sun. .nflfifimurunflfl. . v3.0: 4.111.... _ .a .. II. I 11 ‘ .e Iv:’t!.)-.!. 9... : ..ar..s$un.|. . . .. . . . 2 .u‘.‘lu\vv.b. lv-I....l_....;.f.«u.t...r ..| . . ‘ . ‘ ‘ . . h... It Duh}..- :1: a. - a... ‘A ‘Izi.r:vfi, v II Vv .r.. H. .. ‘ . . ‘ o. s; .30. n1. . I! 4!]. ‘2: o layerivvfl .4 .00.. vi: . . . , .. v!.’l.< ItifircllO-Icl‘ l... . ~ . x . a h . . . . 1‘, v0.9.0.0 ...v!rov ‘lnl no! I: . , . ‘ ...Ilvialb$o’l .1 .‘ . . . Z . . . I ‘ . . .. . 0:70.. 2‘31131 .tr‘... V . A... y . . » y . . a. 11.4.! ‘ Io.v4 . . . .. . ‘ 3a 1 . .v..e , . . I 0: 31v Ive-l 00f. Ek’n‘ '13-. .0. . 7-0 a a y) .b; 107‘: ~ 1.! . ‘.1A.¢t)vv.11| I; I. . :w‘; l. A . .II AZ): . n .5 ‘V to ‘Inlu (7' tug-I! . . (IV? 11") iOu¢ “‘QO’Y a \K 7 3.... lie VDI lrkifo'3vut- : .13. Q... v'.v....|. .17“: :Ofi: .1. . :13...) .e' lnisbib ! .1. p.‘ 1.3.} u. . . uh I 10 percent >25 percent Urbanization -> Literacy -> Media Exposure -> Participation The key to this progression was empathy: "the capacity to see oneself in the other fellow's situation" (ibid:50). Empathy made for a mobile personality and thereby made change possible. Lerner's study highlights several themes common to modernization theory: a bipolar classification of society (modern vs traditional) modified with an intermediate category; the use of indices of modernity-~urbanization and literacy; and the importance given to personality types. Strengths of Lerner For Lerner, the social processes of urbanization and mass media exposure are crucial to modernization. Such a social-process approach focuses on change. Its virtue lies 4r- .U -unu ’- 15 in its efforts to establish relationships between variables and particularly between changes in sets of variables (Huntington 1976:50). Another strength of Lerner's research is that he was able to attempt empirical construction and causation modeling using some 2,000 interviews from six Middle Eastern countries he had gathered. This is in marked contrast to many critics then and now who have only attempted to theorize relationships. It has been suggested that the quantitative empiricism of North American social science research was partly instrumental in the thrust of Lerner's work (Rogers and Singhal 1989:20). Weeknesses in Lerner Although Lerner intended his model to be of universal application, some cracks in the seams were, on hindsight, emerging even in his book. For example, in Egypt and Iran, literacy was making for an unemployed class of college graduates in cities. In Syria, Lerner mentioned the need to raise rural standards to "decelerate movement off the land" and "to support industrial expansion" (ibid;267). In short, he was actually discouraging urbanization so as to encourage development, a process that went against his own model. Then there were the Bedouin who did not want to be modernized in Lerner's sense of the word. There were therefore even in Lerner's book, exceptions to the model he created. In a test of the model some 10 years later, Schramm and l6 Ruggels reported that they failed to find the basic relationship of urbanization to literacy. Further, they could not find one pattern of mass media growth but instead found that the pattern varied widely by region and culture (1967). Lerner himself later noted that the pattern of urbanization->literacy->mass media exposure was reversed in the developing countries (1976:290). Along with explosive population growth and mass media exposure, those in rural areas migrated to the urban areas only to lead unproductive lives. He has also noted the failures of development in many parts of the world. He said that just 15 years after Point IV, development programs have stalled and even reversed; stable traditional societies have fallen apart; the revolution of rising expectations has led to the revolution of rising frustrations (Lerner 1967). He blamed the failures on projects that were too small, the politicization of development policies both in the US aid agencies and in the developing countries and anti-Western rhetoric, which he considered an obstacle to development. Finally, he has also blamed people--"[t]hey don't do what, on any rational course of behavior, they should do" (1958, vii). Rogers and Singhal say there were internal and external forces. The internal causes were traditionalism, widespread poverty, illiteracy, a burgeoning population, the lack of an adequate infrastructure to tap the resources, and a rigid ' l7 and often corrupt bureaucracy inherited from colonialism. The external causes were the international terms of trade, economic imperialism of multinational corporations, and the vulnerability and dependence of developing countries (1989:23-24). Criticisms Others criticisms are grouped into several several broad areas below: Concept The first is conceptual. There are several components to this. The major set of criticism is on the ambiguity in the definition of modernity. Huntington has pointed out that modernity and tradition are essentially asymmetrical concepts, which means they cannot be contrasted (1976:35). Theorists, however, define modernity as a virtue where the modern ideal is set forth and the leftover is defined as traditional. This is not a true dichotomy and is therefore not a useful conception. Another set of criticism is that the theory implies the elimination of tradition. This is not necessarily the case as So points out that a Chinese could well take a herbal drink in the afternoon and an aspirin in the evening (1990:56). Tipps points out that colonial domination has produced hybrid practices that cannot be considered modern nor traditional (1976:77). Further, some traditional 18 practices may be used to aid development. Also, the focus by its own design, ignores external factors. An alternative explanation in the form of dependency theory (discussed more fully below) has been developed to counter this neglect. It is ironic that the politics of the context should be criticised because Lerner's model has a political goal--participatory democracy-~as one of the ultimate goals of development. Another criticism is that modernization is seen as a linear process heading in the general direction of progress, and even irreversible. This conceptualization is in contrast to earlier explanations that saw society as moving in a rise and fall pattern. Empirical evidence in the inner cities of developed countries has shown that urbanization does not necessarily lead to literacy, that the presence of mass media does not always help, and that development can be reversed. Finally, the theory assumes the hypodermic needle model of the media where the effects of the media are direct and powerful. Around the time that Lerner was completing his book, Katz and Lazarsfeld (1955) were completing their description of the two-step flow theory of mass communication. The two-step flow theory suggests that mass media influence is mediated and indirect and much less powerful than the influence exerted by the opinion leaders. 0n hindsight, Lerner had the data to incorporate the change. In his study of Turkey, he mentions the grocer of Balgat who 19 has access to media and therefore can imagine a world different from the one he lives in. But this knowledge makes him different from other villagers and the grocer is disdained by the village chief. When Lerner returns a few years later, however, the grocer had died. But, in a sense, he had been resurrected by the chief's sons who were now grocers. The villagers now recognized the grocer as a man ahead of his time. The emulation of the grocer by the chief's sons suggests that personal influence had played a part in their change. Ideology The second area of criticism is ideological. Typical is the suggestion by Tipps that "[t]he idea of modernization is primarily an American idea, developed by social scientists in the period after the Second World War and reaching the height of its popularity in the middle years of the 19605. "American society tended to be viewed as fundamentally _consen5ual, combining an unmatched economic prosperity and political stability within a democratic framework. Such social problems as might exist, moreover, were treated not as endemic but rather as aberrations which could be resolved by normal political processes within existing institutions. After two decades of turmoil, the postwar tranquility of prosperity and stability seemed no mean accomplishment. The future of modern society now seemed assured; only that of the 'developing areas' appeared problematic" (Tipps 20 1976:71). He suggests that optimism led scientists to believe that modernity was an unmixed blessing. Modernization, therefore, does not reflect the state of the Third World but instead reflects "a particular phase in the development of a single society, that of the United States" (Tipps 1976:73). Along a similar line, others have said that modernization does not draw a distinction between itself and Westernization. That is, modernization sees a modern society as a Western society. There is no model of a modern non- Western society. To such criticisms, essentially of being ethnocentric and the then-vogue phrase "Ugly American", Lerner defends his model as being of universal application. Lerner (1967) replies that the model was not a Western model but a model developed in the West that is globally applicable. Further, other studies had borne out the model (ibid, viii). He also counters that it is the ethnocentricism of developing countries, expressed in wars, that hindered development (1958:viii). He says, "One plain lesson is that many of the Asian peoples do want materialism in the sense of better homes, better food, better hygiene, better education, better lives for their children . . ." (emphasis in original 1967:114). To some extent, Lerner is right. Ethnocentricism is not the monopoly of the Western world. There have been ethnic wars then and now. Further, universal indicators of 21 development such as literacy and longevity point to the Western world as being more developed. And critics who point out the absence of a non-Western model have not done better; they have not drawn up an Eastern or a Southern "counter- model." Methodology. The third area of criticism is in methodology. Instead of doing before and after studies of developing countries, modernization studies have taken time-slices and compared them with developed countries. It is thus assumed, for instance, that 20th century Middle East is developmentally the same as 18th or 19th century Europe (So 1990). It also assumes that the traditional societies have more in common with each other than with modernity. Finally, where the model does not fit, "chance" factors are used to account for the deviation. Res t The criticisms have led to divergent results. One school has attempted an outright rejection of modernization. This school, dependency, will be discussed in greater depth below. Another school has attempted to revise modernization through rectifying or eliminating the more glaring deficiencies. This is a new pluralism or multiplicity paradim, also known as Another Development. They are 22 discussed in depth below. Whitacker suggests that it may be more useful to have limited but empirically grounded generalizations than to construct alternative models at the same high abstraction level of modernization theory (1967). The attraction of this approach is that its implications may be practicable. By and large, however, his call has not been taken up. One school along this line is that the modernization theory is applicable to Asia. As Pye puts it: modernization has a closer empirical fit with Asia than Africa or Latin America (Pye 1985:3). Apter has observed that confidence in modernization theory has "declined everywhere except perhaps in China, Japan, or Taiwan (where it is still accepted that innovation, increasing productivity, and growth will result in integrative change, more and better jobs, and greater choice despite temporary setbacks and the need for adjustments) (1987:17). Pye says that Asia had elite-guided development; Japan and China wanted to gain strength and security by adopting modernizing technologies. He says Latin American states were not newly emerging but had evolved their own distinctive political and social systems. He says that postcolonial Africa lacked nationalism and earnest commitment to modernization that were characteristic of Asia. On this final point, his case is weak because if those two elements were key, then it is surely a matter of working them into 23 the model, rather than creating an exception. So (1990) has suggested that there is a new school of modernization studies with several distinctive elements: they do not treat modernization and tradition as mutually exclusive concepts; their methodology is focused not at the abstract level but on concrete cases; development is not viewed as linear or unidirectional after the path of the West; and they consider also external forces. Interestingly enough, two of the three studies he cites are of Asian cases, Hong Kong and Japan. The third is a case study of Iran suggesting that modernization does not necessarily lead to secularization (1990:63-78). These cases, however, are not directly applicable to development communication. Perhaps the most elaborate statements of the changes are from Rogers (1976a, 1976b, 1989) and Servaes (1991). Although stating that they reject the'modernization paradigm, they have in fact incorporated several key elements in the modernization paradigm. A discussion of this pluralism or multiplicity paradigm follows later. 24 Dependencia3 Introduction The decline of the modernization theory and the failure in its application to Latin America in the late 19605 gave rise to dependencia. The school, however, has had many proponents each stressing a slightly different theme. Therefore in this part of the review, the discussion will be confined to the more salient parts of the theory. Latin American development had begun promisingly. After World War II, countries there embarked on a policy of import substitution. The policy was pressed upon governments not by elites but by the business and labor classes for three reasons. First, much of the imports were luxury goods demanded by the elites. Second, local manufacturers were asking for protection from foreign competition. Third, nationals were asking for guaranteed jobs. It was thought then that only a complex industrial society could fully benefit from modern technology to reach high standards of living, enjoy a stable democracy and be truly independent (Kahl 1976:2-3). Background Lending theoretical and analytical weight to the above was perhaps the most sophisticated and influential of 3 In this part, dependencia will be used to describe the dependency school. Blomstérm and Hettne (1984:77) consider the word "theory" unsuitable because "a number of different dependency theorists have received their theoretical impulses from different sources." Further, in the late 25 scenarios painted by Raul Prebisch, who while still under 30, was under-secretary of finance in the Argentine government from 1930 to 1932 and later at the Central Bank. The period coincided with the worldwide depression. Argentina suffered a sharp decline in volume and prices of exports and a shrinkage in foreign exchange. Prebisch said it was in this setting that he was converted from neoclassical economic orthodoxy to protectionism (Meier and Seers 1984:175, 178). But Prebisch's formulation of a theory of industrialization through import substitution in the 19405 was not just a rationalization of protectionism. It was to answer two deeper questions that had arisen from the Argentine crisis in the 19305: "Why was it necessary for the state to play an active role in development? Why was it that policies formulated at the center (the developed world)' could not be followed at the periphery (the undeveloped world)?" (ibid:176). In 1949, Prebisch became the executive secretary of the United Nations Economic Commission of Latin America (ECLA in English, CELAP in Spanish) set up in Santiago, Chile. He could now put his theory into practice. Prebisch's doctrine divided the world into two: Center, the developed and industrialized world, and Periphery, the undeveloped and agricultural world. The structures of 19605 and early 19705, a number of scholars wanted a distinct theory but the idea has since been abandoned. 26 production were said to be different; the Center was homogenous in productivity and diversified in output while the Periphery was heterogeneous in productivity (both high and low productivity existed side by side) and specialized in output, mainly primary products. Classical economics suggests that the international division of labor would benefit both. Prebisch challenged the assumptions behind such a conclusion. His basic thesis was that the problem of development in Latin America principally was reliance on export of primary products in which terms of trade had been deteriorating. That is, prices of agricultural products were falling in relation to prices of industrial products. Not much could be done about world prices so the ECLA suggested two courses of action: import substitution (which it called "inwardly directed development") with better economic planning and development of Latin American common market. These required active government planning as import substitution needed tariffs to protect local industries and the development of a common market required the cooperation of governments (Prebisch 1950, 1963). Prebisch also suggested other reforms in, inter alia, banking, taxation and land. But there was resistance to some of these reforms so not all of them could be implemented (Baer 1969). Nevertheless, the ECLA program was an initial success. In the late19405 and early 19505, the economies of 27 Argentina, Brazil and Mexico boomed. But in the mid 19505, signs of trouble loomed. Import substitution had indeed reduced the import of consumer goods. But it increased the imports of capital goods. Countries became more dependent on multinational corporations to fill in the gaps of technology and capital. The emphasis on industrialization led to the neglect of agriculture and thereby to increased imports of food (Seers 1981). Improvements in health care lowered infant mortality so population growth surged. As the rural areas could not absorb them, many moved to the cities. The cities, however, were industrialized using labor-saving technologies from the West. According to Kahl, the proportion of workers in the industrial sector was half that of the developed countries at a comparable stage of industrialization (1976:6). Economic growth was also being distributed unequally between rich and poor. Such negative effects of industrialization were discussed by ECLA economists under the rubric of "noneconomic concomitants of economic development" (Kahl 1976:7). Or as the president of Brazil, General Garrastazu Médici said, "The economy may be doing well, but the majority of the people are still doing poorly" (ibid). In the late 19505 to early 19605, some economists working for the United Nations began to question the 28 development policies. Gunnar Myrdal said that so-called "backwash" effects from trade might lead to underdevelopment (1957). Dudley Seers argued that the economic development policies implemented were not universally applicable but only worked in advanced industrial capitalist countries (1983). Then in the early 19605, some of the Latin American countries suffered economic downturns serious enough to threaten a breakdown in social order and eventually did lead to military rule to restore order (Kahl 1973:1-17; So 1990:91-94). A dependentista, Fernando Henrique Cardoso wrote Empresario Industrial e Desenvolvimento Econémico (Industrial Entrepreneur and Economic Development) with the thesis that special conditions of Brazilian development were leading the country into a crisis that would lead to a coup. The title page of the book carried the date "April 1964" though the book had been written earlier. On the first day of April 1964, a right-wing coup overthrew the Brazilian government. The multitude of problems converted the ECLA "structuralists" into dependentistas. Palma points out that the dependentistas were helped by the spread of Marxist critiques of imperialism then (Palma 1981). Dependency thus is in part a reaction to modernization theory and in part an attempt to explain the Latin American condition. The classical definition of dependencia is given by Dos Santos who states that dependence is not an effect but an 29 integral and deterministic part of development (1969:73). His formal definition states: Dependence is a conditioning situation in which the economies of one group of countries are conditioned by the development and expansion of others. A relationship of interdependence between two or more economies of between such economies and the world trading system becomes a dependent relationship when some countries can expand through self—impulsion while others, being in a dependent position, can only expand as a reflection of the dominant countries, which may have positive or negative effects on their immediate development (1970:289-290). Three schools In a detailed analysis of dependency, Palma writes that there are three schools of thought: those who hold that developing countries can never develop (which some have called the catastrophic school); those who say that developing countries can develop but only in a dependent relationship; and those who concentrate on obstacles to development using the dependency analysis (Palma 1981). Among dependentistas in the first school, the catastrophic, are the "father" of dependencia, Paul Baran (1957) and André Gunder Frank (1967, 1969). Baran, a Cambridge economist, argued that the British, as colonial master, deliberately underdeveloped its colony India. Thus was introduced the idea that underdevelopment was not just a the result of non-action but a possible outcome of planned action. 30 This theme was reiterated by Frank. One of the best known of dependentistas, Frank is not a Latin American but a German who was trained as an economist in Chicago and taught in Michigan State University from 1957 to 1961 (Contemporary Authors 28:181). He was a supporter of the Cuban revolution and a committed socialist (Harrison 1988:78). In 1964, Frank was a consultant to the ECLA and thereby became one of the first to work with an alternative theory of Latin American economic development. (Mirroring the center-periphery relationship dependency described, scholars from the United States such as Frank got more recognition for the theory than their Latin American counterparts. According to Latin American scholars, Frank "produced the least number of ideas on dependency but monopolised all the adulation and limelight related to it" (Somjee 1991:54). Frank has been described as "one of the most intellectually influential" but also "one of the most polemical and simplistic" (Chirot and Hall 1982:83).) Frank (1967, 1969) critiques modernization theory on several counts. He argues that first, the theory offers only an internal explanation. Colonialism, for example, was a common phenomenon in the Third World. Yet this was not considered in the formulation of modernization theory. Second, he argues that modernization assumes that the {Third World is undeveloped because something is wrong with {it-mlack of achievement motivation, traditional and :resistant culture, overpopulation. 31 Third, he argues that modernization ignores the history of the Third World and thereby assumes that the path of development of the Industrial World is directly comparable. Frank borrows from Prebisch's model of center and periphery and developed a model where the national cities of the Third World are satellites to the metropolises of the First World. The national cities in turn are metropolises to the satellite cities of the Third World. Among Frank's many hypotheses (see So 1990:97-98) is that satellites develop the most when their ties to the metropolises are weakest. Frank noted the autonomous development of Latin American industries during World War II. Frank therefore concluded that modernization was the cause of underdevelopment. Among the theorists in Palma's second school of dependency, which focuses on problems of development, is Theotonio Dos Santos. According to Dos Santos, the structure of dependency takes three forms: colonial, financial- industrial and technological-industrial. Dos Santos laid particular stress on the industrial arguing that industrialization is dependent on technological innovations and equipment from the West. In many cases, Third World countries cannot pay the multinational corporations for the technology and so must offer special incentives to attract them. This means that the multinational corporations have an advantage over domestic companies (Dos Santos 1971). 32 What are the impacts? First, the economy is divided between the traditional agricultural and the modern technological instead of being treated as an integrated whole. Second, those in the modern technological sector enjoy a much higher income than those outside, thus increasing income disparity. Third, because there is a limited internal market, such industries look to exports and this in turn causes a further neglect of the internal market. Dos Santos therefore concludes that it is not the lack of integration into the world economy that is the cause of underdevelopment. Instead, as So (1990:102) summarizes succinctly: "it is the monopolistic control of foreign capital, foreign finance, and foreign technology at national and international levels that prevents underdeveloped countries from reaching an advantageous position, resulting in the reproduction of backwardness, misery, and social marginalization within their borders." The third school, as categorized by Palma, is a compromise of the first and second schools mentioned above and combines elements of them. Criticisms of the two schools mentioned above therefore apply to this third school and will not be repeated here. Recently, some in the first, ie catastrophic, school, appeared to have a change of heart in the light of developments in the Eastern bloc and other command 33 economies. Frank, for example, now appears to concede that development in the periphery may be possible (1984). Essence of dependency Dependencia, with its many shades and variations, has not always been captured successful by writers. Cardoso notes: "The most general and formal of Gunder Frank's works are received as though they were his best, the formal definition of dependency furnished by Theotonio dos Santos is appended, the problematic of subimperialism, and marginality is sometimes inserted, one or another of my works or of Sunkel is footnoted, and the result is a theory of dependency--a strawman easy to destroy" (Blomstdrm and Hettne 1984:70). Cardoso himself sees it as an outlook more than a theory (Kahl 1976:176). Blomstdrm and Hettne (1984:70-76) have listed several dimensions to sketch out the "ideal-typical" dependentista. First, the obstacles to development are seen as external, not individual-psychological (such as the lack of achievement motivation) or national (such as the absence of entrepreneurs). There have been attempts to explain the relationship of internal influences but the greater importance has been placed on the external. Second, dependencia sees underdevelopment as part of the result of polarization in the world system where wealth and resources are transferred from the periphery to the center. 34 Third, dependencia thus treats development and underdevelopment as incompatible concepts. Fourth, dependencia sees the phenomena of surplus transfer as a general and global process though the analysis is primarily concerned with the periphery. National and even regional variations are downplayed. Fifth, as dependencia was rooted in economics, it emphasizes economic analysis. Finally, as the cause of underdevelopment was the West, dependentistas see a cut in the linkage as an automatic spur to development. (So 1990:104-105; Blomstbrm and Hettne 1984:71-76). Policy implications Because dependentistas argue that underdevelopment is caused by linkages to the Western world, they urge a withdrawal from the world economy as the path to growth. Frank recommended Castro's Cuba as the "nondependent" model. Further, as the theory has roots in Marxism and neo- Marxism, they also recommend revolution. The title of Frank's book, Latin America: underdevelopment or Revolution (1969), bears this out. Advantages The advantages of dependencia are that first it considers external factors in the development of a country. 35 This contrasts it with the modernization theory, which did not consider a common history of colonialism in many Third World countries. Second, it accounts for the intranational and international power structures and the close interdependence between them (Lee 1979:32). Third, by conceptualizing development as leading to underdevelopment, it forced a rethinking of the benefits and trade-offs in development. Criticisms Dependencia began to come under attack in the mid 19705. Blomstdrm and Hettne (1984:92-93) point out that empirical research began discovering flaws in the dependencia. Perhaps the best known of empirical studies was that performed by Bill Warren (1973). He found that in many Third World countries the outlook for development was good, that potential obstacles were more likely internal than external, that imperialism facilitated industrialization, and that the Third World's dependence on the First was declining. These findings were a direct frontal assault on the school. Around that time too, some dependentistas began criticizing the theory. (It is instructive to note that the criticisms came from within. Modernization theorists 36 apparently ignored the debate and emerged with a new paradigm discussed below.) One of them was Colin Leys, who has been described as one of the best known dependentistas in Kenya (Blomstérm and Hettne 1984). Leys in 1977 wrote an article that called the theory "no longer serviceable and must now be transcended." Leys said dependencia was, first, repetitious and theoretically stagnant. Second, it was silent on certain fundamental problems it could not neither solve nor even formulate. Third, it was not having a practical impact on the Third World but was instead being co-opted by developmentalists in the First World (1977). Other criticisms are categorized below. Methodology If modernization theory is seen as tainted by Western ideology, dependencia is seen as tainted by Marxist ideology. Indeed, while Palma notes that there is no single Marxist theory of underdevelopment, the various strands of dependencia may be grouped under the heading of Marxist- inspired (1981:63). Dependencia is partly clouded by rhetoric and emotion: some dependentistas have paid a personal price for their views, having been ostracized or thrown in jail by governments. Critics also observed that dependentistas adopt a deductive approach where cases are made to conform to 37 theory. Such instances as Australia, New Zealand, Switzerland and Canada, which do not fit neatly into the model, are treated as anomalies, deviant and accidental (Lee 1979:33). Further, in attempting to formulate a generally applicable model, they have ignored some of the historically-specific characteristics of individual countries. In their reaction to modernization, which focuses on the internal dynamics of states, dependentistas have ignored internal factors such as the absence of entrepreneurs, cultural patterns such as the exclusion of women (Barnett 1988:43) and corruption (Myrdal 1970:230- 2231). Finally, Seers has noted that while dependentistas write of the dangers of cultural dependence, ironically they draw on Marxism, a European doctrine (1981:15). Usefulness of dependency as a scientific concept Sanjay Lall (1975), who is generally sympathetic to dependency, said the theory was not useful for two reasons. First, it failed to identify characteristics of dependent economies not found in non-dependent. Second, it failed to show how the characteristics adversely affect development. Marxist theorists have criticized dependency for not fitting in class analyses and the state (So 1990:132). 38 overemphasis on economic determinism Others have noted that dependencia overexaggerates the power of the center such that the periphery countries are deemed to be "passive victim[s] of capitalism from without" (Trimberger 1979:128). It has also been argued that however strong the dominanting force, the evidence is that nevertheless there are opportunities for the periphery to deVelop (Portes 1976:79). Further, the overemphasis on economics neglects the social, political and cultural dimensions of development. There have been some attempts, both direct and indirect, to incorporate these dimensions. For example, Varis and Nordenstreng (1974) showed that many countries were dependent on programming from the West, principally the United States. Along a similar line, the MacBride Commission Report (1980) called for a more balanced flow of news from and'to the West. These efforts, however, are piecemeal in the overall scheme of dependencia. Myrdal suggests that the lack of development in Latin America during a period that saw development in Asia, may be traced to the colonial past. Myrdal argues that different colonizers have left different legacies. Britain, when it was pulling out of its colonies, was moving toward a liberal system of rule with a firmer and less corrupt government. In contrast, when Spain and Portugal where pulling out from Latin America, the two former powers were shadows of their greatness (1970:433-434). 39 Questionable policies The policy recommendations are the most controversial aspect of dependencia. Recent history strongly suggests that withdrawal from the world economy may not just be unrealistic but harmful. The fastest growing economies in East Asia and parts of Latin America have been the ones most exposed to international linkages. On the other hand, insulated economies, such as those in the Eastern bloc, have stagnated. The socialist model as applied by the former Soviet bloc, especially, has also been shown to be a failure economically--the very area where dependencia is supposed to be at its strongest. Berger (1976:88) had pointed out, before the command economies turned more capitalistic, that the policy recommendation of a socialist economy would merely exchange one form of dependency for another. Finally, the recommendation of a revolution is not predicated on any empirical finding. The perspective does not answer the question: what happens after the revolution? The need for education, technology, and other aspects of development remains unchanged. Conclusion Despite the various shortcomings listed above, dependencia has played an important role in development. 