Africa Region Working Paper Series No. 90 Kenya Export Prospects and Problems Francis Ng Alexander Yeats October 2005 Kenya Export Prospects and Problems Africa Region Working Paper Series No. 90 October 2005 Abstract Although Kenya was once viewed as being among the African countries with the most favorable growth prospects, the last two decades witnessed significant declines in many measures of economic performance and social standards. As a result, Kenya's share of world trade is now less than one-half its average level in the early-1980s. How can Kenya halt, and then reverse, these negative trends? While the answer to this question has multiple dimensions, trade policy certainly can play an important positive role. This report examines reasons why Kenya and other African trade did not provide the "engine of growth" that it did elsewhere. During the last quarter century many Sub-Saharan African countries non-oil exports either declined in absolute terms, or expanded at a slower pace than world trade. Evidence suggests African countries, including Kenya, experienced serious supply constraints that limited their ability to capitalize on opportunities of international production sharing in foreign markets. Inappropriate governance policies, and a general unfavorable commercial environment, were largely responsible for Africa's supply problems. In addition, a decomposition of recent trade changes into supply and demand factors shows that, with two important exceptions (cut flowers and apparel exports) Kenya experienced a general erosion of its US and EU import market shares. These findings support the conclusion of the recent Africa Competitiveness Report that concludes Kenya is at a competitive disadvantage vis-à-vis more than one half the other African countries surveyed. Finally, the authors suggest that the diversifications of Kenya's exports away from traditional products must have a very high priority in practice. For the developments in regional markets, it would be more advantageous for Kenya to pursue the trade liberalization on an MFN basis, rather than through the exchange of regional preferences. Kenya also need to set itself up to attract private investment, and that means a clean regulatory environment, a judicial system that works, proper police enforcement and corporate law, capacity building, and development and maintenance of infrastructure necessary to support manufacturing activity. The Africa Region Working Paper Series expedites dissemination of applied research and policy studies with potential for improving economic performance and social conditions in Sub-Saharan Africa. The series publishes papers at preliminary stages to stimulate timely discussions within the Region and among client countries, donors, and the policy research community. The editorial board for the series consists of representatives from professional families appointed by the Region's Sector Directors. For additional information, please contact Momar Gueye, (82220), Email: mgueye@worldbank.org or visit the Web Site: http://www.worldbank.org/afr/wps/index.htm. The findings, interpretations, and conclusions in this paper are those of the authors. They do not necessarily represent the views of the World Bank, its Executive Directors, or the countries that they represent and should not be attributed to them. Authors'Affiliation and Sponsorship Francis Ng, Economist, Trade Team of Development Research Group, The World Bank fng@worldbank.org Alexander Yeats Consultant, Trade Team and Africa Region, The World Bank ayeats@msn.com Acknowledgement The authors, wish to acknowledge the valuable contributions of Christiane Kraus and Manuel De la Rocha who commented extensively and helped on data problems during the preparation of this paper. The views expressed in this report are those of the authors, and do not necessary reflect the views of The World Bank Group, its Executive Directors, or its staff. ii Table of Contents EXECUTIVE SUMMARY ............................................................................................................. i Trade Data Issues ........................................................................................................................ i Implications of the Marginalization of Africa ............................................................................. i Diversification! A Priority Policy Issue......................................................................................ii Can Kenya Compete? .................................................................................................................ii Opportunities in International Production Sharing ..................................................................iii Developments in Regional Markets ...........................................................................................iii Implications of the Commercial Environment in Kenya............................................................ iv I. INTRODUCTION....................................................................................................................... 1 II. DATA CONSIDERATIONS AND CONSTRAINTS............................................................... 5 A. On the Secular Deterioration in the Quality of Kenya's Trade Data..................................... 9 B. On the Availability of African Trade Statistics...................................................................... 9 C. Analyzing Kenya's Export Performance: What Can Be Done? .......................................... 14 Annex 2.1a. On the General Accuracy of African Trade Statistics.......................................... 15 III. IMPLICATIONS OF THE MARGINALIZATION OF AFRICA......................................... 21 A. Africa's Production and Supply Problems........................................................................... 22 B. Long-Term Demand Prospects for Traditional Exports....................................................... 23 C. Changes in the Relative Importance of East African Countries........................................... 25 IV. THE GEOGRAPHIC DIRECTIONS OF KENYA'S TRADE.............................................. 25 A. The Destinations of Kenya's Exports .................................................................................. 26 B. The Origins of Kenya's Imports........................................................................................... 28 V. THE EFFECTS OF AGOA ON KENYA'S EXPORTS ......................................................... 29 Annex 5.1a. A Note on AGOA and Kenya's Transportation Costs for Exports to the United States......................................................................................................................................... 31 VI. IMPLICATIONS OF THE COMPOSITION OF KENYA'S EXPORTS ............................ 36 A. Does Kenya Have Dynamic Exports?.................................................................................. 40 B. Have Kenya's Exports Become More "Complementary"?.................................................. 44 C. Has Kenya Diversified Its Exports?..................................................................................... 46 D. Implications of An "Export Prospects" Index...................................................................... 49 VII. COMPETITIVE FACTORS AND THE CHANGE IN KENYA'S EXPORTS................... 51 A. How Important is Kenya as a Supplier of US Garment Imports?........................................ 54 VIII. INTRA-INDUSTRY TRADE AND PRODUCTION SHARING ..................................... 62 A. How Big is Production Sharing in Kenya?......................................................................... 64 B. Implications of Production Sharing Outside Africa............................................................. 68 IX. DEVELOPMENTS IN REGIONAL MARKETS.................................................................. 72 iii X. PROSPECTS FOR EXPORT DIVERSIFICATION............................................................... 80 A. Evaluating Kenya's Commercial Environment .................................................................. 83 B. Identifying Products for a "Successful" Diversification...................................................... 