CORPORATE GOVERNANCE THROUGH BOARD …



CALL ME MAYBE:THE IMPACT OF TELECOMMUNICATIONS ON ECONOMIC GROWTH IN THE ASEAN REGIONMcRey Banderlipe IISchool of Economics, De La Salle University______________________________________________________________________________ABSTRACTThe importance of telecommunications plays a very important role in achieving one of the targets in the Millennium Development Goals of the United Nations, which is to increase the access of people in both fixed-line and mobile phone telecommunications, measured in terms of teledensity. This paper attempts to determine if the teledensity in both modes of telecommunications, as well as the aggregate of these two modes, can contribute to economic growth, particularly in the ASEAN region where there is a promising venue to conduct business and to harness the benefits of investing in such facilities. Using a panel data analysis from seven ASEAN countries from 1992 – 2010 employing the LSDV-1 process and penetration rates as a proxy for teledensity, the findings revealed that fixed-line penetration rates are positively related to economic growth measured in terms of GDP per capita, while mobile and total penetration rates are not significantly associated with economic growth considering that mobile telecommunications were introduced later than fixed line telephones. It will still need a considerable length of time to fully realize the benefits of telecommunications. Furthermore, this study recognizes the need for better and proper use of telecommunications facilities and to ensure its accessibility in the remote areas of the region. Such policies are important to ensure that the benefits of telecommunications access can be fully achieved.Keywords: telecommunications, teledensity, economic growth, telephones IntroductionThere is no doubt about the need for people to communicate with one another. Having a conversation over a cup of coffee or a meal, writing lectures in the blackboard, delivering a talk or a speech, publishing a book or a newspaper, doing sales talk using the telephone, and sending a letter via snail mail are just some of the ways where one can communicate to another person. Perhaps it could be their friend, students, audience, readers, and receiver of the call or mail to whom information is intended to be conveyed. With the advent of contemporary modes of communication using mobile phones, computers, laptops, PC tablets, the e-mail, and the Internet, the way people communicate has changed abruptly. The blossoming of the generation elevated communication to a higher level where there is stronger focus towards the transfer of real-time information between senders and receivers. Such introduction of new modes of communication technology also changed the way economic activities take place because of the availability of the needed information that facilitates effective decision making.For a country to sustain an economy where information is very important, it has to develop and to maintain an information generation and transfer system that promotes the interaction among the members of the domestic economy and of the international community. Telecommunications would be one of the ways to support the need for this mechanism. Prior research suggests that investment in telecommunications increases growth dividend (Roller and Waverman, 2001), combats poverty (Calderon and Serven, 2004, as cited in Negash and Patala, 2006), and promotes economic expansion (World Bank, 1991, as cited in Negash and Patala, 2006). However, the validity of these arguments is still in question up to this day.In the 8th plenary meeting of the United Nations General Assembly last September 8, 2000, the 189 countries, including the 147 heads of state and government, signed the Millennium Declaration. This declaration signifies the pledge of support from both developed and developing countries to forge a partnership towards the creation of “an environment – at the national and global levels alike – which is conducive to development and to elimination of poverty” (United Nations, 2000). To represent such partnership, the United Nations released the set of Millennium Development Goals (MDG) and targets, the most recent version of which was published on January 15, 2008. Indicators for monitoring progress are identified side-by-side with the MDGs, all of which must be achieved by 2015, the year of deadline set by the United Nations.The United Nations believes that the emergence of an information superhighway should serve as a channel for economic growth to address the eighth Millennium Development Goal (MDG). Following the resolution adopted by the United Nations General Assembly in 1997 declaring access to communication as a basic human right (Salazar, 2006/2007), it was no surprise that one of the goals of the MDG, particularly Target 8.F, is to make available the benefits of new technologies, especially in information and communications. In the Official List of MDG Indicators, achieving target 8.F can be done by cooperating with the private sector to ensure that the benefits arising from new technologies can be channeled to everyone to foster economic development. Table 1 presents the details of target 8.F.Inspired by the landmark papers of Roller and Waverman (2001) and Waverman, Meschi, and Fuss (2005) and Sridhar and Sridhar (2009), this study aims to investigate how telecommunications affect economic growth, particularly within the economies in the ASEAN region. When the Millennium Declaration was signed, academic literature on achieving target 8.F became extensive by discussing the impact of mobile telecommunications on the economic growth of developing countries (see articles of Biancini, 2011; Gruber and Koutroumpis, 2011; Lam and Shiu, 2010; Chakraborty and Nandi, 2009; Sridhar and Sridhar, 2009; Shiu and Lam, 2008; Melamed, 2007; Negash and Patala, 2006; Waverman, et al., 2005; Datta and Arwal, 2004; and Brock and Sutherland, 2000).Table 1MDG indicators on the benefits of new technologies as targeted in Goal 8Goals and Targets(from the Millennium Declaration)Indicators for Monitoring ProgressGoal 8: Develop a global partnership for developmentTarget 8.F: In cooperation with the private sector, make available the benefits of new technologies, especially information and communications.Fixed telephone lines per 100 inhabitantsMobile cellular subscriptions per 100 inhabitantsInternet users per 100 inhabitantsNote: This is an excerpt from the original document. For the official list of indicators, the reader is advised to visit the UN Millennium Development Goals Official List of MDG Indicators link /unsd /mdg /Resources/Attach/Indicators/OfficialList2008.pdfNote, however, that the previously mentioned papers do not totally make reference to Target 8.F of the MDG, particularly in 8.1 and 8.2, as some of them were written prior to the revision of the indicators in 2008. Our study aims to provide significant contribution by measuring the impact of telecommunications in the ever-changing economic dynamics of the ASEAN member nations towards meeting the Millennium Development Goals set by the United Nations. From the speech of the Prime Minister Go Chok Tong of Singapore during the Third ASEAN Telecommunications and IT Ministers meeting in 2003, he iterated the need for ASEAN industries to re-direct the future of Information and Communications Technology (ICT) since ICT development fosters the move towards achieving economic recovery and job creation for many people. Moreover, such undertaking contributes to sustained growth while bringing more closely the different economies in the region.Statement of the ProblemThe importance of telecommunications infrastructure on economic growth has been indispensable. The products of telecommunication lead to the increase in the demand for goods and services used to produce them. In addition, economic returns on telecommunications infrastructure outweigh the cost of searching for and exchanging information since it fosters direct communication despite increased distances. Such benefit transcends to other sectors that causes spillovers and externalities that contributes to the growth of other sectors and of the economy on a broader perspective. In the ASEAN region, the need to embrace technology would support continuous improvement in terms of productivity, efficiency, competitiveness, and the quality of the lives of people. This is the only way, according to Prime Minister Go Chok Tong (2003), for all of us to reap the benefits of a truly connected ASEAN region. In response to this call, this paper will seek to answer the following research question: Does teledensity/penetration rate of mobile and fixed telecommunications affect economic growth among countries in the ASEAN region?Review of Related LiteratureTowards attaining the ultimate goal of development and poverty eradication, economists have started to find out the potential sources for economic growth. The neoclassical growth model proposed by Solow (1956) attempted to answer the question of possible convergence over time in the levels of per capita income and output. Such prognostication inspired economic researchers to empirically test the determinants of per capita growth, including the article on increasing returns and long-run growth (Romer, 1986). Several papers identified in the introductory part of the study by R?ller and Waverman (2001) were clustered in terms of how human and physical capital is accumulated, how spillovers affect external economies, and how industrial innovation can serve as the locomotive that will steer economic growth.The prevailing questions on what could be the major sources of economic growth have plagued economic researchers over time. In cross-country analysis of economic growth, Datta and Agarwal (2004) questioned the possible convergence of per capita incomes across countries in the long run, coupled with identifying the possible sources of growth for different economies. The presence of this “guiding hand” that links per capita income and per capita production served as the inspiration behind the works of Barro (1991) and Barro and Sala-i-Martin (1991) which determined the factors of long-term growth emanating from the expanded neoclassical growth theory by Solow (1956). According to Chakraborty and Nandi (2011), the advances in economic literature augmented Solow’s growth theory by providing possible explanation on the sources of technological change. Such progress resulted to observed faster economic growth especially when the government undertakes investment in public infrastructure. Consistent with the Solow (1956) convergence hypothesis expanded in Barro and Sala-i-Martin (1992), greater attention was given to telecommunications investment as a component of public investment and its ability to generate economic gains since it creates spillover effects through network externalities to impact economic growth than investing in alternatives forms of infrastructure.Prior to the signing of the Millennium Declaration in 2000, a number of studies have already examined the potential impact of telecommunication facilities on economic growth. Snow (1989) argued that the policies constructed to liberalize ASEAN-US trade and direct foreign investment in services should be aimed towards export-led growth, allocative efficiency, and technology transfer, since information is treated as a service and resource in the world economy. In addition, the recent advances in technological progress resulted in the fall in the prices of telecommunication services pricing and increased consumption of these services (Cronnin, Collerlan, and Gold, 1997). As a factor input, substituting other inputs using telecommunications would generate substantial and cost-effective savings to the economy particularly in resource requirements. Such economic phenomenon of technological progress is evident since there is an increased bias towards skilled workers and the use of weightless technologies, as coined in the paper of Cameron (1998).Before telecommunications technology was introduced, searching for information takes an arduous process, and its impact on growth is not yet visible. This could be probably explained in Brock and Sutherland (2000) where no significant influence is exhibited between telecommunications and output growth in the Former Soviet Union republics although there is perceived output