40 First, it has had a lasting impact on theory. It was the Third World's first contribution to development theory, a significance that should not be underestimated. Jagdish Bhagwati (Meier and Seers 1984:197), commenting on the contribution of Raul Prebisch, said: . . . [AJmong the colonial attitudes which afflicted our societies in those days was the belief that fundamental thinking required that one belong to the center, not the periphery. . . . Prebisch . . ., among a few key figures, helped to shatter that myth decisively. Second, it has had an impact on development strategies, particularly in Latin America and parts of the Caribbean and Africa. Third, it has played important ideological and sociological roles in Latin America during the 19605 (Blomstrdm and Hettne (1984)), the effect of which still continues. ' Dependencia provides both an easy and a difficult answer, both a spirit of optimism and a spirit of pessimism. The easy answer is that it offers a utopia--if only the linkages to the developed world are cut. But this is also a difficult answer because it is an unrealistic, and perhaps unrealizable, option; the result then is deep pessimism. It has been suggested that radical writers are good at diagnosis but poor at therapy (Lee 1979:63). Perhaps the last pronouncement is particularly 41 instructive. It is made by a scholar of Asian descent and it is in Asia that dependency has been resoundingly rejected, even though one of the first dependency studies was of India (Baran 1957). As noted earlier, Apter has observed that while confidence in the modernization paradigm is declined everywhere, it is still strong in East Asian countries such as China, Japan and Taiwan (Apter 1987:17). The confidence does not appear misplaced; countries such as Hong Kong, Singapore and South Korea have flourished economically using the modernization paradigm. The dependency analysis could not have worked in city-states such as Singapore and Hong Kong because self-sufficiency is unrealizable. This issue will be discussed in further detail below. The newly-industrializing countries in the Far East, who were initially considered good examples of dependent economies, have now become competitors to the West and began to invest there (Somjee 1991:56). In this writer's view, it is accurate to say that dependencia is on the decline. To most Western development economists, the conclusions of dependencia are "overdrawn . . and [may] be questioned on both theoretical and empirical grounds" (Chirot and Hall 1982:92). The terms of trade apparently do not deteriorate continuously but fluctuate; Prebisch appears now to have captured a slice of reality (ibid 93). The evidence is increasingly mounting against it so much so that some writers now say that 42 dependencia is not a "precisely articulated theoretical model" but "a conceptual framework, a set of concepts, hypothesized linkages, and an optic that attempts to locate and clarify a wide range of problems" (Emphasis added. It is curious how one can clarify if one is not precise.) (Villamil 1979:2; Fagen 1977). Among dependentistas, there have been two courses of action. So (1990) suggests that some dependentistas have attempted to refine a more sophisticated model. But the example he uses is of Cardoso, who had not associated himself closely with many of the dependentistas but had called some of them "vulgar Marxists" (Kahl 1973). Then there has been criticism from theorists, principally Immanuel Wallerstein (1979), who believe that development should be analyzed from a "world system" perspective rather than the more parochial national angle. This world-system approach, as it is known, is often seen as heir to dependencia (Blomstrdm and Hettne 1984). 43 IneeWorld-Svstem Perspective Intnoduction First, the newly-industrializing countries from the Far East (South Korea, Taiwan, Hong Kong and Singapore) were growing economically and challenging even some developed countries world in some sectors. It was difficult to cast their success merely in terms of "dynamic dependence," "manufacturing imperialism" or "dependent development" (So 1990: 170). Second, the command economies in the Eastern bloc were showing signs of crisis and there were questions raised as to the wisdom of severing links to the West (ibid.). Third, the US was showing signs of weakening as a world superpower with a budget and trade deficit in part created by its involvement in Vietnam (ibid.). Building on dependencia, which may account for why some have considered it as an extension of the school, Immanuel Wallerstein formulated an explanation that considers these "anomalies" and labeled it the world-system theory. Beekgnound Wallerstein began his career by studying development problem in post-independent Africa (Wallerstein 1964, 1967). With the failures of development in the late 19605, his formulation of a development theory was strongly influenced. by the dependentistas. Indeed, he has borrowed many key concepts from them--unequal exchange, core-periphery 44 exploitation and the world-market (1979). Probably because of this heritage, many scholars consider the world—systems perspective to be part of the dependency school (Lee 1979; Chirot and Hall 1982; Blomstrdm and Hettne 1984). More recently, however, Wallerstein has moved beyond the domain of the neo-Marxist dependency school. It has been suggested that the shift in orientation may be attributable to the influence of the French Annales school and its leader Fernand Braudel (So 1990: 172). Braudel argued that history should be viewed as an all- embracing or "global discipline." That is, history should not be subordinated to any discipline but should be viewed in the "totality of the social force" (ibid). Braudel also argued that history and the social sciences should be synthesized through a long-term perspective. Thereby, history would "move away from the 'uniqueness of events' (eventism), and the social sciences would gain a historical perspective lacking in much of its attempts to formulate trans-historical theory" (Kaye 1979:409). Braudel played a key role in shifting the focus of history from a recording of periods to an orientation toward problems. Thus he asked "big" questions such as "What is capitalism?" and "What accounts for the failure of France ever to have dominated the European world?" (So 1990:173). Wallerstein, after his success in expounding the world- system perspective, became director of Fernand Braudel 45 Center for the Study of Economies, Historical Systems and Civilization at the State University of New York at Binghamton. Besides producing working papers and holding an annual conference, this international enterprise publishes a journal called Review that urges "the primacy of analysis of economies over long historical time and large space, the holism of the socio-historical process, and the transitory (heuristic) nature of theories" (ibid 170). From its roots in sociology, the theory has been extended to anthropology, history, political science, and urban planning (ibid 171). The Perspective Wallerstein sees the world-systems perspective not so much as a theory but as "a protest against the ways in which social scientific inquiry was structured for all of us at its inception in the middle of the nineteenth century" (1987:309). Specifically, he challenges several assumptions embedded in traditional social science research (So 1990:173-180). Disciplines Wallerstein rejects the boundaries set up by disciplines in the social sciences, arguing that the differences within disciplines were often greater than the differences between them. He says that the cleavages in the social sciences emerged only in the 19th centry and that the chundaries are an obstacle to scientific inquiry (1979; g1 46 1987). History and social science Wallerstein also calls for a synthesis of history, traditionally viewed as a generalization of the particular, with the social sciences, traditionally viewed as a generalization of the universal. He fuses them into what he calls in his own words, inelegantly, historical social science (1979). He writes: One cannot talk about (analyze) any particular set of occurrences without using concepts that imply theorems or generalizations about recurrent phenomena. Thus all "history" is based on "social science." However, conversely, not only is all "social science" a set of inductions from "history," but there are no generalizations which are ahistorical, that is, universal (emphasis in original. 1979:ix). Unit of analysis Wallerstein argues that the unit of analysis should not be the society or the state but the historical system. It is a term consistent with the integration of history and social science. The term divides the link between society and states, which were considered opposing concepts in the 19th century (So 1990:176). Society then was the sum of the manners and customs of a group of people and considered deeper than state. In modern terminology, this would be culture. (ibid 176-7). Accordingly, the unit of analysis is the capitalist world economy. All phenomena is to be explained in that light. Thus internal political or class struggles in a state 47 are seen as "efforts to alter or preserve a position within the world economy which is to the advantage or disadvantage of particular groups located within a particular state" (Wallerstein 1975:16). Progress through development Wallerstein says the traditional social sciences, including developmentalists, see human history as "progressive and inevitably so" but he argues that this is not necessarily the case (1987:322). He writes: It is not at all certain that there has been a linear trend---upward, downward or straightforward. Perhaps the trend line is uneven, or perhaps indeterminate (1987:322-3). That scholarship and politics should not mix Wallerstein argues that the gap between social science and politics should be narrowed. He writes (1979:x): Every choice of conceptual framework is a political option. Every assertion of "truth," even if one qualifies it as transitory truth, or heuristic theory, is an assertion of value. All good scholarship is polemic (but not all polemic is good scholarship). Assumptions In his work, Wallerstein made two key assumptions. First, he assumes, without stating it baldly, that trade is necessarily bad for a country. Most Western economists adopt the view of David Ricardo that there is mutual benefit to be derived from the production of goods in which countries have comparative 48 advantages and then trading them (Chirot and Hall 1982:87). However, there is another school of thought argued by Friedrich List in 1841 that it would be better to promote infant industries that could not compete in the short-term because of the long-term advantages in technological know- how (Chirot and Hall 1982). The debate still continues but Wallerstein adopts the stance that Ricardo has lost. The second assumption, stated explicitly this time, is that development is a zero-sum game. He says: If in the next 30 years, China or India or Brazil were in a true sense to "catch up," a significant segment of the world's population elsewhere in this world-system would have to decline as a locus of capital accumulation (1988:2022). The experience of the Asian Tigers--South Korea, Taiwan, Hong Kong and Singapore, however, militates against this view. These countries have grown economically precisely because of the economic prosperity in China. Some' implications of this assumption are discussed further below. The tneory The use of the historical system as the unit of analysis is the key to Wallerstein's conception. He argues that human history has seen three forms: mini-systems, world empires, and world economies. According to Wallerstein, all societies began as mini- systems, which are entities that have a self-contained division of labor, and a single cultural framework. Such 49 systems no longer exist today but were found in simple agricultural and hunting-gathering societies. They are called "mini" because they are small in space and time, with a life-span of about six generations. Their basic logic is "reciprocity" of exchanges. Wallerstein says there are not as many minisystems in the past as thought because once a minisystem paid a tribute as "protection costs" to an empire, it could no longer claim to be self-contained (1979:5; 1987:317). The mini-system then evolved into the world-system, which is "a unit with a single division of labor and multiple cultural systems". There are two varieties of this: one without (called world-economies) and one with a common political system (called world-empires) (1979:5). Without a common political system, world-economies are unstable and tend toward either disintegration or conquest by a group and thereby transformation into a world-empire. China, Egypt and the Roman Empire are examples of world- economies that, with common political systems, became world- empires. The basic logic of world-empires is "the extraction of tribute from otherwise locally self-administered direct producers (mostly rural) that is passed upward to the centre and redistributed to a thin but crucial network of officials" (1987:317). The European colonial powers of the 19th century are not world empires but "nation-states with colonial appendages operating within the framework of a world-economy" (1979 : 5-6) . 50 World-empires were the dominant form of historical- system between 8000 B.C. and 1500 A.D. These were vast political structures encompassing a wide range of cultural patterns. The basic logic was the extraction of tribute from otherwise locally self-administered direct producers that was passed up to the center and redistributed to a network of officials. From 1500 A.D., the capitalist world-economies emerged. These are vast uneven chains of integrated production structures with multiple political structures. Political energy is used to distribute accumulated surplus in favor of those able to achieve monopolies in the market networks. The operation of the market therefore created incentives to increase productivity and attain economic development (19740). The capitalist world-economies then expanded to cover the entire globe, absorbing in the process all existing mini-systems and world-empires. By the late 19th century, only the capitalist world-system existed. Wallerstein defines capitalism as "production for sale in a market in which the object is to realize the maximum profit" (1974a:399). This requires the cooperation of the entire world; where there is free competition, it is because the core allows it only when it is to its advantage (1974b:1). Under this conception, uneven development is a basic component of capitalist development. 51 Semiperipnery To the core and periphery as understood by the dependency school, Wallerstein added a third position: the semipheriphery. According to him, this is not merely a position on the continuum but one that arises both inductively and deductively (1974a). The semiperiphery serves two functions: political and politico-economic. First, a bimodal perspective raises the issue of why the exploited majority do not overwhelm the exploiting minority. He suggests that the semiperiphery therefore plays a role in ensuring political stability by deflecting the anger of an otherwise polarized world. Second, as the developed world loses its competitive edge through the rising wages of workers in the developed world and technological progress, they shift some of their production to countries that are in the semiperiphery (1974b: 69-70). “ Core, periphery and semiperiphery refer to positions in the world economic system (1972:96). There can and has been movement in the position of countries into and out of the categories. Thus modern Japan has risen to the core while Spain has fallen to the semiperiphery. Mevenent fpon Eeripnery te Semiperipheny Wallerstein suggests three strategies of development for a country in the periphery. First, in a period of economic contraction, countries 52 should "seize the chance" and adopt import-substitution industrialization (1974bz9-10). Wallerstein says this strategy accounted for the expansion of Russia and Italy in the late 19th century. Second, the country could adopt what he calls "semiperipheral development by invitation" where multinational companies are invited to invest in the country (1974b:12). Unlike the first strategy, this is done during a period of expansion. ’ Third, a country could adopt a path of self-reliance. Wallerstein offers Tanzania as a model in this respect but cautions that the country may have been buffered from external pressures because of its poverty. Wallerstein also cautions that small countries will have greater difficulty being self-reliant because of their size (1974b:14). Movement {gem Semiperipherv to Core “ The key to entry into the core is to increase the size of its market enough to "justify an advanced technology, for which it must produce at a lower cost than existing producers" (1974b:15). Wallerstein lists several ways: conquest, increasing the cost of imported goods, dencreasing the cost of exported goods, increasing wages, and finally manipulating the tastes of consumers. According to Wallerstein, the United States, Canada, New Zealand and Australia had high wages before industrialization; their distance from the then more 53 developed world offered some protection for industries (1974b:16). However, he observes that if Russia, which is a large country, just barely made it, the prospects for other countries must be worse (ibid). Advantages The world-system perspective explains the rise of Western societies and the poverty of non-Western societies on a long-term and large—scale global basis. Instead of looking at even a decade or two, the perspective looks at centuries. Wallerstein has studied the period from 1450 to 1750 to develop his thesis. Two other researchers using the perspective studying colonialism have covered the period 1415 to the present (Bergesen and Schoenberg 1986). Thus the perspective goes beyond a slice of reality to analyze historical dynamics. ' Such a perspective directs attention to new research agenda on a global scale. The questions the perspective raises are not raised by modernization nor dependency theorists. The perspective has also been said to lead to a "more parsimonious interpretation of statemaking, international conflict and social change" (Lee 1979:32). Policy impiications Wallerstein and many other prominent world-system theorists share a common optimism in the belief in socialism 54 as an answer to development problems. So Wallerstein rejects Great Britain and the United States as models for development. But so is the Soviet Union similarly rejected because it was operating in a capitalist world (1979:133). The world-system perspective's goal is for a truly egalitarian democracy worldwide. He therefore argues for a movement world-wide (not a coordinated world-wide movement) instead of one at the national level (1988:2021). Key is Wallerstein's call for an attack on surplus at the point of production. He says: Suppose that anti-systemic movements concentrated their energies everywhere--in the OECD countries, in the Third World countries, and yes, in the socialist countries as well on efforts defined as retaining most of the surplus created. One obvious way would be to seek to increase the price of labor or the price of sale by the direct producers (1988:2022). Wallerstein argues that such increases would be more effective today than before because the world's resources are reaching the limits of exhaustion. In the call for a more equal distribution of resources, the world perspective has provided theoretical and ideological support for the new international economic order (NIEO) (Chirot and Hall 1982:81). Criticisms In general, it has been difficult to mount a frontal attack on the world-system perspective because of the encyclopedic knowledge needed. Perhaps the strongest attempt 55 in this regard has been made by Robert Brenner who demonstrates that Wallerstein has reversed the causality of dependence and backwardness. Brenner (1976) shows that the economic backwardness of eastern Europe (primarily Poland) in the late medieval and early modern period did not arise from dependence but instead was the cause of dependence.4 Along a similar line, Perry Anderson (1974) concluded that Poland in the late middle ages was a vast underpopulated area with predominantly poor 50115, a backward agricultural technology, and a fragile, decaying urban network before the grain trade with the West began (emphasis in original. Chirot and Hall 1982:98). The emphasis on the world system has tended to be carried so far as to neglect internal factors such as class and political struggles, as Marxist scholars have noted. This neglect of historically specific development weakens the'world system perspective as an attempt to be a comprehensive theory. For example, Amin (1974:3) writes that "[n]ot a single concrete socio-economic formation of our time can be understood except as part of this world system." 4 First, the availability of forced labor discouraged the landlords, the ruling class, from introducing agricultural improvements. Second, the landlords by continuing to increase the extraction of surplus from the peasants limited the home market for industrial goods. Third, direct and powerful controls over the movement of peasant labor eventually suffocated industries. Finally, the landlords usurped the functions of merchants by acting as middlemen and encouraged industrial imports from the West. All these resulted in dismal productivity and unequal distribution of income, which together led to dependence on trade in primary products to the West. 56 This makes the countries out to be pawns of the world system, leading in turn to the error of teleological explanation. That is, explaining the behavior of states by the presumed needs of a larger system (Koo 1984:37). Critics have noted that Wallerstein commits historical teleology by using historical events to explain the origins of the world economy and then insisting that these historical events had to happen because the world economy required it (Skocpol 1977; Zeitlin 1984). Thus Zeitlin (1984:228) quotes Wallerstein: The world economy was based precisely on the assumption [whose?] that there were in fact these three zones and that they did in fact have different modes of labor control. Were this not so, it would not have been possible to assure the kind of flow of the surplus which enabled the capitalist system to come into existence. (emphasis added by Zeitlin). Zeitlin (ibid) says the statement argues that the world economy originated because of its consequences. “ Marxists have also noted that Wallerstein neglects to account for the role of technology. Chirot and Hall (1982:98) noted that this was a curious omission considering that this was one of the strongest Marxian contributions to economic development. Brenner (1977) argues that the neglect was necessary to prove that Poland was not much behind the most advanced parts of Western Europe in the 15th century. Also, the lumping of precapitalist societies into minisystems and world-empires has been described as being "perilously close to the ahistorical eurocentrism that 57 characterized modernization theory" (Chirot and Hall 1982:99). China as a world-empire underwent technological change and economic growth from the Han to the Ming dynasty. But another world-empire, Egypt, remained largely stagnant from the time of the Pyramids to the Macedonian conquest 2,300 years later (ibid). By lumping all world-empires under one head, the perspective does not explain the differing reception and results of the entry of capitalism. Wallerstein's assumption that capitalist economic development is a zero-sum game has also been questioned. Kuznets (1971) and Bairoch (1977) have shown that this is not the case and that the growth rate of poorer countries after World War II was higher than the historic growth rate of core countries. The ideological bias of the world-system toward socialism has been questioned as it has led to blind spots. Besides ignoring the question of core and periphery among socialist states, world-system theorists do not answer the question of how socialism can increase labor productivity in the long-run (Chirot and Hall 1982). Finally, the policy recommendation of Wallerstein (1988) is weak. Wallerstein does not elaborate how "an attack on surplus at the point of production" is to be carried out beyond the concentration of energies everywhere. It appears to be a call to raise prices through some cartel arrangement among producing Third World countries. Wallerstein's call in 1988 postdates several attempts by 58 Third World countries to do precisely that. There have been cartels attempted in such commodities, besides oil, such as tin and rubber. All of them, however, have met either limited success, as in the case of oil, or dramatic failures, as in the case of rubber. The world-system perspective has made important contributions to the field of development theory as outlined above. Perhaps because it is an offshoot of dependencia, theorists who ignored dependencia have also ignored the perspective, choosing instead to develop a perspective called variously pluralism, multiplicity and Another Development. This perspective is discussed next. 59 The PluralitylMultiplicity Paradign (a.k.a. Another Development) Introduction The failure of the modernization paradigm to predict the path and consequence of development did not go unnoticed by its theorists. As early as 1967, empirical research had rejected Lerner's thesis of modernization through the mass media (Schramm and Ruggels 1967). Meanwhile, Third World intellectuals were looking to base their development on self-reliance rather than foreign assistance, coining slogans such as "Development without modernization." They endorsed the goals of development and the equitable distribution of the benefits, such as better health, better housing, and better education, but they rejected the adoption of Western-derived institutions. Development and modernization were to be deliberately chosen values of each country (Berger et al 1973:173,174). Against this backdrop, some development theorists did adopt the dependency and world-system perspectives. Others attempted to change the paradigm. One of the instrumental figures in the change was Everett Rogers who in 1976 called for a new paradigm to replace the old. (This period also marked the rise of the New World Information Order talks, the consideration of UNESCO's "Draft Declaration on Fundamental Principles Governing the Use of the Mass Media in Strengthening Peace and International Understanding and in Combating War 60 Propaganda, Racialism and Apartheid" and the formation of the MacBride Commission (McPhail 1981:98-100).) The paradign Rogers (1976b) said that a distillation of the literature yielded four main elements in the modernization theory, which he called the dominant paradigm. These were: a focus on economic growth both as a measure of and the goal of development; reliance on capital-intensive technology; centralized planning for development; and an almost exclusive stress on internal causes of development. Field experience and critiCisms, however, indicated that a new paradigm should include elements such as: integration of the traditional with the modern systems of communication; self-reliance at both the local and national levels in development; popular participation; and equality in the distribution of information and social benefits (Rogers 1978). Rogers' paradigm is based in part on these new elements as well as based on a reaction to the elements in the dominant paradigm. Instead of economic growth, Rogers suggested a focus on the quality of life and equality of distribution. There was, in part, a recognition that the distribution of communication tended to reinforce existent economic and social inequalities and provide unequal access. Capital-intensive technology, almost invariably imported, was to be avoided in favor of labor-intensive methods and "traditional" and "modern" systems should be 61 integrated. This incorporated the recognition that technologies such as labor-saving devices that were suitable for the West could inflict more harm than good on developing countries and that "traditional" systems and technology were therefore not always bad. Instead of a top-down approach, he urged self-reliance in development and popular participation in planning and execution at the village level. Here, Rogers, along with many in the field, was recognizing that some development projects had failed as the locals were not consulted and therefore valuable local knowledge missed. Finally, instead of focusing only on internal causes, he pointed out the need to look at external causes (Rogers 1976a). This final point takes into consideration the criticism of dependentistas that the modernization paradigm tends to reject the possibility that deep structural factors might impede development. “ The two paradigms are contrasted in the following Table 2.1 below. Rogers' paradigm is a reflection of the change in emphasis in economic development. At the start of the 19705, mainstream economists began putting greater emphasis on equity instead of merely growth (Lewis 1987:6). The shift is not new in that as early as 1961, the Indian Planning Commission was attempting to fight poverty, as opposed to merely encouraging growth, by implementing a minimum-needs program (ibid). 62 Table 2.1.--Alternatives to the Dominant Paradigm Main Elements in the Alternatives to the Dominant Paradigm Dominant Paradigm 1. Economic growth 1. Equality of distribution 2. Quality of life 2. Capital-intensive 1. Greater emphasis on technology labor-intensive strategies 2. Integration of "traditional" and "modern" systems 3. Centralized 1. Self-reliance in development planning 2. Popular participation in decentralized planning and execution (eg at village level) 4. Internal causes of 1. Internal and external undevelopment causes of undevelopment Source: Rogers 1976b:50-51. Several influences appear to have been at work behind the pluralism paradigm. Rogers appears to have been influenced by a Pakistani economist, Chaudry Inayatullah. In 1964, at a conference where Lerner and Schramm were also present, Inayatullah had called for a non-Western model of development. In a criticism then that seems, on hindsight, prescient, Inayatullah said the modernization paradigm presupposes that societies that have not developed technologically by Western standards were "sterile, unproductive, uncreative and hence worth liquidating" (1967:100). He noted that the limited measures of industrialization and urbanization ignored non-material areas of creativity. Development, he said, should be voluntary; the process should be through innovation rather 63 than imitation. He suggested that development be defined as "a process through which a society achieves increased control over environment, increases control over its own political destiny, and enables its component individuals to gain increased control over themselves" (1967:101). Inayatullah's definition bears some resemblance to Rogers' definition of development of 1976 which is "a widely participatory process of social change in a society, intended to bring about both social and material advancement (including greater equality, freedom, and other valued qualities) for the majority of the people through their gaining greater control over their environment." This is in contrast to Rogers' earlier definition (1971) of development as "modernization at the social system level" which occurs when "new ideas are introduced into a social system in order to produce higher per capita incomes and'levels of living through more modern productibn methods and improved social organization" (11). There also appear to be traces of the "Buddhist economics" written about by E. F. Schumacher in 1973 in his book "Small is beautiful." In one chapter, Schumacher uses Buddhism, although he said it could just as well have been another religion, to point out the inadequacies of economics. Schumacher said that labor from a Buddhist or religious standpoint, was intended to be a liberating experience in that it was supposed to enliven man and release his creative energies. Full-employment would 64 therefore be a goal of Buddhist economics. In the pluralism' paradigm, this idea is reflected in the preference for labor-intensive over capital-intensive industrialization. Schumacher said meeting needs not through imports but through local production was less likely to lead to exploitation of limited physical resources in faraway places. In the paradigm, this is reflected in the emphasis on local production. Schumacher also said modernization as practiced needs to heed religious and spiritual values (1973:62). This point has been resurrected recently through scholars using Buddhist development projects as models of how the new paradigm works (Dissanayake 1984, 1991). More recently, five elements of the paradigm have been suggested by Jayaweera (1987a:xvii): 1) modernization as recommended by the West is neither ' practical nor desirable; “ 2) Third World societies should aim to satisfy "basic needs"; 3) fundamental reforms in the structures of international trade and monetary institutions are a necessary condition of development; 4) similarly, fundamental structural reforms within Third World societies themselves, such as land reform, opportunities for political participation, decentralization, etc., are a prerequisite for development; 65 5) reliance on foreign aid and capital-intensive technology must give way to self-reliance and appropriate technology; the bias for industry must give way to a greater commitment to agriculture; and 6) development is unthinkable except within a framework of culture. Jayaweera's approach attempts a middle path between dependencia and the traditional modernization paradigm. It considers both internal and external factors with an emphasis on control by the individual over self and environment. Most recently, these ideas have been expounded more fully by Jan Servaes, who calls the paradigm "one world, multiple cultures" or "multiplicity". Servaes summarizes the paradigm thus: ' "1. All nations are, in one way or the other) dependent upon one another. Consequently, internal as well as external factors may influence the development process. 