Digital Divide and Its Impact on Economic Growth in SAARC Countries

7/30/2019 Digital Divide and Its Impact on Economic Growth in SAARC Countries

1/14

Shamim & A. Mamun / IJAR-BAE, 02, 02 (2013) 14 26

www.setscholars.org/index.php/ijarbae

P a g e

1 4

V V oo lluu mm ee 00 22 ,, 22 00 11 33 IIss ss uu ee 00 22

SETScholarswww.setsscholars.org

I n t e r

n a t i o n a

l J o u r n a

l o

f A p p

l i e d

R e s e a r c

h i n

International Journal of Applied Research in

ISSN: 1839-8456Perth, Australia


EE xxeeccuu tt iivvee EE dd iitt oorr -- iinn --CC hh iiee f f

Md. Al MamunCC hh iieef f ,, EE dd iitt oorr iiaa ll BBooaa rr dd Md. Abdul Hannan Mia, PhD

EE dd iitt oorr iiaa ll BB ooaa rr dd MM eemm bb eerr ss Shaokun Coral Yu, PhD

Northern Illinois University, USA Dr. Iqbal Hossain, Post-Doc in Econometrics,Imperial College, UK M Kabir Hassan, PhD

University of New Orleans, USA Md. Abdul Hannan Mia, PhD

University of Dhaka, Bangladesh &University of Toronto, Canada .

Md. Shariat UllahUniversity of Dhaka, Bangladesh &

Ritsumaiken University, Japan .Saalem SadequeUniversity of Western Australia, Australia .Rami M. Ayoubi, PhDDamascus University, Syria Ondrej Zizlavsky PhDBrono University of Technology, Czech

Republic Rarshid Mehrdoust, PhDUniversity of Guilan, Iran

Md. Moazzam Hossain

Curtin University, Australia Jai Ganesh PhDInfosys Labs, Bangalore, India

Md. Mohan Uddin, PhDUniversity Utara Malaysia, Malaysia

Mahfuja Malik University of Boston, USA

Md. Anisur RahmanUniversity of Dhaka, Bangladesh .

Md. Mamun Habib, PhD, AIUB, Bangladesh Kazi Sohag, UKM, MalaysiaFazeela Jameel Ahsan, PhDUniversity of Colombo, Sri Lanka

Rajeshkumar U. Sambhe, PhDJawaharalal Darda Institute of E&T, India.

V V oo lluu mm ee 00 22 ,, 22 00 11 33 IIss ss uu ee 00 22

EE dd ii tt oorr iiaa ll CC oomm mm eenn tt

SSttr r uuccttuur r iinngg LLaatteenntt N N aattuur r ee oof f PPllaannnniinngg CCoomm p peetteenncciieess oof f BBuussiinneessss OO p peer r aattoor r ss aanndd iittss IImm p paacctt oonn BBuussiinneessss PPeer r f f oor r mmaannccee:: EEvviiddeennccee f f r r oomm SSr r ii LLaannk k aa

I I J J A A R R-- B B A A E E , , 0022 , , 0022 ( ( 22001133 ) ) 0011 1133.. K K eenn g g aa t t hhaa r r aa nn.. N N

DDiiggiittaall DDiivviiddee aanndd IIttss IImm p paacctt oonn EEccoonnoomm iicc GGr r oowwtthh iinn SSAAAAR R CC CCoouunnttr r iieess

I I J J A A R R-- B B A A E E , , 0022 , , 0022 ( ( 22001133 ) ) 1144-- 226 6 .. M M d d .. S S hhaa mmiimmuul l I I s s l l aa mm aa nnd d M M d d .. A Al l M M aa mmuunn..

DDeetteer r mm iinnaannttss oof f WW oor r k k --lliif f ee BBaallaannccee oof f WW oomm eenn PPr r oof f eessssiioonnaallss:: EEvviiddeennccee f f r r oomm BBaannggllaaddeesshh

I I J J A A R R-- B B A A E E , , 0022 , , 0022 ( ( 22001133 ) ) 227 7 336 6 .. K K oohhiinnuur r A Ak k t t eer r

EEmm p p iir r iiccaall AAnnaallyyssiiss oof f GGHHAAR R CCHH MMooddeell iinn VVaalluuee aatt R R iissk k EEssttiimm aattiioonn:: EEvviiddeennccee f f r r oomm TTeehhr r aann SSttoocck k EExxcchhaannggee

I I J J A A R R-- B B A A E E , , 0022 , , 0022 ( ( 22001133 ) ) 337 7 --446 6 .. F F aa r r z z iinn R Ree z z aa eeii , , A Ammiir r Y Y eek k ee z z aa r r ee , , aa nnd d l l iinnaa k k aa vviiaa nniinnee j jaa d d

AAcccceessss ttoo f f iinnaannccee aanndd FFiir r mm SSiizzee I I J J A A R R-- B B A A E E , , 0022 , , 0022 ( ( 22001133 ) ) 447 7 --5588.. H H uumm y yr r aa J J aa bbeeeenn B Br r ii s s t t y y

CC aa llll f f oorr PP aa pp eerr VVooll.. 0022,, II ssssuu ee 0033,, 22001133..