growth upon the dissolution of the union. This was explained by the difficulty experienced by these new former USSR countries to identify sources of financing for its telecommunications facilities. Telecommunications, back then, are still facing the hard times at the time of its adoption. Alleman, et al. (1994, as cited in Negash and Patala, 2006) added that, in developing countries, economic growth is not totally experienced despite the adoption of telecommunications technology because most telecommunication firms are identified to be owned and controlled by state governments, and like any public institution, they must compete for budget allocations to sustain their operations and constant development. Government authorities, at that time, might have been unaware of the potential economic benefits of allocating a large chunk of their budget to telecommunication investments.However, as telecommunication continuously evolves in the 1990s, a noticeable growth of consumption took place that signified the advent of a new technological revolution. This increased consumption delivered promising benefits not only for convenience of users but also contributed to the advancement of developed and developing economies. The landmark paper of Roller and Waverman (2001) estimated a micro model for telecommunication investment with a macro production function. The results of their study revealed a positive causal link between telecommunications infrastructure and aggregate output, measured by GDP, using a set of simultaneous equations. This is attributed to the presence of network externalities and critical mass that leads to increased returns on growth levels of 21 OECD and a sample of non-OECD countries that were included in the study. It was concluded that higher growth effects are visible in developed OECD states in contrast with the less-developed non-OECD states.To expand the study, Waverman et al. (2005) (WMF) modeled the impact of telecommunications in less-developed countries since during this period of study, the fixed-line communications are now being supplemented, if not replaced, by the increased usage of mobile phones. The series of simultaneous equations adopted was found to experience difficulty in terms of estimating the growth impact and robustness to changes in the sample and its specifications. It is hypothesized that less-developed countries will have slower penetration rates for mobile telecommunications than those countries that are already traversing the path of growth.Post-Waverman studies further provided substantial evidence on the impact of telecommunications and economic growth. For one, Negash and Patala (2006) stated that the GDP budget allocation for telecommunication investment is not enough to realize its full benefits; hence they proposed a community-focused framework to encourage economic activities in developing countries with the aid of telecommunications. In 2007, Melamed replicated the WMF study in Canada and revealed that the country’s GDP would have increased by 1% as a result of the increased penetration rate of mobile telecommunications if the benefits were already realized. The problem is that there is high cost of mobile telephone services that prevents them to increase mobile telephony subscriptions.Shiu and Lam (2008) noted the presence of unidirectional relationship between GDP and telecommunications development in China, but subdividing into eastern and western regions, the significance of such relationship maybe attributed to the affluence of the eastern region; thus they posed an argument that there are other ways that could stimulate growth in the western region. In a study of Sridhar and Sridhar (2009), a 3SLS system of simultaneous equations highlighted the positive association between the mobile and fixed-line phones and national output when control variables for capital and labor are instituted. This conclusion was made after an analysis of 24 developing countries, majority of which are located in Africa. However, a bi-directional relationship between GDP and teledensity as the measure of telecommunications development is expected to exist, unless when mobile telecommunications and economic growth is measured separately (Lam and Shiu, 2010). The same unidirectional relationships are also evident in the papers of Biancini (2011) and Gruber and Koutroumpis (2011).From this literature review, it is important to direct our attention to the focal point of this paper, the ASEAN Region. This paper found the importance of providing an overview on the evolution of this vibrant and dynamic sector that is a promising avenue for economic growth among the countries in Southeast Asia.Telecommunications in the ASEAN RegionThe ASEAN Region underwent the process of taking bold movements towards working as an integrated group that aims to be an outward and forward looking so that they would realize the effects of dynamic development while living in a community of caring societies. This took place when the Heads of States and Governments assembled in Kuala Lumpur, Malaysia in December 1997 to promulgate the ASEAN Vision 2020 that envisages an economic region that harnesses stability, prosperity, and competitiveness through liberalized flow of goods, services, investment, and capital. This vision was supported by the crafting of the Hanoi Plan of Action in 1998. Such plan of action is a series of steps towards strengthening international cooperation, economic integration, as well as establishing infrastructures for the promotion of science and technology, information technology, and human capital development (Trong Vu, 2009). Consistent with the ASEAN Vision 2020, former Singapore Prime Minister Go Chok Tong (2003) stated that since the thrust for regional integration is not new given by this vision, there is a need to create the ASEAN Economic Community that will push forth the economic integration of the different ASEAN countries. Through a stronger information and communications technology (ICT) pillar, having such mechanism facilitates the accessibility of faraway markets, the growth of various industries, as well as the boost in the economic sector of the region. Towards attaining this goal, the following courses of action are proposed by Prime Minister Go Chok Tong (2003):Focus on providing people access to pervasive and reliable information and communications infrastructure;Develop people’s literacy and comfort in using ICT services;Work towards harmonized regulations so that telecom operators can participate fully in a liberalized ASEAN market;Build linkages to countries outside of the ASEAN region;Embrace technology to improve the productivity, efficiency, and competitiveness of the economies and the quality of the people’s lives; and Protect information and communication networks from intentional harm and degradation.These courses of action are deemed to form the very important task of the ASEAN member countries to work towards a liberalized and protected telecommunications sector that will chart the new direction of the ASEAN region towards becoming one of the strongest economic regions in the world.Given the manifold objective of telecommunications development in the ASEAN region, this paper found the importance of providing an overview of the telecommunications sector in some members of this region, as extracted from the paper of Trong Vu (2009).IndonesiaTwo phases of telecommunications development took place in this country. When Indonesia adopted Telecommunications Law No. 3 from 1989-1999, state-owned companies (Telkom and Indosat) were partially privatized, and build operate transfer (BOT) investment agreements with foreigners were forged. It was during this time when foreigners were allowed direct equity investments in value-added services to raise network teledensity (measured as the number of fixed line subscribers per 100 people in their country) of 2.5% in 1992. However, most of the firms quit the market at the onset of the Asian Financial Crisis of 1997.The issuance of Telecommunications Blueprint and the adoption of the Telecommunications Law No. 36 marked the second phase in Indonesia’s telecommunications sector. It became witness to the return of foreign investors and the welcoming of KPN, Deutsche Telecom, and Telekom Malaysia which pushed forth privatization and competition in the industry. At present, the major players in Indonesia are PT Telkom (Telkom), Telkomcel, Indosat, and Excelcomindo.MalaysiaBeing one of the top economies in the world in terms of purchasing power parity, the country positioned itself quietly in advancing technology, having established more progressive telecommunication facilities resulted from the consolidation of networks due to competition. Mobile penetration in Malaysia has surpassed the 85% mark in the first half of 2007 and has already 21 million subscribers during that time. Presently, Telekom Malaysia, Maxis Communications, Celcom, and DiGi Telecommunications are the competing giants in this industry.PhilippinesCombined efforts of the government and the private sector were aimed towards expanding the national fixed network in the country. However, there exists difficulty in extending its fixed-line telephone networks resulting only to a 5% teledensity compared to the targeted 12% in 2002. In contrast, mobile phone penetration rate is much higher, surpassing the 50% mark in 2007 and grows at an annual rate of 26%. Investors believe that there is a promising ground for telecommunications considering that this sector contributes 10% to the country’s GDP. The only challenge that this country has to face is for the government to provide guidance on foreign direct investment of telecommunication companies and to manage competition rather than solely issuing licenses prior to operation. Philippine Long Distance Telephone Company (PLDT), Smart Communications, Globe Telecom, and Digital Telecommunications Philippines, Inc. (Digitel) are the key players in the Philippine telecommunications industry.SingaporeAs a highly developed country, Singapore is known for its high quality and progressive regulated telecommunications industry and is hailed as one of the world leader in telecommunications, with 98% of the households having fixed-line telephone networks. Singapore was one of the first countries in the world to adopt a fully digital telephone network and has established its strong presence in the regional telecommunications market through SingTel, StarHub, and MobileOne (M1). Liberalization has allowed the entry of new operators even if SingTel dominated the entire industry in the country.Thailand Despite the presence of uncertainty regarding the economic prospects given the range of events that took place in the country for last decade, Thailand’s telecommunication sector displayed vitality and a lot of energy as it registered high growth rates in the mobile telecom market. The National Telecommunications Commission (NTC) was established in 2004 to implement reforms in the market that once showed signs of tardy improvements. Although the momentum was lost during the military coup overthrow of Thaksin Shinawatra’s government in 2006, the mobile penetration rate was at 67% while the subscriber growth rates had an annual increase of 35%. TOT (formerly the Telephone Organization of Thailand) Corporation, True Corporation, Advanced Info Services (AIS), and DTAC (Total Access Communication) are the present-day competitors in the industry.VietnamWhen Vietnam became a full member of the ASEAN in 1995, it began setting ambitious targets in the telecommunications sector with some shortcomings. However, limiting the level of competition as well as the continuous improvement in its economic climate, Vietnam saw the growth of its telecommunications industry over the years. In addition, when Vietnam officially became a member of the World Trade Organization, it opened more opportunities towards deregulation and liberalization of the sector resulting to a more competitive telecommunications market in the country up to this date. Vietnam Post & Telecommunications (VNPT), Viettel, MobiFone, and VinaPhone are the leaders in the telecommunications industry of this country. Efforts towards Economic Integration and Telecommunications DevelopmentTrong Vu (2009) stated that the ASEAN members organized various initiatives to face the challenges and to identify opportunities brought about by the