2. development has to be studied in a global context, in which both the Center and the Periphery, as well as its interrelated sub-divisions, has [sic] to be taken into consideration. 3. More attention should be paid to the content of development, which implies a more normative approach. 4. There is no universal model for development. Each society must develop its own development strategy" 0“? 'U 5” o ' F 2'15. AI‘:F cloudi- 92" has i 1 NM; 9:. u rm F “Ash ‘ n‘p ‘- Ho. uUb‘ Ass: tar: h.) 11.: (1990:39). 66 He says the central idea behind the new paradigm is that there is no universal development model. Development is "an integral multidimensional and dialectic process that can differ from society to society" (1990:38). Development is a relative problem and should be defined as need-oriented, endogenous, self-reliant, ecologically sound and based on participatory democracy and structural transformations. Servaes has drawn up a list of 27 items for the ideal- typical model of multiplicity, contrasting the new paradigm, which he calls organic, with the old, which he calls mechanistic. The items are listed in Table 2.2 below. Table 2.2.--Contrasts between the old and new paradigms Overall Objectives Mechanistic Model Organic Model Motive to cooperate People need to be helped Charity People are able to help themselves Empowerment Assumption about target group People lack skills and resources to develop selves They are helpless People have skills to develop selves They can be mobilized Attitude to problems Problem-solving Problem-posing Attitude to participation Means to achieve ends Never-ending process Objective of policy makers and researchers, Implementation of project objectives Striving for a common vision and understanding of self- development Table 2.2.--continued 67 Learning relationship Teacher-student: know-all vs know nothing Everyone is teacher- student at the same time; all have something to share Valuation of knowledge Western knowledge is superior Traditional knowledge is as relevant Actors Agent of change Policy maker or researcher People themselves People seen as Targets, objects 1‘ Subjects, actors "Leadership" position Project leader Coordinator, animator, facilitator Selection of "Leaders" Appointed by Preferably higher authority by people chosen "Leadership" Decision making, Cooperation, qualifications management, delegation, authoritative receptive, ' adaptability to new circumstance Relationship with Expert-counterpart; Shared people and colleagues authority centered leadership; shared responsibility Table 2.2.--continued Policy and planning 68 Design criteria Productivity and economic growth Needs and for well-being criteria formulated by people Organizational Hierarchical, Horizontal, structure vertical two-way Type of work Technical-economic Education- organizational Approach to work Executing tasks Listening to people; facilitating Organization of work Formal, static Informal, dynamic Mode of communication Monologue, Dialogue consultation Communication Projects Type of media used Mainly mass media Mixed and integrated media; Interpersonal communication Direction of ideas Top-down, one-way Bottom-up, and information two-way Implementation and Evaluation Planning format Blueprint Open-ended Project approach Process approach Change seen as Improvements Transformation Time perspective Short term Long term Effect of absence of Project activities Development leader slow down process continues Initiative for By funding agency Usually evaluation or higher authority initiated by people themselves 69 Table 2.2.--continued Type of solution Symptom curing Aimed at elimination of root causes Evolutionary change Structural change Source: Servaes 1991:77-78. Advantages The new paradigm is an improvement over modernization theory because it considers some of the criticisms of dependencia, particularly those relating to external factors. This makes development theory and policy richer and more holistic. Instead of mere economic growth, the new paradigm also attempts to look into other factors of development that determine quality of life. Second, the paradigm by considering local context is more culturally sensitive. Lerner has consistently argued that the criticism of ethnocentricity in the modernization paradigm is not valid as his model is based on historical happenstance of development in the West (1967; 1976). Even allowing for his defense, it is clear that not everything that has accounted for development in the West has worked when transplanted to developing countries. Third, the need to listen brings humility to development agents and thereby reduce, if not eliminate, mistakes (Oshima 1976). 70 Policy implications To the extent that the pluralism paradigm is a reaction to the dominant paradigm, its policy implications are diametrically opposite those previously recommended. Instead of an emphasis on big media, pluralism puts as much as, if not more, emphasis on "little media" such as puppetry, local dances and plays. Instead of an emphasis on urban projects, pluralism looks at rural projects. Where the policies in the dominant paradigm are effected through the rich and powerful, the pluralism paradigm aims to effect change through the poor and powerless. Consistent with the paradigm, many of these implications are at the project or individual level. From a macro perspective, the major implication of the pluralism or multiplicity paradigm is that a developing country cannot merely ape the West. In the more extreme form of the paradigm, which stresses a culture-specific approach, there is no model to follow, not even that of a developing country (Eisenstadt 1976). Development agencies and officials therefore have to consider local knowledge instead of merely forcing down their own ideas. Another implication is that the indicators of development must consider the impact on individuals and their quality of life, not just macro-level economic statistics. 71 Criticisms Although it has been more than 15 years since Rogers proposed the "new" paradigm, there has been little criticism of it. Dependentistas and world-system theorists appeared to have ignored it. Perhaps it is because this paradigm is less intellectually stimulating than dependencia in opening up new avenues of research. Inayatullah, who in 1964 proposed a model similar to the new paradigm, has criticized himself in a conference in 1976 and his criticisms are incorporated below. Methodology A major problem with the new paradigm is that it is normative and not empirically grounded. It may be inevitable that development is normative. But the theories and approaches in modernization, dependency and world-system were grounded in historical and social research. Coupled with the fact that the new paradigm is a reaction to failures in modernization, the paradigm is able to give a laundry list of practical policies for development but is a weak stab at a theory of development. Significantly, among the many authors Servaes says has influenced him is Guy Gran, who wrote a book on a participatory approach to development. Gran himself appears to see the participatory approach as "an operational response," rather than a theoretical reply, to development (1983:19). The result is that the paradigm does not quite knit 72 itself neatly into a logical, consistent and coherent whole. Instead, it incorporates many elements, some conflicting with others. Inayatullah, for instance, has criticized himself for not establishing a balance among the development goals of increased control over the environment, over national destiny and over individual lives. A desire to control the environment could lead to its destruction, the subjugation of weaker nations by stronger ones to control resources and the regimentation of individuals (1976:59). The paradigm implies that developing countries have the freedom to choose their paths of development when in reality there are constraints "imposed by the historical evolution of these societies and their contemporary internal and external environments" (Inayatullah 1976:58). Usefulness as a concept ' The paradigm suffers from a dialectical inconsistency: it makes an absolute case for cultural relativism. That is, the only absolute sure thing in development is the paradigm and the norms that flow from it. The logical conclusion for some is that everything else has to be culture-specific, a conclusion that Lerner has tried to avoid (1976:62). This contradiction within the paradigm is evident in studies and development approaches. Nordenstreng and Schiller (1979) compliment those who adopt the plurality paradigm for noting some of the criticisms from dependency but observed that this has not influenced their 73 conceptualization and that many of the notions of the modernization paradigm are still extant. Others have noted that "[t]here is at present a manifest disjuncture between general theory, where the world-system perspective has become dominant, and the myriad lower-level focused 5tudies--national, local, and thematic-- based on the earlier modernization model" (Portes and Walton 1981:13). In a survey of 224 studies that examined the role of media in development, Jo Fair found little difference between studies conducted from 1958 to 1975 and those from 1976 to 1986. They varied little in theoretical framework, were cross-sectional and seldom examined media content. Some even used Lerner's model as late as 1984 (Fair 1989). Indeed, Inayatullah notes that it is possible to argue that the modernization paradigm was equally concerned with increasing control over the environment, national“ determination and the individual (1976:59). Conceptually, it is unclear as to whom the grass-roots refers to. Is it the village chief or the villagers? Development researchers almost invariably recount initial "skepticism and resistance from village chief or villagers, and sometimes both, even to eventually successful projects. According to the paradigm, development should be halted when villagers resist the innovations being introduced. But in practice, researchers have simply tried to change the villagers minds through persuasion. 74 Finally, it is hard to romanticize the role of grass- roots development after Pol Pot when millions of Cambodians were killed. Just as abuses can occur from the top in a top- down approach, so abuses can occur from the bottom in a bottom-up approach. Policy It is clearly difficult to implement the bottom-up approach. Rogers, who had earlier noted a subculture of peasantry (1969), has not explained what happened to that subculture under the new paradigm. Perhaps an ideal case where the paradigm would apply is that mentioned by Marcus Linear (1985) of a Dutch farmer who went on his own to Ghana specifically to teach the Africans new farming techniques. At first, the elders in the village welcomed him, expecting more Western aid. Instead, the farmer, Johan Spee, merely listened to them. Later Spee introduced a few minor changes to improve crop output and reduce harvest time. According to Linear, Spee encountered resistance from the elders and the aid agencies and had to leave. However, word of his work spread and Spee was later invited back to Ghana by another village to demonstrate his farming techniques (1985:220-229). The above scenario, however, still requires a top-down method. In fact, by its very definition, development requires a top-down approach. If the natives knew how to develop, why would not they want to develop themselves? 75 Spee's experience indicated that the natives did not know better farming methods. They had merely carried on the methods used by their forefathers and did not even consider the possibility that there were better ways to do what they had been doing. Many instances of development also appear to have succeeded only with the support, at least tacitly, of top officials (Schramm and Lerner 1976). Indeed, Arnold Pacey has shown that the top—down approach, where innovations are introduced by central officials or a central bureaucracy, has worked in such diverse places as Persian (current Iran) and China and as far back as the AD 9005 (Pacey 1990:1-19). Also, it is not always the case that a bottom-up approach works. This is especially true in modern inventions--one either knows it is there or one does not. Further, for modern interactive media such as telecommunications, where the usefulness of the medium increases with the number of users, the top-down approach has worked. In Singapore, a top-down electronic data interchange (EDI) system was constructed and software was then developed for the trading and shipping community. The system today is widely used and praised. In Hong Kong, a similar system from the ground up by the trading and shipping community failed to take off. I Finally, the policy implications of the paradigm are difficult for development officials to accept: it says that every country has to start from scratch without reference to 76 any other country's development. If so, development studies have nothing to offer to developing countries. If there is no model to follow, how can one know if an innovation will succeed or fail? How does one explain the benefits of, say, telecommunications investment? Conclusion The new paradigm specifically rejects the modernization paradigm. The rejection is fundamental, down to the philosophical level. While modernization was said to have reflected the Protestant ethic, the new paradigm reflects the Buddhist ethic. Buddhist development projects have been used as models of how the new paradigm works (Dissanayake 1984, 1991). The strong pragmatic orientation, however, has not been supported with an equally strong theoretical underpinning. The'result, as Inayatullah has pointed out above,'is that development agencies must grapple with balancing contending forces. For instance, balancing control over the environment and exploitation of it, balancing self-determination and too much individualism. The goal of self-reliance, too, appears similarly difficult to realize. The new paradigm cites problems of imported technology. But some dependence on imported technology is unavoidable. This is true especially of such new technologies in telecommunications and computing. Only the First World has them. 77 Also, the evidence from collapsed closed economies in the Eastern bloc suggests that self-reliance and development are conflicting goals. Those countries most closed in their attempt to be self-reliant were least developed. In sum then, pluralism may be too expansive in its claim of being a paradigm. Certainly, Fair's research shows that the dominant paradigm is still alive. Kuhn has defined a paradigm as the generally accepted view of scientists on fundamental facts, that determines the object of research and how research should be conducted (1962). Clearly, there is much in pluralism to be clarified and much more need to be done before a consensus can be reached. To the extent that pluralism has more to say in practical than in theoretical implications, it is half a theory. It is debatable if half a theory is better than none. ' Kuhn (ibid) argues that turmoil in theory is“an indication of dissatisfaction with the explanations and a movement toward resolution. If Kuhn is right, the emergence of development theory may lie just ahead. 78 Development Theorv and East Aeie The current turmoil in development theory has been called a period "between conventional wisdoms" (Lewis 1987:4). It has been argued that the paradigms in modernization, dependency and world-system are extant, with researchers adapting them to suit modern context (So 1990). East Asia, however, does not appear to have felt this turmoil. Instead, the modernization paradigm, or some variant of it, appears to have been adopted by the newly industrializing economies (NIEs). As noted earlier, Apter has observed that confidence in the modernization paradigm has declined everywhere but in East Asian countries such as China, Japan and Taiwan (Apter 1987:17). Dependencia and the world-system perspective were rejected by Asian intellectuals (Blomstdrm and Hettne 1984:6, Pye 1985:3). This may be attributed largely to the development success of the Asian NIEs, which spoke louder than any possible theorizing. South Korea, Taiwan, Singapore and Hong Kong achieved annual growth rates of 9.3 percent for more than a decade from the 19705. Contrary to the predictions of dependencia, growth was attained through outward-looking export-oriented industrialization. Further, the growth provided rapid labor absorption and did not increase inequality to the level as predicted by dependencia (Lee 1981:12). In essence, these countries subscribed to theories and ideologies that viewed economic growth and institutional 79 modernization as good in themselves. Marxist critics in Latin America, however, deride such an outlook as desarrolismo--developmentalism (Berger et al 1973:161). To be sure, economic prosperity in the NIEs has brought problems. But the surest sign of success is that many countries are adopting the model of opening their economies and inviting more foreign investments. Lucian Pye observed that the modernization paradigm had a closer empirical fit with Asia than with Africa (particularly Senegal and Tanzania) or Latin America or the Caribbean (Guyana, Trinidad and Jamaica) where dependencia was accepted (1985:3). Pye theorized that postcolonial Africa lacked nationalism and earnest commitment to modernization that were characteristic of Asia. (Pye's theorizing on this point raises the question of why the two elements of nationalism and commitment could not be incdrporated into a new theory.) At that same time, Latin America states were not newly emerging but had evolved their own distinctive political and social systems. Asia, however, had elite-guided development; Japan and China wanted to gain strength and security by adopting modernizing technologies (ibid). So East Asia wooed, instead of shooed, multinationals (ibid:4). Recently, China has set aside special economic zones for MNCs; India, which had attempted to develop on the paths of self-reliance, is also opening up to foreign investments. 80 How satisfactory is the new pluralism-multiplicity- Another Development paradigm in explaining East Asian development? To the extent that the pluralism paradigm of self-reliance and bottom-up development adopts dependencia, it militates against the evidence. One could stretch the paradigm by stressing the culture-specific approach. That is, one could argue that because the pluralism paradigm accommodates the particular context of each country, it is therefore able to "explain" each of their development. But, as pointed out earlier, this is not a methodologically satisfactory approach. Among other things, by explaining everything, the paradigm is therefore incapable of being tested and explains nothing. Is there therefore no development theory except those from economics? In the late 19605, Benjamin Higgins wrote, "Sociologists, psychologists, political scientists, anthropologists are flocking to the development field. . . . Yet we seem as far from a general theory as ever. Indeed, the flood of new knowledge seems to make generalizations of a kind useful for policy purposes in all developing countries more and more hopeless. . . . It is my view that no Newton will appear, no breakthrough will occur. Development is not that kind of problem" (Higgins 1968:843). Thus far, Higgins has been right. It may be that the development of a development theory, at least from the sociological approach, is difficult. It has been suggested that no one single body or theory about development is 81 likely to develop because the field engenders "political conflict and sharp intellectual divergences" (Bernstein 1973:2). This author, however, feels that it is possible at least to winnow out "bad" theories by studying instances of development failures and, more importantly, success. A development strategy is unlikely to work if it is inconsistent with demonstrated theory or if it does, the theory has to be modified or abandoned. Even if consistent with demonstrated theory, a development strategy is unlikely to work if it is applied in the wrong context (Geiger and Geiger 1976:23). A case study of Singapore's development is therefore instructive. The tendency is to dismiss Singapore as a special case, a deviance from the norm. A robust development theory, however, should be capable of explaining this "special case." As will be shown next, Singapore's development strategy may be loosely classified as falling under the modernization paradigm. 8 2 Development Theory and Telecommunications Introduction Telecommunications role in development has risen in status. Back in the 19605, Schramm criticized India's policy to allocate two-tenths of one percent of outlays to developing the broadcasting system, but 14 times as much to posts and telegraphs. He said, "This is a type of policy which the developing countries would do well to review. The growth of modern communication must go along closely with other modernization. It can lag slightly behind without greatly delaying the general pattern of advance. On the other hand, if it runs slightly ahead it can, to a degree, hurry the general rate of advance" (1967: 9. Emphasis in original.). This remark was set against the backdrop of the modernization paradigm where the use of the mass media was thought to be key in a country's development. ' Bywthemkeee»197057-however;”thewtelephoneuhadmreplaced Wm-MY'W‘ development -(Parker and Mohammadi 1977). For example, Hudsonmeeid-that-when-ehe aeked natives in_remote nnrthennmCanadauwhatwtheir prioritiesmnesemfrom-amlietflthat“inctudedmbetterwhonsingwand elestrifisation._they replied,-9He.needmsemmunieattons fWWe’EéE‘H-LHAE,” we'IrmsmWwe- need" (Hudson 1984:xiii). Conciously or otherwise, the telephone was being used to resurrect the modernization paradigm. 83 Modern Literature As with early studies on mass media and development, an early study of the telephone showed a correlation between telecommunications and what was called the "wealth of nations" (Jipp 1963). Since then, the evidence from various studies strongly suggests that telecommunications has a direct positive benefit in the economic development of a country. The studies are well documented and summarized in three sources: Telecommunications and Economic Development (Saunders, Warford, and Wellenius 1983), Telecommunications for Development (Pierce and Jequier 1983), and When Telephones Reach the Village (Hudson 1984). The first book, by Saunders et al, is geared toward telecommunications planners; it reviews micro and macro economic studies and examines telecommunications financing and management. The second book, by Pierce and Jequier, is an ITU-OECD (International Telecommunication Union-Organization for Economic Cooperation and Development) project synthesizing all known studies on telecommunications and development. It was brought out in 1983, the year declared by the UN General Assembly as World Communications Year. The objectives of the Assembly included development of telecommunications facilities and equipment, a thorough review of national communications and overall development policies, and the establishment of better national and international coordination machinery (Pelton 1990:17). 84 The third book, by Hudson, was done independently of the other two but in one of those coincidences, was completed at about the same time. It is also a synthesis of studies, with an emphasis on rural development. Until these books that were published in the early 19805, studies on telecommunications and development typically correlated telephone density with economic growth. These studies were analogous to previous studies correlating newspaper circulation with economic growth. For virtually all countries, there is a 70 to 80 percent correlation between telephones per capita and GNP per capita (Hudson 1984:39; Pelton 1990:16). Several of the studies cited in the three books mentioned above go beyond simple correlations to determine the causal link between investment in telecommunications and economic development. The common thread running through these studies is that investment in telecommunications pays off in the form of general economic growth. That is, the benefit is spread throughout the economy, and not limited to the telecommunication authority. This suggests that telecommunication authorities should not focus on direct monetary payback to determine if they should invest in a system. A study in the then Soviet Union in 1977 indicates that each rouble of investment in long-distance telephones led to an eight-fold return in contribution to the GNP (Pierce and .Jéquier 1983). This is close to the estimate of six- to 85 seven-fold investment returns predicted in a study done later by Edwin Parker (1981). The Soviet study found that big enterprises benefited more than small enterprises from the telephone. Further, 90 to 95 per cent of the benefit accrues to the economy as a whole, not just to the telecommunications sector (Pierce and Jéquier 1983). In another study, Jean-Francois Berry suggests that France's underinvestment in telecommunications caused it to lag economically behind Spain. He found that between 1972 and 1977, France's gross national product (GNP) per capita grew by 95 per cent while Spain's grew by 145 per cent. Berry admitted that it was too simplistic to attribute the growth only to investments in telephone, but added that the figures were suggestive of a correlation (Berry 1981). Berry's data when reanalyzed indicated that France's under- investment in telecommunications cost it 2 per cent of its GNP'or around 20 billion francs in 1975 through lbsses caused by poor telecommunications. As an indication of cost- benefit, Berry noted that in 1975 France invested 12 billion francs in its telecommunication system (ibid). He concluded that "telecommunications confers benefits on society which are over and above the private benefits represented or captured by the price system" (ibid 34). A slightly more sophisticated analysis by Andrew Hardy in 1981 suggested that the benefits were inversely proportional to per capita GNP, being greater in developing countries than in developed. That is, the less developed a 86 country, the greater the benefits from telecommunications. Contrary to expectations, Hardy also found that in developing countries the telephone was more important in residences than business, and that penetration in residential telephones preceded economic development (1981). It has been suggested that in developing countries, poor telecommunication links force businesses to use residential telephones for commerce (Hudson 1984:43). Recently, another study analyzing investment in telecommunications and economic growth found that in developing nations, the telecommunications sector precedes national economic growth while the situation was reversed in developed countries (Kim 1992). These findings concerning the causal link between telecommunications and economic growth have not always been replicated. Perhaps this may be due to a difference in the impact of investment in a developed against a less developed country. Or perhaps, it may be because telecommunications entities do use the economy to predict the volume of investments needed in telecommunication infrastructure. In Singapore, a study found that investment in local telecommunications and economic development interacted such that one could not be said to have preceded the other in time (Kuo and Chen 1985). But even where a causal relationship is found, the studies may be criticized. First, these studies overlook the "double counting" inherent in them: investment in 87 telecommunications is counted as part of economic growth. Second, although time-series data are used, the Durbin- Watson statistic, as far as the author is aware, is absent from the analyses. Then there is the critique that ultimately, these studies assume the modernization paradigm, a paradigm that has been discredited in development literature. Further discussion on this issue follows later in this section. International Telecommunications It should be noted that the above studies focused on domestic telecommunications and on investment in the domestic front. This research goes beyond investment to look at the flows itself. It also looks to the international telecommunications arena to test the robustness of the findings of previous studies. ' In the international arena, which is the focus of this study, the best statement of the expected benefits of international telecommunications is summed up in "The Missing Link" report of Sir Donald Maitland: An expanded world teleCommunications network would benefit both developing and industrialised countries. The process of improving and expanding networks in developing countries will create a major market for telecommunications equipment. A more comprehensive world system will increase international traffic to the advantage of the operators. Where information flows so 88 does commerce. More world trade and other contacts will increase understanding. An expanded telecommunications network will make the world a better and safer place (1984, note 4). The Maitland Commission, or more formally, the Independent Commission for World-Wide Telecommunications Development, was established in 1983, the World Communications Year by the ITU Plenipotentiary Conference. It has been described as "the most formidable body yet appointed to promote the worldwide development of telecommunications“ (Pelton 1990:17). The Commission urged governments to accelerate development of telecommunications as it was a prerequisite for social and economic development. The Maitland Commission's recommendations were based on studies of a nation or comparison of several natibns. In fact, virtually all the recommendations could be acted upon by individual countries. Some require action by the ITU, others require cooperation among countries. In the latter case, the cooperation lies in the pooling of equipment purchase. Ween__suggastednthatam WMh internggipnalntelegommunigatioaeseenuhelpmdeveiopment-is tnngngh_gainenén~foreignweuehengeuand.(Hudson and York 1988). The gains in foreign exchange, it is argued, will probably come from those engaged in business activity. 89 Another advantage, it is argued, is that international telecommunications, by virtue of the high profit margins (in the order of 80 to 90 percent) could cross-subsidize residential telephone services (Sharp 1985). Both arguments look attractive: the gains in foreign exchange come from sources that are capable of paying; the benefits to residential services, especially those in the rural areas, will aid in development, as Hardy's research had found. Both arguments, however, are, in the author's opinion flawed. They fail to consider the earlier research on telecommunications--that the benefits are spread throughout the economy. This means that the more efficient the telecommunications services in a country, the greater the benefits. Higher prices, be they for foreign exchange gains or for cross-subsidy, imply inefficiencies. Further, there is the trend worldwide to introduce reduce rates in long-distance telecommunications, which includes “ international traffic. For all the theorizing, there has been surprisingly little done in the area of telecommunications and trade. Saunders et al, without giving precise citations, said the Consultative Committee on Telephone and Telegraph (CCITT) found that "international traffic through time was most closely associated with changes in the volume of international trade" (1983: 81). Moss (1986) showed that there was a close relationship among the internationalization of trade and finance, the 9O centralization of such activities in a few global cities, and the development of the telecommunications infrastructure in these cities. More recently, Kellerman (1990) found that for all countries except the US and the UK, trade explained telecommunications more than incoming tourism. The proposed study is a departure from the earlier studies in that it looks at international, as opposed to domestic, telecommunications, and it looks at services and traffic, as opposed to hardware. Paradign On hindsight, it should have been obvious that the telephone can play a significant part in development. Back in 1967, Schramm listed, among the several conditions of effective communication that support national development, two-way communication. Citing the Katz and Lazarsfield study (1955), Schramm advocated the radio-forum and other devices that offered feedback (Schramm 1967). Later, Lerner and Schramm observed: ". . . most media, well used, can contribute to development. . . . And any medium is likely to be more effective when combined with two-way communication of one kind or another" (1976:343. Emphasis added.). But, as pointed out in the beginning of this chapter, Schramm criticized the Indian government for allocating more funds to education and telecommunications than to the traditional mass media. The concept of the telephone as a contributor to development came later, after the pluralism paradigm was 91 unveiled. From the start, it appeared that the telephone was a technology almost too good to be true. Unlike traditional mass media, the telephone was found to offer economic benefits that is maximized when implemented in the rural areas. This appears to be universally true in countries in the First World, in Africa, and in the Soviet bloc (Pierce and Jéquier 1983). In general, the quality of life is improved as death-threatening emergencies are responded to, families kept in touch, and communication with the outside world maintained. As a technology, it should therefore fit the pluralism/multiplicity paradigm well. In reality, however, it fits unevenly. First, telephones in Third World countries have been run by government or government-sponsored authorities on a top-down basis. Second, despite the evidence that the benefits to the poorer are greater than the benefits to the richer, especially in developing nations, many governments still give priority to the urban areas. This may be attributed to pork-barrel politics, and the fact that the telecommunications system is usually run as a revenue source. Third, there is evidence that some leaders love large expensive projects such as telecommunications because of the potential to hide bribes. In sum, there are extraneous factors that can inhibit even a potentially useful development tool such as the telephone. The problem, however, is not easily resolved. Hudson, 92 for instance, in the introduction to her review of studies, was uncertain as to where the telephone stood in development theory (1984). The dependency and world-system perspectives are not helpful; under them, nations would have to sever links with the industrialized world, from which telecommunication technology originated. Having fewer links undermines the usefulness of the telephone. The pluralism/multiplicity paradigm is, at best, uneven in application. The paradigm is strongest in the emphasis on the rural. The rest of the paradigm, however, offers little guidance as to how telecommunications should be deployed. As an imported technology, it needs top-down management from those with the expertise. It is hard to see how Servaes can achieve the approach under the "organic" model--for technicians and peasants to learn from each other, and for them to see traditional knowledge to be just as relevant as Western knowledge. " Understandably therefore, in considering the role of telecommunications in development, the tendency has been to use the dominant paradigm. Journal articles writing of new technologies and development are particularly prone to using the old paradigm. For example, a recent article about satellites in development communications, said: "telephone, data and TV transmissions, made easier by satellites, are now seen as prerequisite for national development" (Flournoy 1991:41. Emphasis added.). The author, without mentioning who sees 93 satellite transmissions as prerequisite of national development, then pointed to the potential for disseminating health, education, social services. Substitute satellites with airwaves and "telephone, data and TV transmissions" with radio and one is back to the era of Lerner's study. That is, telecommunications has now replaced the mass media in Lerner's model. The replacement is ironic. As noted above, Lerner's modernization paradigm was criticized for being a simplistic linear model and for adopting the discredited magic bullet conception that the mass media have direct uniform and powerful effects. There is, however, no ready paradigm for telecommunications to fit. Perhaps it is this lack of a paradigm that accounts for the resilience of the dominant paradigm in telecommunications use in development theory. ' To some extent, this research will test that“paradigm. If international telecommunications does cause economic development, the dominant paradigm role in development will be bolstered. On the other hand, if economic development causes international telecommunications, the dominant paradigm's role will be questioned. CHAPTER THREE znensinqepore Contenp Introduction Currently, more than half of Singapore Telecom's revenue is derived from the international market. Singapore has the highest call volume per capita in the world (Singapore Telecom 1992). Volume would be even higher if telephone calls to Peninsular Malaysia, the nearest neighboring country, were counted. Instead, for historical reasons, calls to Peninsular Malaysia (as opposed to East Malaysia, which is on the island of Borneo) are billed as "trunk calls." This may be thought of as the equivalent of long-distance calls in the United States. (Trunk calls are not measured and are not switched through the international gateway. Plans are afoot, however, to move the trunk calls to an international gateway (Lim 1993).) 