IIJJAAR R --BBAAEE
http://www.setscholars.org/index.php/ijarbaehttp://www.setscholars.org/index.php/ijarbae


2/14



P a g e

1 5

DDiiggiit t aa ll DDiivviidd ee aann dd IIt t ss IImm pp aacct t oo nn EEccoonn oomm iicc GGrr ooww t t hh iinn SSAAAARRCC CCoouu nn t t rr iiee ss Md. Shamimul Islam 1, Md. Al Mamun 2* 1Md. Shamimul Islam is a Senior Lecturer in MIS in the Department of Business Administration at ASA University Bangladesh. He can bereached at [email protected]*Md. Al Mamun, Senior Lecturer in Finance at East West University, Dhaka-1212, Bangladesh is the corresponding author and can bereached through email: [email protected]

Article History ABSTRACTReceived : 13-01-2013Revision Received: 10-03-2013

Accepted : 12-04-2013 Available Online : 22-04-2013

Keywords : Digital Divide, ICT, Internet,ICT4D, SAARC.

JEL Classification :R58, Z13, O33, I38.

In this information age, digital divide is a significant area of concern for sustainabledevelopment and even growth of a country. Because of this concern this paperstudied the status of digital divide among SAARC countries. An exploration on thedigital divide among the SAARC countries reveals that there exists a significant gapbetween the member countries which in turn can cost the face of growth anddevelopment as observed in the static and dynamic regression models. Among theSAARC countries Maldives has the lowest digital divide i.e. has achieved highest ICTfacilities and thereby has the highest GDP per capital income. In contrast MaldivesBangladesh is at the bottom among the countries in ICT facilities and its GDP percapita income is one of the lowest as well.

Citation : Md. Shamimul Islam (2013). Digital Divide & Its Impact on Economic Growth in SAARC Countries, IJAR-BAE 2(2): 14 26.

Copyright: @2013 Md. Shamimul Islam. This is an open access article distributed according to the terms of the CreativeCommon Attribution (CCC) 3.0 License under PKP (Public Knowledge Project) of Simon Fraser University, Canada.

1.0 Introduction

OECD (2001) referred the term digital divide as the gap between individuals, households, businessesand geographic areas at different socio-economic levels with regard to both the opportunities to access toinformation and communication technologies and to their use of the internet for a wide variety of activities.

At the start of globalization after 1990 there has been a tremendous evolution and growth of Informationand Communication Technologies (ICT). Indeed ICT consolidated the growth and development of manymultinationals corporations (MNCs) by efficiently handling the flow of information and establishing thecontrol over distant subsidiaries. The developments in hardware, software, and electronics have lead toaffordable general purpose technologies (Bresnahan & Trajtenberg, 1995) that can be used in everyspace of life including the business and economics. ICT have been recognized as a major contributor insocial and economic development ( OECD, 2004 ). Numerous empirics including WEF (2009) , Fontenay &Beltran (2008) suggest that ICT is also a means for economic development. This highlights theimportance of studying of digital divide so as to recommend measures to accelerate the face of growthand development especially for developing economics.

IJAR-BAEVol. 02. Issue 02. Article No. 02

Full-length Original Open Access Research Paper www.setscholars.org/index.php/ijarba e

International Journal of Applied Research in Business Administration and Economics 02, 02 (2013) 14 26


3/14



P a g e

1 6

As ICT and the economic development are correlated, hence it determines the ability of achievingcompetitive advantage at individual, organizational or government levels and enables them to do theirjob in more efficient and effective manner. Therefore, the unequal access to ICT may have seriousimplications for economic growth, human development, and the creation of wealth (ITU, 2006) . For thisreason, the inequalities or digital divide has been emerged as a hot topic in sustainable growth anddevelopment literature in national and international arena (ITU & UNCTAD, 2007) and (World Bank,

2009) . In fact at the World Summit on the information society, it was declared that the global challengefor the new millennium is to build a society where everyone can create, access, utilize and shareinformation and knowledge; enabling individuals, communities and peoples to achieve their full potentialin promoting sustainable development and improving their quality of life (WSIS, 2005) .

Thus at political level the world leaders have also become concerned about digital divide. It is trulyamazing to see that both American and European recovery plans from the global economic crisis haveconsidered ICT as part of their strategic actions. For instance, the European Commission has earmarked 1 billion of extra spending for investment in broadband, with special attention to the improvement of high-speed connections in rural regions (European Commission, 2009) . Contrasting this picture of thedeveloped world, in South Asia such initiatives from the political level is rare to find and any effectiveplans to minimize the gap of digital divide to expedite sustainable growth is only in political rhetoric i.e.the digital Bangladesh slogan in Bangladesh.

Established in 8 th December 1985, South Asian Association for Regional Cooperation (SAARC) as anorganization of South Asian nations comprising Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, andSri Lanka, and Afghanistan though reiterated the commitment of reducing the income inequalities withinthis region by adapting the SAARC declaration of 2011 i.e. commitment to alleviate poverty and reduce income inequalities within the societies and reaffirming their resolve to improve the quality of life and well-being of their people through people-centered sustainable development ; but in real ity suchcommitment can only be considered as an utopian fairy tale. This is because SAARC is still considered asone of the poorest regional economic blocs and no effective step have been taken to improve the incomeinequalities especially through the avenues of reducing the digital divide within this region as compared

with the developed world. For example back in 1994, SDC ( SAARC documentation centre ) has beendeveloped to meet the information needs of the scholars in the SAARC region, to harness the applicationsof information technology in information management in Member States, to create traditional knowledgedigital libraries, e-journal access facility, networking the focal points for SAARC member states, todevelop the web enabled databases of SDC information and to provide the other information related toits activities, programmes, meetings, events, etc. and finally to facilitate in furthering the economic, social,and industrial development of the member states by providing timely access to relevant and accurateinformation. But its functioning becomes standstill and its effectiveness in reaching to its objectiveremained a remote possibility even today.