91 1. The Survey Approach ....................................................................................................... 91 2. Implications of Low Income Countries' Comparative Advantage................................... 93 3. Fast Growing Exports ....................................................................................................... 96 4. The Economic Implications.............................................................................................. 99 XI. PRIORITY POLICY ISSUES.............................................................................................. 100 A. Trade Data Issues............................................................................................................... 100 B. The Implications of Kenya's Commercial Environment ................................................... 101 REFERENCES ........................................................................................................................... 102 BOXES Box 1.1 The Structure of Kenya's Exports: Then and Now!.......................................................................4 Box 2.1 Have One and One-Half Billion Dollars in Kenya's Exports Been "Lost"...........................6 Box 2.2 Inconsistencies in Uganda and Kenya's Import Statistics.............................................12 Box 6.1 Africa's Recent Experience with Export Processing Zones..........................................38 Box 6.2 Dimensions of the Export Concentration Problem in Africa..........................................47 Box 7.1 The Development of Kenya's Exports of Cut Flowers................................................54 Box 7.2 Recent Developments in EU 15 Markets for Tropical Beverage Products..........................56 Box 7.3 Long-Term Effects of the Removal of MFA Restrictions.............................................58 Box 9.1 Characteristics of Kenya's 2003 Trade with the Republic of South Africa.........................73 Box 9.2 Trade Barriers Facing East Asian Exporters at the Start of Their Industrialization Drive.........80 Box 10.1 Examples of Successful National Diversification Strategies.........................................81 Box 10.2 Income Levels and the General Commercial Environment in Africa...............................85 Box 10.3 Can Kenya Compete? Implications of the Africa Competitiveness Report........................89 Box 10.3 Implications of Industry Level Labor Intensity Indices...............................................93 TABLES Table 2.1 Comparisons of Kenya's Reported Exports with the Imports of Trading Partners...............7 Table 2.2 Discrepancies Between Reported Kenyan Exports and Partner Country Imports.................8 Table 2.3 Product Specific Discrepancies Between Kenya's Reported 2002 Exports and Partner Countries' Imports..............................................................................11 Table 2.4 COMTRADE Data Availability for Selected African Countries...................................13 Annex Table 2.1a Partner Country Comparisons of Kenya and Tanzania's Total Imports and Exports................................................................................................16 Annex Table 2.2a Partner Country Data Comparisons for Kenya's Ten Largest Exports To Tanzania................................................................................................17 Annex Table 2.3a Comparisons of 2002 Exports Reported by Uganda to UN COMTRADE With the Matched Imports of African Trading Partners.............................................18 Annex Table 2.4a Comparisons of 2002 Exports Reported by Tanzania to UN COMTRADE With the Matched Imports of African Trading Partners.............................................19 Annex Table 2.4a Comparisons of 2002 Exports Reported by Kenya to UN COMTRADE With the Matched Imports of African Trading Partners.............................................20 iv Table 3.1 The Relative Importance of Sub-Saharan Africa and Other Regions in World Trade...........22 Table 3.2 Changes in the Relative Importance of Kenya, Tanzania and Uganda in World Trade.........27 Table 4.1 The Geographic Destinations of Kenya's Exports; 1975 to 2003.................................28 Table 4.2 The Geographic Origins of Kenya's Imports; 1975 to 2003.......................................29 Table 5.1 The Composition of Africa's AGOA Eligible Exports to the United States.....................30 Table 5.2 The Origins of Africa's AGOA Eligible Exports to the United States.............................31 Annex Table 5.1a Nominal Transportation Costs for Kenya's Major Two-Digit HS Product Exports to the United States: Vessel and Air Freight Shipments Combined.....................34 Annex Table 5.2a The Level and Variation in Nominal Transportation Costs for Kenya's Air Freight Exports.......................................................................................35 Table 6.1 The Structure of Kenya's Exports by Major Product Categories..................................37 Table 6.2 Kenya's Largest Two-Digit SITC Export Products in 1976 and 2003............................39 Table 6.3 Dynamic and Declining Products in Kenya's Recent Exports to the European Union..........41 Table 6.4 Dynamic and Declining Products in Kenya's Recent Exports to the United States.............42 Table 6.5 Complementarity Indices for Kenya's Trade with Selected Countries............................44 Table 6.6 Concentration Indices for Kenya's Exports...........................................................46 Table 6.7 The Export Prospects Index for Sub-Saharan African Countries..................................49 Table 7.1 The Influence of Demand and Competitive Changes on Kenya's Exports to the EU...........52 Table 7.2 The Influence of Demand and Competitive Changes on Kenya's Exports to the US...........55 Table 7.3 The Relative Importance of Kenya as a United States Clothing Supplier........................60 Table 8.1 Intra-Industry Trade Ratios for Kenya and Other African Countries..............................62 Table 8.2 Sib-Saharan Africa 2003 Trade in Parts and Components..........................................65 Table 8.3 Changes in Major African Countries Trade in Parts and Components...........................68 Table 8.4 The 2003 Composition of Kenyan and Other Sub-Saharan African Countries Imports of Parts and Components (values in $000)...........................................................69 Table 9.1 The Largest Products in Kenya's Exports to Uganda and Tanzania in 1976 and 2003.........72 Table 9.2 The Largest Products in Kenya's Imports from Uganda and Tanzania in 1976 and 2003......74 Table 9.3 Dynamic and Declining Products in Kenya's Recent Non-Oil Exports to Uganda..............76 Table 9.4 Dynamic and Declining Products in Kenya's Recent Non-Oil Exports to Tanzania............77 Table 10.1 Kenya's Scores for Major Policy Variables Influencing the Commercial Environment.......86 Table 10.2 The Relative Position of Kenya's WSJ-Heritage Overall Commercial Environment Index Compared to Averages for Major Developing Country Groups...........................88 Table 10.3 The Importance of Labor Intensive Exports as a Catalyst for Growth in Newly Industrialized Developing Countries..................................................................94 Table 10.4 The Ten Fastest Growing Labor Intensive Manufactures in All Sub-Saharan African Countries' (Exclusive of Republic of South Africa) Exports.......................................95 Table 10.5 Fast Growing Exports for Which African Countries Increased World Market Shares.........97 v ABBREVIATIONS AND ACRONYMS AGOA African Growth and Opportunities Act of the United States COMESA Common Market for Eastern and Southern Africa COMTRADE United Nations Commodity Trade Statistics EAC East African Community EPZ Export Processing Zone FDI Foreign Direct Investment GSP Generalized System of Preferences HS Harmonized Trade Classification System IMF International Monetary Fund NICS Newly Industrialized Countries OECD Organization for Economic Cooperation and Development RPED Regional Program on Enterprise Development RSA Republic of South Africa SITC Standard