1 Singapore therefore represents an extreme case in international telecommunications. Conceptually at least, a study of Singapore as an extreme case should evidence the effects of international telecommunications more readily. History Singapore's links with the international community have long historical roots. The openness of the Singapore economy to international forces is probably the main reason for the high volume of international traffic. 94 9 5 The Republic of Singapore is one of the smallest countries in the world. It consists of a group of islands occupying a total area of about 250 square miles, 10 percent of which was reclaimed from the sea. This area is about the size of Greater Boston or Greater Los Angeles. Nearly all three million residents live on the main diamond-shaped island, which is 26 miles from east to west at its widest and 16 miles from north to south at its longest. The main island lies at the southern tip of the Malaysian Peninsula and is 85 miles north of the equator. Fran—1:5.fifirliest_recorded times, Singaporewheevbeen‘a’ trading_center.-There-is-evidence that itmwas'oae—efwthem three capitals of the Srivijayan empire_hasedmin13nmarzamin ._1___yfln_wwwwu- . ‘Ehenlthneengury_AD (Lee 1973:2). By 1330, it had become an important communication node on the trade route between the Indian Ocean and the South China Sea, appearing in Chinese, Javanese, Vietnamese and Arabian records (Chew 1991). Singapore's modern history, however, began on January 30, 1819 when its founder Stamford Raffles established it as a British maritime base because of its natural advantages-~a deep harbor, sheltered anchorage and plentiful fresh water. There were at most 150 residents then, most of them Malay natives eking out a living through fishing (Lee 1973). W More“ E6" “be "' a"strateg1t“seaport*‘1n the'regtonehpuring the Napoleonic War, the French had ruled Holland and the British had used the ports of the Dutch colony Indonesia. After the War, Indonesia was returned to 96 the Dutch as part of the peace settlement. The Dutch, however, then tried to exclude the British from the region. Singapore was therefore intended to be a British foothold right next to Dutch Indonesia, as well as a center for trade links with the surrounding countries from which resources could be drawn. The British brought to Singapore some uniquely Western benefits that soon attracted migrants with hopes of making their fortune here (Lee 1973). The formalized British legal system offered Asian merchants, financiers and manufacturers better protection of private property and fewer restrictions on commercial activities than the "usually extortionate and often capricious administration of justice by Chinese magistrates and Malay sultans." The port grew so quickly that in 1824, the British bought the entire island and the 58 offshore islets, a total area then of 225.6 square miles (Geiger and Geiger 1976:4,5). " Singapore soon became the administrative and communication center for trade in South-east Asia. Its position as the main British colony in the area was strengthened in the early 18305 when with two other British posts on the Malay peninsula, Penang and Malacca, it was incorporated as the Straits Settlements. The British Governor of the Straits Settlements ruled from Singapore (Lee 1973). Fprhmuch of its_nistnryTmSingapeseis~economy—wae—high1y dependentflon its entggRQEWBOII.IItsflfanetion"wa5"E6"Ttnk"the 9 7 hinggglands_of1Malaya_and_lndenesianwithithe—worldnahrough thsmmmi .iunCtiMiWa—d-im... WM transmutation -and-emenieation canterT’From the import and export activities grew supporting services such as banking, legal, and accounting services. Singapore also nurtured English as this was the language of commerce. So compared“with‘otherwcountries in .Ehe_regionemsingaporeuhad.a more literate populationwwith W9£§_¥§:ied.skills~ It has been noted that, as with any successful entrepot port, it was "outward oriented, attentive to changes elsewhere in the world that could affect its own interests, and inclined to import innovations from other countries that could help it protect or improve its competitive position" (Geiger and Geiger 1976:8). Singapore's early development as a commercial center was accelerated by the arrival of steam-powered boats in 1845 (which helped stamp out piracy in the region), the Australian gold rush of the early 18505, the opening of the Suez canal in 1869 and the introduction of the telegraph in the 18705 (Lee 1973:3). The regional importance of Singapore was enhanced by the Japanese when they defeated the British and occupied the island for three and a half years during World War II. A local historian noted that the "irrelevance of colonial boundaries meant that Syonan (Singapore) became a convenient focal point for official and private business extending into the Peninsula and beyond" (Thio 1991:96). of the British shattered, 98 After World War II, with the myth of the invincibility and the international trend of freeing the colonies, Singapore was allowed greater self- rule. Unlike most other developing countries, however, Singapore did not then relish the prospect of independence. Economists deemed it too small to be economically viable on its own. At the time of self-government in 1959, there were other compounding problems: unemployment was rising and the outlook for entrepot trade, which had been stagnating, was not promising. 1959, and following the recommendations of development theorists, the Singapore government embarked on industrialization aimed at import substitution. (See Table 3.1.) So, upon taking office in Table 3.1.--Singapore's stagnant entrepot trade 1953-1962 Year Total Trade % +/- Imports % +/- Exports % +/— 1953 5,673 3,201 2,652 1954 5,705 0.6* 3,024 -5.5* 2,681 1.1* 1955 7,220 26.6 3,851 27.3 3,369 25.7 1956 7,358 1.9 3,929 2.0 3,429 1.8 1957 7,570 2.9 4,092 4.1 3,478 1.4 1958 6,880 -9.1 3,740 -8.6 3,140 -9.7 1959 7,348 6.8 3,908 4.5 3,440 9.6 1960 7,555 2.8 4,078 4.4 3,477 1.1 1961 7,272 -3.7 3,963 -2.8 3,309 -4.8 1962 7,453 2.5 4,036 1.8 3,417 3.3 Source: Singapore Yearbook of Statistics, various years. * Percent change from the previous year. The ruling party still had no confidence that Singapore could survive on its own and so worked toward a political union with its neighbor, Malaya. The idea was that with a 9 9 hinterland to buy from and sell to, Singapore would have an enlarged market. Se,_inw£963Tm6ingapeve~nergefl”with Malaya thfi. the. Federation 02— Malaysia ‘.~ ' Bub -there‘"'w"'a‘s disagreemefit“among~the.leadersmof-singapore and”mataya and seeinflflugust”1965)“SingaporowwaSHasked“to'withdrewnfcom 331.151a (Lee 1973:10). Once again the threat of economic doom loomed large. Prime Minister Lee Kuan Yew actually wept at the press conference announcing the dissolution of the merger. Singapore at first continued the policy of import- substitution but soon found this to be not feasible as the Singapore market was too small. Another problem arose in 1966 when the British Labour government announced that it was to close its military base and withdraw from Singapore over the following four years. The military base was estimated then to employ 30,000 civilians directly and 8,000 indirectly (together forming 16 percent of the workforce) and to account for 14 percent of the gross national product ‘(Lee 1973:45-46, 96; Buchanan 1972:86-87). So from 1966, Singapore changed direction, switching to labor-intensive export-oriented industrialization (Tan and Ow 1982:283). The aim of the industrial policy during this period was job creation. The policies adopted were largely based on the recommendations of a 1959 United Nations' Industrial Survey Mission. This was the team that the United Nations had sent at the request of the Singapore government to help it build 100 a manufacturing base. The mission proposed the establishment of a non-political government organization, the Economic Development Board, to spearhead the industrialization drive (Lee 1973:17, 27-29). The industrialization plan aimed at attracting multinational corporations to set up operations on the island and bring their expertise. Coincidentally, this was a period in which banks and manufacturing concerns from developed nations were looking abroad to establish bases of operations. The Singapore government also hoped that the large foreign companies would subcontract work to the small local companies (Lee 1976:7). The plan succeeded. By*197r7~ MNCs~accounted*fer—asupercent_OfuellifinmsTué3—peroentwof-.. enp1oyment7fl75~percent_offltotaluualueiaddedy_endealmeee—4s~~ lperoent—of—menefaeturingflexportsvwexcleding~re~exports (Tan and Ow 1982:283). In short, soon after the implementation of the'industrial policy, MNCs were a dominant forcerin the economy. Other chance factors helped Singapore's growth. World trade boomed in the late 19605. Unexpectedly, entrepot trade also began a period of expansion. Oil exploration in the region increased and Singapore soon became the management headquarters and the center for producing and servicing equipment. The escalation of the Vietnam War provided further opportunities. Up to 1972, South Vietnam was Singapore's largest customer for refined oil. The external factors, coupled with increased public expenditure in social 101 services and infrastructure, enabled the government to create jobs. By 1972, the unemployment rate was half that of the mid-19605 (Geiger and Geiger 1976:163-166, 170). The period of industrialization was a period of fast economic growth as shown in Table 3.2. Table 3.2.--Indigenous GNP Per Capita Year Amount (SS) Change From Previous Year (%) 1960 1,252 1961 1,327 6.0 1962 1,369 3.2 1963 1,495 9.2 1964 1,466 -1.9 1965 1,613 10.0 1966 1,740 7.9 1967 1,819 4.5 1968 1,949 7.2 1969 2,189 12.3 1970 2,405 9.9 1971 2,761 14.8 1972 3,206 16.1 1973 3,849 20.1 1974 4,491 16.7 1975 4,888 8.8 1976 5,496 12.4 1977 6,011 9.4 1978 6,303 4.9 1979 6,892 9.3 1980 8,343 21.1 1981 9,854 18.1 1982 11,085 12.5 1983 12,533 13.1 1984 13,599 8.5 1985 13,044 -4.1 1986 12,824 -1.7 1987 13,733 7.1 1988 15,424 12.3 1989 17,093 10.8 1990 18,865 10.4 1991 20,031 6.2 Sources: Singapore Department of Statistics, Monthly Digest of Statistics, various years. 102 By the mid-19705, the government had solved the problem of unemployment and was discouraging enterprises that wanted to set up simple production processes using large numbers of unskilled or semi-skilled workers (Goh 1977:199). In 1979, discouragement of labor-intensive manufacturing was made part of a new phase of industrial policy, which some have called the "Second Industrial Revolution." Wages were allowed to rise 20 percent a year between 1979 and 1981. Capital-intensive and higher value-added high-technology and information industries were encouraged at the expense of labor-intensive industries, which did indeed either fold or 2 move to neighboring countries. But not as many of the desired industries were attracted as was wanted principally because Singapore did not have enough of the skilled labor necessary for the high-technology industries. Partly because of this, Singapore's economy suffered a decline in 1985--its first in more than two decades. The "high-tech" policy has since been abandoned. The vision that information industries would be the next wave of economic development in Singapore, however, has remained. A key industry within the definition of "information industries" is the telecommunications industry. Singapore's telecommunications authority, Telecommunications Authority of Singapore, is working with the Economic Development Board both to woo investments and to invest abroad (Kuo, et al, 1989). 103 Development Paradign The modern history of Singapore sketched out above does not indicate any clear development paradigm. Before World War II, Singapore was a British colony and treated accordingly. Development was centered on meeting the demands of colonial master Britain. Unlike other colonies, Singapore had little by way of natural resources to be exploited. Its strategic location was used as a staging post for trade in the region and so the British built up Singapore's infrastructure. Singapore may therefore have benefited more from colonialism than resource-rich colonies such as India. Bearing in mind also that Britain wanted Singapore as a seaport for strategic purposes, Singapore cannot therefore be considered the periphery in the sense of an exploited colony. ' After World War II, Britain granted more autonomy to its colonies, weakening the direct colonial links to the point of self-government for the colonies. In its attempt to be self-reliant and in common with development thinking then, Singapore's economic development policy was import- substitution. To the extent that the import-substitution policy is derived from dependencia, Singapore was testing the theory. In a matter of years, however, dependencia was found wanting. Singapore's small market forced it to abandon import-substitution and adopt an export-oriented stance. The export-oriented stance adopted has little to do 104 with the plurality paradigm. Decisions were made and implemented top-down with little popular participation. Initially, the government encouraged labor—intensive industries to be established. Unemployment eased and a tight labor market emerged. So Singapore turned to more capital- intensive industries from the mid—19705, the very period that the plurality paradigm was being elucidated. In the 19805, the Singapore government embarked on a campaign to encourage greater use of technology in all aspects of society. This too militates against the plurality paradigm. Perhaps the closest of any paradigm is Wallerstein's characterization of a country like Singapore as a semi- periphery. Singapore may be considered part of the semi- periphery even in colonial times but development theory then had not evolved the concept. The island even then was a go- between for the colonial master and the outlying colonies. ' The characterization, however, does not sit easily for at least three reasons. First, as shown earlier, Wallerstein assumes trade is bad. In Singapore, trade is a major economic sector and is essential for the economic wellbeing of the country. Second, a small country such as Singapore tests the limits of Wallerstein's world system theory. As noted earlier, Wallerstein himself cautions that small countries will have greater difficulty being self-reliant because of their size (1974b:14). But he does not answer the question of whether self-reliant is therefore limited or cannot even 105 be practiced. Singapore has chosen the latter, holding that self-reliance cannot be practiced in the small economy. Finally, Wallerstein neglects the role of technology in development. In Singapore, however, technology is afforded a very important role. As outlined above, technology is to be the springboard for the next level of development in Singapore. In sum, no theory has sufficiently explained the development of Singapore. In fact, the former deputy prime— minister, who holds a doctoral degree in economics, wrote in the preface to a book that he was not into the theory but the practice of development (Goh 1977). Perhaps Singapore is too small to be self-reliant, a state that all the later paradigms have recommended. Perhaps as Singapore developed, its needs change, outpacing the paradigm themselves. Plurality in its insistence on popular participation and little media, for instance, seems more suited forrrural villages than urban metropolises. Finally, technology, especially those employed in commerce, is embraced in Singapore, but eschewed by the paradigms. Without technology, especially in communication, Singapore's role as a staging post for the region would diminish. A classic example of the use of technology is telecommunications. §ingepene Telecommunications As a trading post, Singapore had been among the first cities in East Asia to have telecommunication links with the 106 Western world. Telephone services were first provided by a private company, Oriental Telephone and Electric Company and regulated by a government agency called the Telecommunications Department. In 1953, the phone company was taken over by a government body, the Singapore Telephone Board (STB). In 1974, the government agency and STB were merged to form the Telecommunication Authority of Singapore (TAS). In 1982, contrary to the worldwide trend of separating postal and telecommunications services, TAS was merged with a government agency, the Postal Department, to be established as a statutory board, which is a quasi- government company set up under law. The entity so established was called Singapore Telecom (Kuo et a1 1989). At the time of writing, the company had plans offer the public its shares and be listed on the Singapore stock market. It is widely expected to be a blue-chip stock. ' Because of Singapore's open economy, international telecommunications has played a dominant role in Singapore telecommunications as far back as 1966. Revenue from international and other services in 1966 amounted to S$8 million (US$5 million), more than the $57 million (US$4.4 million) revenue from the national telephone service (Singapore Telephone Board 1966). Today, international telecommunications continues to contribute more than half the revenue of the operating company Singapore Telecom (Singapore Telecom, various years). CHAPTER 2003 Research Design Introduction This chapter begins by stating the hypotheses to be researched. It then explains the data collected and the limitations inherent in them. The variables used in the study are then defined. Finally, a major part of the chapter is devoted to justifying the use of economic indicators despite their limitations. Hypotheses This dissertation investigates the magnitude and direction of the impact of international telecommunications on economic development. As previous research has identified international telecommunications to have the greatest impact on trade and tourism, economic development will be considered only to the extent that it includes these sectors. Further, as it is generally assumed that the financial services sector is also affected, this was also studied. International teleCommunications is identified in this study as a causal variable in economic development. For the purposes of this study, it is not a necessary or sufficient cause. It is probably best described as a facilitating or contributing cause because development of the trade, tourism and financial services sectors has taken place without international telecommunications. What is meant therefore is 107 108 that international telecommunications facilitates growth above and beyond what is attributable to other causes. The following are the hypotheses tested. H1. International telecommunication correlates strongest with trade, tourism and financial services among all the sectors of the Singapore economy. H2. International telecommunication causes trade development more so than vice versa in Singapore. H3. International telecommunications causes tourism development more so than vice versa in Singapore. H4. International telecommunication causes financial services development more so than vice versa in vSingapore. '- H5. International telecommunication causes more growth in trade development than tourism development or financial services development in Singapore. H6. International telecommunication causes more growth in tourism development than trade development or financial services development in Singapore. H7. International telecommunication causes more growth in 109 financial services development than trade development or tourism development in Singapore. Limitations and Key Assumptions The above approach considers but does not adopt the dependency paradigm, which argues for self-sufficiency. Instead, implicitly it holds that the more linkages to the world, the better for development. This assumption is backed up by strong evidence that a more open outlook to external communication linkages improves the economic well-being of a nation. The major limitation of this study is that the data collected are restricted to international telecommunications whereas previous research has been almost exclusively focused on domestic telecommunications. Strictly speaking therefore, the implications of this research will refer only to the international arena. ' Research on domestic telecommunications has tended to imply that the same benefits that accrue to domestic telecommunications would accrue to international telecommunications. To the extent that the findings in this study diverge from that found in domestic telecommunications, they contradict conventional wisdom that the two arenas are similar. A frequent criticism of a study such as this is that it is confined to only one country--in this case, Singapore. However, many of the conditions that are theorized to be 110 present in the telecommunication-economic development relationship--telecommunications-intensive, open trade and business policies--are to be found in Singapore. The criticism in this context, therefore, is a two-edged sword. Data were obtained from several published sources. The Singapore Monthly Digest of Statistics was the source for economic indicators and data on international outgoing telegram, telex and telephone calls by volume and month. The data were available on a computer database called Public Access Time-series (PATS) and the relevant data were downloaded on to a diskette after a search. The digest and the database are maintained by the Department of Statistics, the government body charged with the task of collecting and compiling such information. ' For this study, GDP instead of GNP was used.“The difference between the two economic measures lies in their territorial coverage. GDP covers the domestic economy, which is the purpose of this study. Three levels of aggregation of data were used. First, data for all the sectors used in computing GDP were collected. Next, data for trade and tourism data were collected. Finally, data for trade and tourism between Singapore and various countries were extracted and compiled. Telecommunications data are restricted to outbound international telephone calls made on the public switched 111 network. Calls on leased lines are excluded. International telephone traffic in this study is measured by the number of calls. Although it has been suggested that it would be better to use minutes (Staple and Mullins 1989b), such data were not available for this study. Most of the telecommunications data were also from the computer database. Those that fell outside the scope of the Digest of Statistics were obtained from the following publications: International Telecommunications Union Yearbook of Common Carrier Statistics, AT&T's World's Telephone, Singapore Telecom Annual Report and Singapore Telecom's Statistical Bulletin. The quality and scope of the data varied. For example, Singapore Telecom's Statistical Bulletin provided international telephone calls broken down monthly for outgoing for the top 20 countries by call volume only from 1976 to 1985. ' Most of the other variables are in dollar value at current prices unless otherwise indicated. The obvious divergence from the use of dollar value is in visitor traffic where the unit is a person. Qpenetionai Definitions International telecommunications in this study refers to international telephone traffic switched through the public network flowing to a party outside the country. It covers voice and data transmissions such as faxes but excludes telegram, telex, maritime, television and radio 112 transmissions. Voice and data transmissions through leased lines are also excluded. Where the specific variable is referred to, the term International Telephone will be used in the uppercase. The various economic indicators are defined according to official classifications by the Department of Statistics as placed in Appendix 1. As above, where the specific variable is referred to, the term will be used in the uppercase. The theme of telecommunications and economic development is open to two interpretations: first, telecommunications as a means to attract business to or retain business in a country, and, second, telecommunications as a factor of production (Jurgen and others 1989). As this study does not look into the role of telecommunications in attracting or retaining businesses, the'second interpretation is adopted and telecommunications is taken to be as a factor of production. The development of a country has political, social and economic dimensions. Rogers (1976) sees development as involving the equality of distribution of wealth and socioeconomic benefits, decentralization of decision-making, planning and execution, self-reliance and independence. Development by this notion is not limited to economic wealth. Economic indicators such as GDP (gross domestic product) and GNP (gross national product) per capita are therefore not the best measures of development as defined. 113 Some quality of life index would be ideal. Ideally, therefore, a quality of life index would be the best measure of the national impact of international telecommunications. As the rest of this chapter will show, however, such an indicator is difficult to arrive at and therefore development has been defined by the traditional economic indicators of GDP and various economic sectors. Definitions of Economic Development Development is necessarily a normative concept. It has been variously defined as economic growth, structural economic change, autonomous industrialization, capitalism, and self-determination of the nation and individual (Harrison 1988:154). In the 19505 and 19605, under the modernization paradigm, economics loomed large. During this period, developmentalists focussed on the increase of goods and'services produced by a country, and therefore'GDP, as indicators of development. This is illustrated by Rogers who initially defined development as "modernization at the social system level" that occurs when "new ideas are introduced into a social system in order to produce higher per capita incomes and levels of living through more modern production methods and improved social organization" (1971:11). By the 19705, the emphasis had shifted to redistribution and other "softer" goals such as improvements in health, welfare, education and the general quality of 114 life. This is illustrated, again by Rogers who in 1976 modified his definition of development to "a widely participatory process of social change in a society, intended to bring about both social and material advancement (including greater equality, freedom, and other valued qualities) for the majority of the people through their gaining greater control over their environment" (1976a:215). That definition is in line with the attempt during the period to move away from merely equating development with economic growth to defining development in terms of "distributive justice and general human fulfilment" (Dissanayake 1984:43). Taking this track, others have defined development to be "the deliberate attempt to reduce the discrepancy between conditions of the rich and the poor" (Narula and Pearce, 1986:xviii). The idea is not new. In 1969, Dudley Seers said that the'questions to ask about a country's development are: What has been happening to poverty? What has been happening to inequality? What has been happening to employment? He wrote: "If all three of these have declined from high levels, then beyond a doubt, this has been a period of development for the country concerned. If one or two of these central problems have been growing worse, especially if all three have, it would be strange to call the result 'development' even if the per capita income doubled" (1969:2). Others have pushed this idea further, and in effect have given up the attempt at definition. They have instead 115 gone on to define the goals of development as "increased economic output, improvement of health and life expectancy, a reliable civil service, an appropriately educated population, the achievement of governable urban areas" (Moore, 1979:1). To measure these less easily quantifiable factors, it is not enough to use GDP alone as it would assume a trickle down of wealth. Indicators of development have therefore been used. The indicators are used partly for comparison and partly to measure equality of income distribution (Hicks and Streeten 1981). Among the various approaches of measuring development, three are particularly relevant. They are: first, adjustments to the GDP to try to capture welfare, second, the use of social indicators and, third, the use of composite indices to measure and compare development. . . 1. Adjustments to the GNP In the first approach, the GNP is adjusted to "to capture some of the welfare aspects of development and to improve international comparability" (Hicks and Streeten 1981:54). One method of doing this is to use "purchasing power parities" where the output of the country is converted into local purchasing power. The Economist magazine, for example, measures the number of hours a worker needs to work to buy a MacDonald's Big Mac hamburger as an indicator of purchasing 116 power. Using purchasing power parities, a study by Kravis et al (1978) suggested that the Indian GDP per capita should be increased by 3.5 times. But using purchasing power parities encounters a major problem: some countries would spend more on certain services than others because of differences in location or climate. For example, comparability between Canada and Singapore is reduced because Canada has snow and Singapore is in the tropics-~differences that would affect expenditure in areas such as clothing, transportation and housing. Furthermore, this method is demanding as a wide range of goods and services must be compared. Another method, propounded by Nordhaus and Tobin (1972), suggests that "regrettable necessities" such as defense and "disamenities" such as crime and pollution be subtracted from the GDP while leisure activities should be added to create a "Measure of Economic Welfare." In their study, the US GNP doubled because of the weighting given to leisure, a weighting hard to defend. This method of measuring development runs into the problem of using GNP for something it was not intended: to capture welfare. 