Indeed rising inequality in many parts of the world means a growing digital divide and Fontenay &Beltran (2008) argued that such societies have not been able to take advantage of the ICT resources theycould muster, hence have fallen short of their growth potential. Therefore it is necessary to examine thestatus of digital divide among SAARC countries and uncover the impact of digital divide on the economicgrowth of the member states.

2.0 Literature review

It is very difficult to say when the concept of digital divide has been originated (Gunkel, 2003) . However,literature on this subject generalizes that the term emerged in the mid-1990s by the former Assistant Secretary for Communications and Information of the United States (US) Department of Commerce, LarryIrving Junior (Dragulanescu, 2002) . As far as the connotation is concerned, a number of connotations of digital divide are available (Rao, 2005) . As mentioned earlier, OECD (2001) argued that digital divide

refers to the gap between individuals, households, businesses and geographic areas at different socio-economic levels with regard to both their opportunities to access ICT and to their use of the internet for awide variety of activities.


4/14



P a g e

1 7

Though in recent times there has been exponential growth in the development and use of ICT, thetechnologies those are playing a decisive role in improving almost every aspect of our societies (WB,2009) , including business transactions, communications, economics, and politics (OECD, 2004), yet thereexists significant gaps in digital divide across the world. Regarding the analysis of digital divide twoapproaches can be considered. The first one is the consideration to international level, i.e. digital divideacross cross sectional countries. The second one is the consideration at intra-national level, or within acountry. In both types of measures of digital divide, gaps can occur regarding access to ICT between

regions, or groups of individuals, when there exist characteristics of different nature (Ono & Zavodny,2007) and (I. f. S.Unesco, 2003) .

Carlsson (2004) studied the effects of the ICT in the economy by comparing the potential of thesetechnologies to the so- called general-purpose technologies (GPT ) such as the transportation andcommunications technologies in the 19th century i.e. the Corliss steam engine or the electric motor,which in the past revolutionized the economy. The study concluded that ICT appears to have an evengreater impact on the economy since it affects the service industries (e.g. health care, government, andfinancial services) even more profoundly than the goods-producing industries; these service sectorsrepresent over 75% of global GDP. Jalava and Pohjola (2008) showed that the ICT contribution toFinlands GDP between 1990 and 2004 was three times greater than the contribution of electricity from1920 to 1938. However the growth impact of ICT has not been found uniform. For example, Cruz-Jesus,Oliveira, & Bacao, (2012) and Vicente & Lopez (2011) working with EU data has argued that digitaldivide exists among European countries as well.

With such astonishing revelation of empirical works the present study has been conducted once it hasbeen found that there is dearth of scientific works on this subject matter considering the entire SAARCregion. Moreover earlier studies on digital divide in India (Rao, 2005) , in Bangladesh (Rahman, Nahar,Islam, & Akhter, 2006) , in Pakistan (Ahsan, 2009) , and in Srilanka (Gamage, 2006) are quite inadequatedue to the elementary nature of methodologies applied.

3.0 Measuring the digital divide

In order to measure the digital divide the present study employed multiple indicators to represent theICT development of a country. According to the recommendations of the (OECD, 2009) , the variablesthose should be used to measure the digital divide vary with the goals of the research. For instance, if someone wishes to measure the internal or domestic digital divide s/he should drill down the ICT le velindicators by groups such as gender, age, income, education, geographical place, and so on, which aremore likely to uncover the disparities between categories. However to measure the digital divide amongcountries, the indicators should refer to the aggregated national reality. Since the objective of this studyis uncovering the digital divide within the SAARC, it follows the second recommendation.

4.0 Data & methodology

Keeping consistency with the other related studies i.e. Cruz-Jesus, Oliveira, & Bacao (2012), Vicente &Lopez (2011) , and ilan, Bolat, & Cokun (2008) as well as following the recommendations from theOECD and the European Commission, this study uses five variables to investigate the subject matter of interest. Data have been collected from the World Bank for the year 1995 to 2011. The variablesconsidered have been the following: cell subscriber (%), fixed broadband (%), internet server permillion, internet user (%), and telephone line user (%) and GDP per capita.

Regarding the methodology the current study has initially employed descriptive statistics to understandthe nature of digital divide among member countries. Upon investigating the actual digital divideamongst the SAARC countries, the study has used various panel methodologies to answer the questionwhether such difference in the digital progression has an impact on the economic growth in SAARC. Thusthe research question that has been addressed in this study is, if there is a digital divide in SAARC

member countries and if such divide exists, what is the growth implication of such digital divide in thesecountries.


5/14



P a g e

1 8

To address the second question initially panel unit root tests proposed by Levin et al. (2002), and Breitung (2000) have been used to understanding the nature of data set. Upon understanding thecharacteristics of the data set, the study has avoided the vector error correction based on Johanssoncointegration test in favor of static and dynamic panel models.

The conventional panel techniques are widely used in growth literature 1 to capture the steady-staterelationship between the dependent and independent variables. These are based on fixed effect

estimator to explore the relationship between ICT related variables and economic growth within eachcountry. This estimator has common slopes and variance and specific intercept. The rationale behindusing this estimator is to control for all possible unobserved characteristics of each country in the study.The basic model for panel data study has been presented in the following equation (01):

)01.......(..................................................0 it it it ICT GDP

GDPit is the dependent variable (GDP per capita) observed in i th country and t represents time periods.Moreover, it ICT represents the variables measuring the digital divide and is the coefficient of thesevariables, while it is the error term.