increased penetration of information and communications technologies. In the 33rd ASEAN Ministerial Meeting in 2000, it was affirmed that although such efforts are not intended to undertake numerous researches on ICT, it simply encouraged all members to embrace technology to remain competitive and to stay running as a vibrant and dynamic economy while improving the lives of the people.True to its mission, the ASEAN organized annually the ASEAN Telecommunications Ministers Meeting (TELMIN) and the ASEAN Telecommunications Senior Officials Meeting. These meetings were conducted: 1) to develop, to coordinate, and to implement work plans in the ICT with emphasis on telecommunications; 2) to promote the participation of the public/private sectors, as well as regional/international and non-government organizations; and 3) to establish working groups that will carry out the intended work plans. Appendix 1 presents the summary of the TELMIN meetings from 2001 to 2009.In addition, the members of the ASEAN region signed: 1) The e-ASEAN Framework Agreement; 2) The Ministerial Understanding on ASEAN Cooperation in Telecommunications and Information Technology; 3) The Vientiane Action Program on Telecommunications and IT Sector; 4) The ASEAN Sectoral Integration Protocol for e-ASEAN; 5) The Brunei Action Plan on Enhancing ICT Competitiveness; 6) The Siem Reap Declaration on Enhancing Universal Access of ICT Services in ASEAN; 7) The Bali Declaration in Forging Partnership to Advance High Speed Connection to Bridge Digital Divide in ASEAN; and 8) The Vientiane Declaration on Promoting the Realization of Broadband across ASEAN. These declarations manifested their solidarity in working towards reaping the benefits from the development, usage, and implementation of telecommunications technologies, notwithstanding the need fostering regional integration to promote economic growth and development in the region. Appendix 2 summarizes the main points of these agreements signed by the ASEAN member countries.Given the presence of limited studies measuring the impact of telecommunications on economic growth in this region, we found it imperative to perform an attempt to contribute to this expanding field of study. This prompted the researcher to consider the ASEAN region as a suitable locale for this paper given its promising future as it took preliminary steps towards economic advancement through the initiatives and declarations that they have signed in the recent years, harnessing the power of telecommunications infrastructure and recognizing its capability to influence other growing sectors of their respective economies. Note. Data obtained from the World Bank database (data.)Figure 1Mobile phone teledensity in ASEAN countries and the world, 1990-2011Figure 1 presents the teledensity of mobile phone subscriptions in the ASEAN Region, Timor-Leste, and the World for the years 1990 – 2011. As can be seen, Singapore, Vietnam, Malaysia, Thailand, Brunei Darussalam, Indonesia, and the Philippines exceeded the world average teledensity, indicating a huge number of inhabitants who has mobile phones. Other countries have yet to surpass the world’s batting average for mobile teledensity.Note. Data obtained from the World Bank database (data.)Figure 2Fixed-line telephone teledensity in ASEAN countries and the world, 1990-2011Figure 2 presents the fixed-line teledensity of ASEAN countries, Timor-Leste, and the World for the years 1990 – 2011. In contrast to Figure 1, only Singapore and Brunei Darussalam exceeded the world teledensity for fixed-line telephones. We also note that for some time, Vietnam surpassed the world teledensity from 1993 – 2003, while the Philippines exceeded the world’s teledensity mark in 2009. Thus, it can be inferred that more inhabitants from Southeast Asia are geared towards the use of mobile phones because of convenience and compactness in terms of the services that it provides to subscribers.One of the limitations of this paper is that it will not cover the “density” of Internet users. Given the limited data furnished by reliable data sources, the researcher opted not to include Internet usage for the moment. However, the researcher believes that internet facilities also provide mechanism to facilitate communication between distant parties; notwithstanding the integration of the Internet in telecommunication services. It is hoped that future studies would consider this variable to provide a comprehensive purview of economic growth using the complete of indicators identified in Target 8.F.Framework of the StudyThis study aims to determine the impact of telecommunications on economic growth among the ASEAN member countries. Prior to the conduct of this study, this paper identified the theoretical underpinnings that will serve as the foundation in formulating our approaches to answer our research problem.Growth Most studies on telecommunications and economic growth were rooted in the neoclassical growth theory proposed by Solow (1956) who attempted to support the belief in the convergence theory that over time, per capita production will move together with per capita income which is an evidence of economic growth. In his model, Solow identified the determinants of long term per capita growth such as labor, capital, and knowledge. Emphasis is placed on knowledge because Romer (2006) argues that since knowledge comes in various forms, many of these types of knowledge played important roles in economic growth. The numerous scientific discoveries, the invention of applied technologies and new products, and the subsequent improvements in the design of these technologies and products after invention would have different effects on growth at different lagged periods. Whatever knowledge accumulation mechanisms are undertaken, it is understood that these endogenous factors would contribute to the economic work Externalities and Spillover EffectsSeveral studies were geared towards identifying endogenous sources of technological change, under which telecommunications fall into place, as identified in Chakraborty and Nandi (2011). This supports the claim of Roller and Waverman (2001) that telecommunications infrastructure, resulting to the creation of the information superhighway, is clearly different from transportation and other superhighways due to the presence of network externalities. The network externality theory states that more utility is derived by users when there is growth in the number of users of a particular infrastructure. In the case of telecommunications, there exist network externalities due to the presence of spillover effects. When people use fixed-line or mobile telecommunication systems, the costs of doing business falls, the more efficient firms will operate because people are able to communicate despite increased distances. Thus, firms can engage in more productive activities since the information intensity of the production process increases. The productive units of a firm can, therefore, produce better goods and services towards increasing the output of the firm. When taken collectively, the output of all the firms in the economy will increase, indicating positive signs of economic growth.Transaction Costs and Spillover EffectsThis paper also rests on the premise of Madden and Savage (2000) regarding the reduction of transaction costs while doing business. Transaction costs are costs incurred when economic agents find other agents to exchange goods and services, reach an agreed price for such exchange, and fulfill the stipulations of the exchange agreement. Since agents are not fully aware of the prices at all times, they must canvass for the most favorable price in which transactions can finally take place.Telecommunications fulfill this role of reducing transaction costs since such infrastructure reduces costs of acquiring and transmitting information, making firms and factor markets efficient. Cognizant with the spillover effects, telecommunications infrastructure can service information-intensive sectors such as finance, trade, tourism, transportation, and other sectors identified in Cronnin et al. (1993). This is because the increase in information flows assists in the integration of domestic and international markets and enhances national income by increased competition market efficiency, and increase in entrepreneurial talent for human capital development and research. These benefits, according to Madden and Savage (2000), create efficient production mechanisms that can flow through all sectors of the economy.Social Overhead CapitalLastly, this paper utilizes the social overhead capital (SOC) concept identified in Waverman et al. (2005). SOCs are expenditures on education, health and social services, and public infrastructure. Telecommunications forms a crucial part on infrastructure spending, for the presence of networked infrastructures pave way for a speedy flow of information. Given the faster rate of information flows and lower costs arising from inflation, it therefore leads to the creation of what the Chairman of the US Federal Reserve Board Mr. Alan Greenspan (as indicated in Waverman, et al., 2005) calls as the “New Economy.” This new economy is a result of externalities associated with networked computers that allows competition and new means of organizing a more enhanced production process that increases output and that leads the way towards the path of economic growth.MethodologyOperational Framework and Model SpecificationThis study utilized one of the structural models in Sridhar and Sridhar (2009) that traverses from the micro-level demand for and supply of telecommunication infrastructure to aggregate changes in penetration rates to the macro production function in which GDP is determined by traditional inputs including capital and labor stock, and teledensity. As a limitation, this study does not attempt to provide a general explanation on the determinants of national output; thus, measures such as openness to trade and deficits are not included. We used a macro-production function consistent with the input-output in which, similar to the challenges posed in Roller and Waverman (2001) and the application made in Waverman et al. (2005), telecom investment is treated to be endogenous. The model is intended to demonstrate relationships between telecom penetration and growth, to identify factors that determine the supply of and demand for telecommunication services, as well as to determine factors that may possibly influence the change of penetration rates for telecommunication facilities.Panel data analysis will be conducted on the following with their corresponding specifications:(1)where:= natural logarithm of real GDP (at constant 2000 US$) of country i in year t;= proxy for expenditure (natural logarithm of gross national expenditure at constant 2000 US$) for country i in year t;= total labor force participation rate (expressed in decimal terms) of country i in year t; = penetration rate (PENR, measured as teledensity divided by 100) of mobile phones (CPPENR), fixed-line telephones (FLPENR), and total of mobile and fixed-line telephones (TPENR) using teledensity (TEL, measured as the number of subscribers per 100 inhabitants) for mobile phones (CPTEL), fixed-line telephones (FLTEL), and total of mobile and fixed-line telephones (TTEL) for country i in year t, and= a variable that captures the essence of time trend (1 = 1992, 2 = 1993,…, 19 = 2010).The complete dataset, found in Appendix 3 of this paper includes Ccode, a variable that identifies the country where a particular data originated. Hence, it has the following indexed values: 1 = Brunei Darrusalam, 2 = Indonesia, 3 = Malaysia, 4 = Philippines, 5 = Singapore, 6 = Thailand and 7 = Vietnam. This variable, however, will not be used for statistical testing since it is only intended for identification purposes. We used the per capita figures for GDP and Gross National Expenditure to even out the disparity of output and expenditures per person. This is to eliminate the undue advantage of people who can largely benefit from GDP and national expenditures. The paper also computed the natural logarithm of the variables except t to control the peculiarities of observations among different countries since the values of the variables differ largely depending on the country and the time the data was observed. This working equation was prepared to determine whether