2. Social indicators The second approach is to use measures such as health, nutrition, housing, income distribution, and other aspects of social and cultural life to indicate the level of development. Several international agencies have used this 117 method (UN 1975; OECD 1977; AID 1976; UNESCO 1977). The key advantage of this method is that it is concerned with ends as well as means. The disadvantage of the method is that there is no conceptual framework for unifying the indicators. Further, the data needed tend to be unreliable, and are particularly weak in those countries that probably need the information most for development (Hicks and Streeten 1981). Just how close economic and social indicators are correlated is uncertain. Studies on the correlation between economic and social indicators have produced conflicting results. Some studies have shown a high correlations (McGranaham, et al 1972; United Nations 1975; Sheehan and Hopkins 1978) while others have found weak correlations (Morawetz 1977). Hicks and Streeten suggest in Table 4.1 below that the contradictory conclusions arose from the use of differing indicators, differing sources of data, differing country samples as well as "differing interpretations of resultsh (1981:58). Using the World Bank's Social Data Bank, they showed that there was a correlation between social indicators and GNP of r2=0.5. The correlation rose to r3=0.71 when economic indicators were used. But when the social indicator data are disaggregated into samples of developing and developed countries, the correlations for both dropped to r2=0.25 and r2=0.18 respectively, showing that the data are sensitive to aggregation. Hicks and 118 Streeten concluded that "studies which examine only social variables for developing countries are apt to discover a poor relationship, while those that consider economic and social variables for all countries are likely to find better relationships" (1981:59). They suggest that the correlation between social indicators and GNP is weak because of the "ceiling effect." That is, beyond a certain level of GNP, it is impossible to achieve, in linear fashion, literacy growth of more than 100 percent and life-expectancy of much more than 70 years. The limits are reached by the more developed countries, which may explain why social indicators are better correlated in developing countries (Hicks and Streeten 1981:59). Table 4.1.--Correlation of indicators with GNP per capita (1970). Coefficients of determination (r3) ’ All countries Developing “Developed Social indicators 0.13 Life expectancy 0.53 0.28 Calories (% of need) 0.44 0.22 0.02 Infant mortality 0.42 0.34 0.25 Primary enrolment 0.28 0.24 0.05 Literacy 0.54 0.47 0.16 Avg persons per room 0.58 0.08 0.29 Homes w/o piped water (%) 0.74 0.13 0.36 Average 0.50 0.25 0.18 Economic indicators Newsprint use 0.79 0.20 0.46 Automobiles 0.85 0.59 0.46 Radios 0.43 0.14 0.07 Electricity use 0.67 0.30 0.24 Energy use 0.82 0.28 0.49 Average 0.71 0.30 0.34 Source: Hicks and Streeten (1981:58). 119 3. Composite indices The third approach develops composite indices by combining several social indicators into a single index of human and social development. The difficulty with such indices though is that they run into problems of scaling, weighting and data collection, problems similar to that in the second approach of using social indicators. The UN Research Institute for Social Development (UNRISD) performed substantial research during the 19605 to develop better social indicators, including composite indicators. Jan Drewnowski and Wolf Scott (1966) under the UNRISD developed the "Level of Living" index, which is defined as "the level of satisfaction of the needs of the population as measured by the flow of goods and services enjoyed in a unit of time" (1966:1). They divided needs into "Basic" and "Surplus over Basic" or "Higher" needs. The basic'needs included some indicators for which data were difficult to obtain: the amount of leisure time available, the number of people who had private savings, and the quality of housing. The index was therefore difficult to apply. Even in their sample of 20, some shortcuts were needed. Their work ceased after 1966 (Hicks and Streeten 1981). There have since been attempts to formulate social indicators of developments. Perhaps the best publicized social indicator is the Physical Quality of Life Index by M.D. Morris (1979). She argued that indicators of 120 development should concentrate on outputs or results rather than inputs. He therefore measured literacy, infant mortality and life expectancy, weighted them equally, and arrived at an index. Critics, however, noted that the term "quality of life" was a misnomer. What Morris really measured was effectiveness in reducing mortality and raising literacy, in effect, the "quantity of life." Second, the weighting system was arbitrary and could not be defended (Hicks and Streeten 1981:61-62). Two studies in the 19705 using social indicators found that they had a high correlation to per capita GNP. The first, by D. V. McGranahan et al (1972), examined and found a high correlation among 73 social and economic indicators. Through elimination, they devised a "Development Index" based on 18 "core indicators" that included nine social and nine economic characteristics. The resulting index was highly correlated (r3=0.89) with GNP per capita. A few years later, a study by the UN Economics and Social Council (1975) analyzed development by ranking 140 countries with seven indicators other than GNP. There were two social indicators (literacy and life expectancy) and five economic indicators (energy, manufacturing, share of GDP, manufacturing share of exports, employment outside agriculture, number of telephones). An overall rank was created by giving equal weight to each indicator. The study found that the overall ranking closely matched ranking by 121 GNP. The problem with indicators of development that are highly correlated with GNP is that one is tempted to simply use GNP, especially as the indices are difficult to create. Hicks and Streeten have argued for a composite index to supplement, not replace, GNP. They suggested six essential and non-exhaustive measures: nutrition, basic education, health, sanitation, water supply and housing, and related infrastructure (ibid). Nothing, however, has come out of their proposal. Summary In sum, there are good reasons for using GDP/GNP and their components to analyze development. First, as noted above, social indicators appear to suffer from a ceiling effect in more developed countries. Although Singapore is not'a developed country, its GDP and standard of living is neck-and-neck with Hong Kong, behind only Japan's in Asia. Second, there is often a high correlation between the more complex social indicator and the simpler GDP/GNP. This militates against the use of the complex indicators. Third, by the 19805, it was clear that without an increase in the economic performance of a country, there would be less to redistribute. So the use of GDP is seen today as useful for indicating quality of life. CHAPTER FIVE Analysis of Quarterly Data Method As originally conceived, the method to be used for determining causality of international telecommunications and economic development was the cross-lagged correlation. This was the method Hardy used in his 1980 study and had been used in other communication studies such as agenda- setting (Shaw and McCombs 1974). The idea is that if A causes B, then the correlation should increase if prior values in A are used to correlate with B. This is best illustrated in the following examples. Table 5.1 shows the usual manner in which ordinary least squares (OLS) correlation is performed. Trade in one year is correlated with International Telephone traffic in the corresponding year. Table 5.1.--Correlation between Trade (tr) and International Telephone (tc) traffic. ‘ Trade I 1980tr I 1981tr I l982tr I l983tr I 1984tr I 1985tr I TC traff I 1980tc I l981tc I l982tc I 1983tc I 1984tc I l985tc I Table 5.2.--Correlation between Trade (tr) and time-lagged International Telephone (tc) traffic Trade I 1980tr I l981tr I 1982tr I l983tr I 1984tr I 1985tr I TC traff I I 1980tc I l981tc I l982tc I 1983tc I 1984tc I 122 123 Table 5.3.--Correlation between time-lagged Trade (tr) and International Telephone (tc) traffic. Trade I I 1980tr I l981tr I l982tr I l983tr I 1984tr I TC traff I 1980tc I l981tc I 1982tc I 1983tc I 1984tc I l985tc I In Table 5.2, International Telephone traffic is regressed against Trade with a lag of one period. If International Telephone traffic causes Trade, then one would expect a higher R2 here compared with the R2 in Table 5.1. In Table 5.3, Trade is regressed against International Telephone traffic with a lag of one period. If International Telephone traffic causes Trade, one would expect a lower R2 here compared with the R2 in tables 5.1 and 5.2. If, on the other hand, Trade causes International Telephone traffic, then the correlation should be highest in Table 5.3, second highest in Table 5.1 and lowest in Table 5.2. In the above tables, causality does not refer to an exclusive relationship where other variables must be excluded. It does not mean the only cause. Rather, the causality refers to a contributory cause. The requirements for causality in the sense used above between two variables A and B are therefore as follows: 1. there must be a relationship between A and B; and 2. A must be causally prior to B. Qata Data for the performance of the entire Singapore 124 economy are released quarterly. Table 5.4 shows the contribution each sector to GDP at constant 1985 prices from March 1975 to March 1992. Table 5.4.--Average percentage contribution of various sectors to GDP at constant 1985 prices (March 1975 to March 1992). Sector Percent of GDP Manufacturing 27.29 Financial Business 23.38 Commerce 18.28 Transport and Communication 12.58 Other Services 11.62 Construction 8.09 Utility 2.02 Agriculture and Fishery 0.87 Quarry 0.24 Source: Author's calculations from data by the Department of Statistics. A preliminary test of the correlation was run using the quarterly data of economic activity. As the data were an aggregation intermediate between yearly and monthly, it was expected that a preliminary analysis could highlight problems common to all three levels of aggregation. Indeed, the "eyeballing" of the raw data Suggested that a straightforward correlation would not be appropriate. Figure 5.1 shows that International Telephone traffic has been growing significantly, at a seemingly exponential rate. Such a curve suggests that the data should be log- transformed. (An advantage of using natural log transformations is that it shows the elasticity in the 125 regression relationship, where regression may validly be used.) Regressions were therefore performed first on raw untransformed data and then on log-transformed data. Even with untransformed data, OLS regressions yielded high R25, as evidenced by Table 5.5. Of the 10 sectors, eight had adjusted-R25 higher than 0.80. Further, in all but one instance, significance was at the <0.001 level. The one exception-~Construction--had significance at the 0.003 level. Table 5.5.--Relationship between untransformed quarterly International Telephone by call volume and various economic sectors (in Singapore dollars) in decreasing order of R2. Sectors qu-adj t-ratio** Commerce .903 25.16 Manufacturing .887 23.10 Utility .882 22.55 Agriculture .878 -22.11 Finance .872 21.55 GDP .868 21.13 Transport .855 20.02 Other Services .819 17.58 Construction .108 3.04* Quarry .047 30.21 Alternate rows have been made bold for easier reading. *p<0.003. **p<0.001 level unless otherwise indicated. When OLS regressions were performed on log-transformed data, the R25 rose for all sectors except Agriculture. (See Table 5.6.) The two tables, however, have a low Spearman's rho rank-order correlation--0.318. The low statistic shows that the data, and by implication the results, are susceptible to transformation. Con vol (millions); GDP (billions 5:) Figure 26 24 22 20 14 12 10 126 5.1 International Telephone Traffic and GDP at 1985 pflces CD' _ Tewphone 1975 [1977 I1979 I19D1 I19D3I19D5 I19E7 I19E9 I1951 I 1976 1978 1980 1982 1984 1986 1988 1990 1992 26 24 22 20 18 16 14 12 127 Table 5.6.--Relationship between the natural log of International Telephone and the natural log of various economic sectors in decreasing order of R2. Sectors R-sq qu-adj t-ratio* DW Finance .991 .991 85.97 0.30 other Services .991 .991 85.40 1.24 Utility .986 .986 69.43 0.72 GDP .985 .985 67.10 0.51 Transport .980 .979 56.75 0.12 Commerce .959 .958 39.48 0.68 Manufacturing .926 .925 28.90 0.56 Agriculture .656 .651 -11.32 0.09 Construction .488 .480 7.99 0.04 Quarry .422 .414 7.00 0.08 Alternate rows have been made bold for easier reading. *p<0.001 These high R2 and significance, however, are not grounds for rejoicing. A key assumption in OLS regression is that the data are not autocorrelated. That is, the residuals are not related among themselves. OLS estimation disregarding the presence of autocorrelation will lead to underestimation of variance and overestimation of R2 (Gujarati 1988:364). The degree of autocorrelation is measured by the Durbin-Watson statistic, which determines if the residuals are stationary. If the residuals are stationary, then they are not autocorrelated. If they are not stationary, then the residuals are autocorrelated and OLS cannot be taken as a good method for estimation and prediction. 128 The Durbin-Watson (DW) statistic, d is defined as: The above equation may be looked upon as the sum of squares of the first difference of residuals divided by the sum of squares. (The first difference is the difference between the value of a period and the value in the earlier period. See Appendix B for a more detailed explanation.) Ideally, the DW statistic should be at 2. If as in above, the DW statistic tends to zero, the residuals are not stationary and using R2 in such a context will lead to spurious inferences. The DW statistic is such that there is an upper and lower limit, usually designated dU and dL respectively. There is a zone of indecision between the upper and lower limits such that if the test statistic falls between two, one cannot conclude if autocorrelation does or does not exist. The situation is illustrated in Figure 5.2. Figure 5.2.--Durbin-Watson d statistic Positive Zone No Zone Negative Auto-. of Auto- of Auto- correlation Indecision correlation Indecision correlation ‘— I_ 129 Where the test statistic falls into the "zone of indecision, it has been suggested that the upper limit of the d-statistic be used as the critical value (Gujarati 1988:379). The critical values of the DW statistic for the current data set with 68 observations are listed in Table 5.7. Table 5.7.--Critical Values of the Durbin-Watson statistic Significance 1% 5% :3L 1.420 1.577 dU 1.478 1.636 Source: Gujarati (1988:686-689) In the regressions presented in tables 5.5 and 5.6, none of the variables had significant DW statistics. With log-transformation, the DW statistic increases but not to a significant level. The above analysis raises the question of whether the conclusions of Hardy's 1980 study were based on spurious relationships. It is not possible based on the information available to decide one way or the other. Certainly, there is no indication that Hardy computed the DW statistic. On the other hand, the data Hardy used were annual, not quarterly or monthly, and a recent study of telecommunications and economic development in South Korea suggests that the problems of autocorrelation may not be as serious when annual data is used (Kim 1992). Nevertheless, 130 it will undoubtedly be worthwhile to reanalyze the Hardy study with an eye on the DW statistic. Whv Doe§ Autocorrelation Occur? Six reasons have been given for the existence of autocorrelation such as that in the above data set (Gujarati 1988:356-9). The first three reasons are beyond the control of the current study. 1. The regression has excluded important variables. In the current analysis, only two variables are intended to be studied at one time. 2. The regression uses incorrect functional forms. The current study assumes a linear or, at most, a log-linear relationship. Other functional forms may fit better but they will complicate the study; finding the "right" functional form alone could be the basis of an entire study. For uniformity and ease of both application and understanding therefore, the simple regression is used in this study. 3. The manipulation of data has contributed to autocorrelation. "Massaging" of data through seasonal adjustments may impose on the data a systematic pattern that might not have been there before. Some key data, particularlyGNP, are known to be seasonally affected and 'therefore seasonal adjustments are necessary. In the present study, it is assumed that there are similar seasonalities and charts of the data do suggest seasonal variations. For consistency of analysis, seasonal adjustment was applied to 131 all the data in this study. The analysis was carried out on both seasonal and non-seasonal data for comparison. The next three reasons may be inherent in the data under study. Autocorrelation arises because the error terms are not random but are related to the variables under consideration. 4. The two series could have a "cobweb" phenomenon. This is when the supply of_one product reacts to the price of another product. 5. The two series may have inertia or sluggishness in that the series move together. This means the data are interdependent. 6. The two series may have lags, where one series lags behind another. In the current study, this would mean that the leading series is "causing" the second series. The latter two reasons will leave behind non-random residuals, which explains the high autocorrelation extant. Consequences of OLS Estimation Disregarding Autocorrelation Using ordinary least squares regression in the presence of autocorrelation will lead, first, to an underestimation of variance, which leads next to an overestimation of R2. This leads to Type I errors, where the null hypotheses are rejected when they should not be. The usual t and F tests of significance are likely to give misleading conclusions about the statistical significance of the estimated regression coefficients (Gujarati 1988:364). As the current study 132 compares R25, their overestimation could lead to spurious results. The Solution: Time Series Analysis How is it possible to achieve high R25 and high significance if a key assumption of OLS is violated? The answer lies in the fact that the data are time-series data. That is, they share the common element of time. As far back as the early 19005, if not before, statisticians found that time-series sometimes yielded what they called "non-sense correlations" (Yule 1926). An example of such a nonsense correlation found was between rainfall and inflation. As in the analyses of telecommunications and various economic sectors above, analyses of "nonsense correlations" found high R2 and significance but low DW values (Hendry 1986). Indeed, it has been shown that where two time-series are unrelated and the variance in each series is not constant, that is non-stationary, the R2 of the regression tends to unity (Yule 1926; Banerjee et a1 1986). The answer to avoiding such spurious estimates is to use time-series analysis. The basic logic behind time-series analyses is to transform the series and make it stationary so that linear regression is possible. A time series is said to be stationary if there is no systematic change in mean (that is, there is no trend), if there is no systematic change in variance, and if seasonal or other periodic 133 variations have been removed (Chatfield 1989:10). Seasonal adjustments are necessary in the data sets. For example, GDP on a quarterly basis is known to be affected by seasonal trends. Figures 5.3 and 5.4 plot GDP against time to illustrate the effect of seasonal influences and the effect of smoothing respectively. Data were smoothed using the X-11 procedure in the statistical software package SAS. The X-11 procedure is the same procedure used by the US Census Bureau in making seasonal adjustments. The major function of the procedure is to pick out a seasonal trend and to remove it. Other adjustments in X-11 are minor, such as adjusting for the number of days in a quarter or month. Regressions were then performed using smoothed and log- transformed smoothed data and the results tabulated in tables 5.8 and 5.9. They showed that R25 were higher for smoothed than raw data, indicating that seasonal influences introduced some random errors that should be removed for analysis. The presence of random errors is corroborated by the DW statistic. It is higher for the regressions pre-Xll adjustment. Bhfions (SS) Billions (ss) ’ 10— 134 Figure 5.3 GDP Before Seasonal Adjustments Constant 1985 Prices 17 16 e 15 — 14 e 1.3 — 12 ~ 11- 1975 I 19177 I 19.79 I 19‘81‘I 1983 I 19.85 I 1987 I 1989 I 19.91 I 1976 1978 1980 1982 1984 1986 1988 1990 1992 . .Figure 5.4GDP After X11 Seasonal Adjustments Constant 1985 Price 17 ISI- 15'- 14 — 13'- 12'- 11— 1935 I 1977 I 1979 I 1981 I 1983 I 1985 I 1987 I 1989 I 1991 1976 1978 1980 1982 1984 1986 1988 1990 Table 5.8.--Comparison of Untransformed Data Pre- and Post- X11 Seasonal Adjustments. 135 Pre-Xll Post-X11 Sectors R-sq DW R-sq DW Commerce .904 0.35 .916 0.07 Manufacturing .888 0.46 .911 0.07 Agriculture .879 0.47 .894 0.25 Utility .884 0.10 .889 0.02 Finance .874 0.04 .874 0.03 GDP .869 0.07 .873 0.02 Transport .857 0.02 .857 0.01 Other Services .822 0.08 .824 0.03 Construction** .121 0.06 .121 0.04 Quarry .061 0.11 .062* 0.06 Alternate rows have been made bold for easier reading. *p<0.04. **p<0.003. ***p<0.001 unless otherwise indicated. Table 5.9.4-Comparison of Natural Log Transformed Data Pre- and Post-X11 Seasonal Adjustments. Pre-Xll Post-X11 Sectors R-sq DW R-sq DW Other Services .991 1.24 .994 0.49 Utility .986 0.72 .990 0.16 Finance .991 0.30 .991 0.24 GDP .985 0.51 .988 0.08 Transport .980 0.12 .981 0.06 Commerce .959 0.68 .969 0.14 Manufacturing .926 0.56 .942 0.09 Agriculture .656 0.09 .667 0.04 Construction .488 0.04 .489 0.03 Quarry .422 0.08 .427 0.06 Alternate rows have been made bold for easier reading. *p<0.001. Table 5.9 again shows that despite higher R25, all the variables do not have significant DW statistics. Time-series analysis suggests that with an apparently exponential curve of International Telephone traffic and a relatively flat GDP 136 pattern, regression cannot be run on untransformed data and may not be used even on log-transformed data. The usual next step in time-series analysis is to analyze the percentage growth rates. In equation form the variable becomes transformed thus: X = lnXt - lnxeqj where X is first log transformed and then differenced. Table 5.10 shows the outcome of the regressions of International Telephone traffic growth rate against the growth rate of various economic sectors. Table S.10.--The relationship between International Telephone traffic and various economic sectors' growth percentages after X-ll seasonal adjustments. Sectors qu qu—adj p< DW Transport .133 .120 .002 1.52 GDP .082 .069 .018 1.81 Finance .036 .021 .123 1.62 Manufacturing .035 .020 .128 1.74 Commerce .025 .010 .198 1.68 Utility .021 .006 .241 1.63 Construction .004 .000 .588 1.58 Agriculture .003 .000 .684 1.61 Quarry .002 .000 .741 1.60 Other Services .000 .000 .864 1.60 Alternate rows have been made bold for easier reading. In Table 5.10, the R25 are not as high as before, but there is ground for rejoicing as the key DW statistic is significant for all variables at least at the .05 level. As noted earlier, apart from significance, the key concern is 137 the direction of causality. As causality has been defined above to mean contributory causality, the low correlations (0.133 and less) are not a cause for concern in this analysis. Proving Causality Of the time-series tests to prove causality, probably the easiest to use and understand is the Granger test (Granger 1969). The method checks if lagged terms of the independent variable offers additional explanatory power when the lagged terms of both the dependent and independent variables are included in the regression. The idea is similar to that used in cross-lagged correlation: if the independent variable causes the dependent variable, then lagged terms of the independent variable must offer additional explanatory power when included in the regression. In the present study, the variables are run with four lags in the following form: (Equat1ZnA11)+ let_1 + 82xt_2 + B3Xt_3 + B4Xt_4 + error and xt= A2 + cl)(t_1 + czxt_2 + c3xt_3 + c4xt_4 + (Equationogt-l + D2Yt_2 + D3Yt_3 + D4Yt_4 + error. Ho: X does not cause Y Dl=D2=D3=D4=0 using the F-test. (where t-1 is a lag of 1, t-2 is a lag of 2, t-3 is a lag of 3 and t-4 is a lag of 4.) 138 The number of lags is arbitrary and as many may be added as desired. But adding more lags decreases the explanatory power of the regression. econometric convention where the data are quarterly, lags were used. In line with four The question is whether the additional variable and its lags in Equation 2 add significantly to Equation 1. That is, whether Equation 2 increases the R2 significantly. The F- statistic is used for the test and is derived thus: - 2 2 Rr)/4 R u (l-Rzu) / Phone Causes Re-Export (Current) 3.0531 0.025 Phone Causes Re-Export (Constant) 2.1033 0.096 Phone! Causes Re-Export (Constant)! 2.2262 0.082 Trade (Constant)! Causes Phone! 2.0967 0.098 Import (Constant)! Causes Phone! 2.0967 0.098 Table 6.9 shows three relationships: first, that International Telephone "causes" Re-Exports, second, that seasonally-adjusted Trade at constant prices causes International Telephone and third that seasonally-adjusted Import at constant prices causes International Telephone. Assuming the relationships above to be true, coefficients of regressions will yield the impact of growth of one variable on the other. Table 6.10 below shows the result of a 1 percent growth of the independent variable on the dependent. 158 Table 6.10.--The relationship between International Telephone and various economic sectors' growth percentages. Causal Sectors Growth in International Telephone 1% Increase % Trade (Constant) 0.18115 Import (Constant) 0.14569 Int'l Phone 0.1953 % Growth in Re-Export (Constant) The findings here are in contrast to those in Chapter 5 where the growth of 1 percent in the causal economic sectors caused growth of more than 1 percent in International Telecommunication. Table 6.10 shows that a 1 percent growth at the subsectoral level will cause growth of between one fifth (0.2) and one seventh (0.14) of 1 percent in the dependent variable. Qiscuesion and Coneinsion The findings for Trade and Import are consistent with that in Chapter 5, which found that Commerce (the sector that embraces Trade) causes International Telephone traffic. This suggests some internal validity. The relationship between International Telephone traffic and Re-Export is interesting because of the presence of the latter variable in current and constant prices and adjusted and non-adjusted forms. (It should be noted here that seasonally adjusted Re-Export at current prices is absent as it was not stationary even after log- 159 transformation and differencing.) It would appear that International Telephone traffic takes on great importance when coordination is needed, such as in the coming and going of goods for Re-Export. The degree of causality, however, is less significant than that at the aggregated level presented earlier in Chapter 5. The degree of causality in this chapter is smaller than that in Chapter 5 where it was found that the economic sectors caused International Telecommunication growth greater than unity. (The exception was the impact of International Telecommunication on Transport, where it was less than unity. Nevertheless, it is greater than the 0.19 in this chapter.) Exactly what this means is not clear from the current analysis. It does suggest, however, that the level of aggregation of data may have some impact on the results of analysis. One clear thread from this study is that constant prices dominate in the causal relationship over current prices. Table 6.9 shows the dominance well. Constant prices have a deflator that is typically equivalent to the inflation rate. The relationship with constant prices may be better understood if it is borne in mind that International Telephone traffic is measured by the number of calls and not the length of the call. If one phone call is all that was needed to deal with a lot of goods 10 years ago, there is no reason to believe that more phone calls are needed for that same lot now. The value of the lot would likely have 160 increased and perhaps the length of the call would have increased with the increased value. But the value of the lot is discounted by constant prices and the length of call is ignored in the data. Conspicuous by its absence in Table 6.9 is tourism. Neither Visitor Arrivals nor Plane Landings had significant relationships. The variable that comes closest is Plane Landings, with an F-value of 1.8366, which is significant at p>.15. One possible reason for the absences is the eruption of the Gulf War from late 1990 to early 1991. Visitor arrivals fell while International Telephone traffic remained the same during the period. Figure 6.1 shows the fall-off in visitor arrivals. Seasonal adjustment appears to increase the likelihood of finding significant results. Figure 6.2 shows that there is very little seasonal variation in International Telephone traffic. In contrast, GDP and many of the visitor variables have marked seasonal variations. This is clear in Figure 6.1 showing total visitor arrivals. (Thousands) 161 Figure 6.1 Total Visitor Arrivals March 7989 to 1992 29 28 — 27 .— 26 - 25 _ 24 — 23 b 22 I I I I I I r MAR89 I SEPBQ I MARQO I 59°90 I MARQI I SEPQI I MAR92 I JUNBQ 05689 JUNQO DECQO JUNQI DECQI . JUN92 Figure 6.2 International Telephone Traffic March 7969 to 1992 95 90 _ 55 _ 50.. 75 _ Thousands m o. o 01 0 0| 3 I l T7 i 01 O I 45 J 4O MAIIIQGQ SEPIQBQ I MARIQQO I SEPIQQO I M48199: JUNIQBQ 050989 JUNIQQO DECIQQO JUNIQQI 5521991 MARI992 DE C199! 162 On the whole, the findings in this chapter are consistent with that found earlier. The absence of Visitors as a significant variable, however, runs counter to various hypotheses that a good telecommunication system is necessary for tourism. CHAPTER SEVEN Analysis of Trade and Visitor Flows Introduction This chapter looks at outgoing International Telephone traffic to various countries and compares them with data in trade and tourism. The data, unfortunately, are available only from 1976 to 1985. Various attempts to obtain more up-to-date information from Singapore Telecom, including an appeal through the minister in charge, failed to yield any result. A Singapore Telecom spokesman said that there were two reasons for this reserve. First, other telecommunication authorities, both regional and international, were known to be using such information for their own competitive purposes. Second, the authority is preparing for an initial public offering of its shares in late 1993. For fear of giving away any "financial advantage," all information were being closely guarded. (Foo 1992). Caveats Several caveats about the data are therefore in order. First, unlike chapters five and six, which used recent data, the current data set is more than seven years old. Second, while the data are supposed to reflect International Telephone traffic between Singapore and various countries, the top two destinations are either absent or have very little data available. The top 163 164 international destination for Singapore, like many countries, is its next-door neighbor Malaysia. Calls to Malaysia, however, are considered trunk calls as they are not routed through an international gateway. The analogy to the American situation would be to consider calls to Malaysia long-distance instead of international traffic. The treatment of calls is a throwback to the period when Singapore was part of Malaysia or governed as such by the British. The top international destination is the next closest neighbor, Indonesia. Analysis for flows into and out of this country was curtailed for lack of data. International Telephone traffic data were available but data on trade were absent. This is another historical quirk. When Singapore agreed to join Malaysia to form a federation in the early 19605, Indonesia objected. Trade between Singapore and Indonesia collapsed. (As a matter of interest, the drop in trade was so severe that the X-11 seasonal adjustment could not handle the drastic change. Under the adjustment, trade was negative!) Since then, trade between the two countries has not been publicly reported. For several other countries, data collection began much later. Fortunately, the data were either absent or in sufficient quantity as to make analysis possible. Thus, although Singapore Telecom reported the call volume to the top 20 countries for each of the years under study, only data for 14 countries could be analyzed. 165 Third, it should be noted that the year that the data set ended--1985--was the year of a recession in Singapore, the first in living memory for many Singaporeans. The last time that happened was in 1961. This means that for half the period of the data set, Singapore was on an economic decline. Although the analysis may be valid for Singapore, caution should be exercised in comparing the results with those of other countries. Finally, although the data set was available on a monthly basis, it had to be aggregated to quarterly. The autocorrelation for monthly data was extremely high. After several months of attempts, it was concluded that the analytical method needed to obtain meaningful results was beyond the scope of this dissertation. The result of the aggregation to a quarterly basis was to reduce each variable to 37 data points, or fewer, from the more than 100 they started out with. With these caveats, the analysis was performed on the top 14 countries in outgoing calls from Singapore. From 1976 to 1985, Singapore Telecom released data on the top 20 countries but these 14 were consistent enough in the top rankings so that analysis was possible. Data for other countries had gaps when they were not among the top 20, making analysis difficult. The 14 countries are: Indonesia, Philippines, South Korea, Australia, France, Italy, Taiwan, West Germany, USA, Hong Kong, India, Japan, Netherlands and United Kingdom. 166 In the data sets below, Visitors refer to tourists and their country of residence. A data set of say, Taiwan Visitors thus refers to tourists who are residents of Taiwan. Arrivals and Departures refer to embarkation and disembarkation at the airport; Arrivals and Departures exclude transit passengers but include Visitors entering Singapore by air. Analysis and Results As with the earlier chapters, the data were first smoothed with X11 adjustments, then logged and then differenced. It was necessary to log and difference the data to make them stationary. Then, as in the previous chapters, the F-test was used to determine causality. The full results of the analysis are presented in Appendix C. Significant findings are presented in the tables below. It should be noted that both adjusted and non-adjusted data were used in this analysis because the seasonality of the data is not known. Seasonality is very marked for the data from European countries but less so for that in other countries. In general, the results of analysis from non- adjusted data should be treated with caution because of possible correlations with the seasonal instead of the underlying trend. 