International Trade Classification SSA Sub-Saharan Africa UN United Nations UNSO United Nations Statistical Office WTO World Trade Organization EXECUTIVE SUMMARY Although Kenya was once viewed as being among the African countries with the most favorable growth prospects, the last two decades witnessed significant declines in many measures of economic performance and social standards. For example, the annual growth in gross domestic product per capita averaged 2.3 percent over 1960-1990, but recently turned negative. The ratio of trade to GDP also fell sharply, from about 71 percent in 1995 to its current 56 percent level. As a result, Kenya's share of world trade is now less than one-half its average level in the early-1980s. In this environment, many key social indicators worsened Life expectancy at birth fell from about 57 years in 1990 to 45 years in 2002, school enrollment declined from 1990 to 2001, along with the share of GDP spent on education. This deterioration in economic and social standards raises a question of major importance. How can Kenya halt, and then reverse, these negative trends? While the answer to this question has multiple dimensions, trade policy certainly can play an important positive role. The experience of other countries shows a vigorous expansion of exports can be a major catalyst for raising incomes and living standards. An important question is what constraints kept Kenya, and other Sub-Saharan African countries, from achieving the sort of export lead growth observed in East Asia, Latin America, and other developing country regions. This report examines reasons why African trade did not provide the "engine of growth" that it did elsewhere. Trade Data Issues Any analyses of Kenya's export performance must confront some very serious data problems. The export and import statistics Kenya has been reporting to the United Nations have little or no utility for empirical analyses, or for policy formulation. Contrary to UN guidelines, Kenya fails to include shipments from its export processing zones (EPZs) in its official trade data. In addition, major discrepancies in Kenya's trade data occur for other reasons. As such, the statistics compiled by Kenya grossly understates actual trade values. International Monetary Fund Direction of Trade (DOT) data indicate Kenya's 2001 and 2002 trade statistics are under-reporting exports by as much as one and one- half billion dollars. Over 1998 to 2002 the value of unreported exports appears to be about $1 billion higher. The problems relating to the accuracy of Kenya's export statistics are not solely of recent origins. Statistical comparisons indicate there has been a secular deterioration in data quality that extends back to the early or mid-1980s. As such, partner country import statistics must be used, where possible, for analyses of Kenya's export performance. Given concerns about the quality of other African countries statistics the approach must be applied with considerable caution when analyzing changes in intra- regional trade. Implications of the Marginalization of Africa During the last quarter century many Sub-Saharan African countries non-oil exports either declined in absolute terms, or expanded at a slower pace than world trade. Evidence suggests African countries, including Kenya, experienced serious supply constraints that limited their ability to capitalize on opportunities in foreign markets. Inappropriate governance policies, and a general unfavorable commercial environment, were largely responsible for Africa's supply problems. However, a related point is that many of Africa's traditional exports face unfavorable long-term global demand prospects that resulted in low (or even negative) growth and declining real prices. Competition from synthetics, like plastics, had a serious negative effect on demand for some metal ores, while Kenya's exports of hard ii fibers (like sisal) were displaced by synthetic fibers (like nylon). If Africa's marginalization is to be halted the region must diversify into new, more promising, exports, and also significantly improve the internal "commercial environment" in most countries. However, there is little evidence that any of these crucial changes are occurring in Kenya or elsewhere. Global trade shares for Kenya, Tanzania and Uganda fell by almost fifty percent from their level in the early 1980s to the present. There is no evidence that the marginalization of these East African countries differed substantially from that of Sub-Saharan Africa as a whole. However, evidence suggests that recently adopted programs in favor of Africa, like the US African Growth and Opportunities Act (AGOA) may have had a positive influence on Kenya's exports of apparel and clothing. Diversification! A Priority Policy Issue Countries are often advised to avoid a concentration of exports in a narrow range of similar goods that may be negatively affected by new more efficient competitors, by the development of substitutes, or by factors adversely affecting demand. Like most other Sub-Saharan African countries Kenya's exports are highly concentrated, largely due to coffee and tea. Coffee exports were severely affected by a recent collapse in international coffee prices. However, a loss of European market share to Brazil and Vietnam further reduced the recent value of Kenya's coffee exports. The diversification of Kenya's exports away from traditional products has a very high priority. However, several empirical measures of export concentration (such as a simple count of the number of products exported) suggest little progress has been made in this direction. Statistics show a high export concentration in products with very poor growth and real price change prospects is a major problem for almost all Sub-Saharan Africa countries. Very high short-term price and export revenue volatility for some products (like coffee) have weakened African countries capacity for rationale development planning. Assuming that recent trade growth rates for specific products continue, a key question is how favorable, or unfavorable, are Kenya's overall export prospects. Relevant information is available from an index that establishes a concordance between the share of a product in a country's total exports, and the rate of growth of that product in world trade. This "export prospects" index suggests that Kenya should expect its exports to grow at about one-half the rate of growth in global trade. Unless it diversifies into new, more promising product lines, Kenya will continue to be marginalized. Can Kenya Compete? Any aggregate analysis of changes in Kenya's exports could be misleading if the influence of underlying factors, like changes in demand and import market shares, are not properly accounted for. A decomposition of recent trade changes into supply and demand factors shows that, with two important exceptions (cut flowers and apparel exports) Kenya experienced a general erosion of its US and EU import market shares. These findings support the conclusion of the recent Africa Competitiveness Report that concludes Kenya is at a competitive disadvantage vis-à-vis more than one half the other African countries surveyed, and is in the bottom twenty percent in expanded comparisons involving non-African countries. While apparel exports to the United States are of major importance to Kenya, US import statistics show Kenya is a very marginal supplier compared to China, Mexico, Bangladesh, or even Vietnam. iii Furthermore, Kenya's apparel exports are highly concentrated in a limited number of product lines, which makes the sector very vulnerable to competitive changes that might occur, or to developments following the full phase out of MFA trade restrictions. Recent United States surveys of executives in major apparel producing firms indicate Kenya, and other African countries, will likely experience major export losses as a result of the MFA phase out. Opportunities in International Production Sharing International production sharing, which often involves the importation and further assembly of parts and components, can significantly broaden the range of new products a country, like Kenya, can successfully export. In the absence of production sharing, Kenya would have to master