teledensity, measured in terms of penetration rates, has greater power to influence GDP since prior literature utilized either of these variables. It is similar to the Cobb-Douglas production function model since our model deals with the elasticity of output as influenced by the labor, capital, and knowledge. Note that the measure of knowledge in this model is through technological progress that includes the effect of telecommunications. The study initially estimated six (6) regressions for teledensity and penetration rates separately for mobile phones, fixed-line telephones, and the sum of mobile and fixed-line telephones. Since the penetration rate utilizes teledensity, this paper focused on penetration rates as the proxy for telecommunications since the initial testing for both variables yielded the same results. However, for descriptive statistics purposes, teledensity was used to provide a more realistic description of the initial data. The descriptive statistics also used the actual Gross Domestic Product per capita (GDPPC) and Gross National Expenditure (GNEPC) instead of natural logarithms and Labor Force Participation Rate (LFPR) expressed in percentage (%) instead of decimal terms. The statistics are presented in Appendix 3.Our a-priori expectation for these variables is that they will exhibit a positive relationship with GDP per capita as a measure of economic growth. It is expected that higher the expenditure, the teledensity and penetration rates of telecommunications, labor force participation rate, and the essence of time will contribute to the increase of GDP over the period since prior literature have confirmed the association of these variables with economic output.Data for the StudyAll the relevant data for this study were extracted from the World Bank database (data.) which was conducted last December 4, 2012. Initially, extraction was made utilizing the International Technological Union (ITU) World Telecommunications Indicators for 2011 database. However, due to the absence of relevant data for some countries, it generated an unbalanced panel of only 69 observations from 1990-2003. Since the construct of this study is to provide insights for as many ASEAN countries as possible for a considerable longer period using a balanced panel of observations, the dataset from ITU was dropped.The researcher gathered the data for the ten (10) ASEAN member countries and Timor Leste from 1992 to 2010. We started with the year 1992 because the World Bank database provided complete observations for most countries and it was during this period when the popularity of mobile telecommunications became visible. The refinements made from the raw data prompted the researcher to eliminate four countries for this study: Cambodia, Lao PDR, Myanmar, and Timor Leste. Cambodia and Lao PDR showed complete set of observations prior to the year 2000 while most of the data for Myanmar were not available even up to 2010. Timor Leste was excluded because limited data was made available by the World Bank as this country established its sovereignty only in 2002. The World Bank’s description of the variables of interest is shown in Table 2. All other relevant variables were computed as presented in the legend preceding equation (1) and in Appendix 3.Because we have already identified our working model in estimate the function of economic growth using penetration rates of fixed-line and mobile phone telecommunications using N = 7 cross sectional observations for ASEAN countries and T = 19 years (from 1992 – 2010), a balanced panel was established since each cross-sectional unit has the same number of time series observations. It is therefore imperative for us to perform the panel data analysis using the methodology identified in Gujarati and Porter (2009) using Stata12 software.Table 2Description of variables from the World Bank databaseVariable NameIndicator Name from the World Bank DatabaseVerbatim DescriptionGDPPCGDP (Constant 2000 US$)GDP per capita is gross domestic product divided by midyear population. GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. Data are in constant U.S. dollars.GNEPCGross National Expenditure (constant 2000 US$)Gross national expenditure (formerly domestic absorption) is the sum of household final consumption expenditure (formerly private consumption), general government final consumption expenditure (formerly general government consumption), and gross capital formation (formerly gross domestic investment). Data are in constant 2000 U.S. dollars.Note: GNE per capita is not available in the World Bank Database. To obtain GNEPC, we divided the Gross national expenditure by the population of the country during the year of observation.LFPRLabor Force Participation Rate, Total (% of total population ages 15-64)Labor force participation rate is the proportion of the population ages 15-64 that is economically active: all people who supply labor for the production of goods and services during a specified period.CPTELMobile Cellular Subscriptions (per 100 people) Mobile cellular telephone subscriptions are subscriptions to a public mobile telephone service using cellular technology, which provide access to the public switched telephone network. Post-paid and prepaid subscriptions are included.FLTELTelephone Lines (per 100 people)Telephone lines are fixed telephone lines that connect a subscriber's terminal equipment to the public switched telephone network and that have a port on a telephone exchange. Integrated services digital network channels and fixed wireless subscribers are included.Results and DiscussionThis paper aims to determine the impact of telecommunications on economic growth in the ASEAN region. We first present the descriptive statistics for the dataset compiled for 7 ASEAN countries from 1992 – 2010, followed by the results of the tests conducted to determine the ideal panel data model for this study, after which we present a thorough and insightful discussion about the reported results.Descriptive StatisticsAppendix 4 presents the descriptive statistics for each country involved in this study computed using MegaStat. Note that we used the actual figures for GDPPCi, GNEPCi, and LFPRi as well as for teledensity (CPTELi, FLTELi, and TTELi) where i is the index of Country Code (Ccode). As can be seen, Singapore has the highest mean GDP per capita of US$24,552 and has set the highest GDP per capita of US$32,641. In addition, Singapore also registered the highest per capita Gross National Expenditure amounting toUS$23,536. This could be attributed to the highly developed status of Singapore, being the economic hub of Southeast Asia. It was followed closely by Brunei Darussalam, Malaysia, and Thailand. In terms of labor force, Vietnam has the highest labor force participation rate at 82.8% on an average basis.Singapore registered the highest teledensity in mobile and fixed line telephones, having an average teledensity of 69 subscribers per 100 inhabitants for mobile phones and 43 subscribers per 100 inhabitants for fixed line telephones, compared to Indonesia with only 17 mobile phone subscribers and 8 fixed line telephone subscribers. The advanced telecom facilities of this country and its intense economic and market activities would have contributed to such high teledensity. Brunei Darussalam, a progressive country although small in terms of land area and population, came second; while Malaysia placed third. The coefficient of variation for CPTEL and TTEL are highest in Vietnam which can be explained by rapid economic growth of which increased teledensity may be a contributing factor. For our next set of analyses, we transformed GDPPC, GNEPC, and LFPR into natural logarithm forms, while we transformed CPTEL, FLTEL, and TTEL into CPPENR, FLPENR, and TPENR, respectively by dividing the teledensity figures by 100. We also expressed labor force participation rate in decimal form instead of percentages. The succeeding sections highlighted the results for determining the best econometric model using the Na?ve, Fixed Effects, and Random Effects processes.Na?ve ModelsWe ran three na?ve models using equation (1). The na?ve model assumes that all coefficients for a given subject are constant over time. Thus, we pooled all 133 observations (7 countries for 19 years) and estimated a grand regression with lnGDPPC as the dependent variable, without reference to the cross-section or time-series nature of data. The results of the na?ve regressions using Stata12 are presented in Appendix 5, where it was observed that in both the CPPENR and TPENR models, lnGNEPC is identified to be significant at α = 0.001, while LFPR is significant at α = 0.05. For the FLPENR model, lnGNEPC, LFPR, and FLPENR is significant at α = 0.001, α = 0.01, and α = 0.05, respectively. Consistent with the prior expectations stated in Gujarati (2009), the R-squared of all the models are high for these na?ve regressions.In addition, we performed tests of plausibility and robustness in each of the na?ve PENR models. As can be seen from Appendix 5, all models do not exhibit multicollinearity and heteroscedasticity but they showed traces of autocorrelation since it fails to account for the unobserved heterogeneity of the variables as well as possible specification errors. In this regard, we need to resort to other estimation techniques: The Fixed Effects Model (FEM) using Least Squares Dummy Variable (LSDV) process, and the Random Effects (REM) model.LSDV-1The LSDV-1 model utilizes the following specification:(2)where = 1, 2, 3, …, 7 (N) and = 1, 2, 3,…, 19.This model assigns dummy variables for each of the cross-section units; that is, countries, to account for the observed heterogeneity of these units. So as not to fall into the dummy variable trap, we do not assign a dummy variable for the base country, i.e. Brunei Darussalam, being the first identified country in our study. Furthermore, the model assumes that while intercepts differ across cross-sectional units, they are identified to be time-invariant. For this study, we determine whether the differences in descriptive characteristics of each ASEAN country involved in this study (such as GNE, LFPR, CPTEL, FLTEL, and TTEL) would affect economic growth, measured in terms of GDP per capita. Appendix 6 shows the results of the LSDV-1 model tested for CPPENR, FLPENR, and TPENR. In CPPENR and TPENR models, both lnGNEPC and t showed statistical significance with a p-value less than α = 0.001 among the variables of interest, excluding the LSDV-1 country dummy variables introduced, and are consistent with our a-priori expectations. For the fixed line model, FLPENR also exhibited statistical significance at α = 0.01. We cannot, however, make the final interpretations until the best model is identified.LSDV-2The LSDV-2 model utilizes the following equation specification:(3)where = 1, 2, 3, …, 7 and = 1, 2, 3,…, 19 (T).This model accounts for structural change over time by considering the time factor in the evolution of technological progress in the telecommunications industry which can affect economic growth. Here, we accounted for the unobserved heterogeneity across time periods. With the advancement of telecommunications technology, the old technologies are dropped out and new technologies are welcomed, resulting to a more competitive market for fixed-line and mobile telephone subscriptions as prices become more reasonable and affordable for businesses and households. So as not to fall into the dummy variable trap, we do not assign a dummy variable for the base year (i.e., 1992).Appendix 7 shows the results of the LSDV-2 model tested for the mobile phone, fixed line and total penetration rates. For the mobile phone model, all variables of interest showed statistical significance with a p-value less than α = 0.001 among the variables of interest, excluding the LSDV-2 time dummy variables introduced. It is interesting to note that lnLF is not consistent with our a-priori expectations, as it exhibited a negative coefficient when regressed with lnGDPPC. For the fixed line model, only lnGNEPC and LFPR exhibited statistical significance at different levels while TPENR is significant at α = 0.05, among the variables of interest and are consistent with our a priori expectations. Again, we cannot make any conclusion as of this time until we arrive at the final model for this study.LSDV-3To account for the unobserved heterogeneity across