167 Philippines Table 7.1.--Causal factors for Philippines F Value p Air Arrival Causes Phone 2.3229 .0885 NB. Arrivals, and Departures, include those coming for business. Visitors are tourists. Some caution is in order when interpreting Table 7.1, which shows that Arrivals by Air was found to cause International Telephone traffic. Non-adjusted data were used to arrive at the conclusion. South Korea South Korea was another country that lack data present in other cases. Here, Visitor traffic was not available. The causal factors are listed in Table 7.2 below. Table 7.2.--Causal factors for South Korea F Value p Phone! Causes Domestic Export! 2.3667 .0841 Phone! Causes Export! 2.4305 .0780 Import Causes Phone 2.5710 .0663 International Telephone traffic contributes to the growth of Exports and Domestic Exports. Imports were found to cause International Telephone traffic growth but it should be noted that the finding is of non-adjusted data. 168 France The data on France yielded a highly statistically significant finding for Export causing International Telephone traffic growth as shown in Table 7.3. The finding was true for both adjusted and non-adjusted data. The finding of Visitors by Air and Land causing International Telephone traffic growth is interesting but again, it should be noted that these are for non-adjusted data. Table 7.3.--Causal factors for France F Value p Export Causes Phone 5.8903 .0022 Export! Causes Phone! 3.7523 .0179 Visitors (Air) Causes Phone 2.5532 .0822 Visitors (Land) Causes Phone 4.4037 .0146 Italy The data on Italy showed that Domestic Export and Total Export caused International Telephone traffic growth. The verdict for Visitors by Sea is mixed. Table 7.4 shows causality going both ways, particularly for adjusted data. Such bi-directional causality is called feedback. It should be noted, however, that Visitors by Sea from Italy have been on the decline since 1979 while International Telephone traffic has been on the increase. This result may therefore be spurious. 169 Table 7.4.--Causal factors for Italy F Value p Domestic Export Causes Phone 5.0068 .0050 Domestic Export! Causes Phone! 2.8680 .0472 Export Causes Phone 3.7060 .0188 Export! Causes Phone! 2.4111 .0708 Phone Causes Visitors (Sea) 3.8873 .0232 Visitors (Sea)! Causes Phone! 3.8764 .0235 Phone! Causes Visitors (Sea)! 3.2964 .0398 Taiwan The data on Taiwan shows International Telephone traffic growth to that country to be caused by Imports and Visitors by Land from it. The number of Visitors by Land from Taiwan, however, is small. Toward the end of 1985, it was between 500 and 700 a month. Table 7.5.--Causal factors for Taiwan F Value p Import! Causes Phone! 2.5246 .0699 Visitors (Land) Causes Phone 3.3151 .0391 Visitors (Land)! Causes Phone! 2.4745 .0890 West Germany The data from West Germany showed that International Telephone traffic growth caused growth in Domestic Exports, Total Exports and Total Imports. These were true both with non-adjusted and adjusted data. The findings from West Germany are instructive as the country is one of the largest trading nations in the world. 170 The verdict for visitors is mixed. The number of Visitors by Land and Sea has leveled off after 1980 so interpreting causality here should be cautious. Table 7.6.--Causal factors for West Germany F Value p Phone Causes Domestic Export 2.5392 .0688 Phone! Causes Domestic Export! 4.6966 .0068 Phone Causes Export 4.1657 .0116 Phone! Causes Export! 7.0167 .0008 Phone Causes Import 4.4000 .0092 Phone! Causes Import! 2.4187 .0791 Visitors (Land) Causes Phone 2.3456 .1017 Phone! Causes Visitors (Land)! 3.7416 .0264 Visitors (Sea) Causes Phone 3.0017 .0527 USA Of all the variables, only Visitors by Land were found to cause International Telephone traffic growth for the USA. This finding is reminiscent of Kellerman's finding that International Telephone traffic correlated with tourism for the USA (Kellerman 1990). Table 7.7.--Causal factors for USA F Value p Visitors (Land)! Causes Phone! 2.6522 .0744 Hong Kong Several relationships were found to be statistically significant, as presented in Table 7.8. The most interesting 171 is the case of Imports, which shows feedback. The finding is consistent for both adjusted and non-adjusted data. Similar consistency was found for Departures by Air and Visitors by Sea. Table 7.8.--Causal factors for Hong Kong F Value p Air Arrival! Causes Phone! 2.3344 .0873 Air Departure Causes Phone 2.5338 .9308 Air Departure! Causes Phone! 2.7175 .0560 Import Causes Phone 2.8865 .0462 Phone Causes Import 3.7283 .0183 Import! Causes Phone! 4.8461 .0059 Phone! Causes Import! 2.3255 .0882 Phone Causes Visitors (Sea) 3.5215 .0323 Phone! Causes Visitors (Sea)! 2.5613 .0815 India There was a sharp upturn in International Telephone traffic to India from May 1983. A senior Singapore Telecom official said that this was due to the opening of a new international gateway in India around that time (Lim 1993). Imports were found to cause International Telephone traffic growth. Table 7.9.--Causal factors for India F Value p Import Causes Phone 3.4873 .0333 Import! Causes Phone! 3.4519 .0344 172 Japan International Telephone traffic growth was found to cause Import growth. This was consistent for non-adjusted and adjusted data. Arrivals by Air, however, were found to cause International Telephone traffic growth. Table 7.10.--Causa1 factors for Japan F Value p Air Arrival! Causes Phone! 2.8325 .0491 Phone Causes Import 3.3457 .0277 Phone! Causes Import! 3.2690 .0301 The Netherlands The significant relationships found on The Netherlands used data that were not adjusted. These results should therefore be treated with caution. Growth of Arrival and Departure by Air and Visitors by Land were found to cause growth in International Telephone traffic. Table 7.11.--Causal factors for Netherlands F Value p Air Arrival Causes Phone 4.9514 .0053 Air Departure Causes Phone 4.9254 .0055 Visitors (Land) Causes Phone 2.4375 .0925 United Kingdom The data on the United Kingdom yielded the most interesting results on visitors: International Telephone 173 traffic growth was found to cause growth in all types of Visitor traffic. Arrivals by Air, however, were found to cause International Telephone traffic growth. Domestic Exports were also found to cause International Telephone traffic growth. Table 7.12.--Causal factors for United Kingdom F Value p Air Arrival Causes Phone 2.2782 .0933 Domestic Export! Causes Phone! 2.3127 .0996 Phone Causes Visitors (Air) 2.7187 .0696 Phone Causes Visitors (Land) 3.1915 .0439 Phone! Causes Visitors (Land)! 3.6573 .0285 Phone! Causes Visitors (Sea)! 5.0333 .0090 Phone Causes Visitors (Total) 3.0266 .0514 Phone! Causes Visitors (Total)! 2.4356 .0926 No statistically significant results were found for Australia, Canada and Indonesia. The case of Indonesia is disturbing in that only data for Arrivals by Air (Air Arrivals) and Departures by Air (Air Departures) were available for analysis. Discussion The results of the analysis above are re-drawn and presented in Table 7.13. 174 Table 7.13.--Summary of Significant Relationships Domestic Export Causes Phone Italy (IT); Domestic Export! Causes Phone! IT; United Kingdom (UK) Phone Causes Domestic Export West Germany (WG) Phone! Causes Domestic Export! South Korea (SK); WG Import Causes Phone SK; Hongkong (HK); India Import! Causes Phone! Taiwan (TW); HK; India (IN) Phone Causes Import WG; HK; Japan (JP) Phone! Causes Import! WG; HK; JP Export Causes Phone France (FR); IT Export! Causes Phone! FR; IT Phone Causes Export WG; SK Phone! Causes Export! WG Visitors (Air) Causes Phone FR Phone Causes Visitors (Air) UK Air Departure Causes Phone HK; Netherlands (NT) Air Departure! Causes Phone! HK Air Arrival Causes Phone Philippines (PH); NT; UK Air Arrival! Causes Phone! HK; JP Visitors (Land) Causes Phone FR; TW; WG; NT Visitors (Land)! Causes Phone! TW; WG Phone Causes Visitors (Land) UK Phone! Causes Visitors (Land)! WG; UK Visitors (Sea) Causes Phone WG Visitors (Sea)! Causes Phone! IT Phone Causes Visitors (Sea) IT; HK Phone! Causes Visitors (Sea)! IT; HK; UK Phone Causes Visitors (Total) UK Phone! Causes Visitors (Total)! UK Emphasis added. Imports, Exports, and Domestic Exports The findings overall seem to suggest a correlation between world trade volume. The most striking finding is that International Telephone traffic growth causes growth in Imports. The result is consistent for West Germany, Hong Kong and Japan. After the United States, West Germany and 175 Japan are the next largest trading countries in the world. Hong Kong is the second largest trading country in Asia, after Japan. This seems to suggest that the size of world trade by that country both reflects and is reflected by International Telephone traffic. This observation is consistent with the findings for Domestic Exports and Exports. Significant causal relationships also exist for France, United Kingdom, Italy and South Korea. During the period under study--1976 to 1985--all these countries were among the world's largest trading countries. Conspicuously absent is the United States. The absence, however, is consistent with the finding of Kellerman (1990) that for the USA, International Telephone traffic correlated with tourism, not trade. In the case of West Germany, the results consistently show that International Telephone traffic growth causes Imports, Exports and Domestic Exports growth. But it is the only country with a causal relationship among the three variables and International Telephone traffic. South Korea and Italy provide interesting contrasts. In the case of South Korea, International Telephone traffic growth causes both Exports and Domestic Exports growth. The causal relationship is vice versa for Italy. Hong Kong is interesting in that the relationship between Imports and International Telephone traffic shows feedback. 176 The causal relationship between Imports and International Telephone traffic for India and Taiwan is also interesting. Both these countries have low trade volume. Taiwan's trade overtook Singapore's only in 1987. India's world trade volume has yet to match Singapore's (IMF 1989). The results suggest that International Telephone traffic and trade do have significant relationships, the direction of which depends on the importance of trade for the destination country. As Singapore is a major trading country itself, this result raises the question of the validity of the statement. Would such a relationship hold for countries that are less important trading centers? Visitors and Air Travellers The link between International Telephone traffic growth and foreign visitors is less clear, perhaps because the definition of Visitors and Arrivals and Departures are themselves not clear. The results should therefore be treated with caution. The United Kingdom and Hong Kong stand out. International Telephone traffic growth causes visitors from the United Kingdom to increase across the board. The causal relationship is for Total Visitors as well as Visitors by Air, Land and Sea. Conventional wisdom would hold that International Telephone calls are more important for air travel than land or sea travel. The findings from data on Hong Kong are more consistent 177 with that logic. International Telephone traffic growth causes growth in Air Arrivals from and Air Departures to Hong Kong. It may be that the common link is the colonial heritage shared by Hong Kong and Singapore with the United Kingdom. Economic geographers argue that communication links tend to follow historical links (Castells 1989). The results for Visitors by Sea may be spurious. Sea travel has been on the decline at least since 1980 for virtually all countries in the study. Conclusion There is a more consistent relationship between International Telephone traffic and trade than between International Telephone and tourism. Of course, consistency per se does not necessarily indicate the strength of the relationship. But the consistency is consonant with that of Kellerman's (1990) study in which international telecommunications was found to correlate with trade for all countries studied except the USA. For the exception, USA, international telecommunications was found to correlate with tourism, a finding marginally replicated here. One possible explanation for this anomaly in the case of the USA is that telecommunications data for the financial sector was not captured. The data set above was up to 1985-- a period when telecommunications liberalization was sweeping the USA but not other parts of the world. The greater 178 flexibility afforded American companies meant that companies, especially in the financial sector, could move their telecommunications system to leased lines and private networks. Individuals, however, had to use the public switched networks still. As the above data set captures public switched network traffic, telecommunications is seen to have a relationship with the tourism industry, which has more individual users, than with the financial industry, which has more corporate users. One interesting finding is that outbound International Telecommunication traffic growth was shown to cause Imports growth from West Germany, Japan and Hong Kong. After the United States of America, West Germany and Japan are second and third in total world trade volume. Hong Kong is the largest trading country in Asia after Japan. The finding suggests that International Telecommunication with the three major trading countries creates wealth. This finding, however, is the reverse of the finding at the sectoral level that Total Imports caused International Telecommunication traffic growth. The "opposite" finding cannot be reconciled by the current data set: data for the sectoral level are current to 1992; data for international flows are only up to 1985. More data for international flows will be needed to reconcile the difference. Overall, the absence of a universal pattern of telecommunication flows between countries suggests that the 179 question of causality depends on both the calling and the called parties--telephones are after all an interactive medium. This raises the question of whether there will be a mirror image in causality should be study be replicated in the countries studied. That is, if International Telephone traffic growth causes Imports growth from West Germany, would a study from West Germany find that Exports cause International Telephone traffic growth to Singapore? That is one of many possible areas of research from the suggestive, if tentative, findings in this chapter. CHAPTER EIGHT Concineion Tests of Hypotheses The study began with several hypotheses (Chapter 3) the results of which are elaborated upon more fully below. H1. International telecommunication correlates strongest with trade, tourism and financial services among all the sectors of the Singapore economy. Result: Partly rejected and partly failed to reject. If one were to use the conventional method of correlating economic sectors with International Telecommunication, one would find a high correlation indeed. Significance is better than .001 in all cases except the Construction sector, which is significant at .003. Table 8.1 below shows that the three sectors of trade, tourism and financial services ranked high in their correlations with International Telecommunication. If one takes GDP as the aggregate average of correlation, these three sectors are "above average." Their low and insignificant Durbin-Watson statistics, however, indicate that these relationships have extremely high autocorrelations. This means the R2 is overestimated. Table 8.1 therefore is, in all probability, a table of spurious results. 180 181 Table 8.1.--Correlation of International Telecommunication by call volume and various economic sectors. Sectors qu-adj** DW Visitors (Total) .920 1.00 Commerce .903 .35 Manufacturing .887 .42 Utility .882 .10 Agriculture .878 .47 Finance .872 .04 GDP .868 .07 Transport .855 .02 Other Services .819 .08 Construction .108* .06 Quarry .047 .11 *p<0.003. **p<0.001 level unless otherwise indicated. The data therefore had to be transformed to minimize autocorrelation. The transformations necessary for a statistically valid conclusion were seasonal adjustments, log-transformation, and differencing. These steps . transformed the data to percentage growth rates. Correlations of these growth rates are presented in Table 8.2. Transport and GDP had statistically significant correlations. Finance, Manufacturing and Commerce (the trade sector) followed in order but were not statistically significant. Tourism, as measured by the total number of visitors, dropped to the middle of the ranking of sectors. 182 Table 8.2--Relationship of growth rates of International Telecommunication and various economic sectors (after seasonal adjustments). Sectors qu qu-adj p< DW Transport .133 .120 .002 2.27 GDP .082 .069 .018 1.81 Finance .036 .021 .123 1.79 Manufacturing .035 .020 .128 1.74 Commerce .025 .010 .198 1.93 Utility .021 .006 .241 1.96 Total Visitors .007 .001 .289 2.57 Construction .004 .000 .588 1.58 Agriculture .003 .000 .684 2.07 Quarry .002 .000 .741 1.60 Other Services .000 .000 .864 1.60 A surprise inclusion among these sectors is Manufacturing, which ranks ahead of Commerce. Manufacturing is not a sector normally considered to use much of telecommunications, let alone international telecommunications. After all, as in the developed countries, the occupational structure in Singapore is moving away from manufacturing (Kuo and Chen 1984). The high correlation between International Telecommunication and Manufacturing may be attributable to the unique context of Singapore where there are many international manufacturing operations. If this is so, it would imply that there may be contextual factors to be considered in any attempt to develop a "grand" or "unifying" theory of telecommunications and economic development. 183 H2. International telecommunication causes trade development more so than vice versa in Singapore. Result: Rejected. The analysis showed that Trade growth, at an aggregate level and at constant prices, caused International Telecommunication traffic growth. This finding was unexepected as it the Trade sector is telecommunications intensive. Telecommunication costs account for five to 10 percent of total costs in the Trade sector (Lim 1993). The above finding must be qualified in that at the subsectoral level, International Telecommunication was shown to cause Re-Export growth. H3. International telecommunications causes tourism development more so than vice versa in Singapore. Result: partly rejected. No significant causal relationship was found between tourism--whether by air, land or sea--and International Telecommunication at the aggregate level. The analysis showed that the relationship of International TelecommuniCation is stronger with the flow of goods than people. 184 At the country level, there were some causal relationships found. But the results were decidedly mixed. Sometimes International Telecommunication caused Tourism growth; at other times, it was vice versa. It is suggested that perhaps there are contextual factors that explain the relationship at the country level. H4. International telecommunication causes financial services development more so than vice versa in Singapore. Result: both null and alternate hypotheses rejected. The results of the analysis at the aggregate level were not statistically significant one way or the other. The bias is in favor of International Telecommunication growth causing Finance growth. Further tests, at the subsectoral level, were not possible because the data were absent. Financial institutions are heavy users of telecommunications, with five to 10 percent of total costs going to telecommunications (Lim 1993). Foreign exchange dealing rooms display conspicuous batteries of telephones. So the absence of a significant causal relationship is puzzling. Three reasons have been suggested for the lack of a significant result. The first is that the Finance industry is a heavy user of private circuits. Therefore, not all the 185 data for the Finance industry are captured by measurement only of the public switched network. The second reason is that the relationship may be one of "instantaneous causality." The idea is that the impact of the telephone call is felt right at that moment in the Finance industry. The test cannot account for instantaneous causality. The third reason is that the definition of Finance industry, at least in Singapore, includes such businesses as detective and modeling agencies. The detectives may be used as debt collectors; but the models are baffling. H5. International telecommunication causes more growth in trade development than tourism development or financial services development in Singapore. Result: rejected. H6. International telecommunication causes more growth in tourism development than trade development or financial services development in Singapore. Result: rejected. H7. International telecommunication causes more growth in financial services development than trade development or tourism development in Singapore. 186 Result: rejected. Based on previous research, which showed that investment in telecommunications causes economic growth, it had been assumed that International Telecommunication would cause economic development. As the above analysis shows, however, the various economic sectors were more frequently the causal factor. The results do not stand alone; there is some internal validity among them. The three sectors in order of importance are Trade, tourism and Finance. To recapitulate, this study has looked at the causal link between the flow of International Telecommunications and economic growth at three levels--at the national and sectoral levels, specifically trade, and at the level of international flows. At the national aggregated level, it has been shown that the economic activity causes international telecommunications growth. As the measure of economic activity is a measure of wealth, this may be termed the "wealth effect." To paraphrase Hemingway, the rich are different because they make more phone calls. On hindsight, there was some evidence of the causality of economic factors in at least one paper written earlier. Staple and Mullins had written: "The rate of increase in the demand for international services nevertheless generally 187 fluctuates according to the business cycle in the country where the telecommunication originates" (1989a:25; italics added for emphasis). The sentence implies that the demand in telephone traffic is caused by the business cycle. This is an implication contrary to previous findings. The only sector whose growth was shown to be caused be International Telecommunication was Transport and Communication. This is the very sector into which telecommunications falls. It is suggested that the existence of the relationship may be attributable to double counting. If so, it could explain previous studies that found telecommunications investment to cause economic growth. In those instances, investment in telecommunications infrastructure was used to predict GDP growth. Those are cases of double counting because telecommunications investment is part of GDP and so telecommunications investment conceptually will contribute to GDP. At the sub-sectoral level of trade, the wealth effect still holds. Imports and Overall Trade caused international telecommunication. There is therefore some internal validity in the results. An interesting and unexpected finding was that international telecommunication caused Re-Exports. This activity is the entrepot trade, where goods are broken up into smaller consignments for shipment to other countries in the region. It is analogous to hubbing activity in telecommunications where several messages are sent to a 188 central location to be re-sent to other sites. This unexpected finding may be due to Re-Exports needing more coordination than any other type of trading activity. Conceptually, twice as much communication is needed. Finally, at the international bilateral flows level, no consistent pattern of causality emerged save one exception. The exception was that outbound International Telecommunication traffic growth was shown to cause Imports growth from West Germany, Japan and Hong Kong. After the United States of America, West Germany and Japan are second and third in total world trade volume. Hong Kong is the largest trading country in Asia after Japan. The finding suggests that International Telecommunication with the three major trading countries creates wealth. This finding is the reverse of the finding at the sectoral level that Total Imports caused International Telecommunication traffic growth. The "opposite" finding cannot be reconciled by the current data set: data for the sectoral level are current to 1992; data for international flows are only up to 1985. More data for international flows will be needed to reconcile the difference. Analysis using data for the sectoral level truncated to 1985 was not performed because comparability would be limited. The conclusions from analysis of a truncated data set would only be comparable with that of another similarly truncated data set. 189 The lack of a pattern from the international flows data set suggests that the volume of outgoing calls depends not just on the originating country but the destination country as well. On hindsight this is obvious. Telecommunication after all is a two-way traffic. Serendipitous Findings The first major surprise of this study is in its methodological implications: in future time-series analysis of telecommunications-GDP correlations, the Durbin-Watson statistic must be calculated. The Statistic, a measure of autocorrelation, has been extant since 1970 but not widely used. Conclusions from studies that do not use the statistic may be seriously misleading. Another methodological implication is the use of constant prices. The study found more relationships, which also tended to be more stable and consistent, when constant prices were used. It does make some intuitive sense because if X minutes is needed to transact Y ton of goods 10 years ago, there is no reason why more calls would be needed to transact the same volume of goods today. Inflation, however, may increase the value of goods; this means that a deflator will uncover any relationship. At the level of anlysis, there was greater consistency of results when the data were aggregated than when the data were disaggregated to the sectoral and sub-sectoral levels. It is not evident from the results of the analysis why this 190 should be so. If anything, the aggregated data set should contain more random "errors" than the sectoral level data set. The relationships should therefore be more evident at the sectoral and sub-sectoral levels. Comparative analysis with other countries' data is needed to determine if the result is peculiar to Singapore. Turning to the data themselves, it would appear that International Telephone traffic is fairly insensitive to seasonal forces. Seasonal increases and decreases are in the order of 1 to 2 percent. The exceptions are February and March. Typically, February is the month when the Chinese New Year falls and the seasonal decline is about 8 percent. Some Chinese-owned businesses close for as long as two weeks. March would therefore be catchup time for these businesses. Looking next at bilateral data turned up two countries with interesting insights into possible influences of international telecommunications flows: India and the Philippines. Indian international telecommunication ranks poorly by one yardstick of service, the answer/seizure ratio. A call is successfully answered when someone at the called number picks up the telephone. (If no person or answering machine answers, the call is considered not answered.) Seizure refers to a successful handing over to the operator of the international telecommunication service in the destination country. That is, when the call passes the halfway point. In a satellite link, a call is considered seized when the satellite picks up the call and Call Volume (Thousands) 191 Figure 8.1 International Calls to India (Monthly) 1976-85 60 50)— 30— go_ I I I I I APR76 APR7'6 I APRBO I A8882 I APR64 APR77 A8879 APRBI APR83 Years 192 hands it over to the destination carrier. In developed countries, the answer/seizure ratio is 70 percent at best. In less developed countries, the ratio can drop to as low as 15 percent. That is, only 15 percent of the calls that go through end up with someone at the other end answering. India was one of those countries (Lim 1993). Figure 8.1 shows international calls to India from 1976 to 1985. Between April and October 1983, call volume jumped 200 percent and has not looked back since. According to a senior official at Singapore Telecom, India opened an international gateway thereby reducing the bottleneck of calls (Lim 1993). Also, India installed more automatic exchanges and lines between 1983 and 1984 (India 1985). The data therefore suggest that Indian services for international telecommunication grew with greater availability of circuits and access. Together, the data suggest that demand for international telecommunication from Singapore to India was suppressed by poor service on the Indian side of the circuit. This again suggests that there are contextual factors in the destination country that could explain the flows in the originating country. Again, on hindsight, this finding should be obvious as telecommunications is an interactive medium. The other country with interesting data is Philippines, where the data suggest travel was substituted by telecommunication. The data set covered the period of the crisis leading up to the People Power Revolution and the 193 overthrow of Marcos and suggests that there was a substitution effect between telecommunications and travel during the crisis. This is consonant with use of other media in a crisis. (The series, unfortunately, did not cover the actual period of the overthrow of Marcos.) Figure 8.2 shows, among other things, international calls to the Philippines growing in an essentially exponential curve. The data set has been seasonally adjusted so spikes may be considered to be out of the ordinary. The steepest spike is at the end of the series. The next steepest spikes are in 1983, the year in which returned exile Benigno Aquino was assassinated. He was killed on 21 August. The decline in travel to and from Singapore by Philippine nationals had begun in mid 1982. It was suggested to the author that the decline in travel may be due to a levy imposed by the Singapore government to discourage the use of foreign workers (Lim 1993). In April 1982, the Singapore government imposed a tax on employers who hired foreign workers. If, as was suggested, the levy was effective, than arrivals of Philippine nationals should decrease while departures increase. But as Figure 8.2 shows, that was not the case. Both arrivals and departures followed very similar, almost identical, paths. The decline in travel, accompanied by an increase in international telecommunication, is suggestive of substitution during the Not to scale (linearly transformed) (Thousands) Figure 8.2 45 4O 35 30 25 20 I5 70 194 International Calls to Philippines Substitution Effect l _. Phon _Arrlv l lepartures 88878I88877I88878IA8879I88880I 88881IAP882I88883IA8884I 06T78 0CT77 06r78 0CT79 06780 OCTBI 06T82 06r83 06T84 Years 195 period of crisis. That is, there is a transportation- telecommunication trade-off during the period of crisis. This substitution was demonstrated during the Gulf War when international videoconferencing increased by 60 percent as companies discouraged corporate travel (Quillinan 1991). On a related noted, the senior Singapore Telecom official interviewed said a study he had conducted during Operation Desert Storm in 1991 showed no substitution effect. That is, while travel to Singapore tailed off, there was not a jump in outbound international call volume. The official attributed this to a significant (about 10 percent) of calls being made by tourists in hotels (Lim 1993). While not large, the lack of contribution from that sector was enough to stunt growth. It should also be noted that even if there had been some substitution effect, the increase in videoconferencing would not have been captured in the current data if the videoconference originated from outside Singapore. As noted earlier in Chapter 6, International Telecommunication continued to grow during the Gulf War while tourist arrivals fell. Growth in International Telecommunication during the period, however, was not at the pace projected. that At the time of writing, no data on the period in question was available for analysis. One question Kellerman asks is whether international telecommuniCation is caused by or causes the flow of people or goods. Overall, the analysis of international flows 196 suggests that it is goods, which includes services. Alternative Explanations for Findings As this is a field study, there are influences on the data that cannot be controlled. The first group of influences are those that may have contributed to the exponential growth of International Telecommunication in Singapore. There may be two factors behind the exponential growth. The first is that rates have fallen dramatically. The average rate at the time of writing in early 1993 is two- thirds that in 1979 (Lim 1993). The second reason may be technological. According to one study, the number of facsimile machines doubled between 1988 and 1989 while total outbound traffic increased 42 percent (Staple 1990:12). Staple seems to suggest that fax machines make it easier and cheaper to send larger amounts of information than a voice conversation. The impact of the fax machine will probably be better understood through studies of other countries that have not had these new technologies. A more serious challenge of sorts to the above conclusions is a study completed recently by two Singapore academics. Chow and Choy, in attempting to develop indicators for the Singapore economy, found that International Telephone traffic leads GDP by three quarters (1993). There was no attempt, however, at theorizing the relationship. 