93 74.04 59.23 65.38 63.59 55.93 50.69
F-value
(0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

CONCLUSION AND RECOMMENDATIONS

Edu. in each and every sense is very important for accelerating econ.
growth and reducing Pov.. Edu. helps in reducing Pov. and improving
the socio-economic indicator of the country.

CONCLUSION

The present research was designed to explore the linkages between

education, poverty and econ. growth in selected South Asian countries by
utilizing panel data for the period of 1995-96 to 2010-11. Fixed Effects
Estimation Technique(FEM) was applied in the analysis. The FEM has an
advantage because it control for the country-specific characteristics that
are time invariant. Overall i.e., for all selected countries, the average score
of GDPI has been found to be 52. The average GDPI score of Bangladesh
and Pakistan has been found to be below than that of the overall averages
score of GDPI of these countries. The average score of GDPI for Sri lank
and India is found to be above than that of the overall averages score of
GDPI. Pakistan is at the lowest in terms of her average GEI score.
Pakistan is in dire need to raise her average score of GEI to compete with
other countries of the region.

The effect of physical capital and education has been found to be positive
and statistical significant on econ. growth of these countries. Poverty has
significant negative effect on econ. growth, while the population has
negative but insignificant effect on econ. growth of these countries.
Poverty seems to be very extensive in the countries being studied. The
population of these countries seems to be less productive.

RECOMMENDATIONS

On the basis of discussions and findings, this study recommends that:

a. Each sample country, especially Pakistan should give more priority to

enhance the education level of their masses so that the econ. growth of
these countries, especially the econ. growth of Pakistan may be further
accelerated.
b. The investment in education provides at least constant returns to scale
and helps in achieving sustainable econ. growth. The low productivity

150|
 JournalofQualityandTechnologyManagement

growth is mainly attributed to the insufficient and low quality

education of the labor force. The worker training programs should be
promoted by increasing investment in education.
c. Poverty reduction policies must be the part and parcel of every
economic policy and plan of these countries. Government of these
countries should make such policies that reduce poverty both in the
short run and long run.
d. These countries should reduce their population or must introduce
such policies that engage their population in a more productive uses,
so that inactive population may positively contribute in the way of
econ. growth and hence econ. development of these countries.
e. This study also recommends that the linkages among education,
poverty and econ. growth by including other factors other than
physical capital and labour force may further be checked and
generalized.

REFERENCES

Afzal, M., Farooq, M. S., Ahmad, H. K., Begum, I & Quddus, M. A. (2010).
Relationship between school education and economic growth in
Pakistan: ARDL Bounds Testing approach to co-integration.
Pakistan Economic and Social Review, 1(48), 39-60.
Afzal, M., Rehman, H., Farooq, M, S., & Sarwar, K. (2011). Education and
economic growth in Pakistan: A cointegration and causality
analysis. International Journal of Educational Research, 50, 321-335.
Afzal, M., Malik, M., E., Begum, I., Sarwar, K., & Fatima, H. (2012).
Relationship among education, poverty and economic growth in
Pakistan: An econometric analysis. Journal of Elementary Education,
22(1), 23-45.
Babalola, S. J. (2011). Long-run relationship between education and
economic growth: evidence from Nigeria. International Journal of
Humanities and Social Science, 1(14), 123-128.
Baltagi, H. (2005). Econometric analysis of Panel data (3rd Edition). John
Wiley and Sons Inc., USA.
Barro, R. J. (1991). Economic growth in a cross section of countries.
Quarterly Journal of Economics, 166(2), 407-444.
Barro, R. J. & Lee, J. W. (2010). A new data set of educational attainment
in the world, 19502010, NBER working papers 15902, National
Bureau of Economic Research, Inc.
Barro, R. J. (1996). Democracy and growth. Journal of Economic Growth,