countries and time periods, we use the LSDV-3 model in the following form: (4)where = 1, 2, 3, …, 7 (N) and = 1, 2, 3,…, 21 (T).This model assigned dummy variables for both countries and years covered, excluding the base country and year, to account for the unobserved heterogeneity across space (countries) and time (years). Intercepts of these models are identified to be time- and space- invariant.Appendix 8 shows the results of the LSDV-3 model tested for the mobile, fixed line, and total penetration rates. In all models, only lnGNEPC exhibited statistical significance at α = 0.001, with the signs of their coefficients consistent to our a priori expectations. We also find CPPENR, FLPENR, and TPENR to be significant at α = 0.05. In all LSDV models, the ANOVA tests suggest that all the variables, when taken collectively, shows statistical significance with p-values less than α = 0.001, thereby indicating its strength to determine the impact of telecommunications variables on economic growth, measured in terms of GDP.Determining the Best LSDV ModelTo determine the best model among the previously-discussed models, we use the Wald’s Test or the restricted F-test that assesses if the restrictions made on each of the preceding models are valid. Reducing the dummy variables indicating the time- and space- variant restrictions of LSDV models to zero, it then results to the original na?ve model that was identified in equation (1). We set the following hypothesis:Ho: The restricted model is valid.Ha: The restricted model is not valid; hence the unrestricted model is valid.The Wald’s Test statistic is computed as follows: (5)where:= residual sum of the squares of the restricted model;= residual sum of the squares of the unrestricted model; and= degrees of freedom of the unrestricted model. Since the null hypothesis sets m restrictions, the restricted model must have fewer variables than the unrestricted model. Using the F-distribution table, we determine whether the computed F-statistic and p-value by Stata12 lies in the critical region of rejecting the null hypothesis, indicating that restrictions are not valid. Following is a discussion of the tests to be conducted, as well as the summary results for all tests.Na?ve vs. LSDV-1In this test, the Na?ve model is the restricted model while the LSDV-1 model is the unrestricted model arising from the restrictions of cross-sectional variables. The hypothesis for this test is:Ho: γi = 0, where i = 2, 3…, NHa: Restrictions are not valid.with the following m restrictions and degrees of freedom from the Unrestricted LSDV-1 model:m = N - 1 = 7 -1 = 6; and DFUR = NT – (N – 1 + k) = 7(19) – (7 – 1 + 5) = 122.Na?ve vs. LSDV-2In this test, the Na?ve model is the restricted model while the LSDV-2 model is the unrestricted model arising from the restrictions of time series variables. The hypothesis for this test is:Ho: δt = 0, where t = 2, 3…, THa: Restrictions are not valid.with the following m restrictions and degrees of freedom from the Unrestricted LSDV-2 model:m = T - 1 = 19 -1 = 18 – 1 (dropped constraint) = 17; and DFUR = NT – (T – 1 + k) = 7(19) – (19 – 1 + 5) = 110 + 1 (dropped constraint) = 111.Na?ve vs. LSDV-3In this test, the Na?ve model is the restricted model while the LSDV-3 model is the unrestricted model arising from the restrictions of both cross-sectional and time series variables. The hypothesis for this test is:Ho: γi = 0 and δt = 0, where i = 2, 3…, N and t = 2, 3…, THa: Restrictions are not valid.with the following m restrictions and degrees of freedom from the Unrestricted LSDV-3 model:m = (N – 1) + (T – 1) = (7 – 1) + (19 -1) = 26 – 1 (dropped constraint) = 23; and DFUR = NT – (N + T – 2 + k) = 7(19) – (7 + 19 – 2 + 5) = 116 + 1 (dropped constraint) = 105.LSDV-1 vs. LSDV-3In this test, the LSDV-1 model is the restricted model while the LSDV-3 model is the unrestricted model arising from the restrictions on time series variables. Since only the time series dummy variables are involved, we are only running a single test instead of the usual three for all penetration rate variables. The hypothesis for this test is:Ho: δt = 0, where t = 2, 3…, T.Ha: Restrictions are not valid.with the following m restrictions and degrees of freedom from the Unrestricted LSDV-3 model:m = (T – 1) = (19 -1) = 18 – 1 (dropped constraint) = 17; and DFUR = NT – (N + T – 2 + k) = 7(19) – (7 + 19 – 2 + 5) = 116 + 1 (dropped constraint) = 105.LSDV-2 vs. LSDV-3In this test, the LSDV- 2 model is the restricted model while the LSDV-3 model is the unrestricted model arising from the restrictions on cross-sectional variables. Since only the cross-sectional dummy variables are involved, we are only running a single test instead of the usual three for all penetration rate variables. The hypothesis for this test is:Ho: γi = 0, where i = 2, 3,…, NHa: Restrictions are not valid.with the following m restrictions and degrees of freedom from the Unrestricted LSDV-3 model:m = N - 1 = 7 -1 = 6; andDFUR = NT – (N + T – 2 + k) = 7(19) – (7 + 19 – 2 + 5) = 116 + 1 (dropped constraint) = 105.Appendix 9 presents the critical F-statistics for each of the tests using Stata12. Summarizing the results, we can determine from Table 3 which is the ideal model to employ for this research. Given the critical F-statistics and the p-value of each test, it can be concluded that the null hypothesis has to be rejected, except for the LSDV-1 vs. LSDV-3 test and the Na?ve vs. LSDV-2 for FLPENR. However, we focus on the first confrontation because both LSDV-1 and LSDV-3 is significant in most instances. But since the time trend t variable is not significant in all cases, there arises confusion as to which among LSDV-1 and LSDV-3 have to be used. Thus, we need to perform additional procedures to determine the best model for this study. In the meantime, we use both LSDV-1 and LSDV-3 in our analyses.Table 3Summary of Restricted and Unrestricted test resultsRestrictedUnrestrictedVariableCritical FP-ValueDecisionNa?veLSDV-1CPPENR33.400.0000***LSDV-1FLPENR35.310.0000***LSDV-1TPENR33.650.0000***LSDV-1Na?veLSDV-2CPPENR1.920.0227*LSDV-2FLPENR1.670.0600Na?veTPENR2.080.0125*LSDV-2Na?veLSDV-3CPPENR9.480.0000***LSDV-3FLPENR9.070.0000***LSDV-3TPENR9.850.0000***LSDV-3LSDV-1LSDV-31.170.3043LSDV-1LSDV-2LSDV-324.410.0000***LSDV-3 Note: *** - significant at α = 0.001;* - significant at α = 0.05. Random Effects and the Breusch-Pagan Poolability TestTo express the so-called “ignorance” on the (true) model to be employed on the disturbance term, as stated in Gujarati and Porter (2009), we need to run another set of regression using the Random Effects Model. This aims to account for the individual differences in the intercept values of each country reflected in the stochastic disturbance term. Appendix 10 presents the regression results using the Random Effects Model. The results are at its best use they are compared with Na?ve model to determine which among the two is better for analysis. Such poolability test will determine the presence of variation in the stochastic disturbance term, indicating the presence of unobserved heterogeneity in the model. Using the Breusch-Pagan Poolability Test (Breusch-Pagan Lagrangian Multiplier Test for Random Effects), we test the following hypothesis:Ho: Na?ve model is better.Ha: Random Effects Model (REM) is better.Table 4Result of the Breusch-Pagan Poolability TestThe null hypothesis will be accepted when the results of the test reveal that there is no unobserved heterogeneity in the model caused by the variation in the error term. Rejecting the null hypothesis indicates the unobserved heterogeneity and hence, there is a need to incorporate the random effects in the model. Given the results of the poolability test in Table 4, a p-value less than α = 0.001 strongly indicates that REM is a more suitable model than the Na?ve model. Hence, we will employ the Hausman test to determine the better model between the Fixed Effects Model, represented by LSDV-3, and the Random Effects Model.Fixed Effects Model vs. Random Effects ModelTo finally determine the ultimate model to be employed in this study, we used the Hausman Specification Test to analyze whether there is a relation between the unique errors and the regressors as well as to test whether there is a significant difference between the fixed and random effects estimators. We use the following hypothesis:Ho: Random Effects Model (REM) is better.Ha: Fixed Effects Model (FEM) is better.The null hypothesis is to be accepted when the test reveals that these unique, random errors are not correlated with the regressors of the model. Hence, REM will be preferred for use instead of the FEM. The results of the tests in Appendix 11 revealed that using LSDV-1, all three models have exhibited statistical significance with p-values greater than α = 0.001. On the other hand, for LSDV-3, CPPENR and TPENR models generated chi-square values that fail to meet the asymptotic assumptions of the Hausman test. It, therefore, does not allow us to come up with generalizations as to whether REM or FEM using LSDV-3 is better. From the results of this test, it can be said that the Fixed Effects Model using the LSDV-1 process is the most appropriate model to be employed in this study. Perhaps the time factor does not strongly influence penetration rates of mobile and fixed telecommunications unlike the difference in the characteristics of the countries in the ASEAN region.Testing for Plausibility and RobustnessWith the results of this test, we can now test for plausibility and robustness of the LSDV-1 model. This model is identified to be suitable because it accounts for the unobserved heterogeneity across countries included in this study. We notice that differences in the characteristics of the ASEAN countries would influence teledensity and penetration rates and are more likely to be indicative of a country’s productivity and economic growth rather than the influence of time trend.Appendix 12 presents the results of the test for multicollinearity employing the Variance Inflation Factor (test). As can be seen for all models, lnGNEPC has the highest VIF value compared to other variables of interest. We also observed high VIF and tolerance values in variables that define the country where a particular observation came from. This result suggests the need for corrective measures to eliminate a high degree of correlation among the variables.In terms of heteroscedasticity, we employed the Breusch-Pagan/Cook-Weisberg test with the null hypothesis that there exists constant variance in the error term. However, the results for CPPENR, FLPENR, and TPENR models (see Appendix 13) reveal otherwise. Thus, with p-values less than α = 0.001, the results of the test indicates the commission of this violation of non-constant variances in all the three models. It, therefore, suggests that the robust option be employed by these models to correct the problem.In terms of autocorrelation, we used the Wooldridge test for autocorrelation in the panel data with a null hypothesis that there is no first-order autocorrelation among the error terms of the dataset drawn for this study. But, with the p-values less than α = 0.001, the results in Appendix 14 indicates that strong existence of autocorrelation in all the three models and corrective measures must be undertaken to make them robust and plausible. Thus, having committed all these violations, it can be inferred that the estimation process carried out by OLS are not efficient. Hence, the robust option for our models is to employ the generalized least squares estimation process since it utilizes first-order difference transformation that will correct the first-order autocorrelation, as well as reducing the standard errors of the regression estimates.Table 5Corrected regression for CPPENR using GLSTable 5 presents the regression results for the CPPENR model using Generalized Least Squares estimation process. As can be seen, lnEXP, LFPR, and t were identified to be the significant variables, with p-values less than α = 0.001 for lnGNEPC and α = 0.05 for LFPR and t. In addition, lnCPPENR did not manifest statistical significance. We interpret the model such that a one percent increase in expenditures results to the 1.07 percent increase in GDP per capita, whereas the passage of time results to an increase in GDP per capita by .006 percent. The result for labor force participation rate is quite disturbing