4.01 Panel fixed and random effect model

The fixed effects model has constant slopes but intercepts differ according to the cross-sectional units.For i countries i 1 dummy variables are used to designate the particular country. Another fixed effectspanel model where the slope coefficients are constant, but the intercept varies over individual/ countryas well as time. Fixed effect model with differential intercepts and slopes can also be used however;inclusion of lot of variables and dummies may give inconclusive result since so many dummies may result the problem of multicollinearity. There is also a fixed effects panel model in which both intercepts andslopes might vary according to country and time. This model specifies i-1 country dummies, t-1 timedummies, the variables under consideration and the interactions between them.

In the random effects model the intercept is assumed to be a random outcome variable, whereas therandom outcome is a function of a mean value plus a random error. Two way random effects model isused for estimation purpose. Swamy (1971) suggested the random effects model and Swamy and Arora(1972) and Swamy et al. (1988a, 1989) suggested an extension of the random effects model as follows:

iiit it iit

Var and E where

N iiT t x B y

i0.

)02........(....................,.......1),(,.......1,

This model is a generalized, groupwise heteroscedatic model. Moreover, for selecting the best modelthere are different test like F test, Hausman Specification Test, Breusch-Pagan Lagrange Multiplier test,which have been applied to decide the best fitting model.

4.02 The panel ARDL approaches

As the study aims at identifying the long-run and association between ICT variables and economic growthas well as their short term impact, the appropriate technique to be followed is autoregressive distributedlag ARDL (p, q) model by Pesaran and Smith (1995) following maximum likelihood (MLE) basedestimation . There are three different approaches of panel ARDL proposed by Pesaran (1995) i.e. themean group (MG), the pooled mean group (PMG) and the dynamic fixed effect (DFE). The study has usedall the three approaches.

The first form of panel ARDL is known as mean group (MG) introduced by Pesaran & Smith (1995) . This

model estimates the long run parameters for the panel from an average of the long run coefficients fromARDL models for each cross sectional units. Thus in a panel set up of the like the following:

1 For example, see Barro and Sala-i-Martin (1992) and Mankiew et al. (1992).


6/14



P a g e

1 9

)03....(..........1, it it it iii X Y it Y ,the long run parameter i for country i is iii 1/ while MG estimator for panel is

N

iii N 1

1

and N

iii a N

a1

1 . Thus it is clear that this model estimates separate regressions for each countries and then

calculate parameters as unweighted means of the estimated coefficients for the individual countries

without imposing any restriction; so all coefficients are allowed to be heterogeneous in the long-run andin the short-run.

However, the necessary condition for the consistency and validity of this approach is to have asufficiently large time-series dimension. Moreover the cross-country dimension should also be large(about 20 to 30 units according to Pesaran et al. (1999 ) at I (0) or I (I) order integration. With short timeseries data this model however may give misleading result.

Though Johansen (1995), Philipps & Hansen (1990) argue that a long-run relationships exist only in thecontext of cointegration among integrated variables, but Pesaran and Shin (1999) counter suchassumptions by presenting MG and PMG with better efficiency. The application of PMG and even MGestimators do not require cointegration tests. Since the mixed integrated order amongst the variables donot affect the efficiency of the estimation hence conventional stationarity check is no longer required.Moreover, this model is appropriate for the panel with large N and T dimensions thus there is no issue of short panel or long panel in the estimation.

PMG also allows short-run coefficients including the intercepts and the speed of adjustment to the longrun equilibrium to be heterogeneous country by country, while the long-run slope coefficients arerestricted to be homogeneous across countries. Thus this study chooses PMG model suggested byPesaran (1999) in the following format:

q

jit jit ij

p

j jit ijiit y y

0

/

1)04(...........................

Where, i = 1, 2,.., N represents cross sectional unit t = 1, 2, 3, ..T represent time (annual), j is thenumber of time lag, y it = GDPPC it , xit = independent variables like telephone line users it , internet users it, cell phone subscribers it etc. and finally i is the fixed effect. By re-parameterization, the above equationcan be written as:

)05(...........................1

0

*1

11 *

q

jit jit ij

p

j jit ijit iit iiit y y y

.1,........,,2,1,

,1,........,,2,1,

,),1(1

1

1

01

*

*

q j

and p j

p

jmimij

p

jmimij

p j ij

p j iiji

Where,

Now by grouping the variables in levels further, Eq. (5) is rewritten as an error correction equation:

)06(...........................)(1

0

*1

11 *

q

jit jit ij

p

j jit ijit iit iiit y y y

Where i = -( i/ i) defines the long-run or equilibrium relationship among yit and x it . In contrast *ij and*ij are short run coefficients relating growth to its past values and other determinants like xit . Finally , the

error-correction coefficient i measures the speed of adjustment of yit toward its long-run equilibriumfollowing a change in xit . The condition i


7/14



P a g e

2 0

)07........(..............................341

622

k S n

JB

Where n is the number of observations (or degrees of freedom in general); S is the sample skewness, andK is the sample kurtosis.

Jarque-Bera (1980) test suggests that we can reject the null hypothesis of all variables whose p valuewould be less than 0.05 and we can say the distribution is normal. In case of India we can reject the nullhypothesis for all variables and can conclude that distribution of variables about India is normal. On theother hand the distribution of data about Bangladesh are also normal except one variable i.e. internet server per million whose p value is 0.00. In case of Afghanistan, all the variables are normally distributedexcept two variables such as fixed broadband % and internet server per million. The data distribution forPakistan, Srilanka, Maldives, and Bhutan are normal except one variable while most of the data set about Nepal shows skewed distribution. However, in general the distributions of all the variables for all thecountries are mostly normally distributed.