entire production processes for a good in order to become a viable exporter. Production sharing provided a major stimulus to trade and growth in East Asia, the Caribbean, and even between industrial countries themselves. However, evidence suggests that Kenya, and other African countries, failed to fully capitalized on opportunities that exist in this activity. By importing components for further assembly, Kenya may have the potential to access international markets for low and medium-technology products like office machinery, telecommunications equipment, or some consumer electronics. In the Far East, low wage underdeveloped countries like Indonesia, Thailand and the Philippines provide useful examples as to how this transition has been accomplished. Developments in Regional Markets A general problem often encountered in efforts to establish regional trade arrangements is that the goods African countries export are often quite different than those others import. If the profile of Kenya's exports is changing in ways that more closely match other African countries imports this would indicate a growing potential for expanded trade. An important related question is whether Kenya's exports are becoming more similar to OECD members imports ­ a development that would signal growing interdependence. However, a "trade complementarity" index shows that such positive changes are not occurring. Furthermore, the values of the index are currently so low that few opportunities exist for a meaningful expansion of intra-regional African trade. These findings suggest it would be more advantageous for Kenya to pursue a trade liberalization on an MFN basis, rather than through the exchange of regional preferences. Some recent analyses of the direction of Kenya's exports concluded there has been a positive change the relative importance of other East African countries as markets. This assertion is open to question on three counts. First, International Monetary Fund statistics indicate that East Africa's import shares are highly erratic, but now are at about the same level as they were in the mid-1970s. Second, most previous analyses failed to account for the importance of refined petroleum products that now constitute approximately one-half (by value) of Kenya's exports to Uganda and Tanzania. Third, as noted (see "trade data issues"), Kenya's export statistics are seriously underreporting trade with the United States and European Union. As a result, the relative importance of East African markets is greatly overstated. Finally, although the available statistics are of uncertain quality, it appears that Kenya's non- oil exports to East Africa are highly concentrated in terms of their product composition, and have grown relatively slowly. iv Implications of the Commercial Environment in Kenya Policy makers in a country like Kenya previously had difficulty in determining how their domestic commercial environment compared with those implemented elsewhere. Several recent efforts to compile comprehensive cross-country indices of the quality of governance and commercial policies now provide relevant information. These statistics suggest that the domestic environment makes Kenya relatively less attractive for foreign investment compared to many other developing countries. Less than 35 percent of all Latin American countries have a commercial environment inferior to that in Kenya, while two-thirds of the East Asian countries have a superior business climate. The commercial cost of dealing with crime in Kenya appears to be higher than in any other African country, as is the problem of official corruption. This highly unfavorable local business climate is having a major retardation effect on foreign investment, which in turn constrains Kenya's trade and growth. Kenya needs to set itself up to attract private investment, and that means a clean regulatory environment, it means a judicial system that works, proper police enforcement and corporate law, capacity building, and the development and maintenance of infrastructure necessary to support manufacturing activity. And it means that corruption must be stamped out. Without these, private investors simply will not invest. KENYA Export Prospects and Problems* Francis Ng and Alexander Yeats I. INTRODUCTION Although Kenya was once viewed as being among the African countries with the most promising prospects, the last two decades witnessed significant declines in many measures of economic performance and social standards. For example, the annual growth in gross domestic product per capita averaged 2.3 percent over 1960-1990, but turned negative at the beginning of the current decade. The ratio of trade to GDP also fell sharply, from about 71 percent in 1995 to its current 56 percent level. As a result, Kenya's current share of world trade is now less than one-half its average level in the early-1980s. In this environment many key social indicators worsened Life expectancy at birth in Kenya fell from about 57 years in 1990 to 45 years in 2002, school enrollments have declined from 1990 to 2001, along with the share of GDP spent on education (World Bank 2003, Table 2.2). This deterioration in economic and social standards raises a question of major importance. How can Kenya halt, and then reverse, these negative trends? While the answer to this question clearly has multiple dimensions, trade policy certainly can have an important positive role. The experience of other countries clearly shows a vigorous expansion of exports can be an important catalyst in raising incomes and living standards.1 This has been well documented in the case of East Asia, as well as in other regions. From 1975 to 2001, East Asia's share of global exports expanded more than three-fold (to just under) 19 percent and doubled from 1985 to 2001. Intra-regional Asian exports, expressed as a share of world trade, experienced and even sharper expansion rising more than six-fold during 1975-2001 (Ng and Yeats 2003). In this dynamic environment per capita incomes for countries like Malaysia, Singapore, Hong Kong, Republic of Korea and Thailand grew at rates more than double that for the global average. The reasons for East Asia's success are complex, but the regions general adoption of relatively open export oriented growth strategies are often cited as playing an important positive role. These strategies focus on the creation of a domestic commercial environment that helps national enterprises capitalize on opportunities in foreign markets. An important question is what constraints kept Kenya, and other Sub-Saharan African countries, from achieving the sort of export lead growth observed in East Asia, Latin America, and other developing country regions. Explanations range from the fact that most African countries are located in the tropics, and climate may have a retardation effect on development (Sachs 2001), or that many African countries are land-locked which causes major commercial and logistical problems. Others stress the importance of governance problems in Africa and the unattractive commercial environment that these conditions create (Collier 1995 or Ng and Yeats 1999).2 However, many explanations of Africa's marginalization in * This paper was prepared for the background study of Kenya Diagnostic Trade Integration Study under Integrated Framework in Winter 2004. 1See Dollar (1992), Edwards (1993), or Easterly and Levine (1994) for empirical analyses that establish a clear direct relationship between improved export performance and the achievement of higher rates of economic growth. 