since a one percent increase in LFPR results to a decrease in GDP by .30 percent while a one percent increase in lnCPPENR decreases GDP per capita by 0.28 percent. Since we are expecting that increasing labor force should support economic, a counterintuitive result warrants some explanation.Table 6Corrected regression for FLPENR using GLSTable 6 presents the regression results of the final lnFLPENR model using GLS. From this table, lnGNEPC, LFPR, and t exhibited statistical significance at α = 0.001, α = 0.01, and α = 0.05, respectively. The variable of interest, FLPENR, was identified to be significant at α = 0.05, indicating that fixed-line telephone penetration rate/teledensity contributes to economic growth in a significant way. Interpreting the results, we can say that a one percent increase in expenditures results to a 1.04 percent increase in GDP per capita. In addition, the passage of time contributes to the GDP per capita by 0.003 percent. On the other hand, a one percent increase in labor force participation results to a .20 percent decrease in GDP per capita, resulting in a counter intuitive relationship, which has to be explained as well. Most importantly, a one percent increase in fixed-line penetration rate increases GDP per capita by .263 percent.Table 7Corrected regression for lnTPENR using GLSTable 7 presents the regression results of the final model for lnTPENR. From this table, it can be seen that lnGNEPC and LFPR are significant at α = 0.001 and α = 0.05, respectively. TPENR is identified to be insignificant; this could be explained by the effect of mobile phone penetration rate when we combined them together with fixed line penetration rates. Thus, a one percent increase in expenditure results to an increase in GDP per capita by 1.07 percent. However, a one percent increase in labor force decreases GDP per capita by .0.293 percent. All of the models have been corrected for possible violations and hence, considered to be plausible. However, the inconsistency of the labor force participation rate to support economic growth with prior literature reviewed in this study, warrant sufficient explanation. It is also noticeable that only fixed line telecommunications contribute to the per capita GDP instead of mobile phone telecommunications. Perhaps the countries the ASEAN region have not yet fully realized the benefits of mobile telecommunications and high teledensity brought about by increased subscription to mobile phones.To cap off the tests conducted for panel data, we tested for the hypothesis that the residuals of the FEM LSDV-3 model are contemporaneously correlated; hence we employed the Breusch-Pagan test for Contemporaneous Correlation of Residuals. We developed 7 equations for each cross-sectional unit and performed the regression together with the variables of interest. Should contemporaneous correlation exist among residuals, it is best to employ Seemingly Unrelated Regression Equations for our model. This was validated by our study as evidenced by the results in Appendix 15, indicating that the best policy is to use Seemingly Unrelated Regression to avoid micronumerosity by achieving full complement in our cross-sectional and time series observations.For the control variables, empirical evidence suggests that increase in expenditure stimulates economic growth. Such is expected to happen when the government invests on public infrastructure, and economic and social services that will benefit people and their communities. Surprisingly, the negative relationship between labor force and economic growth signifies the presence of stiff competition on job hunting that there is a high incidence of unemployment among developing countries. It suggests that the government should continue its work in mitigating its problem on unemployment through the provision of decent and stable employment to the labor force so that economic growth may finally be felt. From the variables of interest, only the penetration of fixed line telecommunications has significant relationship with economic growth. It is undeniable that businesses would still resort to the use of fixed-line telephones because traditional economic factors dictate the prominence of the use of fixed landlines, as indicated in Sridhar and Sridhar (2009). It is because basic services provided by landlines have lower rental ceilings that make landline telephones more affordable to the businessmen and households. Furthermore, since these landlines are installed within the premises of the firm, people are restricted to manage their calls related only to business matters to ensure their productivity in effectively managing communications. In this way, businesses can manage their operations effectively due to the proper use of fixed-line telephone facilities.What could probably explain the insignificance of mobile and total penetration rates on the per capita GDP as our measure of economic growth? Even if we control for capital and labor in the model, the results subscribe to the idea of Shiu and Lam (2008) that the continuous improvement and development in mobile telecommunications alone may not be sufficient enough to stimulate economic growth, considering that there are other “complementary” factors like promoting a more stable economic climate conducive to conduct business, the creation and maintenance of public infrastructure suitable for transportation superhighway, providing greater access to quality education, and sufficient training of manpower to encourage people to make the best and proper use of telecommunications in conducting business transactions.Moreover, as most of the countries in South East Asia are identified as developing countries, perhaps these countries are facing difficulties in allocating their resources to fund different investments to support economic growth. This is the rationale behind the results of the study of Negash and Patala (2006) that developing countries have yet to fully realize the benefits derived from investment in telecommunications, because most developing countries have vast rural communities that installing telecommunications facilities are way beyond its practicality. This is consistent to the premise of Roller and Waverman (2001) that investment in telecommunications must first reach a critical level where significant benefits of economic development can be fully achieved. Recognizing the fact that if telecommunications can accelerate economic activities, Negash and Patala (2006) stated that it must reach the rural areas as well. Thus, there must be a better way that mobile telecommunications must be accessible by members of this particular vicinity as they provide the sources of different factors of production that result to the generation of economic output. Furthermore, there must be a better way of exploiting network externalities. Shiu and Lam (2008) posited that penetrations rates may be high in urban areas but not in remote rural areas, where telecommunication services are still hard to reach. As countries continue to develop their telecommunication infrastructures, they still have to recover those sizeable investments; hence, there is a need to impose higher tariffs and fees that makes telecommunication services not affordable in some places in the region. This paper believes that providing affordable telecommunication services, both fixed-line and mobile, are important to achieve the so-called “critical mass” in which the benefits of telecommunications can be fully achieved.Technically speaking, we also cannot discount the fact that there is a need to produce a more rigid statistical scrutiny for the model we employed in this study. Most of the literature cited in this paper utilized various structural forms. For one, Waverman et al. (2005) resorted to a simpler cross-section analysis of data after it was found out that the use of simultaneous equations would not work because of endogeneity issues. In addition, Negash and Patala (2006) utilized a simple trend analysis for the 87 countries studied. Moreover, Datta and Agarwal (2004) enhanced the panel data analysis by using a more dynamic panel data to avoid omitted variable bias for single equation cross-sectional regression, while Grouber and Koutroumpis (2011) adopted the 3SLS generalized method of moments in the estimation process. Lastly, since most literature uses GDP as a measure of growth, perhaps the use of the GDP growth rate itself could be a more viable of measure since GDP actually measures the aggregate output of a particular economy in a given time. Other endogenous variables may also be considered for inclusion in the model, which was not accounted for by this paper given the limited time. Perhaps we could also use another measure of economic growth. Considering that this discourse is a contribution to the still limited compendium of studies about the impact of telecommunications and economic growth in the ASEAN region, the use of more stringent and better statistical techniques must be carefully considered in future research. Summary, Conclusions and Policy RecommendationsThis paper attempted to determine the impact of telecommunications on economic growth in the ASEAN region. The idea behind this study emanated from the theories of network externalities and the spillover effects that reduce transaction costs brought about by the use of telecommunications as a channel to support economic growth. In response to Target 8.F set by the United Nations in its Millennium Development Program, we tested the impact of the continuous increase of penetration by fixed-line and mobile telecommunications, as a result of increased teledensity over the number of inhabitants, on economic growth which is measured in terms of the country’s GDP per capita. The Fixed Effects Model using LSDV-1 process is determined to be the plausible model for this study to account for the unobserved heterogeneity associated with differences in the characteristics of the countries and of the time periods brought about by the continuous surge of GDP, national expenditure, labor force penetration, and teledensity in both mobile and fixed-line communications.Drawn from a panel data of 7 countries from 1992 – 2010, the results of the study revealed that the penetration of fixed line telecommunications is positively related to economic growth for the countries located in this region. Separately, we identified mobile telecommunications and total penetration to be not significantly associated with economic growth. This is because most of the countries in the ASEAN region are developing countries; thus, the benefits of increased investment in telecommunications and teledensity have yet to be realized over time, particularly in mobile telecommunications which were introduced only for more than twenty years. Such investment in telecommunications must be complemented by other factors such as providing quality education and training towards the better and proper use of these facilities, the continuous investment on public infrastructure to support the mobility of the transfer of goods and services, and the continuous upgrading and development of rural areas where the factors of production are located.To contribute to the thrust of the United Nations in achieving its Millennium Development Goals in 2015, certain policies have to be implemented to fully achieve the benefits of the investment and proper use of telecommunications especially in developing countries like those located in the ASEAN region. These are consistent with the recommendations of Shiu and Lam (2008), Negash and Patala (2006), and Waverman et al. (2005). For one, there must be a reduction of tariffs and other charges to subscribers of telecommunication facilities to increase the affordability and accessibility of telecommunication services to the rural and marginalized sectors of the country. In addition, there is a need to further improve the business environment where the benefits of telecommunications can be fully maximized. Moreover, continuous training and development towards the better use of telecommunication facilities must be enforced through the aid of educational institutions and the private sector.The liberalization of the telecommunications industry must be fully enforced to stimulate competition among the different investing networks, making telecommunications become competitive in terms of pricing and providing better services to