|151
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

1(1), 1-27.
Bashir, S., Herath, J. & Gebremedhin, T. (2012). An empirical analysis of
higher education and economic growth in West Virginia. Paper
presented at Agricultural and applied economics association
annual meeting Seattle, WA, August 12-14, 2012.
Brempong, K. G., Paddison, O. & Mitiku, K. (2006). Higher education and
economic growth in Africa. Journal of Development Studies, 42(3),
509-529.
Chandra, A. (2010). Does government expenditure on education promote
economic growth? An econometric analysis, MPRA Paper No.
25480, posted 27. September 2010 / 13:18 http://mpra.ub.uni-
muenchen.de/25480/.
Chaudhary, A. R., Iqbal, A. & Gillani, S. Y. M. (2009). The nexus between
higher education and economic growth: An empirical
investigation for Pakistan. Pakistan Journal of Commerce and Social
Sciences, 3(1), 1-9.
Chaudhry, I. S., & Rahman, S. (2009). The impact of gender inequality in
education on rural poverty in Pakistan: An empirical analysis.
European Journal of Economics, Finance and Administrative Sciences,
15, 174-188.
Danacica, D., Belacu, L. & Llie, L. (2010). The interactive causality
between higher education and economic growth in Romania.
International Review of Business Research Paper, 4(6), 491500.
Dauda, R., O., S. (2009). Investment in education and economic growth in
Nigeria: A co-integration approach. Paper presented at 9th Global
Conference on Business and Economics, University of
Cambridge, UK, October 16-17, 2009.
Denison, E. F. (1985). Trends in American economic growth, 1929-1982.
Broking Institutions, Washington, D.C.
Francis, B. & Iyare, S. (2006), Education and development in the
Caribbean: A cointegration and causality approach. Economics
Bulletin, 16, 1-12.
Goode, R. B. (1959). Adding to the stock of physical and human capital.
American Economic Review, 49(2), 147-155.
Hakim, R. A., Razak, N. A. A. & Ismail, R. (2010). Does social capital
reduce poverty? A case study of rural households in Terengganu,
Malaysia. European Journal of Social Science, 4(14).
Islam, T. S., Wadud, M. A. & Islam, Q. B. T. (2007). Relationship between
education and GDP growth: A multivariate causality analysis for
Bangladesh. Economics Bulletin, 3(35), 1-7.

152|
 JournalofQualityandTechnologyManagement

Kakar, Z. K., Khilji, B. A., & Khan, M. J. (2011). Relationship between

education and economic growth: A time series analysis. Journal of
International Economic Research, 11(1), 27-32.
Krueger, A. B. & Maleckova, J. (2003). Education, poverty and terrorism:
Is there a causal connection? Journal of Economic Perspectives, 4(17),
119144.
Krugman, P (1994), The Myth of Asias Miracle. Foreign Affairs, 73(6), 62-
78.
Lucas, R. E. (1988). On the mechanics of economic development. Journal of
Monetary Economics, 22, 3-42.
Mankiw, N. G. Romer, D. & Weil, D. N. (1992). A contribution to the
empirics of economic growth. Quarterly Journal of Economics, 107,
407-37.
Moav, O. & Neeman, Z. (2008). Bling Bling, human capital, and poverty.
Paper presented at European Summer Symposium in
International Macroeconomics (ESSIM), Tarragona, Spain, 20-25
May 2008.
Morote, E.S. (2000). Higher education, employment and economic
Growth: Mexico and Peru. Retrieved from
www.americanprofessor.org/documentation/ lasapaper.doc.
Norton, S. W. (2010). Human capital, institutions and world poverty.
Paper presented at Western Economics Association, International
Meeting, Vancouver, B.C., July 2004.
Parmani, R. (2009). The role of education in economic growth in East Asia:
A survey. Asia-Pacific Economic Literature, 23(1), 1-20.
Podreca, E. & Carmeci, G. (2002). Does education cause economic growth?
Working Papers Dises No. 96B, University of Trieste.
Raja, N. (2005). Humanization of education in Pakistan through Freires
concept of literacy. Asia Pacific Education Review, 1(6), 1-6.
Reham, M. M., & Peng, L. (2012). Long run effect of education on GDP:
An empirical evidence from BCIM Countries. International Research
Journal of Finance and Economics, 82, 95-106.
Romer, P. M. (1989). Human capital and growth: Theory and evidence.
NBER Working Papers 3173, National Bureau of Economic
Research, Inc.
Romer, P. M. (1990). Endogenous technological change. Journal of Political
Economy, 5(98), 71-101.
Saloki, M. (2006). Relationship between education and GDP growth: A bi-
variate causality analysis for Greece. Harokopio University,
Athens Greece.

|153
Education,PovertyandEconomicGrowthinSouthAsia:APanelDataAnalysis

Shahbaz, M., Iqbal, A. & Butt, M.A. (2011). Testing causality between
human development and economic growth: A panel data
approach, International Journal of Education Economic and
Development, 2(1), 90-102.
Smith, A. (1776). An Inquiry into the nature and Causes of wealth of
nations. (Indianapolis: Liberty Fund, 1981), ed. by R. H. Campbell,
Andrew S. Skinner, and W. B. Todd, contrib. by Edwin Cannan1,
265-266.
Solow, R. M. (1956). A contribution to the theory of economic growth.
Quarterly Journal of Economics, 70, 65-95.
UNDP, Human Development Report, (various issues), New York: Oxford
University Press.
Veron, E. A. & Fabre, A. (2004). Education, poverty and child labour.
Econometric Society 2004. Far Eastern Meetings 738, Econometric
Society.
World Bank (1993). The East Asian Miracle: Economic growth and public
policy, New York: Oxford University Press.
World Bank, World Development Indicators (WDI) (Various issues).

154|