2According to the World Bank (1997, p. 32) "Governments enormous impact on development is well illustrated by contrasting economic performance of developing countries in Sub-Saharan Africa and East Asia. In 1960 incomes per capita in much of East Asia were only slightly higher than in Africa. Governments in the two regions were also 2 world trade also acknowledge that the widespread adoption of so called "import substitution" policies had important negative effects. Import substitution growth strategies often erect high trade barriers against (often far more efficient foreign producers) so local concerns can (purportedly) have uncontested access to domestic markets in order to achieve economies of scale and other production efficiencies. Unfortunately, evidence suggests that import substitution often promoted the development of inefficient local monopolies accompanied by the severe misallocation of resources.3 Empirical analyses of other countries experience clearly shows expanding exports and the adoption of more "open' commercial policies can provide an important stimulus to economic growth and development (see Kravis 1970, Keesing 1967 or Nash and Thomas 1991). As such, the purpose of this report is to identify existing constraints to the expansion of Kenya's exports, and also to determine what policy initiatives are needed to change the composition of this exchange. The analysis focuses on considerations such as the following; · Previous studies document the "marginalization" or long-term diminished relative importance of Africa in global trade. To what extent, if any, has Kenya's experience differed from that of other Sub- Saharan African countries. · Kenya's exports are concentrated in a limited number of agricultural commodities like tea and coffee. How favorable, or unfavorable, are the growth prospects for these types of products? Relevant information is available from recent World Bank analyses of longer-term changes in real commodity prices, as well as projections of future price changes for these products. · Does the evidence indicate Kenya is retaining international competitiveness for its major exports, or does a loss of import market shares account for the lack of dynamism in its trade. If Kenya is experiencing a deterioration in its international competitiveness what countries and factors are responsible. To what extent have Kenya's exports been negatively affected by sluggish international demand for its traditional products, or the failure to diversify into new product lines with more favorable growth prospects. · Is Kenya becoming more interdependent with the global economy? Increasing interdependence is generally viewed as having positive effects including greater stability of export earnings and more favorable longer-term trade prospects. This interdependence could manifest itself through increased similar in size, although not in composition. African governments were already spending more on consumption, primarily on public employment. By the mid-1990s, however, incomes in East Asia were more than five times those in Africa and government consumption in Africa, relative to GDP, had ballooned to one-and-a-half times that in East Asia. The sources of this divergence are complex, but it is widely believed that the superior performance of the state in East Asia ­ the limits it set on its own growth, the soundness of the policies it adopted, and the effectiveness with which it delivered services ­ made a powerful contribution to the growing gap in the quality of life experienced by the average citizen in these two parts of the world." 3Import substitution strategies were generally formulated in the 1950s and 1960s when there may have been more compelling reasons for their adoption than there are at present. Prior to the Kennedy Round of the multilateral trade negotiations in the late 1960s, OECD trade barriers facing exports from developing countries often were quite high ­ tariffs averaged close to 20 percent, and often escalated sharply ­ and multiple nontariff barriers (including so called "voluntary" export restraints) were used by industrial countries (see Box 9.2 that follows). This situation changed markedly due to the major liberalization of trade barriers that occurred in the Kennedy, Tokyo, and Uruguay Round. In addition, various measures such as the generalized system of preferences (GSP), the Lome Convention, and the recent United States African Growth and Opportunities Act (AGOA) were all specifically intended to improve market access conditions for developing countries. The rationale for import substitution policies is greatly weakened by these significant improvements developing countries received in access to OECD markets. 3 intra-industry trade, a growing "complementarity" between the types of goods Kenya exports and those its trading partners import, or through increased participation in international production sharing among countries. · To what extent have Kenya's agricultural exports been adversely effected by the policies of other countries. For example, OECD export subsidies for products like cotton and sugar appear to have had an important negative effect on the ability of many developing countries to compete effectively in global markets for these goods. Although the Uruguay Round significantly lowered international trade barriers facing many industrial products these restrictions remain very high on some agricultural goods. These restrictions have a direct negative impact on the poorest strata of many developing countries' societies since increased agricultural exports are one way of alleviating the persistent problem of rural poverty. · What new products might be promising candidates for diversifying Kenya's exports. This question is addressed by identifying goods normally manufactured using highly labor intensive production processes in which low wage countries, like Kenya, are generally believed to have a comparative advantage. An important related question is whether Kenya's export profile has undergone important changes over the longer-term, or whether it has been essentially static (see Box 1.1). Ideally, one would want to see some evolution in the direction toward products with more favorable growth prospects, as was the case for many former commodity exporting countries like Brazil, China, Malaysia and Thailand whose exports shifted rapidly toward manufactures. Finally, it is acknowledged that Kenya must create an appropriate domestic commercial environment to facilitate positive changes in the composition and structure of its major exports, and to attract the required foreign investment needed to support this transition. Key related questions are how attractive is Kenya to foreign investors, and is the commercial environment in Kenya more or less favorable than that in other developing countries? While it previously would have been difficult to address this question, extensive new cross country surveys of the nature of countries' commercial "environment" provide much useful information. These surveys can help a country like Kenya identify specific trade, financial, monetary or fiscal policies that are acting as an impediment to growth. 4 Box 1.1 The Structure of Kenya's Exports: Then and Now! United Nations statistics indicate less than 50 percent of world trade consisted of manufactured goods in the mid-1950s, while their current share is about 75 percent. The present share of commodities and raw materials (approximately 25 percent) is about one-half what is was in the mid-1950s ­ it would now be under 20 percent if crude petroleum were excluded. One factor responsible for this shift in the composition of trade was the conversion of many important commodity producing countries into exporters of manufactures. This started with the Asian "Newly Industrialized Countries (NICS)" in the 1960s and early 1970s, followed by a group of "second tier" industrializing countries, like Malaysia and Thailand, in the late 1970s and 1980s. Today, more than 60 percent (by value) of the total exports of former largely commodity producing countries like Brazil, China, India, Malaysia, Mauritius, Pakistan and Thailand consist of manufactured goods. Is Kenya's export profile evolving in directions similar to the changes that occurred in these former commodity exporting countries? United Nations Commodity Trade (COMTRADE) records extend from the present back to the mid-1960s and can provide useful information on longer-term changes in the level and composition of a country's exports. The statistics shown below report the value and share of Kenya's major one and two-digit SITC product group exports from 1964 to 2003. These data stretch COMTRADE statistics to the limit of currently available information. 