its users. In addition, there is a need for telecommunications facilities to be set up at various community centers so that people from remote areas may benefit from the use of such facilities. This is to shift our focus from the individual perspective to the community perspective since increasing access to telecommunications will reduce the costs incurred per person, and the accessibility of these services can spur economic growth in all areas of the country. The private and other affiliated sectors should be continuously tapped to support the government in its endeavors to provide a more cost-efficient, more accessible fixed line and mobile telecommunications services for everyone to realize substantial benefits from the use of these services. Consistent with Waverman et al. (2005), continuous efforts to eradicate poverty in other aspects should be seriously considered to maximize the benefits of increased competition and speedy penetration of mobile telecommunications in developing countries. Educational institutions must be involved in teaching students how to use telecommunication facilities properly to help them realize the benefits of such services as they advance in age and as they enter the world of work.This study believes that it is never too late for the ASEAN region to achieve the Millennium Development Goals three years from now. Having achieved economic growth realized from the use of fixed-line communications, this is the best time for this region to better understand the power of mobile telecommunications and the entirety of telecommunications and how it can bring the different member economies closer, as well as the other nearby economies, in the continuous response to the global effort of forging partnerships with the different sectors and industries to eradicate poverty by making available the benefits of new technologies, particularly in information and communications. ReferencesBarro, R. J. (1991). Economic growth in a cross-section of countries. The Quarterly Journal of Economics, 408-443.Barro, R. J., & Sala-i-Martin, X. (1992). Convergence. Journal of Political Economy, 100 (2), 223-251.Biancini, S. (2011). Behind the scenes of the telecommunications market: An empirical analysis of the Indian market. Telecommunications Policy, 35, 238-249. doi: 10.1016/j.telpol.2010. 12.011Brock, G., & Sutherland, E. (2000). Telecommunications and economic growth in the former USSR. East European Quarterly, 34 (3), 319-335.Cameron, G. (1998). Economic growth in the information age: From physical capital to weightless economy. Journal of International Affairs, 51 (2), 447-471.Chakraborty, C., & Nandi, B. (2009). Telecommunications adoption and economic growth in developing countries: Do levels of development matter? Journal of Academy of Business and Economics, 9 (2), 51-61.Chakraborty, C., & Nandi, B. (2011). ‘Mainline’ telecommunications infrastructure, levels of development and economic growth: Evidence from a panel of developing countries. Telecommunications Policy, 35, 441-449. doi: 10.1016/j.telpol.2011.03.004Cronnin, F., Colleran, E., & Gold, M. (1997). Telecommunications, factor substitution, and economic growth. Contemporary Economic Policy, 15 (3), 21-31.Datta, A., & Agarwal, S. (2004). Telecommunications and economic growth: A panel data approach. Applied Economics, 36, 1649-1654.Gruber, H., and Koutroumpis, P. (2011). Mobile telecommunications and the impact on economic development. Economic Policy, 387-426Gujarati, D. and Porter, D. (2009). Basic econometrics. Singapore: McGraw-HillKeating, B. (2001). Economic dimensions of telecommunications access. International Journal of Social Science and Economics, 28 (10-12), 879-898.Lam, P., & Shiu, A. (2010). Economic growth, telecommunications development and productivity growth of the telecommunications sector: Evidence around the world. Telecommunications Policy, 34, 185-199. doi: 10.1016/j.telpol.2009.12.001Melamed, L. (2007). The impact of growing mobile telephony penetration in Canada: Lots more Gross Domestic Product. Unpublished paper. Canada: S. Melamed Associates Inc.Negash, S., & Patala, L. (2006). Telecommunication investment in economically developing countries. Proceedings of the 2006 Southern Association for Information Systems Conference, USA, 29-34.Romer, D. (2006). Advanced macroeconomics. USA: McGraw-Hill.R?ller, L., & Waverman, L. (2001). Telecommunications infrastructure and economic development: A simultaneous approach. The American Economic Review, 91 (4), 909-923.Romer, P. (1986). Increasing returns and long-run growth. Journal of Political Economy, 94 (5), 1024-1037.Salazar, L. C. (2006/2007). Competition drives growth: The liberalized telecommunication sector in the Philippines. In R. H. K. Cheng and R. Sen (Eds.), Regional outlook (pp. 122-124). Singapore: Institute of Southeast Asian Studies.Shiu, A. & Lam, P. (2008). Causal relationship between telecommunications and economic growth in China and its regions. Regional Studies, 42 (5), 705-718. doi: 10.1080/00343400701543314Snow, M. (1989). Facilitating ASEAN-U.S. trade and direct foreign investment in information services. ASEAN Economic Bulletin, 6 (1), 31-45.Solow, R. (1956). A Contribution to the Theory of Economic Growth. Quarterly Journal of Economics, 70 (1), 65-94. Sridhar, K., & Sridhar, V. (2009). Telecommunication infrastructure and economic growth: evidence from developing countries. Μехан?зм Ре?улювання Економ?ки, 2, 91-116. The United Nations. (2000). United Nations millennium declaration. Retrieved November 4, 2012 from the United Nations website declaration /ares552e.htmThe World Bank Database. [Data file]. Available from The World Bank Database Web site, data..Tong, Go Chok. (September – October 2003). ASEAN and information technology. Presidents and Prime Ministers, 27-28.Trong Vu, S. T. (2009). Telecommunications sector in ASEAN: ASEAN economic integration and its implications for labour in the region. Retrieved November 30, 2012 from ASEAN Services Employees Trade Union Council Library Website library.php Waverman, L., Meschi, M., & Fuss, M. (2005). The impact of telecoms on economic growth in developing countries. In A. Sarin (Ed.), Africa: The impact of mobile phones (pp. 10-23). Berkshire, London: The Vodafone Group.Appendix 1Summary of nine ASEAN Telecommunications Ministers (TELMIN) meetings from 2001-2009 DateVenueVerbatim HighlightsJuly 2001Kuala Lumpur, MalaysiaFocused on narrowing the digital divide among ASEAN nations; development of skills and expertise, promotion and enhancement of use of information and communications technology, development of local and regional content for the ASEAN region; and ensuring a well-coordinated network security. The members signed the Ministerial Understanding on ASEAN Cooperation in Telecommunications and Information Technology. It also implemented the e-ASEAN Framework Agreement.August 2002Manila, PhilippinesStressed the vital role of telecommunications in the integration of regional economy, in attracting investments into region, and in developing the economies of the ASEAN members. It also established the ASEAN ICT Center to foster ASEAN cooperation in ICT, as well as the agreement to engage China, Japan, and South Korea in ICT cooperation individually and in the ASEAN+3 Framework. September 2003Singapore, SingaporeThe ASEAN ministers expressed their commitment of enhancing regional cooperation in cybersecurity. It also endorsed initiatives to advance ICT integration and trade facilitations, promote universal access to ICT infrastructure and services, and develop ICT skills and competencies. The meeting launched the ASEANconnect, a publicly accessible digital divide database that houses key data statistics and measurement indicators, analysis of initiatives to bridge the digital divide among ASEAN nations.August 2004Bangkok, ThailandPromotion of e-learning culture towards a knowledge-based ASEAN to make ASEAN competitive in the world economy. The ASEAN also established the ASEAN ICT Fund to accelerate the implementation of the ASEAN ICT Work Program.September 2005Hanoi, VietnamEmphasized the importance of cooperation in telecommunication among ASEAN members. It became witness to the Hanoi Agenda on Promoting Online Services and Applications to Realize e-ASEAN and ICT Focus 2005-2010.September 2006Bandar Seri Begawan, Brunei DarussalamImplemented the Brunei Action Plan which outlined a program of action in building ASEAN capacity to enhance the region’s competitiveness in the ICT sector. The ministers agreed to revitalize the e-ASEAN Youth Forum and the e-ASEAN Business Council.August 2007Siem Reap, CambodiaStressed the importance of strong human and institutional capacity and private sector participation to transform the telecommunications and IT sector into an important enabler for economic growth and poverty reduction. The ministers adopted the Siem Reap Declaration on Enhancing Universal Access of ICT Services in ASEAN. It also highlighted the importance and possibility of cooperation with the ASEAN Dialogue Partners such as China, India, Japan, and Korea.August 2008Bali, IndonesiaEmphasized the crucial role of a secure and well-established information infrastructure among ASEAN countries for the region’s economic growth and competitiveness. The ministers adopted the Bali Declaration in Gorging Partnership to Advance High-Speed Connection to Bridge the ASEAN Digital Divide. October 2009Vientiane, LaosAdopted the Vientiane Declaration on Promoting the Realization of Broadband Across ASEAN. The ministers agreed that the vision of the ASEAN ICT Master Plan will identified “Towards an Empowering and Transformational ICT: Creating and Inclusive, Vibrant, and Integrated ASEAN” to being the ASEAN ICT to a higher level and to reinforce the role of ICT for ASEAN integration.Note. From Trong Vu, S. T. (2009). Telecommunications sector in ASEAN: ASEAN economic integration and its implications for labour in the region. Retrieved November 30, 2012 from ASEAN Services Employees Trade Union Council Library Website Appendix 2Summary of agreements and declarations of ASEAN members on telecommunications TitleDateVerbatim HighlightsThe e-ASEAN Framework AgreementNovember 2000The agreement has the following objectives:Promote cooperation to develop, strengthen and enhance the competitiveness of the ICT sector in ASEAN;Promote cooperation to reduce the digital divide within ASEAN Member States and amongst ASEAN Member States;Promote cooperation between the public and private sectors in realizing e-ASEAN; andPromote the liberalization of trade in ICT products, ICT services, and investment to support the e-ASEAN initiative.The Ministerial Understanding on ASEAN Cooperation in Telecommunications and Information TechnologyJuly 2001The agreement has the following objectives:Develop the ASEAN telecommunications and IT sector as a catalyst to foster regional economic integration;Enhance the overall competitiveness of the ASEAN region through a vibrant telecommunications and IT industry; andDevelop the ASEAN Information Society, where its citizens are able to work, communicate and recreate in the knowledge-based economy. The Vientiane Action Program on Telecommunications and IT Sector2004The action program is to leverage on information and communications technology via public-private sector partnerships and strong external linkages, to build a connected, vibrant, and secure ASEAN community by:Striving for universal access to ICT infrastructure and service;Encouraging the development of a pervasive, inter-connected, and secure ASEAN information infrastructure;Strengthening the cooperation and assistance on regulatory policy and strategy issues;Creating digital opportunities through e-government, e-commerce, and e-society initiatives;Enhancing the competitiveness and dynamism of the ASEAN ICT sector by promoting and facilitating trade and investment in ICT services; andDeveloping highly skilled ICT human resources.The ASEAN Sectoral Integration Protocol for e-ASEAN2004In line with the progressive, expeditious, and systematic implementation of the e-ASEAN, the protocol was signed with the following objectives:Building ASEAN capacity, which is a vital component in enhancing ICT competitiveness;Developing the ASEAN Information Infrastructure as the foundation for the sustainable development of an information society;Achieving broader economic and social benefits through