197 In common with this study, the data had also to be transformed through statistical filters. (Otherwise, it would be impossible to explain how International Telecommunication continued to grow in the mid-19805 when Singapore had a contraction in its GDP.) It may be that different statistical filters lead to different results. One conceptual problem with their study is that using international telecommunication as a leading indicator does not necessarily indicate it is a cause. The leading indicator concept works much in the same way as the stock market is an indicator of economic growth. One cannot say that the stock market causes economic growth or decline. Further, Chow herself was attempting to develop a model to forecast telecommunications demand using GDP. But if international telecommunication causes economic growth, the attempt is conceptually unsound. StrengthsLWeaknessesII and Limitations The strength of this study is that it has looked at flows, instead of investment, as a measure of the impact of telecommunications. As indicated earlier, it avoids the problem of double-counting. By avoiding double-counting, the study also suggests that some of the conclusions of previous studies may have been due to double-counting. Also, it used a more sophisticated method to determine causality, thereby improving on Hardy's cross-lagged correlation. 198 On the other hand, Granger causality is not without problems. In some instances, which may not apply here, Granger causality does not equate with causality. For example, analysis using Granger causality would suggest that Christmas cards "cause" Christmas. (This critique would apply to cross-lagged correlation.) It is suggested here that because the intention is to look for contributory cause, as opposed to exclusive cause, the limitations of Granger causality are mitigated. Another limitation is that the data are necessarily limited to Singapore and in some cases, limited--period. The data on flows between countries are especially limited--they are more than eight years old. More recent data may change the outcome. Some of the criticisms may be that Singapore is a small island with unique characteristics and not a "typical" Third World nation, as if there is a typical Third World nation. While that may be true, it should be noted that the revenue from international telecommunication accounts for more than 50 percent of the total revenue of Singapore Telecom. The findings on International Telecommunication traffic should therefore be even stronger. If there is to be a caveat on the data, it would be on that of international flows. Singapore's economic growth rate from 1976 grew till it peaked in 1981. Then the rate began a decline from 1981 till 1985, when it hit the recession. How this may have affected the results is not clear but this may be a point worth noting should a 199 comparison study be performed. impact of Studv The major lesson from this study is that previous studies showing telecommunications as a cause of economic growth may have to be reanalyzed. First, there may be statistical problems with some of the earlier studies showing high correlations between international telecommunication and GDP or GNP. As far as this researcher is aware, there was no measure of the autocorrelation of the results. The various studies on the benefits of telecommunications originating from Stanford University in the late 19705 and early 19805 certainly did not calculate the Durbin-Watson statistic. Second, although this is a study only of international telecommunications, there is some internal validity in the findings that suggest application to domestic telecommunications. Earlier studies had found that investment in telecommunications enhanced economic growth. This study suggests that the element of double-counting may be significant, enough to make a difference in the final outcome. Third and tangentially, the study questions whether it is necessary, as conventional wisdom suggests, to lower prices. If it is truly the wealth effect at work, it means that the rich are able and willing to pay for International Telecommunication services. If so, there is no need to lower 200 prices except perhaps to keep pace with competition from other telecommunications service suppliers. More research needs to be done before any such policy can be acted upon. Conclusion This study has extended research in the use of media in development to international telecommunications. Where previous studies had suggested a linear effect from telecommunications, the finding of a wealth effect suggests that the effect is not linear from telecommunications. To that extent, the result is more consistent with current thinking that development is not a linear path. There is no Magic Bullet to development, even merely economic development. So which paradigm does the study support? None of the development paradigms mentioned above fits the bill. The paradigms suggest that communication is key in development; the study, however, shows that development, or economic development to be more precise, causes telecommunications flows. That is, telecommunications is incidental to development. There is some hint, however, that the relationship between telecommunications and development may depend on the stage of development of a country. It was pointed out earlier that a recent study had found that in developing nations, the telecommunications sector precedes national economic growth while the situation was reversed in 201 developed countries (Kim 1992). The Indian case highlights the impact of a stifled infrastructure--demand is suppressed. When the infrastructure is available, international telecommunications explodes in 200 percent growth in one year. One can only guess at the amount of economic and social well-being that 200 percent growth unleashed. Investment to create an efficient telecommunications system is therefore important. On the other hand, after this demand has been met, provided there are no chokages in the system, the telecommunications system can take care of itself. Whether these findings are restricted to the Singapore case, or whether there is indeed some kind of threshold that Singapore crossed will have to be determined by further research. In sum, the study has not uncovered a grand theory of development. It certainly does not support any one paradigm of development. But it does suggest that there may be differences in the impact of telecommunications depending on the stage of development of a country. RECOMMENDATIONS CHAPTER NLINE Recommendations Beeommendations for Policy The major finding of this study is that, contrary to previous research and to conventional wisdom, International Telecommunication growth is not the cause of but the result of economic activity. Much of the policies to improve telecommunications and increase investment have assumed the conventional wisdom. Clearly, there is evidence that a stifled infrastructure can hold back telecommunications as in the case of India. But there is no conclusive evidence that the Indian economy suffered solely because of it. If International Telecommunication traffic does not cause economic growth, then the major implication is that lower rates will have no great impact on an economy. It may be possible to run a sound international telecommunications system simply by keeping pace with demand. That is, it may be enough to keep rates reasonable compared with that of other telecommunication operators the region, to ensure an internationally competitive answer/seizure ratio, and to keep pace with user requirements. In countries where the telecommunication operator is a bureaucratic organization, these "requirements" may be revolutionary in themselves. But it also means that 202 203 countries such as Singapore, which is privatizing and liberalizing its telecommunication system, need not go too far in, for example, introducing competition. If economic activity causes telecommunications growth, there is no need to lower prices. The wealthy are able and willing to pay. This policy recommendation is not unreasonable: the US has a running deficit of a few billion dollars in telecommunications accounting because of its lower prices. Developing countries may not be able to withstand the loss. There is supporting evidence for the above policy. The senior Singapore Telecom official mentioned earlier said that an in-house study of Japan and the United Kingdom showed that lower international call prices did not increase volume over and above that projected. He said that what was noticeable was a shift in origination of the telephone call to the country with the lower rate (Lim 1993). Recommendations for Future Research Future research could be conducted across countries or in more detail in Singapore. Research across countries will be useful in determining the validity and strength of the relationships uncovered. More developed countries should be compared with the less developed. Kim (1992) suggested that there may be a difference between rich and poor countries in the 204 suggested that the benefits from telecommunications to the richer countries reach a plateau, as it were, where the benefits are harder to derive. In contrast, the poorer countries appear to benefit more. This explanation would fit the data of India, which suggests that poor infrastructure may indeed have suppressed telecommunications demand. Another area is the use of telecommunications during a crisis. The Philippines case is a single instance but it is strongly suggestive that some substitution of telecommunications for travel was present. It also suggests that the use of telecommunications is similar to the use of other traditional media in crisis. If so, this could add to mass communication theory. On a technical note, more sophisticated econometric methods should be used to analyze the available monthly data. The current method could not tame the severe autocorrelation even after several months of attempts. As a result, all data had to be aggregated to a quarterly basis. The analytical methods needed to obtain meaningful results are beyond the scope of this dissertation. APPENDICES APPENDIX A Appendix A: Definitions of Sectors Intnoduction Definitions of the sectors used in the research were obtained from the Singapore Standard Industrial Classification 1990 edition, a publication of the Department of Statistics, a government body under the Ministry of Trade and Industry. The Department maintains a database of statistics from which the data used in the study were obtained. For the purposes of national accounting, the Singapore economy is divided into nine sectors with one sector for "Activities Not Adequately Defined." Data was not available for this sector. The sectors fall neatly into classic primary, secondary and tertiary categories. Where data are in constant dollars, the base year is pegged at 1985. Following are more detailed definitions and descriptions of the types of economic activities covered under each of the nine sectors. Establishments are classified according to their predominant activity. Agrieulture, FishingI Fonestry and Hunting--Labeled "Agriculture" in the research. This is typically considered a primary sector. It covers traditional agricultural and farming activities, livestock production, fish and aquatic animal (such as crocodiles and turtles) farms, and offshore and deep sea fishing. 205 206 Mining and Quarrying--Labeled "Quarrying" in the research. This is typically considered a primary sector. It is essentially the quarrying of granite and extraction of sand, clay and gravel. Manugacturing This is typically considered a secondary sector. This is the industrial sector. It covers the gamut from light industries such as the canning of food and manufacturing of beverages, textiles and apparels, to heavy industries such as chemicals (including petroleum) and the repair of ships, railroad equipment, aircraft and other transport equipment. It also includes the manufacturing of glass and metal products, and electrical and electronic equipment. Printing and publishing are also in this sector. Electricity, Gas and Water--Labeled "Utilities" in the research. This is typically considered a tertiary sector. This is a self-explanatory title. Singapore has agreements with Malaysia to purchase and sell electricity, natural gas and water. Consnruction This is typically considered a secondary sector. This is the building sector and ranges from site preparation such as soil investigation to construction of the building. It includes finishing works such as tiling, plumbing, and installation of fire alarms, awning and solar control films. 207 Real estate management, and architectural and related services such as landscaping and interior decoration are not covered by this sector but by Business Services. Commerce This is typically considered a tertiary sector. This is essentially the trade and tourism sector. It encompasses wholesale trade at the import and export level as well as retail trade such supermarkets and pharmacies. The tourism part of the sector covers hotels, restaurants and food outlets. Related to these, ship Chandlers and air transport suppliers are included too. Transport, Storage and Communication--Labeled "Transport" in the research. This is typically considered a tertiary sector. As the title indicates, there are three broad sub- headings. The first covers air, land and sea transportation. The second covers warehousing. The third covers postal and telecommunications. Financial, Insurance, Real Estate and Business Services-- Labeled "Financial Business" in the research. This is typically considered a tertiary sector. As the title suggests, there are four sub-sectors in this head. The first, financial, covers banks, stock and money broking houses. The second covers both general and life insurance. 208 The third covers real estate management, architectural and technical services but not construction. The final sub-head encompasses a wide range of services. It is intended to cover business services not Classified elsewhere. A listing will indicate the range: legal services, accounting, information technology services, engineering, advertising, management and consulting agencies, equipment rental, news agencies, detective agencies, interior decoration, auctioneering, modeling and debt collection. Community, Social and Other services--Labeled "Other Services" in the research. This is typically considered a tertiary sector. There are four broad areas under this head. The first consists of government and government-related agencies. The executive, legislative and judicial arm of government comes under this head. Salaries are taken as indicators of value- added. Education and medical and health services are also under this head. Under social services are recreational and cultural services such as cinemas, radio and television broadcasting, libraries, zoos, museums and amusement parks. The third consists of personal services run the gamut from hairdressing and laundry, to the repair and servicing of vehicles and consumer goods to escort services, massage parlors and undertaking services. 209 The final head are international and other extra- territorial bodies. They cover foreign embassies and trade bodies, international organizations and foreign armed forces bases. Trade--This combines two components: Imports and Exports. Imports--There is no separate subsector for Imports. Exports--There are two subsectors: Re-Exports and Domestic Exports. 3e-Exports--This is the entrepot sector where goods are imported but not processed, apart from breaking them into small packages and repackaging, before being exported. Domestic Exports—-This subsector covers goods that are processed and exported. Hence refined oil is considered a domestic export even though Singapore does not have a single oil well. Visitons--This sector is intended to cover tourists. The statistics are compiled by the Singapore Tourist Promotion Board for the Totai number of Visitors, and the number of those who arrive by Air and by Sea. 210 Transit--This covers airline passengers who transit in Singapore. Plane Landings--This captures the number of plane landings. APPENDIX B Appendix B: Differencing The first difference means that the residual is subtracted from each other such that the value in an earlier period is subtracted from the latter period. Below is an example of a series that has been differenced once: C1 C2 1 * 3 2 8 --> 5 12 4 7 -5 Note that the second row of C2 contains 3 minus 1 and the second row of C2 contains 8 minus 3. The purpose of differencing is to remove trends that may be present in the series. If there is a trend, then subtracting one value from another should remove it.) 211 APPENDIX C l—v" 1 5'5 o Fren §ingapoze no 15 Counpnies Bold text indicates statistically significant results. Table 10.1.--Indonesia Unrestr Restr F Value Air Arrival Causes Phone Phone Causes Air Arrival Air Arrival! Causes Phone! Phone! Causes Air Arrival! Air Depart Causes Phone Phone Causes Air Depart Air Depart! Causes Phone! Phone! Causes Air Depart! Critical Values F4,22 (p<. 10;p<.05) R-Sq R-Sq 2.219;2.817; .265 .230 .679 .667 .267 .201 .043 .034 .269 .230 .721 .725 .299 .201 .069 .055 0.2619 0.2056 0.4952 0.0517 0.2934 0.0800 0.7689 0.0827 212 213 Table 10.2.--Australia Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .152 .130 0.1427 Phone Causes Domestic Export .132 .082 0.3168 Domestic Export! Causes Phone! .191 .139 0.3535 Phone! Causes Domestic Export! .324 .226 0.7973 Export Causes Phone .196 .130 0.4515 Phone Causes Export .121 .071 0.3129 Export! Causes Phone! .272 .139 1.0048 Phone! Causes Export! .265 .158 0.8007 Import Causes Phone .150 .130 0.1294 Phone Causes Import .315 .205 0.8832 Import! Causes Phone! .229 .139 0.6420 Phone! Causes Import! .343 .242 0.8455 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .224 .148 0.3673 Phone Causes Visitors (Air) .627 .460 1.6790 Visitors (Air)! Causes Phone! .292 .210 0.4343 Phone! Causes Visitors (Air)! .256 .045 1.0635 Visitors (Land) Causes Phone .402 .148 1.5928 Phone Causes Visitors (Land) .859 .833 0.6915 Visitors (Land)! Causes Phone! .318 .210 0.5938 Phone! Causes Visitors (Land)! .209 .083 0.5973 Visitors (Sea) Causes Phone .440 .148 1.9554 Phone Causes Visitors (Sea) .687 .672 0.1797 Visitors (Sea)! Causes Phone! .382 .210 1.0437 Phone! Causes Visitors (Sea)! .659 .647 0.1320 Visitors (Total) Causes Phone .333 .148 1.0401 Phone Causes Visitors (Total) .365 .092 0.6389 Visitors (Total)! Causes Phone! .325 .210 0.6389 Phone! Causes Visitors (Total)! .234 .034 0.9791 214 Table 10.3.--Canada Unrestr Restr F R-Sq R—Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .177 .056 0.8086 Phone Causes Domestic Export .489 .370 1.2808 Domestic Export! Causes Phone! .267 .128 1.0430 Phone! Causes Domestic Export! .207 .420 2.0198 Export Causes Phone .240 .056 1.3316 Phone Causes Export .530 .414 1.3574 Export! Causes Phone! .337 .128 1.7338 Phone! Causes Export! .572 .543 0.3727 Import Causes Phone .190 .056 0.9099 Phone Causes Import .493 .381 1.2150 Import! Causes Phone! .228 .128 0.7124 Phone! Causes Import! .507 .392 1.2830 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .380 .278 0.6169 Phone Causes Visitors (Air) .736 .696 0.5682 Visitors (Air)! Causes Phone! .443 .384 0.3972 Phone! Causes Visitors (Air)! .241 .181 0.2964 Visitors (Land) Causes Phone .307 .278 0.1569 Phone Causes Visitors (Land) .869 .812 1.6317 Visitors (Land)! Causes Phone! .416 .384 0.2055 Phone! Causes Visitors (Land)! .570 .500 0.6105 Visitors (Sea) Causes Phone .382 .278 0.6311 Phone Causes Visitors (Sea) .514 .508 0.0463 Visitors (Sea)! Causes Phone! .443 .384 0.3972 Phone! Causes Visitors (Sea)! .314 .289 0.1367 Visitors (Total) Causes Phone .397 .278 0.7400 Phone Causes Visitors (Total) .294 .213 0.4302 Visitors (Total)! Causes Phone! .483 .384 0.7181 Phone! Causes Visitors (Total)! .303 .230 0.3928 215 Table 10.4.--France Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .751 .679 1.5904 Phone Causes Domestic Export .476 .240 2.4771 Domestic Export! Causes Phone! .420 .280 1.3276 Phone! Causes Domestic Export! .367 .242 1.0861 Export Causes Phone .679 5.890 (.0022) Phone Causes Export .397 .323 0.6750 Export! Causes Phone! .280 3.752 (.0179) Phone! Causes Export! .303 .177 0.9943 Import Causes Phone .741 .679 1.3166 Phone Causes Import .489 .365 1.3346 Import! Causes Phone! .384 .280 0.9286 Phone! Causes Import! .441 .401 0.3936 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .859 .763 2.5532 Phone Causes Visitors (Air) .927 .89 1.9007 Visitors (Air)! Causes Phone! .454 .398 0.3846 Phone! Causes Visitors (Air)! .290 .146 0.7606 Visitors (Land) Causes Phone .891 .763 4.4037 Phone Causes Visitors (Land) .967 .958 1.0227 Visitors (Land)! Causes Phone! .410 .398 0.0763 Phone! Causes Visitors (Land)! .307 .157 0.8117 Visitors (Sea) Causes Phone .792 .763 0.5228 Phone Causes Visitors (Sea) .410 .250 1.0169 Visitors (Sea)! Causes Phone! .456 .398 0.3998 Phone! Causes Visitors (Sea)! .317 .237 0.4392 Visitors (Total) Causes Phone .795 .763 0.5854 Phone Causes Visitors (Total) .258 .101 0.7935 Visitors (Total)! Causes Phone! .501 .398 0.7740 Phone! Causes Visitors (Total)! .189 .008 0.5040 216 Table 10.5.--Hong Kong Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Air Arrival Causes Phone .365 .201 1.4205 Phone Causes Air Arrival .723 .700 0.4567 Air Arrival! Causes Phone! .378 .114 2.3344 Phone! Causes Air Arrival! .233 .060 1.2405 Air Departure Causes Phone .453 .201 2.5338 Phone Causes Air Departure .392 .282 0.9951 Air Departure! Causes Phone! .407 .114 2.7175 Phone! Causes Air Departure! .137 .258 0.8969 Domestic Export Causes Phone .316 .201 0.9247 Phone Causes Domestic Export .389 .223 1.4943 Domestic Export! Causes Phone! .223 .114 0.7716 Phone! Causes Domestic Export! .350 .286 0.5415 Export Causes Phone .350 .201 1.2608 Phone Causes Export .455 .279 1.7761 Export! Causes Phone! .306 .114 1.5216 Phone! Causes Export! .257 .162 0.7032 Import Causes Phone .476 .201 2.8865 Phone Causes Import .643 .401 3.7283 Import! Causes Phone! .529 .114 4.8461 Phone! Causes Import! .404 .152 2.3255 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .522 .438 0.6590 Phone Causes Visitors (Air) .959 .940 1.7378 Visitors (Air)! Causes Phone! .579 .339 2.1378 Phone! Causes Visitors (Air)! .427 .204 1.4594 Visitors (Land) Causes Phone .502 .438 0.4819 Phone Causes Visitors (Land) .905 .877 1.1053 Visitors (Land)! Causes Phone! .420 .339 0.5237 Phone! Causes Visitors (Land)! .393 .241 0.9390 217 Table 10.5.--cont'd Visitors (Sea) Causes Phone Phone Causes Visitors (Sea) Visitors (Sea)! Causes Phone! Phone! Causes Visitors (Sea)! Visitors (Total) Causes Phone Phone Causes Visitors (Total) Visitors (Total)! Causes Phone! Phone! Causes Visitors (Total)! .535 .639 .498 .694 .553 .391 .547 .476 .438 .300 .339 .485 .438 .135 .339 .217 0.7823 3.5215 1.1877 2.5613 0.9648 1.5764 1.7219 1.8535 218 Table 10.6.--India Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Air Arrival Causes Phone .163 .098 0.4271 Phone Causes Air Arrival .412 .353 0.5519 Air Arrival! Causes Phone! .285 .069 1.6615 Phone! Causes Air Arrival! .165 .073 0.6060 Air Departure Causes Phone .239 .098 1.0191 Phone Causes Air Departure .725 .707 0.3600 Air Departure! Causes Phone! .256 .069 1.3824 Phone! Causes Air Departure! .212 .115 0.6770 Domestic Export Causes Phone .179 .098 0.5426 Phone Causes Domestic Export .226 .205 0.1492 Domestic Export! Causes Phone! .153 .069 0.5383 Phone! Causes Domestic Export! .279 .235 0.3356 Export Causes Phone .198 .098 0.6858 Phone Causes Export .183 .146 0.2491 Export! Causes Phone! .161 .069 0.6031 Phone! Causes Export! .292 .248 0.3418 Import Causes Phone .448 .098 3.4873 Phone Causes Import .354 .221 1.1324 Import! Causes Phone! .428 .069 3.4519 Phone! Causes Import! .506 .315 2.1265 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .234 .091 0.7001 Phone Causes Visitors (Air) .725 .692 0.4500 Visitors (Air)! Causes Phone! .312 .062 1.3626 Phone! Causes Visitors (Air)! .242 .178 0.3166 Visitors (Land) Causes Phone .103 .091 0.0502 Phone Causes Visitors (Land) .525 .345 1.3144 Visitors (Land)! Causes Phone! .209 .062 0.6969 Phone! Causes Visitors (Land)! .464 .197 1.8680 219 Table 10.6.--cont'd Visitors (Sea) Causes Phone Phone Causes Visitors (Sea) Visitors (Sea)! Causes Phone! Phone! Causes Visitors (Sea)! Visitors (Total) Causes Phone Phone Causes Visitors (Total) Visitors (Total)! Causes Phone! Phone! Causes Visitors (Total)! .240 .321 .166 .313 .304 .248 .344 .237 .091 .213 .062 .146 .091 .211 .062 .216 0.7352 0.5965 0.4676 0.9166 1.1476 0.1845 1.6120 0.1032 220 Table 10.7.--Italy Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .632 .297 5.0068 Phone Causes Domestic Export .353 .297 0.4760 Domestic Export! Causes Phone! .394 .078 2.8680 Phone! Causes Domestic Export! .271 .172 0.7469 Export Causes Phone .580 .297 3.7060 Phone Causes Export .314 .261 0.4249 Export! Causes Phone! .359 .078 2.4111 Phone! Causes Export! .344 .209 1.1319 Import Causes Phone .407 .297 1.0202 Phone Causes Import .508 .371 1.5315 Import! Causes Phone! .166 .078 0.5803 Phone! Causes Import! .585 .432 2.0277 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .530 .337 1.5399 Phone Causes Visitors (Air) .934 .909 1.4205 Visitors (Air)! Causes Phone! .315 .095 1.2044 Phone! Causes Visitors (Air)! .425 .250 1.1413 Visitors (Land) Causes Phone .463 .337 0.8799 Phone Causes Visitors (Land) .913 .886 1.1638 Visitors (Land)! Causes Phone! .325 .095 1.2778 Phone! Causes Visitors (Land)! .477 .441 0.2581 Visitors (Sea) Causes Phone .448 .337 0.7541 Phone Causes Visitors (Sea) .563 .110 3.8873 Visitors (Sea)! Causes Phone! .555 .095 3.8764 Phone! Causes Visitors (Sea)! .504 .068 3.2964 Visitors (Total) Causes Phone .485 .337 1.0777 Phone Causes Visitors (Total) .399 .311 0.5491 Visitors (Total)! Causes Phone! .335 .095 1.3534 Phone! Causes Visitors (Total)! .275 .051 1.1586 221 Table 10.8.--Japan Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Air Arrival Causes Phone .280 .153 0.9701 Phone Causes Air Arrival .725 .656 1.3800 Air Arrival! Causes Phone! .433 .141 2.8325 Phone! Causes Air Arrival! .474 .321 1.5998 Air Departure Causes Phone .311 .153 1.2612 Phone Causes Air Departure .710 .666 0.8345 Air Departure! Causes Phone! .339 .141 1.6475 Phone! Causes Air Departure! .326 .150 1.4362 Domestic Export Causes Phone .198 .153 0.3086 Phone Causes Domestic Export .275 .151 0.9407 Domestic Export! Causes Phone! .246 .141 0.7659 Phone! Causes Domestic Export! .353 .257 0.7966 Export Causes Phone .195 .153 0.2870 Phone Causes Export .265 .145 0.8980 Export! Causes Phone! .247 .141 0.7742 Phone! Causes Export! .319 .228 0.7349 Import Causes Phone .264 .153 0.8295 Phone Causes Import .446 .109 3.3457 Import! Causes Phone! .334 .141 1.5938 Phone! Causes Import! .433 .096 3.2690 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .117. .175 -0.2463 Phone Causes Visitors (Air) .956 .935 1.7898 Visitors (Air)! Causes Phone! .213 .136 0.3669 Phone! Causes Visitors (Air)! .515 .255 2.0103 Visitors (Land) Causes Phone .209 .175 0.1612 Phone Causes Visitors (Land) .848 .820 0.6908 Visitors (Land)! Causes Phone! .182 .136 0.2109 Phone! Causes Visitors (Land)! .511 .433 0.5982 222 Table 10.8.--cont'd Visitors (Sea) Causes Phone .179 .175 0.0183 Phone Causes Visitors (Sea) .817 .748 1.4139 Visitors (Sea)! Causes Phone! .170 .136 0.1536 Phone! Causes Visitors (Sea)! .538 .457 0.6575 Visitors (Total) Causes Phone .149 .175 -0.1146 Phone Causes Visitors (Total) .505 .213 2.2121 Visitors (Total)! Causes Phone! .175 .136 0.1773 Phone! Causes Visitors (Total)! .564 .343 1.9008 223 Table 10.9.--Netherlands Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Air Arrival Causes Phone .599 .238 4.9514 Phone Causes Air Arrival .885 .866 0.9087 Air Arrival! Causes Phone! .109 .125 -0.0988* Phone! Causes Air Arrival! .202 .154 0.3308 Air Departure Causes Phone .598 .238 4.9254 Phone Causes Air Departure .842 .815 0.9399 Air Departure! Causes Phone! .166 .125 0.2704 Phone! Causes Air Departure! .242 .147 0.6893“ Domestic Export Causes Phone .249 .238 0.0806 Phone Causes Domestic Export .352 .237 0.9761 Domestic Export! Causes Phone! .159 .125 0.2224 Phone! Causes Domestic Export! .406 .389 0.1574 Export Causes Phone .308 .238 0.5564 Phone Causes Export .366 .145 1.9172 Export! Causes Phone! .204 .125 0.5459 Phone! Causes Export! .373 .211 1.4211 Import Causes Phone .282 .238 0.3370 Phone Causes Import .428 .247 1.7404 Import! Causes Phone! .156 .125 0.2020 Phone! Causes Import! .446 .340 1.0523 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .632 .175 4.6569 Phone Causes Visitors (Air) .921 .906 0.7120 Visitors (Air)! Causes Phone! .313 .097 1.1790 Phone! Causes Visitors (Air)! .183 .114 0.3167 Visitors (Land) Causes Phone .500 .175 2.4375 Phone Causes Visitors (Land) .901 .861 1.5152 Visitors (Land)! Causes Phone! .314 .097 1.1862 Phone! Causes Visitors (Land)! .387 .286 0.6179 224 Table 10.9.--cont'd Visitors (Sea) Causes Phone .222 .175 0.2265 Phone Causes Visitors (Sea) .304 .264 0.2155 Visitors (Sea)! Causes Phone! .341 .097 1.3885 Phone! Causes Visitors (Sea)! .463 .384 0.5517 Visitors (Total) Causes Phone .183 .175 0.0367 Phone Causes Visitors (Total) .201 .162 0.1830 Visitors (Total)! Causes Phone! .313 .097 1.1790 Phone! Causes Visitors (Total)! .233 .151 0.4009 * The F-value should not be negative. The negative sign is an indication of problems in the model. In this case, a positive F-value was obtained when the fourth lag was dropped but the result was in the order of smaller than 0.2, a statistically insignificant result. 225 Table 10.10.--Philippines Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Air Arrival Causes Phone .427 .185 2.3229 Phone Causes Air Arrival .626 .692 1.2222 Air Arrival! Causes Phone! .268 .069 1.4952 Phone! Causes Air Arrival! .312 .194 0.9433 Air Depart Causes Phone .284 .185 0.7605 Phone Causes Air Depart .842 .837 0.1741 Air Depart! Causes Phone! .165 .069 0.6323 Phone! Causes Air Depart! .080 .057 0.1375 Import Causes Phone .255 .185 0.5168 Phone Causes Import .412 .549 1.6707 Import! Causes Phone! .147 .069 0.5029 Phone! Causes Import! .513 .487 0.2936 Table 10.11.--8outh Korea Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .226 .121 0.7461 Phone Causes Domestic Export .466 .252 2.2041 Domestic Export! Causes Phone! .354 .269 0.7237 Phone! Causes Domestic Export! .505 .292 2.3667 Export Causes Phone .201 .121 0.5507 Phone Causes Export .379 .181 1.7536 Export! Causes Phone! .302 .269 0.2600 Phone! Causes Export! .181 .432 2.4305 Import Causes Phone .401 .121 2.5710 Phone Causes Import .473 .357 1.2106 Import! Causes Phone! .424 .269 1.4800 Phone! Causes Import! .363 .246 1.0102 226 Table 10.12.--Taiwan ' Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .334 .233 0.8341 Phone Causes Domestic Export .472 .360 1.1667 Domestic Export! Causes Phone! .387 .288 0.8883 Phone! Causes Domestic Export! .395 .356 0.3545 Export Causes Phone .273 .233 0.3026 Phone Causes Export .432 .291 1.3653 Export! Causes Phone! .389 .288 0.9092 Phone! Causes Export! .414 .353 0.5725 Import Causes Phone .338 .233 0.8724 Phone Causes Import .421 .273 1.4059 Import! Causes Phone! .512 .288 2.5246 Phone! Causes Import! .430 .287 1.3798 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .346 .139 1.1869 Phone Causes Visitors (Air) .761 .731 0.4707 Visitors (Air)! Causes Phone! .258 .185 0.3689 Phone! Causes Visitors (Air)! .195 .024 0.7966 Visitors (Land) Causes Phone .543 .139 3.3151 Phone Causes Visitors (Land) .613 .449 1.5891 Visitors (Land)! Causes Phone! .509 .185 2.4745 Phone! Causes Visitors (Land)! .283 .045 1.2448 Visitors (Sea) Causes Phone .333 .139 1.0907 Phone Causes Visitors (Sea) .409 .307 0.6472 Visitors (Sea)! Causes Phone! .397 .185 1.3184 Phone! Causes Visitors (Sea)! .595 .547 0.4444 Visitors (Total) Causes Phone .349 .139 1.2097 Phone Causes Visitors (Total) .247 .137 0.5478 Visitors (Total)! Causes Phone! .267 .185 0.4195 Phone! Causes Visitors (Total)! .240 .061 0.8832 227 Table 10.13.--West Germany Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .386 .259 1.1376 Phone Causes Domestic Export .426 .161 2.5392 Domestic Export! Causes Phone! .278 .252 0.1981 Phone! Causes Domestic Export! .733 .505 4.6966 Export Causes Phone .448 .259 1.8832 Phone Causes Export .493 .109 4.1657 Export! Causes Phone! .261 .252 0.0670 Phone! Causes Export! .670 .249 7.0167 Import Causes Phone .419 .259 1.5146 Phone Causes Import .530 .154 4.4000 Import! Causes Phone! .308 .252 0.4451 Phone! Causes Import! .502 .283 2.4187 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .480 .197 2.0409 Phone Causes Visitors (Air) .691 .680 0.1335 Visitors (Air)! Causes Phone! .383 .094 1.7565 Phone! Causes Visitors (Air)! .260 .184 0.3851 Visitors (Land) Causes Phone .506 .197 2.3456 Phone Causes Visitors (Land) .949 .920 2.1324 Visitors (Land)! Causes Phone! .300 .094 1.1036 Phone! Causes Visitors (Land)! .554 .109 3.7416 Visitors (Sea) Causes Phone .554 .197 3.0017 Phone Causes Visitors (Sea) .671 .645 0.2964 Visitors (Sea)! Causes Phone! .242 .094 0.7322 Phone! Causes Visitors (Sea)! .414 .280 0.8575 Visitors (Total) Causes Phone .425 .197 1.4870 Phone Causes Visitors (Total) .291 .165 0.6664 Visitors (Total)! Causes Phone! .292 .094 1.0487 Phone! Causes Visitors (Total)! .281 .209 0.3755 228 Table 10.14.--United Kingdom Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Air Arrival Causes Phone .353 .085 2.2782 Phone Causes Air Arrival .970 .968 0.3667 Air Arrival! Causes Phone! .306 .071 1.8624 Phone! Causes Air Arrival! .249 .202 0.3442 Air Departure Causes Phone .266 .085 1.3563 Phone Causes Air Departure .872 .859 0.5586 Air Departure! Causes Phone! .173 .071 0.6784 Phone! Causes Air Departure! .184 .061 0.8290 Domestic Export Causes Phone .278 .085 1.4702 Phone Causes Domestic Export .213 .095 0.8247 Domestic Export! Causes Phone! .346 .071 2.3127 Phone! Causes Domestic Export! .157 .083 0.4828 Export Causes Phone .178 .085 0.6223 Phone Causes Export .360 .217 1.2289 Export! Causes Phone! .239 .071 1.2142 Phone! Causes Export! .372 .247 1.0947 Import Causes Phone .189 .085 0.7053 Phone Causes Import .418 .267 1.4270 Import! Causes Phone! .157 .071 0.5611 Phone! Causes Import! .449 .484 0.3731 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .193 .172 0.0976 Phone Causes Visitors (Air) .960 .931 2.7187 Visitors (Air)! Causes Phone! .287 .184 0.5417 Phone! Causes Visitors (Air)! .087 .426 2.2147 Visitors (Land) Causes Phone .425 .172 1.6500 Phone Causes Visitors (Land) .953 .913 3.1915 Visitors (Land)! Causes Phone! .443 .184 1.7437 Phone! Causes Visitors (Land)! .555 .121 3.6573 229 Table 10.14.