1964 Exports 2003 Exports 1964-03 Value Share Value Share Share Product Group (SITC) ($000) (%) ($000) (%) Change Food and live animals (0) 110,163 56.1 937,326 41.3 -14.8 Fruits & vegetables (05) 6,166 3.1 307,892 13.6 10.4 Coffee, tea, & spices (07) 92,323 47.0 486,579 21.4 -25.6 Beverages and tobacco (1) 9 0.0 28,559 1.3 1.3 Crude materials (2) 67,626 34.5 402,691 17.8 -16.7 Textile fibers (26) 43,827 22.3 18,672 0.8 -21.5 Crude vegetable materials (29) 9,509 4.8 309,978 13.7 8.8 Mineral fuels and lubricants (3) 2,948 1.5 262,066 11.6 10.0 Petroleum & products (33) 2,880 1.5 260,358 11.5 10.0 Animal & vegetable oils (4) 82 0.0 12,334 0.5 0.5 Chemicals (5) 5,491 2.8 108,706 4.8 2.0 Manufactured goods (6) 4,021 2.0 172,751 7.6 5.6 Machinery & transport (7) 3,667 1.9 70,046 3.1 1.2 Misc. manufactures (8) 804 0.4 265,142 11.7 11.3 Clothing (84) 39 0.0 209,197 9.2 9.2 Other goods (9) 1,451 0.7 8,345 0.4 -0.3 All goods (0 to 9) 196,261 100.0 2,268,595 100.0 -- Over this (almost) 40 year period Kenya's exports experienced some significant structural changes, but these were nothing like those for countries that made the transition to "successful" exporters of manufactures. The most notable negative share changes occurred for coffee (a 25 percentage point loss) and for textile fibers whose share fell by over 20 percentage points. Evidence suggests the latter losses were largely the result of the replacement of hard fibers, like sisal, by synthetics. Coffee's 2003 trade share may not be representative due to the major decline in international prices for this product that began in 2001-2002. Two of the seemingly positive changes that occurred were for mineral fuels and clothing where export shares increased by about 10 percentage points. The increased share for energy products is the result of increased local refinery operations for foreign produced crude petroleum. The refined petroleum exports are mainly destined for Uganda and Tanzania. Second, the ten point share increase for clothing is largely due to the adoption of United States African Growth and Opportunity Act (AGOA) preferences. On some specific clothing products AGOA provides Kenya with preferences of up to 35 percent over countries facing MFN tariffs. It should be noted that there is major uncertainty as to whether Kenya will be able to maintain its apparel exports to the United States once the MFA quotas against competitors are fully phased out in 2005. 5 II. DATA CONSIDERATIONS AND CONSTRAINTS Key Point The export statistics Kenya has been reporting to the United Nations have little or no utility for analyses of trade problems or performance. Contrary to UN guidelines, Kenya fails to include shipments from its export processing zones (EPZs) in its official trade data. As such, the available statistics grossly understate the value of Kenya's trade with some countries. However, according to the IMF, additional problems such as intentional under-invoicing of both imports and exports, have a further adverse influence on the quality of available statistics. Evidence suggests that Kenya's 2001 and 2002 trade data may be under-reporting exports by as much as one and one-half billion dollars. As such, partner country import statistics must be used, where possible, for analyses of Kenya's export performance. While this approach appears to work satisfactorily for trade with high and middle income countries, serious defects in African partner countries' statistics makes it very difficult to say anything with confidence about the composition of, or trends in, Kenya's intra-regional trade. Before proceeding, it must be acknowledged that there are major trade data problems that seriously limits the potential for empirical analyses of Kenya's export performance. The export statistics Kenya has been reporting to the United Nations have little or no utility for analyses of trade problems or performance. One major problem is that, contrary to UN guidelines, Kenya fails to include exports from its Export Processing Zones (EPZs) in the official United Nations data. As such, the available UN export statistics grossly understates the value of Kenya's trade with many countries.4 However, additional problems such as under-invoicing of both exports and imports, and apparent serious errors in the tabulation of trade statistics, have a further adverse influence on the quality of available data (International Monetary Fund 2000 or UN WIDER 2000). For example, as Table 2.1 shows, Kenya's UN COMTRADE statistics appear to be reporting only about one-tenth of actual exports to the United States (a shortfall of about $180 million), while reported exports to the United Kingdom, Netherlands and France appear to be about $250 million too low. However, the data discrepancies are not confined to the high income countries. In 2002, reported imports by Pakistan and Egypt $262 million were more than four times the value reported in Kenya's export statistics. Overall, Kenya's statistics appear to have understated the value of total exports by about $680 million, although data tabulated by the International Monetary Fund suggests the value of unreported trade may be even higher (see Box 2.1).5 4 Two major data sources are available for analyzing Kenya's trade performance. The first is the International Monetary Fund's Direction of Trade (DOT) database that reports bilateral import and export statistics for almost all developed and developing countries on an annual and quarterly basis. However, DOT's most important negative feature is that the statistics only report import and export totals. This precludes their use for analyses such as those attempting to; (i) examine changes in the structure of a country's trade, (ii) identify relatively fast growing products, or (iii) examine trade trends in broad product groups like foodstuffs, raw materials, or manufactures. To address these types of questions on a cross-country and cross-product basis the only available source of statistics is the United Nations COMTRADE (Commodity Trade) database. In its current Harmonized System (HS) classification scheme, COMTRADE identifies approximately 3,000 individual products at its lowest level of detail. 5 Two points should be noted. First, import statistics are normally reported in cost-insurance-freight (c.i.f.) terms, while export values are normally expressed in free-on-board (f.o.b.) terms. This would normally cause partner country import statistics to be larger than Kenya's reported exports. The IMF estimates that transport and insurance costs normally add about ten percent to the f.o.b. value of a traded good. An offsetting factor is that some countries that are Kenya's trading partners have not reported any import statistics to the United Nations (see Table 1.2). This would tend to lower the total value of reported imports relative to the value of Kenya's exports. 6 Box 2.1. Have One and One-Half Billion Dollars in Kenya's Exports Been "Lost": Evidence from the IMF Although both UN COMTRADE and the International Monetary Fund's Direction of Trade Statistics (DOT) provide data on the total value of a country's imports and exports, important differences may exist in the way this information is compiled. For COMTRADE, the United Nations relies almost exclusively, and without adjustment, on import and export statistics reported by individual member governments. In contrast, the IMF recognizes that various factors such as false invoicing, smuggling, or deficiencies in local customs statistical procedures may seriously bias the trade data submitted to the United Nations. In cases where national trade statistics are highly suspect, the IMF will estimate a country's total imports and exports. These estimates often utilize partner country statistics in various gap filling procedures. The statistics shown below compare 1999 to 2002 trade totals drawn from statistics Kenya reported to UN COMTRADE with corresponding import and export statistics published in the most recent Direction of Trade Annual. The discrepancies between the IMF and COMTRADE import and export totals have been increasing steadily since 1999, and the DOT statistics suggest that Kenya's trade data is underreporting 2001 and 2002 exports by about $1.5 billion. This is roughly double the value of underreported exports ($765 million) that occurred in 1999-2000. In addition, different trends are reflected in the statistics as COMTRADE indicates Kenya's exports declined by about 15 percent over 1999-2002, while the DOT data indicate an increase of about 12 percent occurred. 