wider access to ICT;Facilitating ICT trade and electronic commerce by addressing non-tariff barriers to trade, as well as to lay policy and legal infrastructure for electronic commerce;Exchanging information on, and where appropriate harmonize, policies and regulations to increase ASEAN’s ICT competitiveness and welcoming the implementation of the ATRC 2006-2007 work plan;Engaging the private sector and youths;Forging links with partners and key ICT international organizations to pool our resources and expertise; andStrengthening institutional foundations to achieve the programs elaborated above.The Siem Reap Declaration on Enhancing Universal Access to ICT Services in ASEAN: “ICT Reaching Out to the Rural”August 2007The declaration is the efforts of the ASEAN members to build a connected, vibrant, and secure ASEAN community.The Bali Declaration in Forging Partnership to Advance High Speed Connection to Bridge Digital Divide in ASEAnAugust 2008The declaration aims to deepen and strengthen regional initiatives and activities towards enhancing the infrastructure of the ASEAN information society, and to establish the foundation for ICT applications, services and solutions in the region.The Vientiane Declaration on Promoting the Realization of Broadband across ASEANOctober 2009The declaration is a guide to promote broadband initiatives to enable ICT to become a major empowering and transformative force in the ASEAN Information Society.Note. From Trong Vu, S. T. (2009). Telecommunications sector in ASEAN: ASEAN economic integration and its implications for labour in the region. Retrieved November 30, 2012 from ASEAN Services Employees Trade Union Council Library Website 3Data for the study and descriptive informationCcodeGDPPC GNEPC LFPRCPTELFLTELTTELt119478.10 16,338.65 67.21.5418.0319.571118997.88 16,585.70 67.43.0320.1323.162119065.32 15,735.20 67.65.5421.8527.393119395.52 15,532.07 67.912.3923.5235.914119445.25 18,063.39 68.314.6526.5141.165118685.24 16,151.68 68.814.7725.2039.976118135.13 15,601.04 69.315.7424.9040.647118253.78 14,123.73 69.720.6524.7445.398118350.13 12,563.04 70.129.0524.6253.669118441.35 12,827.92 70.442.7726.4569.2210118749.58 13,429.67 70.344.9823.7968.7811118896.53 12,823.16 70.250.8523.4974.3412118609.15 12,991.12 70.056.8823.3580.2213118311.88 13,145.67 69.964.1423.1087.2414118745.80 13,522.18 69.781.3921.65103.0415118416.91 15,340.54 69.696.9921.07118.0616117722.66 16,189.52 69.4103.6821.00124.6817117092.46 16,034.47 69.3105.3720.56125.9318117225.32 16,870.95 69.2109.0720.03129.10192669.14 615.83 66.60.020.870.8912706.51 647.00 67.50.030.960.9922748.32 704.87 68.20.041.251.2932799.31 784.95 68.90.111.651.7642848.25 831.02 69.50.282.072.3552875.96 878.53 67.80.452.432.8862750.81 697.33 67.20.512.683.1972746.79 682.46 69.61.052.893.9482773.31 691.98 69.41.723.124.8492791.08 718.50 69.23.023.346.36102816.02 738.00 69.05.343.548.88112844.18 752.85 69.28.343.6311.97122875.73 803.07 69.413.514.6218.13132914.60 834.20 69.720.645.9426.58142953.94 864.25 69.727.756.4534.201521003.36 908.94 69.740.178.4048.571621052.43 951.90 69.659.8312.9372.761721089.72 977.83 69.868.9414.6683.601821145.39 1,026.38 69.788.0817.06105.1519CcodeGDPPC GNEPC LFPRCPTELFLTELTTELt32932.02 2,769.75 64.31.0410.8911.94133140.83 3,039.84 64.31.7312.2413.96233344.57 3,314.88 64.32.8314.1717.00333581.95 3,688.02 64.34.8516.0820.93433842.64 3,800.13 64.37.1617.7524.91534022.84 3,991.34 64.39.1819.3928.57633636.47 2,877.51 64.39.8619.6429.50733767.64 2,851.24 64.213.0819.3832.45834005.56 3,235.98 65.421.8719.7741.64933933.94 3,178.06 65.130.8219.6550.471034052.88 3,300.67 64.936.9319.0555.981134194.26 3,398.62 64.644.3918.2462.631234385.97 3,616.07 64.457.1017.3774.471334529.60 3,738.89 64.174.8816.7391.611434706.88 3,952.24 63.873.2116.3389.541534925.77 4,247.56 63.586.3116.08102.391635077.94 4,421.00 63.2100.7716.41117.181734914.91 4,247.86 62.9107.8516.19124.041835184.71 4,697.26 62.8119.2216.10135.32194942.64 956.39 67.20.091.021.1114940.56 972.96 66.80.151.301.4524959.65 977.80 66.80.251.641.8934982.03 1,018.58 67.90.712.042.75441016.31 1,063.88 67.81.352.523.88541045.46 1,080.11 67.71.862.874.72641016.78 1,081.69 68.22.343.375.71741025.55 1,052.97 68.03.773.827.59841048.07 1,068.94 66.58.353.9612.31941055.81 1,120.11 69.315.404.2019.601041071.69 1,166.90 68.119.084.1123.181141102.22 1,188.72 68.727.354.0631.411241153.02 1,209.98 68.039.244.1043.331341185.38 1,233.63 66.340.663.9444.591441225.05 1,221.63 65.449.214.1753.381541283.47 1,248.10 64.864.684.4469.131641314.23 1,307.44 65.475.544.5280.061741307.14 1,299.32 65.882.437.4089.821841383.40 1,382.96 65.985.677.2792.9419516970.76 14,818.63 70.53.7636.6440.401518446.17 16,300.48 70.15.4437.8943.332519767.48 16,770.40 70.36.9639.3446.303520571.46 17,152.94 69.08.7941.0349.824CcodeGDPPC GNEPC LFPRCPTELFLTELTTELt521258.53 17,847.42 70.812.0543.6955.745522305.68 19,052.57 70.123.1245.9069.026521092.56 16,536.27 70.029.1047.2776.377522221.38 17,742.93 70.642.4248.8191.238523814.56 20,754.17 71.070.1049.65119.759522913.32 19,052.70 71.075.1548.92124.0710523658.87 18,849.90 71.182.1647.80129.9611525110.50 17,649.49 71.287.5446.24133.7812527068.97 20,008.62 71.395.9344.64140.5613528388.87 20,025.82 71.5102.7843.23146.0214529925.50 21,169.27 71.7108.5942.03150.6215531247.00 21,522.48 72.0129.2140.61169.8216530131.62 23,536.53 72.3134.4239.31173.7317528949.86 20,835.69 72.3139.1138.94178.0418532640.68 21,407.04 72.8145.1839.24184.431961599.21 1,739.67 81.40.433.073.50161718.03 1,877.46 79.50.703.774.48261858.00 2,040.14 77.61.254.655.90362011.83 2,266.73 78.02.185.848.01462109.05 2,407.49 78.53.066.909.97562057.07 2,151.22 78.63.627.9211.54661819.18 1,605.57 77.83.218.1711.38761877.31 1,659.08 76.73.758.3612.11861943.24 1,775.48 77.44.848.8513.69961962.24 1,789.07 77.811.829.4721.281062042.80 1,868.89 77.815.7310.1425.881162164.30 1,995.73 77.733.3910.1543.541262277.56 2,142.35 77.941.4410.3151.761362359.64 2,287.82 78.146.6810.5557.231462458.52 2,291.95 77.560.5310.5171.041562562.72 2,331.60 78.178.1410.3688.501662608.25 2,428.29 78.290.5810.83101.411762531.23 2,216.82 76.295.9910.49106.481862712.51 2,447.56 77.1103.6210.02113.64197252.58 246.31 84.60.000.220.2217268.29 280.34 84.50.010.360.3727287.14 313.27 84.30.020.610.6237309.41 336.40 84.10.031.051.0847332.94 362.67 83.80.091.581.6757354.51 379.48 83.60.211.751.9667369.24 397.57 83.40.292.262.5577381.10 396.81 83.30.422.703.138CcodeGDPPC GNEPC LFPRCPTELFLTELTTELt7401.55 411.44 83.01.003.234.2397423.83 433.59 82.81.573.835.40107448.60 473.46 82.62.364.887.24117475.97 513.78 82.43.375.418.78127506.94 540.47 82.26.0312.3118.34137543.36 566.59 81.911.5411.2522.78147581.58 614.38 81.722.4710.1932.66157623.96 734.02 81.452.9613.1366.10167656.31 785.36 81.287.1117.18104.29177684.00 798.96 81.2113.0320.05133.08187722.81 849.74 81.2127.0016.36143.3719Legend:NameDescriptionCcodeNumerical code assigned to the country in alphabetical order (1 = Brunei Darussalam, 2 = Indonesia, 3 = Malaysia, 4 = Philippines, 5 = Singapore, 6 = Thailand, 7 = Vietnam)GDPPCGDP per capita (in constant 2000 US$)GNEPCGross National Expenditure per capita (GNE in constant 2000 US$ divided by the population)LFPRLabor Force Participation Rate, Total (ages 15-64, % of total)CPTELMobile Phone Teledensity (No. of subscribers per 100 inhabitants)FLTELFixed Line Teledensity (No. of fixed-line telephones per 100 inhabitants)TTELTotal Mobile Phone and Fixed Line TeledensitytTime Trend; (1 = 1992, 2 = 1993…, 19 = 2010)Note: Data for GDPPC, GNE, LFPR, CPTEL, and FLTEL were obtained from the World Bank Database (data.). All other variables are computed by the researcher. Appendix 4Descriptive statistics for the variables of interest per country1 - Brunei Darussalam?GDPPC1 GNEPC1 LFPR1 CPTEL1 FLTEL1 TTEL1 Mean18,527.2614,940.5169.1745.9722.8468.81Std. Deviation666.831,675.371.0237.682.3336.96Minimum17,092.4612,563.0467.201.5418.0319.57Maximum19,478.1018,063.3970.40109.0726.51129.10Skewness-0.66-0.02-0.790.53-0.230.42Kurtosis0.28-1.27-0.64-1.17-0.60-1.18Coef. Variation3.60%11.21%1.48%81.97%10.20%53.71%2 - Indonesia?GDPPC2 GNEPC2 LFPR2 CPTEL2 FLTEL2 TTEL2 Mean863.41795.2668.9317.895.1823.07Std. Deviation133.20116.710.9827.004.7731.73Minimum669.14615.8366.600.020.870.89Maximum1,145.391,026.3869.8088.0817.06105.15Skewness0.710.42-1.271.611.521.61Kurtosis-0.28-0.720.431.591.371.57Coef. Variation15.43%14.68%1.42%150.94%92.03%137.55%3 - Malaysia?GDPPC3 GNEPC3 LFPR3 CPTEL3 FLTEL3 TTEL3 Mean4,114.813,598.2664.1642.2716.9259.19Std. Deviation653.10561.210.6840.092.4940.19Minimum2,932.022,769.7562.801.0410.8911.94Maximum5,184.714,697.2665.40119.2219.77135.32Skewness0.020.29-0.480.67-0.960.58Kurtosis-0.79-0.810.26-0.990.75-1.01Coef. Variation15.87%15.60%1.06%94.86%14.70%67.90%4 - Philippines?GDPPC4 GNEPC4 LFPR4 CPTEL4 FLTEL4 TTEL4 Mean1,108.341,139.5867.0827.273.7230.99Std. Deviation137.21123.831.2530.691.6832.11Minimum940.56956.3964.800.091.021.11Maximum1,383.401,382.9669.3085.677.4092.94Skewness0.660.25-0.180.850.590.85Kurtosis-0.78-0.87-0.89-0.740.89-0.68Coef. Variation12.38%10.87%1.86%112.54%45.15%103.61%5 - Singapore?GDPPC5 GNEPC5 LFPR5 CPTEL5 FLTEL5 TTEL5 Mean24,551.7819,001.7571.0368.5243.22111.74Std. Deviation4,605.422,235.370.9450.574.1550.53Minimum16,970.7614,818.6369.003.7636.6440.40Maximum32,640.6823,536.5372.80145.1849.65184.43Skewness0.240.09-0.060.060.10-0.12Kurtosis-1.07-0.52-0.04-1.50-1.38-1.50Coef. Variation18.76%11.76%1.33%73.80%9.61%45.22%6 - Thailand?GDPPC6 GNEPC6 LFPR6 CPTEL6 FLTEL6 TTEL6 Mean2,140.672,069.6377.9931.638.4440.07Std. Deviation322.66275.411.0936.812.4738.52Minimum1,599.211,605.5776.200.433.073.50Maximum2,712.512,447.5681.40103.6210.83113.64Skewness0.27-0.231.660.92-1.080.85Kurtosis-0.96-1.355.03-0.69-0.03-0.80Coef. Variation15.07%13.31%1.39%116.39%29.25%96.13%7 - Vietnam?GDPPC7 GNEPC7 LFPR7 CPTEL7 FLTEL7 TTEL7 Mean453.90496.5682.8022.616.7629.36Std. Deviation146.64183.901.1741.026.4646.70Minimum252.58246.3181.200.000.220.22Maximum722.81849.7484.60127.0020.05143.37Skewness0.430.680.031.840.801.73Kurtosis-1.00-0.66-1.322.09-0.761.72Coef. Variation32.31%37.03%1.41%181.44%95.55%159.04%Appendix 5Regression results of the na?ve models for CPPENR, FLPENR, and TPENRMobile Penetration Rates (CPPENR)Regression Test for MulticollinearityTest for HeteroscedasticityTest for AutocorrelationFixed Line Penetration Rates (FLPENR)RegressionTest for MulticollinearityTest for HeteroscedasticityTest for AutocorrelationTotal Penetration Rates (TPENR)RegressionTest for MulticollinearityTest for HeteroscedasticityTest for AutocorrelationAppendix 6Regression results of the LSDV-1 models for CPPENR, FLPENR, and TPENRMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 7Regression results of the LSDV-2 models for CPPENR, FLPENR, and TPENRMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 8Regression results of the LSDV-3 models for CPPENR, FLPENR, and TPENRMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 9Results of Wald’s TestNa?ve vs. LSDV-1Mobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Na?ve vs. LSDV-2Mobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Na?ve vs. LSDV-3Mobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)LSDV-1 vs. LSDV-3LSDV-2 vs. LSDV-3Appendix 10Random Effects Model (REM) regression results for CPPENR, FLPENR, and TPENRMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 11Results of the Hausman Test for CPPENR, FLPENR, and TPENR using LSDV-1 and LSDV-3LSDV-1Mobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)LSDV-3Mobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Note: Because we cannot draw conclusions from LSDV-3 due to the failure to meet asymptotic assumptions in getting the p-value of such test, we find LSDV-1 to be a more suitable model and thus, LSDV-1 will be employed in this study.Appendix 12Tests for Multicollinearity of LSDV-1 models using Variance Inflation Factor and Tolerance testsMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 13Tests for Heteroscedasticity of LSDV-1 models using Breusch-Pagan/Cook-Weisberg TestMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 14Tests for Autocorrelation of LSDV-1 models using the Wooldridge TestMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR)Appendix 15Tests to determine whether to use Seemingly Unrelated RegressionMobile Penetration Rates (CPPENR)Fixed Line Penetration Rates (FLPENR)Total Penetration Rates (TPENR) ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download