Upon concluding the normality nature of the data set Table 01 represents the descriptive statistics of different variables of different countries as well as the Jarque-Bera test statistics for those respectivevariables. This table highlighted mean, standard deviation, and probability of variables. From this table

we see the mean values of different variables of internet users (per 100 people) ranges from 1.06 to7.110 and cell subscriber (per 100 people) ranges from 11.11 to 56.00 etc. The mean values of different indicators indicate that there is a gap among countries. On the basis of internet users per 100 people wesee that Bangladesh has the minimum value (1.06) and on the other hand Maldives has the maximumvalue (7.11). So there is a big digital gap exist between these two countries as far as internet users areconcerned. If we consider the variable like cell subscriber (per 100 people) we see that the gap still exist such as the mean value of cell subscriber of Bangladesh is 12.78 in contrast the mean value of cellsubscriber of Maldives is 56.00. Hence we cannot ignore the digital divide on this dimension too.

The table shows that Maldives has the highest digital facilities measured in all the dimensions consideredin this study while the digital divide in mostly prevalent in Afghanistan. As per India, one of theeconomically influential countries in the world, the level of interest users, cell phone subscribers are farbehind than Maldives and even behind Pakistan. Though the level of internet servers per million in Indiais almost double than that of Pakistan.

Table 01: Descriptive statistics of different ICT indicators of SAARC countries (1995-2011)Countries Variables Mean Std. Dev. Jarque-Bera ProbabilityBangladesh Cell subscriber (%) 12.78 18.63 4.09 0.13

Fixed broadband (%) 0.01 0.02 3.33 0.19Internet server per million 0.08 0.17 38.91 0.00Internet user (%) 1.06 1.58 5.15 0.08Telephone line user (%) 0.61 0.30 1.68 0.43

Afghanistan Cell subscriber (%) 11.11 17.32 5.61 0.06Fixed broadband (%) 0.00 0.00 8.42 0.01Internet server per million 0.12 0.20 10.81 0.00Internet user (%) 1.07 1.51 3.64 0.16Telephone line user (%) 0.17 0.10 1.47 0.48

India Cell subscriber (%) 15.20 23.00 6.59 0.04Fixed broadband (%) 0.21 0.34 6.26 0.04Internet server per million 0.64 0.85 6.14 0.05Internet user (%) 2.54 2.88 5.55 0.06Telephone line user (%) 2.98 0.91 0.89 0.64

Pakistan Cell subscriber (%) 17.73 24.41 2.93 0.23Fixed broadband (%) 0.06 0.12 16.51 0.00Internet server per million 0.28 0.35 3.73 0.16

Internet user (%) 3.91 3.46 2.05 0.36Telephone line user (%) 2.59 0.64 1.37 0.50Maldives Cell subscriber (%) 56.60 63.77 2.21 0.33

Fixed broadband (%) 1.66 2.28 2.86 0.24Internet server per million 14.86 24.50 10.32 0.01


8/14



P a g e

2 1

Internet user (%) 10.03 11.08 2.77 0.25Telephone line user (%) 9.33 2.31 1.34 0.51

Bhutan Cell subscriber (%) 11.29 18.49 5.48 0.06Fixed broadband (%) 0.12 0.31 53.34 0.00Internet server per million 0.52 1.42 20.31 0.00Internet user (%) 3.14 3.75 7.17 0.03Telephone line user (%) 3.12 1.32 1.14 0.57

Sri Lanka Cell subscriber (%) 24.67 31.72 3.35 0.19Fixed broadband (%) 0.28 0.49 14.29 0.00Internet server per million 1.61 1.82 2.88 0.24Internet user (%) 3.33 4.56 7.78 0.02Telephone line user (%) 7.69 6.06 2.28 0.32

Nepal Cell subscriber (%) 19.66 33.13 335.74 0.00Fixed broadband (%) 0.30 0.98 2757.03 0.00Internet server per million 2.33 9.77 10957.95 0.00Internet user (%) 3.33 5.49 935.04 0.00Telephone line user (%) 3.52 3.88 133.50 0.00

5.01 Comparative picture of various ICT status related variables among SAARC countries

The following panel pictures present a comparative picture of digital divide among SAARC countries. Thistwo-way graphical presentation, prepared by making an average over the period of 1995-2011 providesthe clearest idea about the digital divide in SAARC countries.

Panel A represents comparative picture of internet and cell users in percentage among SAARC countries.From this picture we see Maldives enjoy highest position on the basis of two variables and in contrast Bangladesh and Afghanistan have poor position based on these two variables. Bhutan, Pakistan andSrilanka are at the middle stage, though Srilanka has more cell phone users compared to Pakistan, Bhutanand even India. However, Bhutan has more internet users compared to Srilanka.

Panel B represents a comparative picture of internet user in percentage and internet server per millionamong SAARC countries. This measures a dimension of ICT infrastructure in the SAARC countries. In thisdimension Maldives once again enjoys the highest ICT facilities followed by Srilanka, Bhutan, Pakistanand India.

Panel C presents the picture of comparative picture of broadband percentage and internet server permillion among SAARC countries. From this picture it is concluded that there is a huge gap betweenMaldives and other SAARC countries. Even second best country Srilanka is far behind to Maldives.

Finally panel D presents the picture of comparative picture of cell and fixed telephone line user inpercentage among SAARC countries. Here Maldives is the top among counties in terms of cell subscriberswhile Srilanka enjoys superiority in terms of fixed telephone line uses. However, other countries likeIndia, Pakistan, Bangladesh, and Nepal enjoys very poor ICT facilities amongst these countries.