1999-2002 Trade Flow ($000) 1999 2000 2001 2002 % Change KENYA'S EXPORTS Reported in COMTRADE 1,650,926 1,571,001 1,520,162 1,400,378 -15.2 Reported in IMF DOT 2,025,290 1,962,727 2,191,269 2,270,244 12.1 Difference -374,364 -391,726 -671,107 -869,846 Percentage Difference (%) -22.7 -24.9 -44.2 -62.1 KENYA'S IMPORTS Reported in COMTRADE 2,785,628 2,891,381 4,008,028 3,074,653 10.4 Reported in IMF DOT 3,104,493 3,327,594 3,794,368 3,676,080 18.4 Difference -318,865 -436,214 213,660 -601,427 Percentage Difference (%) -11.4 -15.1 5.3 -19.6 Both the partner country comparisons in this chapter, and the DOT data indicate very sizeable underreporting of Kenya's exports is occurring. Estimates based on the International Monetary Fund statistics indicate the underreporting is about $100 million higher than suggested in COMTRADE partner country comparisons. A likely explanation is that the latter is based solely on data for countries that have reported trade statistics to the United Nations. In contrast, the International Monetary Fund numbers probably include estimates of Kenya's trade with some non-reporting countries. 7 Table 2.1. Comparisons of Kenya's Reported Exports With the Imports of Major Trading Partners Kenya's Reported Exports ($000) Partner Countries Reported Imports ($000)* Percentage Difference (%)* Importer 1995 2000 2002 1995 2000 2002 1995 2000 2002 HIGH INCOME COUNTRIES 688,762 574,064 433,652 948,892 848,333 988,559 37.8 47.8 128.0 United Kingdom 197,609 243,972 186,424 253,738 286,270 324,990 28.9 17.3 74.3 Netherlands 86,101 95,511 100,304 120,336 153,541 177,230 39.8 60.8 76.7 Germany 155,171 73,030 35,568 207,786 100,153 81,090 33.9 37.1 128.0 France 46,420 32,021 20,173 71,716 69,588 63,497 54.5 117.3 214.8 United States 49,600 36,232 19,703 108,188 115,258 202,068 118.1 218.1 925.6 Italy 31,832 19,830 17,566 46,017 42,893 40,723 44.6 116.3 131.8 Belgium 43,322 24,172 17,342 30,180 28,312 31,916 -30.3 17.1 84.0 Switzerland 19,511 11,694 11,875 19,147 11,102 14,181 -1.9 -5.1 19.4 Spain 16,643 6,686 9,591 25,245 11,589 16,296 51.7 73.3 69.9 Japan 13,126 18,911 8,955 31,199 21,367 27,761 137.7 13.0 210.0 Sweden 29,427 12,005 6,151 35,340 8,260 8,847 20.1 -31.2 43.8 LOW & MIDDLE INCOME 813,307 644,060 580,290 588,857 705,905 793,564 -27.6 9.6 36.8 Uganda 277,113 226,231 301,739 213,306 286,989 312,727 -23.0 26.9 3.6 Tanzania 216,059 114,495 107,035 69,580 88,804 96,168 -67.8 -22.4 -10.2 Somalia 31,613 33,014 44,335 Na Na Na Na Na Na Rwanda 34,165 23,242 39,960 Na Na 66,703 Na Na 66.9 Congo, Dem. Rep. 41,010 28,037 37,425 Na Na Na Na Na Na India 13,427 17,358 34,509 14,992 19,341 33,390 11.7 11.4 -3.2 Pakistan 110,631 130,807 33,922 123,948 147,965 123,331 12.0 13.1 263.6 Egypt, Arab Rep. 57,993 93,224 25,596 78,671 95,051 139,285 35.7 2.0 444.2 Sudan 21,471 25,531 23,582 9,796 30,356 32,806 -54.4 18.9 39.1 Afghanistan 0 31,549 16,422 Na Na Na Na Na Na Burundi 7,700 5,210 16,379 10,617 7,647 15,646 37.9 46.8 -4.5 United Arab Emirates 9,688 34,203 16,283 Na Na Na Na Na Na Ethiopia 53,289 21,126 11,783 34,793 20,399 17,416 -34.7 -3.4 47.8 Zambia 3,334 2,042 10,072 1,972 2,731 11,704 -40.9 33.7 16.2 South Africa 51,074 7,817 8,831 30,764 6,089 10,214 -39.8 -22.1 15.7 Yemen 6,345 15,773 7,395 6,907 5,976 Na 8.9 -62.1 Na Madagascar 427 1,318 7,294 629 889 Na 47.3 -32.5 Na Algeria 154 180 7,060 19 13 Na -87.7 -92.8 Na Bahrain 1,216 219 6,842 418 533 877 -65.6 143.4 -87.2 * A "na" in the columns shown below indicates the country failed to report any trade statistics to UN COMTRADE in that year. Low and middle income totals exclude any country that had an "na" in any year. Source: Statistics Compiled from UN COMTRADE. 8 Table 2.2. Discrepancies Between Reported Kenyan Exports and Partner Country Imports Kenya ­ United States Trade Kenya ­ European Union (15) Trade Kenya's Reported US Reported Percentage Kenya's Reported EU 15 Reported Percentage Year Exports to US Imports from Kenya Difference (%) Exports to EU 15 Imports from Kenya Difference (%) 1985 60,862 99,581 63.6 456,603 607,111 33.0 1990 24,319 63,235 160.0 345,302 693,019 100.7 1991 39,849 73,455 84.3 480,313 719,592 49.8 1992 39,849 77,986 95.7 445,897 700,816 56.8 1993 48,379 100,431 107.6 503,247 675,890 34.3 1994 63,114 116,264 84.2 628,689 758,423 20.6 1995 49,600 108,190 118.1 649,538 852,383 31.2 1996 55,504 112,494 102.7 687,382 968,284 40.9 1997 57,792 119,859 107.4 676,428 945,294 39.7 1998 68,556 105,026 53.2 605,966 885,303 46.1 1999 37,762 111,916 196.4 553,787 806,233 45.6 2000 36,236 115,261 218.1 527,817 736,300 39.5 2001 40,400 136,902 238.9 549,728 760,461 38.3 2002 19,703 202,072 925.6 404,615 780,074 92.8 Source: United Nations COMTRADE statistics 9 A. On the Secular Deterioration in the Quality of Kenya's Trade Data Key Point The observed problems relating to the accuracy of Kenya's export statistics are not solely of recent origins. Statistical comparisons indicate there has been a secular deterioration in data quality that extends back to the early or mid-1980s. As an example, Kenya's trade statistics may have under-reported exports to the United States by $40 million in 1985, but the under-reporting appears to have widened to about $180 million in 2002. An important question is whether the trade data discrepancies observed in Table 2.1 are confined to Kenya's recent statistics, or do they reflect a longer term problem. For an answer, Table 2.2 compares total exports to the US and EU (15) as reported in Kenya's statistics with the reported imports of the two major trading partners from 1985 to 2002. Although there is some variation, the magnitude of the discrepancies indicates the biases in Kenya's trade data are longer-term in nature and are increasing. For example, as early as 1990 the reported value of United States imports from Kenya was more than two and one half times the corresponding value of Kenyan exports. In the same year, the total value of Kenya's reported exports to the EU were about $350 million lower (or one half) the value of the latter's reported imports from Kenya. These data also suggest that there has been a major recent deterioration in the quality of Kenya's statistics on trade with the US as the 1999-2002 percentage differences in the partner country statistics are among the largest in the table. In any event, Kenya's statistics are highly inaccurate and misleading as to the true level and change in trade with both the United States and European Union. To provide a different perspective on the magnitude of the data problems, Table 2.3 compares Kenya's reported exports of major four-digit SITC products with the reported imports of its trading partners in 2002. These products have been ranked on the reported imports of all Kenya's trading partners. For tea alone, a difference of over $300 million occurs as the partner country imports are about three times the corresponding value Kenya is reporting for exports. Second, the data suggest additional problems are biasing the Kenyan trade statistics since some products with large data discrepancies (like cut flowers, fresh vegetables, and fish fillets) seem unlikely to originate in the export processing zones. In response to a recent International Labor Office (2002) survey Kenya reported that the manufacture of clothing and medicaments, along with the further processing of tea, were the primary activities undertaken in the zones. B. On the Availability of African Trade Statistics Key Point African countries as a group have been among the most deficient in regularly reporting trade statistics to the United Nations Statistical Office. For example, both Uganda and Tanzania failed to provide UN COMTRADE with any import or export statistics from the mid-1970s to the mid-1990s. As result, major uncertainties exist as to changes in both the level and composition of East African intra-trade. In addition, various statistical tests have shown there are extensive technical and quality control problems in the statistics that have been made available. As such, Kenya's regional export performance cannot be adequately monitored using other African countries' data on imports. The nature and magnitude