--cont'd Visitors (Sea) Causes Phone Phone Causes Visitors (Sea) Visitors (Sea)! Causes Phone! Phone! Causes Visitors (Sea)! Visitors (Total) Causes Phone Phone Causes Visitors (Total) Visitors (Total)! Causes Phone! Phone! Causes Visitors (Total)! .475 .530 .396 .775 .254 .492 .280 .418 .172 .267 .184 .473 O 172 .082 .184 .040 2.1643 2.0984 1.3162 5.0333 0.4122 3.0266 0.5000 2.4356 230 Unrestr Restr F R-Sq R-Sq Value Critical Values F4,22 (p<.10;p<.05) 2.219;2.817; Domestic Export Causes Phone .230 .108 0.8714 Phone Causes Domestic Export .289 .229 0.4641 Domestic Export! Causes Phone! .331 .248 0.6824 Phone! Causes Domestic Export! .253 .220 0.2430 Export Causes Phone .236 .108 0.9215 Phone Causes Export .254 .175 0.5824 Export! Causes Phone! .335 .248 0.7195 Phone! Causes Export! .182 .164 0.1210 Import Causes Phone .235 .108 0.9131 Phone Causes Import .558 .384 2.2025 Import! Causes Phone! .300 .248 0.4086 Phone! Causes Import! .530 .392 1.6149 Critical Values F4,15 (p<.10;p<.05) 2.3615;3.0556 Visitors (Air) Causes Phone .547 .387 1.3245 Phone Causes Visitors (Air) .927 .889 1.9521 Visitors (Air)! Causes Phone! .532 .341 1.5304 Phone! Causes Visitors (Air)! .223 .346 0.7053 Visitors (Land) Causes Phone .548 .387 1.3357 Phone Causes Visitors (Land) .443 .299 0.9695 Visitors (Land)! Causes Phone! .614 .341 2.6522 Phone! Causes Visitors (Land)! .437 .304 0.8859 Visitors (Sea) Causes Phone .536 .387 1.2042 Phone Causes Visitors (Sea) .754 .729 0.3811 Visitors (Sea)! Causes Phone! .504 .341 1.2324 Phone! Causes Visitors (Sea)! .539 .326 1.7326 Visitors (Total) Causes Phone .562 .387 1.4983 Phone Causes Visitors (Total) .260 .214 0.2331 Visitors (Total)! Causes Phone! .545 .341 1.6813 Phone! Causes Visitors (Total)! .280 .197 0.4323 LIST OF REFERENCES List of Referencee Agency for International Development. 1977. Socio-economic performance criteria for development. Washington, D.C. February. Amin, Samir. 1990. Maldevelopnent: anatomy of a global failure. Tokyo: United Nations University Press. 1974. Accumulation on a world scale: a critigue of the theprv of underdevelopment. New York: Monthly Review. Anderson, Perry. 1974. Passages from antiguity to feudalism. London: NLB. Apter, David E. 1987. Rethinking development: modernization, dependency, and postmodern politics. Newbury Park, CA: Sage. Aronson, Jonathan, and Peter Cowhey. 1988. When Countries Talk: International Trade in Telecommunications Services. Cambridge, Mass.: Ballinger. Baer, Werner. 1969. The Economics of Prebisch and ECLA. In Latin America: problems in economic development, ed. Charles Nisbet, 203-218. New York: Free Press. Bairoch, P. 1975. The economic development of tne Third World since 1900. Translated by Cynthia Postan. London: Methuen. Banerjee, Anindya, Juan J. Dolado, David F. Hendry and Gregor W. Smith. 1986. Exploring equilibrium relationships in econometrics through static models: some Monte Carlo evidence. Oxford Bulletin of Economics and Statistics 48(3): 253-277. Baran, A. Paul. 1957. The poiitical economy of growt . New York: Modern Reader Paperbacks. Barnett, Tony. 1988. Sociology and development. London: Hutchinson. 231 232 Berger, Peter. 1976. Pvramide of sacrifice. New York: Anchor Books. Berger, Peter, Brigitte Berger and Hansfried Kellner. 1973. The homeless mind: modernization and consciousness. New York: Vintage Books. Bergesen, Albert and Ronald Schoenberg. 1986. Long waves of colonial expansion and contraction: 1415-1969. In Studies of the modern world system, ed. Albert Bergesen, 231-277. New York: Academic Press. Bernstein, Henry, ed. 1973. Underdevelopment & development. England: Penguin. Berry, Jean-Francois. 1981. Comments on the Contribution of Telecommunicatione to Deyelopment: The Case of France and Spain. Marnes-la-Coquette: Association Francaise des Utilisateurs du Telephone et des Telecommunications. Bhagwati, Jagdish N. 1987. Rethinking trade strategy. In Qevelopment strate ies reconsidered, ed. John P. Lewis and Valeriana Kallab, 91-104. New Brunswick, NJ: Transaction Books. Bienefeld, Manfred. 1981. Dependency and the newly industrialising countries (NICs): towards a reappraisal. In Dependency theory: a critical reassessment, ed. Dudley Seers, 79-96. London: Frances Pinter. Black, Cyril E., ed. 1976. Comparative modennizapion: perspectivee on modernization. New York: Free Press. Blanc, G. 1983. Telecommunications For Development: The Impact of Telecommunications On Employnent. Geneve, Switzerland: International Telecommunication Union. Blomstrdm, Magnus and Bjdrn Hettne. 1984. Development theory in transition: the dependency debate and beyond--Third Worid responses. London: Zed. Brenner, Robert. 1976. Agragrian class structure and economic development in pre-industrial Europe. Past and Present. 70:30-75. . 1977. The origins of capitalist development: a critique of neo-Smithian Marxism. new Left Beview. 104:25-92. Casmir, Fred L., ed. 1991. Communication in development. New Jersey: Ablex. 233 Castells, Manuel. 1989. The Informational City. Cambridge, Mass.: Basil Blackwell. Chatfield, Christopher. 1989. The analysis of time-series. New York: Chapman and Hall. Chen, Huey-tysh and Eddie C.Y. Kuo. 1985. Telecommunications and Economic Development in Singapore. Telecommunications Policy (September): 240-244. Chew, Ernest C.T. 1991. The Singapore national identity: its historical evolution and emergence. In A history of Singapore, ed. Ernest Chew and Edwin Lee, 357-368. Singapore: Oxford University Press. Chew, Ernest C.T. and Edwin Lee, ed. 1991. A History of Singapore. Singapore: Oxford University Press. Chirot, Daniel and Thomas D. Hall. 1982. World-system theory. Annual Review of Sociology. 8:81-106. Chow, Hwee Kwan and Choy Keen Meng. 1993. Alternative Approaches to Forecasting the Singapore Economy I. Paper presented at Workshop on Economic Forecasting, Econometric Studies Unit, National University of Singapore, February 10. Contemporary Authors. 1990. 28: 181. Detroit: Gale Research. Daniels, P.W. 1985. §ervice Industries: A Geographic Appraisal. New York: Methuen. Department of Statistics. Public Access Time Series database (PATS). Singapore. . Singapore Monthly Digest pf Statistics. Various years. Desai, A.R. 1976. Need for revaluation of the concept. In Comparative modernization: perspectives on modernization, ed. Cyril E. Black, 89-103. New York: Free Press. Dickey, D.A. and W.A. Fuller. 1981. The Likelihood Ratio Statistics for Autoregressive Time Series with a Unit Root. Journal of the American Statistical Association. 74 (427-31). Dissanayake, Wimal. 1984. A Buddhist approach to development: a Sri Lankan endeavor. In Continuity and change in communication systems: an Asian perspective, ed. Georgette Wang and Wimal Dissanayake, 39-51. Norwood, N.J.: Ablex. 234 Dos Santos, Theotonio. 1973. The crisis of development theory and the problem of dependence in Latin America. In Underdevelopment 9 development, ed. Henry Bernstein, 57-80. England: Penguin. . 1970. The structure of dependence. American Economic Review. 60(22): 231-236. . 1971. The structure of dependence. In Beadinqe in the US imperialism, ed. K.T. Kan and Donald C. Hodges, 225-236 Boston: Extending Horizons. Drewnowski, Jan and Scott Wolf. 1966. The level of living index. Report No. 4. September. Geneva: United Nations Research Institute for Social Development. Duncan K.A., and Ayers J.R., eds. 1988. Teleports end Regional Economic Development. Amsterdam: North- Holland. Dyson, Kenneth H. F. and Peter Humphreys. 1986. The Politics of the Communications Revolution in Western Eunope. London: Frank Cass. Eisenstadt, S.N. 1976. The changing vision of modernization and development. In Communication and chan : the ast ten years--and the next, ed. Wilbur Schramm and Daniel Lerner, 31-44. Honolulu: University Press of Hawaii. Engle, Robert F. and C.W.J. Granger. 1987. Co-integration and error correction. Econometrica. 55(2) March: 251- 76. Estabrooks, Maurice F., and Rodolphe H. Lamarche, eds. 1987. Telecommunications: A Strategic Perspective on Regienal Economic and Business Development. Canada: Canadian Institute for Research on Regional Development. Etzioni, Amitai and Eva Etzioni, ed. 1964. Social change. New York: Basic Books. Fagen, R. 1977. Studying Latin American politics: some implications of a dependencia approach. Latin Amenicen Research Review. Vol XII, no. 2. Fair, Jo Ellen. 1989. 29 years of theory and research and development: the dominant paradigm impact. Gezette. 44:129-150. Feith, Herbert. 1976. South-East Asia and neo-colonialism. In Australia's Northern neighbors: independent or dependent, ed. Edward P. Wolfers, 27-42. Australia: Nelson. 235 Feketekuty, Geza. 1988. International Trade In Servicee: An Overview and Blueprint for Negotiations. Cambridge, Mass.: Ballinger. Flournoy, Don M. 1991. Satellites in development communications. The Journal of Development Communication. (December): 41-48. Foo, Kim Leng, corporate communications officer of Singapore Telecom. Telephone conversation with author, 5 July. Singapore. Frank, Andre Gunder. 1966a. The development of underdevelopment. Monthly Review. 18: 17-31. . 1966b. Capitalism and underdevelopment in Latin America. New York: Monthly Review Press. . 1969. Latin America: underdevelopment or revolution. New York: Monthly Review Press. . 1975. On capitalist underdevelopment. Bombay: Oxford University. . 1984. Critigue and Anti-critigue: essays on dependence and reformism. London: Macmillan. Fukuyama, Francis. 1989. The end of history? The National Interest. Summer: 3-18. Gayle, Dennis John. 1986. The Small Developing §pape. Hampshire, England: Gower Publishing. Geiger, Theodore and Frances Geiger. 1976. The development progress of Hong Kong and Singapore. Hong Kong: Macmillan. Geyl, Pieter, Arnold Toynbee and Pitirim A. Sorokin. 1968. The pattern of the past: can we determine it? New York: Greenwood Press. Goh, Keng Swee. 1977. The ractice o eco omic Singapore: Federal Publications. . 1972. The Economics Of Modernization And Other Essays. Singapore: Asia Pacific Press. Gourevitch, Peter. 1977. The international system and regime formation: a critical review of Anderson and Wallerstein. Comparative Politics. 10 (July): 419-438. Gran, Guy. 1983..Development by people: citizen construction of a just world. New York: Praeger. 236 Granger, C.W.J. 1969. Investigating causal relations by econometric models and cross-spectral methods. Econometrica 37: 424-438. . 1986. Developments in the study of cointegrated economic variables. Oxford Bulletin of Economics and Statistics 48(3): 213-227. Gujarati, Damodar N. 1988. Basic econometrice. Singapore: McGraw-Hill. Hagen, Everett E. 1961. On the theory of social change. Homewood, Ill: The Dorsey Press. Hamby, Alonzo l. 1973. Beyond the new deal: Harry S. Truman and American liperalism. New York: Columbia University Press. Hardy, A. 1980. The Role of the Telephone in Economic Development. Telecommunications Policy (December): 278- 86. Hardy, A. and H. Hudson. 1981. The Role of pne Telepnone in Economic Development: An Empirical Analysis. Washington: Keewatin Communications. Harrison, David. 1988. The sociology of modernization and development. London: Unwin Hyman. Heller, Franics H., ed. 1981. Economics and the Truman administration. Lawrence, KS: Regents Press of Kansas. Hendry, David F. 1986. Econometric modelling with cointegrated variables: an overview. Onford Bulletin of Economics and Statistics. (August): 201-212. Hicks, Norman and Paul Streeten. 1981. Indicators of development: the search for a basic needs yardstick. In Recent issues in world development: a collection of eurvey articlee, ed. Paul Streeten and Richard Jolly, 53-66. Oxford: Pergamon Press. Higgins, Benjamin. 1968. Economic development, pninciples, problems and policies. London: Constable. Hornik, Robert C. 1988. Development communication: information, agriculture, and nutrition in the Third World. New York: Longman. Hoselitz, Bert F., ed. 1952. The progress of uneerdeveleped areas. Chicago: University of Chicago Press. 237 Hudson, Heather E. and Lynn C. York. 1988. The Potential of Telecommunications Investments. Telecommunications Policy September: 272-278. Hudson, Heather E. 1984. When telephones reach the village: the role of telecommunications in rural development. Norwood, NJ: Ablex. Huntington, Samuel. 1976. The change to change: modernization, development, and politics. In Comparative modernization: perspectiyee on modernization, ed. Cyril E. Black, 25-61. New York: Free Press. Hylleberg, Svend. 1986. Seasonality in Regression. Orlando: Academic Press. Inayatullah. 1967. Toward a non-Western model of development. In Communication and change in the developing countries, ed. Daniel Lerner and Wilbur Schramm, 98-102. Honolulu: East-West Center. . 1976. Reconsideration of the Western model. In Communication and change: the last ten yeans - end tne next, ed. Wilbur Schramm and Daniel Lerner, 56-59. Honolulu: East-West Center. India. 1985. New Delhi: Publications Division, Ministry of Information and Broadcasting. Innis, H.A. 1950. Empire and Communications. London: Oxford. INTELSAT, New York University, Economic Development Foundation, and Intergovernmental Bureau for Informatics. 1986. Telecommunications For Development: An International Forum. New York: INTELSAT, New York University, Economic Development Foundation, and Intergovernmental Bureau for Informatics. Intennanionel fieneld Tribune. 1992. Heavy foreign control in Singapore. 25 June, 19. International Monetary Fund. Various years. Direction of Trade Statistics. Washington, D.C. International Telecommunication Union. 1986. Information Telecommunications And Development. Geneve, Switzerland: International Telecommunication Union. . . 1982. General Plan Eor The Development Of The Regional Asian and Oceanian Network. Geneve, Switzerland: International Telecommunication Union. 238 . Various years. Yearbook of Common Carrier Telecommunication Statistics and Telecommunication Statistics. Geneve, Switzerland: International Telecommunication Union. Jayaweera, Neville. 1987a. Rethinking development communication: a holistic view. In Rethinking development communication, ed. N. Jayaweera and S. Amunugam, 76-94. Singapore: Asian Mass Communication Research and Information Center. . 1987b. Introduction. In Rethinking development communication, ed. N. Jayaweera and S. Amunugam, xiii-xix. Singapore: Asian Mass Communication Research and Information Center. Jayaweera, Neville and S. Amunugam, eds. 1987. Betninking development communication. Singapore: Asian Mass Communication Research and Information Center. Jipp, A. 1963. Wealth of nations and telephone density. Telecommunications Journal. July. 199-201. Jurgen, Schmandt, Williams Frederick and Wilson Robert, ed. 1989. Telecommunications Policy and Economic Development. New York: Praeger. Jussawalla, Meheroo and D. Lamberton, ed. 1982. Intepnational Trade Theory and Communications. New York: Pergamon. Jussawalla, Meheroo and C. Cheah. 1987. The Calculus of International Communicatione: A Study In The Political Economy of Transborder Data Flows. Libraries Unlimited Inc. Jussawalla, Meheroo, Dan J. Wedemeyer, and Vijay Menon, ed. 1986. The Passin of Remoteness? nformation ev u 'o In The Asia-Pacific. Singapore: Institute of Southeast Asian Studies. Kahl, Joseph. 1973. Modernization, exploitation and dependency in Latin America. New Brunswick, NJ: Transaction Books. Katz, Elihu and Paul Lazarsfeld. 1955. Personal influence: the paet played by people in the floyeof mass communication. New York: Free Press. 239 Kellerman, Aharon. 1992. The Social Demand for International Telecommunications. In TeleGeography 1992: Global Telecommunications Traffic Statistics and Commentary, ed. Gregory C. Staple, 12-16. International Institute of Communications. . 1990. International Telecommunications Around the World: A Flow Analysis. Telecommunicetions Policy (December): 461-475. Kendall, Maurice. 1976. Time-Series. London: Charles Griffin. Kim, Hansuk. 1992. Causality between the telecommunications sector and the national economy. In Proceedings of the Pacific Telecommunicatigne Council 14th Annual Conference held in Honolulu 12-15 Januar 1992, edited by Mark D. Lofstrom and Dan J. Wedemeyer, 193-198. Honolulu: Pacific Telecommunications Council. Koo, Hagen. 1984. World system, class and state in Third World development: toward an integrative framework of political economy. Sociological Perspectives. 27: 33- 52. Kravis, Irving, Alan Heston, Robert Summers in collaboration with Civitello [and others). 1978. international comparisons of real product and purchasing power. Baltimore: Johns Hopkins University Press. Kuo, Eddie C.Y., Linda Low and Mun-Heng Toh. 1989. The Singapore telecommunications sector and issues affecting its competitive position in the Pacific region. Columbia Journal of World Busineee. (Spring): 59-71. Kuo, Eddie and Tsan Chen. 1985. Telecommunications and economic development in Singapore. Telecommunications Policy 9(3) (September): 240-244. Kuo, Eddie and H. Chen. 1984. Towards an information society: a study of the changing occupational structure in Singapore. Unpublished manuscript. Kuznets, 8.8. 1971. Economic rowth of nations: tota out ut and production structures. Cambridge, Mass: Belknap. Lall, Sanjay. 1975. Is dependence a useful concept in analysing underdevelopment? World Development. Vol 3 No. 11- Lannon, Larry. 1987. Flexibility and Social Values. Telepnony 26 (October): 30-36. 240 Lee, Chin-Chuan. 1979. Media imperialism reconsidered. Beverly Hills: Sage. Lee, Eddy. 1981. Export-led industrialisation in Asia: an overview. In Export-led industrialisation and development, ed. Eddy Lee, 1-22. Geneva: International Labour Office. Lee, Soo Ann. 1973. Industrialisation in Singapore. Melbourne: Longman Press. Lerner, Daniel and Wilbur Schramm, ed. 1967. Communieation and chanqeein the developing countriee. Honolulu: East- West Center. . 1976. Looking forward. In Communication and change: the last ten years - ene the next, ed. Wilbur Schramm and Daniel Lerner, 340-344. Honolulu: East-West Center. Lerner, Daniel. 1958. The Passing of Traditional Society: Modernizing the Middle Eaep. Glencoe, IL: The Free Press. . 1967. International cooperation and development. In Communication and change in the developing countries, ed. Daniel Lerner and Wilbur Schramm, 102-125. Honolulu: East-West Center. . 1976. Toward a new paradigm. In Communication and Chang : the la§t ten years - end tne next, ed. Wilbur Schramm and Daniel Lerner, 60-63. Honolulu: East-West Center. . 1976. Technology, communication and change. In Communication and chan e: the last te the next, ed. Wilbur Schramm and Daniel Lerner, 287- 301. Honolulu: East-West Center. Lesser, B., and L. Osberg. 1983. Telecommunications For Development: The Socio-Economic Development Benefips of Telecommunicationst. Geneve, Switzerland: International Telecommunication Union. Lewis, John P. and Valeriana Kallab, ed. 1987. Developmenp strategies reconsidered. New Brunswick, NJ: Transaction Books. Lewis, John P. 1987. Development promotion: a time for regrouping. In Development strategies reconsidered, ed. John P. Lewis and Valeriana Kallab, 3-33. New Brunswick, NJ: Transaction Books. 241 Leys, Colin. 1977. Underdevelopment and dependency: critical notes. Journal of Contemporary Asia. 7(1): 92-107. Lim, Chong Yah. 1976. South-East Asia: the challenges of economic independence. In Australia's Northern neighbors: independent or dependent, ed. Edward P. Wolfers, 15-26. Australia: Nelson. . 1980. Economic development in Singapore. Singapore: Federal Publications. , ed. 1988. Eolicy options foe the §ingapone economy. Singapore: McGraw-Hill. . 1991. Development and underdevelopment. Singapore: Longman. Lim, Shyong. Vice-President, International Market Development, Singapore Telecom. 1993. Interview by author on April 12. Singapore. Linear, Marcus. 1985. Zapping the Third World. London: Pluto Press. Lippman, Walter. 1955. Essays in the Public Ehilosopny. Boston: Little, Brown. Maitland, Donald. 1984. The Missing Link. Report of the Independent Commission fon World Wide Telecommunications Development. Geneva: ITU. Marx, Karl. 1967. The Communist Manifesto. Harmondsworth: Penguin. . 1974. The Critique of the Gotha Programme. In Kefl Manx: tne Fifet Intefnational end After, ed. D. Fernbach, Harmondsworth: Penguin. . 1976. Capital, Volume l. Harmondsworth: Penguin. Masuda, Yoneji. 1981. The information society as post- industrial society. Washington, D.C.: World Future Society. McClelland, David. 1961. The achieving society. Princeton, N.J.: D. Van Nostrand. McGranahan, D.V., Richard Proust, N.V. Sovani, and M. Subramaniam. 1972. Contents and measurement of socio- economic development. New York: Praeger. McPhail, Thomas L. 1981. Electronic colonialism. Beverly Hills, Calif.: Sage. 242 Meier, Gerald and Dudley Seers. 1984. Pioneers in development. New York: Oxford University Press for the World Bank. Melkote, Srinivas R. 1991. Communication for development in the third world: theory and practice. New Delhi: Sage. Mirza, Hafiz. 1986. Multinationals and the Gromth of The Singapore Economy. London: Croom Helm. Misra, Jay and Byron Belitsos. 1987. Business Telecommunications, Homewood, Ill.: Irwin. Moore, Wilbert Ellis. 1979. World modernization: t e imits of convergence. New York: Elsevier North Holland. Morawetz, David. 1977. Twenty-five years of economic development 1950 to 1975. Washington, D.C.: World Bank. Morris, M.D. 1979. Measuring the conditions of the world's poo . Oxford: Pergamon Press. Moss, Mitchell, ed. 1981. Telecommunications and Productivity, Reading, Mass.: Addison-Wesley Publishing Company, Inc. Myrdah, Gunnar. 1957. Economic theory and underdeveloped regions. London: Gerald Duckworth. . 1970. The challenge of world poverty: e world anti-poverty programme in outline. Middlesex, England: Penguin. Narula, Uma and W. Barnett Pearce. 1986. Development ee Communication: A Perspective on India. Carbondale, Ill: Southern Illinois University Press. Naisbitt, John. 1982. Megatrends: ten new directions transforming our lives. New York: Warner Books. Nisbet, Charles, ed. 1969. Latin America: problems in economic development. New York: Free Press. Nordenstreng, Kaarle and Herbert I. Schiller. 1979. National severeignty and international communication. Norwood: Ablex. Nordenstreng, Kaarle and Tapio Varis. 1974. Television traffic: a one way street? Reports and papers on mass communication 70. Paris: UNESCO. Nordhaus, William and James Tobin. 1972. Economic growth. New York: NBER/Columbia University Press. 243 O'Brien, Rita Cruise. 1979. Mass communications: social mechanisms of incorporation and dependence. In Transnational capitalism and national development: new perspectivee on dependence, ed. José J. Villamil, 129- 143. Sussex: Harvester Press. OECD/DAC. 1977. Socio-economic typologies or Cfiteria and their usefulness in measufing development progress. Paris. Oshima, Harry T. 1976. Old and new strategies--an economist's view. In Commu icat'on nd c a e: the teniyeare - and the next, ed. Wilbur Schramm and Daniel Lerner, 53-56. Honolulu: East-West Center. Pacey, Arnold. 1990. Technology in world civilizetion: a thoueand-year hietory. Oxford: Basil Blackwell. Palma, Gabriel. 1981. Dependency and development: an overview. In Dependency theory: a critical reassessment, ed. Dudley Seers, 20-78. London: Frances Pinter. Pareto, Vilfredo. 1968. The fise and fell of elites. New Jersey: Bedminster Press. Originally published in 1901 in Italian as "Un applicazione di teorie sociologiche" Parker, Edwin B. 1981. Economic and Social Benefits of the Rural Electrification Administration Telephone Lean Program. Stanford: Equatorial Communications. Parker, Edwin E. and Ali Mohammadi. 1977. National development support communication. In Qommunicefiens policy for national development: a comparative ers ective, ed. Majid Teheranian, Farhad Hakimzadeh and Marcello L. Vidale, 167-201. London: Routledge & Kegan Paul. Pelton J.N. 1990. Communications: developing nations faster. Journal of Development Communication. 1(1): 16-19. Petras, James. 1982. Dependency and world-system theory: a critique and new directions. In De endenc a d Marxis , ed. Ronald H. Chilcote, 148-161. Boulder, Colo: Westview Press. Pierce, W. and N. Jequier. 1983. Telecommunications For Development. Geneva: ITU. ' Porat, Marc. 1977. The Information Economy. Washington, D.C.: U.S. Office of Telecommunications. 244 Portes, Alexander and J. Walton. 1981. Labor. class and the international system. New York: Academic Press. Portes, Alexander. 1979. On the sociology of national development: theories and issues. American Journal of Sociology. 82: 55-85. Prebisch, Raul. 1950. The economic development of Latin America and its principal problems. New York: United Nations. . 1963. Towards a dynamic developmenf policy for Latin America. New York: United Nations. Pye, Lucian, ed. 1963. Communications and Political Development. Princeton University Press: Princeton. Pye, Lucian and Mary Pye. 1985. Asian power and politics: the cultural dimensione of authority. Cambridge, Mass.: Harvard University Press. Quillinan, Justin. 1991. Seeing is believing. British Telecom World (UK). March: 38-39. Rogers, Everett M. and F.F. Shoemaker. 1971. Communicatien of innovations: a cross-cultural approach. Beverly Hill: Sage Publications. Rogers, Everett M. 1989. Communication and Development. Chap. in lndia's Information Revolution, Arvind Singhal and Everett M. Rogers, New Delhi: Sage. . 1976a. "Communication and development-- the passing of the dominant paradigm" In Communicetion and development: critical perspectivee, ed. E.M.Rogers, 213-240. Beverly Hill: Sage Publications. . 1976b. "The passing of the dominant paradigm--ref1ections on diffusion research." In Communication and change: the laet ten years - and the next, ed. Wilbur Schramm and Daniel Lerner, 49-52. Honolulu: East-West Center. . 1978. The rise and fall of the dominant paradigm. Journal of Communication. 28(1): 64-69. Rostow, Walt W. 1956. "The takeoff into self-sustained growth." The Economic Joufnal. LXVI (261): 25-48. . 1960. The stages of economic growth: a non- communist manifesto. Cambridge: Cambridge University Press. 245 . 1990. Theorists of economic gnowth from David Hume to the present. New York: Oxford University Press. Saunders, Robert J., Jeremy J. Warford, and Bjorn Wellenius. 1983. Telecommunications and Economic Development. Baltimore: The Johns Hopkins University Press for The World Bank. Sauvant, Karl P. 1986. Trade and Foreign Direct Investment, Boulder, Co1.: Westview Press. Schiller, D. 1982. Telematics and Government. Norwood, N. J.: Ablex Publishing. Schramm, Wilbur and W. Lee Ruggels. 1967. How mass media systems grow. In Communication and change in tne developing countriee, ed. Daniel Lerner and Wilbur Schramm, 57-75. Honolulu: East-West Center. Schramm, Wilbur. 1964. Mass Media and National Development. Stanford, California: University Press. . 1967. Communication and change. In Communication and chan e in the develo in countr'es, ed. Daniel Lerner and Wilbur Schramm, 5-32. Honolulu: East-West Center. Schramm, Wilbur and Daniel Lerner, ed. 1976. Communication and change: the last ten years - and the next. Honolulu: East-West Center. Schumacher, E.F. 1973. Small is beautiful: economics as if people mattered. New York: Harper & Row. Seers, Dudley. 1963. The limitations of the special case. Bulletin of the Oxford Institute of Economics and Statistics. Vol 25. No. 2. . 1969. The meaning of development. International Development Review. 11(4): 2-6. . September 1977. Life expectancy as an integrating concept in social and demographic analysis and planning. Review of Income and Wealth. Vol 23. . 1981. Development options: the strengths and weaknesses of dependency theories in explaining a government's room to manoeuvre. In Dependency theory: a critical reassessment, ed. Dudley Seers, 135-149. London: Frances Pinter. 246 , ed. 1981. Dependency theory: a critical reassessment. London: Frances Pinter. Servaes, Jan. 1991. Toward a new perspective for communication and development. In Communication in development, ed. Fred L. Casmir, 51-85. New Jersey: Ablex. . 1990. Rethinking development communication: one world, multiple cultures. The Journal of Development Communication. I(2) (December): 35-45. Sharp, Eric. 1985. Telecommunications and Economic Development. Telephony. 6 (May): 198-200. Shaw, Donald L. and Maxwell E. McCombs. 1974. The emergence of American political issues: the agenda-setting function of the presen St. Paul, Minn.: West Publishing. Sheehan, Glen and Michael Hopkins. 1978. Basic needs performance: an analysis of some international data. Geneva: ILO. Singhal, Arvind and Everett M. Rogers. 1989. India's Information Revolution. New Delhi: Sage Publications. Singapore Telephone Board. 1966. Annual Report. Singapore Telecom. Various years. Annual Report. Singapore Telecom. 1977-1985. Telecoms Statistical eulletin. Skocpol, Theda. 1977. Wallerstein's world capitalist system: a theoretical and historical critique. American Jounnal of Sociology. 82: 1075-1090. Skousen, Mark. 1991. Economics on trial: lies, myths, and realities. Homewood, IL: Business One Irwin. So, Alvin Y. 1990. Social change and development: modernization, dependency and world-system theofies. Newbury Park: Sage Publications. Somjee, A.H. 1991. Development theony: Cfifigues end explorations. London: Macmillan. Sorokin, Pitirim. 1928. Contemporary Sociological Tpeonies through the first guarter of the twentieth centufy. New York: Harper and Row. 247 Special Secretariat of Informatics of the National Security Council of the Presidency of the Republic of Brazil, United Nations Center on Transnational Corporations. 1983. Transborder Data Flows and Brazil: Brazilian Case Study. New York: United Nations. Spencer, Herbert. 1972. On Social Revolution: Selected Writings. Edited by J.D.Y. Peel. University of Chicago Press: Chicago. Spengler, Oswald. 1928. The decline ef the West: Perspectives of World-History. Authorized translation with notes by Charles Francis Atkinson. Knopf: New York. Staple, Gregory C., ed. 1992. Te e eo h 1 2: l a Telecommunications Traffic Statistics and Commentary. London: International Institute of Communications (IIC). . 1991. Global Teleeommunications Traffic Report. London: International Institute of Communications (IIC). . 1990. The Global Telecommunications Traffic Boom: A Quantitative Brief on Cross-Bonder Markets and Regulation. London: International Institute of Communications (IIC). Staple, Gregory C. and Mark Mullins. 1989a. Global Telecommunications Traffic Flows and Market Structufes: A Quantitative Review. London: International Institute of Communications (IIC). . 1989b. Telecom Traffic Statistics - MiTT Matter. Telecommunications Eolicy. 14(2) June:105-128. Stone, Richard. 1975. Toward a system of social and demographic stafistics. New York: United Nations. Streeten, Paul and Richard Jolly, eds. 1981. Recent issues in wonld development. Oxford: Pergamon Press. Tehranian, Majid. 1990. Technologies of Power. Norwood, New Jersey: Ablex. Telecommunication Authority of Singapore. Various years. Annual_Eeeer_- Singapore- Telecommunication Authority of Singapore Act. Singapore Acts 1985 Revised Edition. Chapter 323. 248 Telecommunications Department. Various years. Annual Report. Singapore. Thio, Eunice. 1991. The Syonan Years, 1942-1945. In A History of Singapore, ed. Ernest C.T. Chew and Edwin Lee, 95-114. Singapore: Oxford University Press. Tipps, Dean C. 1976. Modernization theory and the comparative study of societies: a critical perspective. In Comparative modernization: perspectiyee on modernization, ed. Cyril E. Black, 62-88. New York: Free Press. Toffler, Alvin. 1970. Future shock. New York: Bantam Books. . 1980. The third wave. New York: Bantam Books. Trimberger, Ellen Kay. 1979. World systems analysis: the problem of unequal development. Theory and society. 8: 101-126. United Nations. 1950. The economic eevelopment of Latin Amefica and its principal problems. New York: United Nations. . 1975. Qeveloping eounffies and leyele of development. New York: Economic and Social Council, Committee for Development Planning, United Nations. UNESCO. 1976. The use of socio-economic indicators in development planning. Paris: UNESCO Press. Valenzuela, J. Samuel and Arturo Valenzuela. 1978. Modernization and dependency: alternative perspectives in the study of Latin America underdevelopment. Comparative politics. 10 (July): 535-57. Villamil, José J., ed. 1979. Transnational capitalism and national development: new perspectives on dependence. SUSSEX: Harvester Press . Wagstaff, J.H. 1958. Annual report of the Telecommunications erantment, l2§7. Singapore: Director, Telecommunications. Wang, Georgette and Wimal Dissanayake, eds. 1984. Centinnity eng enange in communication systems: an Asian perspective. Norwood, N.J.: Ablex. 249 Wallerstein, Immanuel. 1972. Three paths of national development in the Sixteenth Century. Studies in Comparative International Development. 7(2) (Summer): 95-101. . 1974a. The rise and future demise of the world capitalist system: concepts for comparative analysis. Comparative Studies in Society and History. 16(4) (September) 387-415. . 1974b. Dependence in an interdependent world: the limited possibilities of transformation within the capitalist world-economy. African Studies Review. 17 (April) 1-26. . 1974c. Tne medefn wofld-system: capitalist agriculture and the origins of the European yprld-economy in the sixteenth century. New York: Academic Press. . 1976. "Modernization: Requiescat in Pace." In The usee of sociology in controverey, ed. Lewis A. Coser and Otto N. Larsen, 131-135. New York: Macmillan. . 1979. The capitalist world eeonomy: essays by Immanuel Wallerstein. New York: Cambridge University Press. . 1987. World Systems analysis. In Social theory today, ed. Anthony Giddens and Jonathan Turner, 309-324. Stanford: Stanford University. . 1988. Development: lodestar or illusion. Economic and Political Weekly. 23(39): 2017- 2023. Warren, Bill. 1973. Imperialism and capitalist industrialization. New Left Review. 81:3-44. Weber, Max. 1930. The Protestant Ethic and the Spifit of Capitalism. London: Allen and Unwin. Whitacker, 0.8. 1967. A dysrhythmic process of political change. World Bolitics. 19(2): 190-217. Wolfers, Edward P., ed. 1976. Australia's Northern neighbors: independent or dependent. Canberra: Australian Institute of International Affairs. Yule, G. Udny. 1926. Why do we sometimes get non-sense correlations between time-series? Journal of tne Boyal Statietical Society. 89: 1-64. 250 World Bank. 1991. World Development Report 1991: The challenge of development. Washington D.C.: Oxford University for the World Bank. World Bank. 1982. IDA In Retrospect: The Firet Tyo Decadee of The International Development Association. Washington D.C.: Oxford University for the World Bank. Zeitlin, Maurice. 1984. The civil wars in Chile (or the bourgeois revolutions that never werel. Princeton, NJ: Princeton University Press. "lllllllllllllllll