Thus a conclusion of this panel diagram clearly highlights that Maldives enjoys the highest ICT facilitiesamongst the SAARC member countries in all dimensions, while the rest of the countries are far behindthan to Maldives in all these ICT dimensions. Therefore, as suggested by studies like WEF (2009) ,Fontenay & Beltran (2008), it is expected that these visible differences should be reflected in thedifferences in the respective level of economic growth among these countries as measured by GDP percapita. This is because a higher level of GDP per capita can only be achieved with higher growth orproductivity and lower level of population. Since most of these countries witnessed a growth inpopulation, thus any increase in GDP per capita should have been due to productivity growth led bytechnology, capital and labor. Indeed ICT plays positively in fostering the productivity of the technology

and labor.


9/14



Panel A: Comparative picture of internet and cell users in percentage among SAARCcountries

Panel B: Comparative picture of internet user percentage and internet server permillion among SAARC countries

BangladeshAfghanistan

Pakistan

India

Maldives

Bhutan

Srilanka

Nepal

0

2

4

6

8

10 20 30 40Cellsub_Cent


Pakistan

India

Maldives

Bhutan

Srilanka

Nepal

0

2

4

6

8

0 2 4 6 8IntSer_PerM

Panel C: Comparative picture of broadband percentage and internet server per

million among SAARC countries Panel D:Comparative picture of cell and fixed telephone line user in percentage

among SAARC countries


Pakistan

India

Maldives

Bhutan

Srilanka

Nepal 0

. 2

. 4

. 6

. 8

1

0 2 4 6 8IntSer_PerM


Pakistan

India

Maldives

Bhutan

Srilanka

Nepal

1 0

2 0

3 0

4 0

0 2 4 6 8TelLin_Cent


10/14



P a g e

2 2

P a g e

2 2

5.02 Impact of digital divide on economic growth in SAARC

Digital divide has become a matter of concern because of its relationship with the economic growth. It has been found that countries with high investment in ICT, have witnessed substantially higher level of economic growth (Kuppusamy et al, 2009). Thus using equation 1, 2 and 6 panel regressions has beenrun to understand the impact of ICT related variables on economic growth in SAARC countries.

However, before moving to the panel regression, the level of integration has been determined by panelunit root test suggested by Im et al. (2003), Dickey and Fuller (1979) and others. The result has beenpresented bellow:

Table 02: Result from panel unit root test (with individual intercept and no trend under first difference)

VariablesAssumes commonunit root process

Assumes individualunit root process

Levin, Lin & Chu t*Im, Pesaran and

Shin W-stat ADF - Fisher

Chi-squarePP - FisherChi-square

GDP Per capita -3.721** -3.457** 38.435** 65.174***Telephone line user (%) -2.811** -2.351** 30.345** 57.741***Internet user (%) 1.891 2.752 7.231 15.972

Fixed broadband (%) 2.862 2.928 13.016 17.058***, **, * refers significance at 1%, 5% and 19% respectively

The result from panel unit root test suggests that not all the variables of interest are integrated at sameorder. This result suggests that the application of Johansen cointegration, Engle ganger or VECM basedcointegration test will not be applicable in this data set. Though cross sectional units are not that largeyet the study has applied panel static models like OLS, fixed effect and random effect model as well aspanel ARDL models like MG, PMG and DFE estimates suggested by Pesaran and Shin (1999).

Using the equation of 1 and 2 the fixed effect and random effect model has been estimated as:

)08(.....................................54

321

it it it

it it it i

percentagebroadband Fixed percntageuser Internet

percentage s subscriber Cell million per servers Internet userslineTelephone DPPC it G

)09(.................................54

321

it it it it

it it it i

percentagebroadband Fixed percntageuser Internet

percentage s subscriber Cell million per servers Internet userslineTelephone DPPC it G

Moreover, following the (Pesaran and Smith, 1995; Pesaran et al. , 1999) methodology for MG, PMG, andDFE models as presented in equation 06, the following equation has been developed.

)10(.......................................................

)()

(

,,10 4,

10 3

,10 2,

10 1

111,4

1,31,21,11,

t i jt iq j

i j jt i

q j

i j

jt iq j

i j jt i

q j

i j jt i

p j

i jt i

t it it it iii

percentageusers Interent Percentage sSubscriber Cell

million per servers Intenet userslineTelephoneGDPPC percentageusers Interent

Percentage sSubscriber Cell million per servers Intenet userslineTelephoneGDPPC it GDPPC

Table 03 and table 04 present the result of the static and dynamic models. In order to test the validity of the impact of ICT on economic growth first the used a less efficient regression model such as OLS, fixedeffect and random effect model following equation 8, 9 and 10.

The result from the panel OLS with no consideration of the panel structure suggests that internet use haspositive and significant long term impact on the economic growth of the SAARC member countries.However, fixed broadband lines and cell subscriber has statistically significant negative long termrelationship with the economic growth of these countries. What explanation do we have for such result?It is probably because most of the cell subscribers are young people and their use in most cases do not

constitute a productive use. Moreover, fixed broad band use has statistically significant negative longterm relationship with the economic growth of these countries as well. The explanation of such outcomeis using broadband is cheap than other internet technologies like WiFi, Wimax, mobile internet etc. Thusin most cases unlimited surfing of the net with a lowest price may not constitute a productive use of


11/14



P a g e

2 3

P a g e

2 3

internet. However, the model explains 96.94% of the variation in the GDP per capita considering the joineffect of all the variables.

Table 03: The long run effect of digital divide on the economic growth in SAARC countriesGDP per capital: Dependent Variables OLS Fixed effect Random effect Long run coefficient Telephone line user (%) 0.205 0.002 0.004

Internet server per million -0.001 0.001 0.000Cell subscriber (%) -0.006*** 0.007*** 0.007***Internet user (%) 0.025*** 0.025*** 0.025***Fixed broadband (%) -0.203*** -0.202*** -0.199***

6.326*** 6.320***Bangladesh 5.908***Afghanistan 6.204***Pakistan 6.117***India 6.148***Maldives 7.567***Bhutan 6.654***Srilanka 6.646***Nepal 5.366***Hausman test 0.000***R-square 0.9694Rho 0.964Cross-section random 0.426Idiosyncratic random 0.124

However, looking at the outcome of the fixed effect model all the variables now shows a long termpositive impact on the economic growth of the SAARC countries except fixed broad band use.Furthermore the calculated rho, which measures variance due to the difference across the panel data, is0.964. Since the rho value is quite high, thus the study concludes that there is a cross sectional or uniquecountry effect on growth. The country dummies of the OLS clearly prove this point.

Finally the random effect model also generates the same result consistent with the OLS and FE model.However, this model shows additional information i.e. Cross-section random of 0.426 and Idiosyncraticrandom of 0.124 respectively while the rho is o.9208. This means that 46.19% variation is due to thecross sectional effect while 13.46% variation is due to time differences .

Furthermore the Breusch and Pagan Lagrangian multiplier test for random effects generatesa 12 452.12 with Prob > 12 0.0000. Thus the test result firmly rejects the null and concludes that random effects are not appropriate. Thus there is evidence of significant differences across countries.Therefore running a simple OLS regression will not suffix. However, in choosing between FE and REmodel, the application of Hausman test fails to make a significant difference between them.

After analyzing the effectiveness of fixed and random effect model in explaining the GDP per capita,dynamic panel models have also been used. The following table 04 reports MG, PMG and DFE estimatesand specification tests of equation (10). The result indicates that the error-correction coefficient i isnegative and significant and fall within the dynamically stable range for MG model. However this is not the case for PMG and DFE estimators. This indicates that there exists a long-run relationship betweenGDP per capita and the ICT related factors. Moreover, this also gives evidences of mean reversion to anon-spurious long-run relationship and therefore stationary residuals, meaning that GDP per capita andICT variables are cointegrated. Moreover, a higher average i's imply greater adjustment process towardsthe long run equilibrium GDP per capita. Before moving further into the discussion, it is important to notethat not all the variables used in OLS, FE, RE has been used in the dynamic estimations. This is because

use of such variables does not generate a result. Moreover, the presence of negative but statisticallyinsignificant error correction in the PMG and DFE method confines the choice to MG model only. This isalso evident in the Hausman test result.


12/14



P a g e

2 4

P a g e

2 4

Table 04: The long run and short run effect of digital divide on the economic growth in SAARC countries VariablesDependent: GDP per capita

Model(mean group)

Model(Pooled mean group)

Model(dynamic fixed effect)

Long run coefficient Cell phone subscriber (%) -0.005 -0.004 0.004Internet user (%) 0.181** 0.074* -0.384Error correction coefficient

Ec -0.105** -0.001 -005Cell phone subscriber (%) 0.002 -0.001 .001Internet user (%) -0.009 0.012* 0.011*** 0.701** 0.0667 0.073

Hausman Test 3.36 (0.186)

Thus the study only focuses on the result from MG model in discovering the potential dynamic effect of the ICT variables on economic growth in SAARC countries. The error correction coefficient suggest that under MG model the speed of adjustment is quite slow i.e. it takes around 10 years to adjust anydeviation in the long run path determined by the relationship between ICT variables and economicgrowth. Regarding the long-run coefficient, consistent with the static models, internet use has been foundto have significant long term impact on the economic growth in SAARC countries while cell phonesubscriber has been found to have very insignificant negative long run relationship with GDP per capita.However such finding is statistically insignificant as well. The long run coefficient of internet use of 0.181suggests that a 1% change increase in the internet use in the SAARC countries can lead to 0.181%increase in the GDP per capita in the region. This is really a phenomenal finding and thus a reduction of digital divide will surely yield more economic growth for the SAARC countries. However, as opposed tothe significant long run relationship, the short run relationship between internet use and GDP per capitahas been found to negative but statistically insignificant.

6.0 Conclusion

The study has highlighted several interesting findings. Firstly , there exists a visible digital divide among

SAARC countries with Maldives has the highest ICT facilities and Afghanistan or Bangladesh has thepoorest ICT facilities in all counts which give rise to the potential of the constituting a visible differencesin the GDP per capita as well. Secondly , under static model differences in GDPPC among the SAARCmember countries could reasonably be explained by the ICT related variables though there aresignificant variation due the cross sectional differences i.e. country specific factors. Finally , amongst theICT related variables internet use has been found to have significant long term impact on the economicgrowth of the SAARC member countries. Therefore a call for action to reduce the digital divide is veryimportant if SAARC wishes to unearth the fullest potential of its economic growth in the years to come.

References

Ahsan, M. (2009). Digital divide and digitization initiatives in pakistan: A birds eye view. Bank, W. (2009). Information and communications for development :Extending reach and increasing

impact. Washington, DC.Breitung, J., (2000). The local power of some unit root tests for panel data. Non-stationary Panels, Panel

Cointegration, and Dynamic Panels.Bresnahan, T., & Trajtenberg, M. (1995). General purpose technologies. Engines of growth? Journal of

Econometrics, 65, 83-108.Carlsson, B. (2004). The digital economy: what is new and what is not? Structural Change and Economic

Dynamics , 245 264.ilan, . A., Bolat, B. A., & Cokun, E. (2008). Analyzing digital divide within and between member and

candidate countries o f European Union. Government Information Quarterly .Cruz-Jesus, F., Oliveira, T., & Bacao, F. (2012). Digital divide across the European Union. Information &

Management .Declaration, a. (2011). seventeenth SAARC summit addu declaration. addu .Dragulanescu, N.-G. (200 2). Social Impact of the Digital Divide in a Central -Eastern European Country.

The International Information & Library Review , 139 151.


13/14



P a g e

2 5

P a g e

2 5

European Commission. (2009). Investing today for tomorrows Europe. Brussels. Fontenay, A. B., & Beltran, f. (2008). Inequality and economic growth: Should we be concerned by the

digital divide? ITS Montreal , 29.Fontenay, a. B., & beltran, f. (2008). Inequality and economic growth: Should we be concerned by the

digital divide? Bourdeau de Fontenay & Beltrn , 1.Gamage, P. (2006). E-Sri Lanka: bridging the digital divide. The Electronic Library .Gunkel, D. (2003). Second thoughts: toward a critique of the digital divide. New Media & Society , 499.

I.f.S.Unesco. (2003). Measuring and monitoring the information and knowledge societies: a statisticalchallenge.

ITU. (2006). Measuring ICT for socia land economic development.Geneva. Worldtelecommunication/ICTdevelopment report 2006.

ITU, & UNCTAD. (2007). World information society report 2007: Beyond WSIS. Geneva.Jalava, J., & Pohjola, M. (2008). The roles of electricity and ICT in economic growth: case Finland.

Explorations in Economic History , 270 287.Jarque, Carlos M.; Bera, Anil K. (1980). Efficient tests for normality, homoscedasticity and serial

independence of regression residuals". Economics Letters 6(3): 255 259. doi:10.1016/0165-1765(80)90024-5.

Kuppusamy, M., Raman, M., & Lee, G. (2009). Whose ict investment matters to economic growth: privateor public? The malaysian perspective. The Electronic Journal on Information Systems in DevelopingCountries , 1.

Lester Henry, P. (n.d.). The Digital Divide, Economic Growth and Potential Poverty Reduction: The Case of the English Speaking Caribbean.

Levin, A., C.F. Lin and C.S.J. Chu, (2002). Unit root tests in panel data: Asymptotic and finite-sampleproperties. Journal of Econometrics, 108(1-24).

OECD. (2009). Guide to Measuring the Information Society. Paris: Secretary-General of the OECD.OECD. (2004). The economic impact of ICT: measurement, evidence and implications. Organization for

Economic Cooperation and Development . Paris.OECD. (2004). The economic impact of ICT: Measurement,evidence and implications.OECD. (2001). Understanding the Digital Divide. OECD Publications, (p. 32). Parish.Ono, H., & Zavodny, M. (2007). Digital inequality: a five country comparison using microdata . Social

Science Research , 1135-1155.Pesaran, M.H. and B. Pesaran, (1997). Microfit 4.0: Interactive econometric analysis. Oxford: Oxford

University Press.Pesaran, M.H., Y. Shin and R. Smith, (1999). Pooled mean group estimator of dynamic heterogeneous

panels. Journal of the American Statistical Association, 94: 621-634.Pesaran, M.H. and R. Smith, (1995). Estimating long-run relationships from dynamic heterogenous

panels. Journal of Econometrics, 68(79-113).Rahman, M. A., Nahar, M., Islam, M. A., & Akhter, R. (2006). The Digital Divide in Developing Countries:

With Special Reference to Bangladesh. Springer Berlin Heidelberg.Rao, S. S. (2005). Bridging digital divide: Efforts in India. Telematics and Informatics , 363.SDC Library. (2013). Retrieved in February 26, 2013, from SAARC Documentation Center:

http://www.sdc.gov.in/objective.phpSingh, D. S. Digital Divide in India: Measurement, Determinants and Policy for Addressing the Challenges

in Bridgingthe Digital Divide.South Asian Association for Regional Cooperation. (2013). Retrieved in February 26, 2013, from

Wikipedia: http://en.wikipedia.org/wiki/South_Asian_Association_for_Regional_CooperationSwamy, P. (1971). Statistical Inference in Random Coefficient Regression Models. New York, Springer-

Verlag.Swamy, P.A. and Arora, A.A. (1972). The Exact Finite Sample Properties of the Estimators of Coefficients

in the Error Components Regression Models. Econometrica , 40, 261-275.Swamy, P. Conway, R. and LeBlanc, M. (1988a). The Stochastic Coefficients Approach to Econometric

Modeling, Part I: A Critique of Fixed Coefficient Models. The Journal of Agricultural Economic Research , 40, 2-10.

Swamy, P. Conway, R. and LeBlanc, M. (1989). The Stochastic Coefficients Approach to EconometricModeling, Part III: Estimation, Stability Testing and Prediction. The Journal of Agricultural Economic Research , 41, 4-20.


14/14


www.setscholars.org/index.php/ijarbae 2 6

2 6

Vicente, M. R., & Lopez, A. J. (2011). Assessing the regional digital divide across the European Union-27.Telecommunications Policy .

World Bank (2009). Information and Communications for Development 2009. Extending Reach andIncreasing Impact. Washington, DC,: World Bank.

WEF. (2009). The globa linformation technology report 2008 2009. Retrieved fromhttp://www.wef.com

WSIS. (2005). Tunis commitment, Second Phase of the World Summit on the Information. Tunis.

IJAR-BAE

Currently Indexed and Hosted in following Digital Libraries

Upcomming Indexing for IJAR-BAE

Digital Divide and Its Impact on Economic Growth in SAARC Countries

Documents