of the problems associated with Kenya's COMTRADE statistics precludes their use for any analysis of its trade prospects and performance. As such, an alternative approach could use partner country import statistics. While this procedure should work for most high and medium income countries, it will produce less than satisfactory results for Kenya's regional trade. The problem is that many African countries failed to consistently report trade data to the United Nations, or 10 there are major statistical inconsistencies in the data they have reported.6 As an illustration, Table 2.4 shows the years that African countries reported trade data to UN COMTRADE. For example, Chad reported trade statistics regularly from 1962 to 1975, but then submitted data for only one year (1995) after the mid-1970s. Somalia reported trade statistics from 1970 to 1982, but failed to report from 1983 to the present. Even when African countries did report trade statistics to UN COMTRADE serious discrepancies may exist between the data at different levels of aggregation ­ a point that seriously reduces their utility (see Box 2.2).7 In addition, international organizations like the IMF (2000) and UN WIDER (2000) allege that Uganda, Tanzania and other African countries are significantly under-invoicing the value of their imports for reasons relating to customs duty and tax avoidance. This would cause their import statistics to understate the value and composition of Kenya's exports. Given the problems connected with Kenya's own COMTRADE data, and the failure of many African countries to report reliable statistics, little can be said with certainty about Kenya's regional trade, or the intra-trade of most Sub-Saharan African countries. An annex to this section provides more information bearing on this point. However, there are additional points to note concerning the use of partner countries' COMTRADE statistics. Specifically, most users of trade data incorrectly assume that products imported by one country from another are actually produced in the latter. This assumption may be invalid for a number of reasons such as the following; · Industrial machinery and other capital equipment produced in Europe or the United States, and originally purchased and used in Kenya, may be later sold to (say) Tanzania. Trade statistics normally would record this as a Tanzanian import from Kenya, not the original country of manufacture. The data could incorrectly suggest Kenya developed a production capacity in these types of goods, even though no such capacity exists. This problem has been shown to have an important negative effect on trade data for other goods like office machinery, cars and transport equipment. · Goods like transport equipment, and other types of expensive capital equipment, are often initially exported from one country to another under some sort of a leasing arrangement. The original importer (say Kenya) might then eventually sublease the item to a third country like Uganda. The transaction would be recorded as an import by Uganda from Kenya, even though the item was manufactured elsewhere. Sub-leasing largely accounts for the sizeable intra-African trade in airplanes as recorded in UN COMTRADE statistics. As an example, in 2002 Kenya's COMTRADE records report $1.4 million in imports of SITC 734.1 "Aircraft, Heavier than Air" from Tanzania. This transaction clearly involves the lease, or sale, of a good originally produced elsewhere since Tanzania foes not have a production capacity for aircraft In short, a key question has to be asked when using African and other 6 Uganda appears to be Kenya's most important African export market, but there are serious inconsistencies in the trade data it submitted to the United Nations. For example, Box 2.2 shows in both 2001 and 2002 the import "totals" Uganda reported for Kenya are more than US$ 150 million higher than the sum of the values of reported four-digit SITC imports from Kenya. As such, considerable uncertainty exists as to the level and composition of Kenya's trade with Uganda. Second, although little is known about its precise magnitude, it is generally acknowledged that a significant volume (perhaps 20 to 40 percent) of many African countries regional trade is not recorded in official trade statistics. To the extent that this underreporting (smuggling) occurs, import statistics will understate the true value of Kenya's African exports. 7 There is a further problem with some African countries trade data that should be noted. In the mid-1970s most countries began reporting statistics in terms of the more precise Revision 2 of the Standard International Trade Classification System and this substantially improved empirical work on trade issues. However, some countries like Sudan, Tanzania and Uganda only began reporting statistics in terms of the Revision 2 classification in the mid- 1990s. As a result, some of our analyses had to employ trade data classified in terms of the Revision 1 system even though it was recognized this information had important analytical shortcomings. 11 countries' statistics. That is, does the country really have a production capacity in the goods that are being reported as traded products? · Africa has more landlocked countries than any other region, and this may make it difficult to identify the true origins of imports, particularly if the goods are even slightly modified in a transit country. Uganda might, for example, incorrectly report some items as imports from Kenya even if it were manufactured in a third country. The resulting import statistics in this case would provide inaccurate information on Kenya's true exports. Table 2.3 Product Specific Discrepancies Between Kenya's Reported 2002 Exports and Partner Countries' Imports. Kenya's Reported Partners' Reported Exports Imports Difference Product (SITC) Value ($000) Share (%) Value ($000) Share (%) ($000) Total Exports (0 to 9) 1,400,372 100.00 2,079,638 100.00 -679,266 Tea (074.1) 140,907 10.06 456,852 21.97 -315,945 Cut flowers (292.7) 101,424 7.24 197,601 9.50 -96,177 Fresh vegetables (054.5) 72,051 5.15 136,056 6.54 -64,005 Coffee (071.1) 35,139 2.51 95,858 4.61 -60,719 Other outer garments (843.9) 80 0.01 63,664 3.06 -63,584 Fish fillets (034.4) 15,661 1.12 51,017 2.45 -35,356 Prepared fruit (058.9) 44,201 3.16 47,125 2.27 -2,924 Trousers (842.3) 224 0.02 34,860 1.68 -34,636 Bulbs and tubers (292.6) 23,201 1.66 30,546 1.47 -7,345 Iron plates (674.9) 26,370 1.88 27,298 1.31 -928 Motor spirits (334.1) 0 0.00 22,739 1.09 -22,739 Portland cement (661.2) 17,212 1.23 22,662 1.09 -5,450 Metallic salts (523.2) 3,176 0.23 22,539 1.08 -19,363 Medicaments (541.7) 6,811 0.49 20,439 0.98 -13,628 Gas oils (334.3) 0 0.00 19,852 0.95 -19,852 Fresh fruit (057.9) 5,451 0.39 18,845 0.91 -13,394 Calf skins (211.2) 2,761 0.20 17,802 0.86 -15,041 Fruit juices (058.5) 7,327 0.52 14,704 0.71 -7,377 Fresh fish fillets (034.3) 6,279 0.45 14,522 0.70 -8,243 Jerseys and pullovers (845.1) 352 0.03 13,454 0.65 -13,102 Prepared vegetables (056.5) 9,023 0.64 12,828 0.62 -3,805 Other vegetable material (292.9) 5,277 0.38 11,603 0.56 -6,326 Edible nuts (057.7) 5,179 0.37 11,351 0.55 -6,172 Beans and lentils (054.2) 1,753 0.13 11,204 0.54 -9,451 Quartz and mica (278.5) 3,162 0.23 11,200 0.54 -8,038 Common salt (278.3) 3,297 0.24 10,511 0.51 -7,214 Other outer garments (845.9) 499 0.04 10,392 0.50 -9,893 Sisal (265.4) 4,770 0.34 10,334 0.50 -5,564 Tobacco stripped (121.2) 20,433 1.46 10,224 0.49 10,209 Maize (044.0) 6,397 0.46 9,648 0.46 -3,251 Source: UN COMTRADE Statistics 12 Box 2.2. Inconsistencies in Uganda's and Kenya's Trade Statistics. Given the necessity of utilizing partner country statistics for analyses of Kenya's export performance, an important question is how reliable are the trading partners import data. One useful test concerns the consistency of the available statistics at different levels of aggregation. That is, are trade totals derived by aggregating import values at different SITC levels equal. That is, the sum of reported three-digit SITC imports should equal the corresponding for one or two digit products. If significant differences exist, the data may seriously bias studies in which they are used. Since Uganda is Kenya's largest export market in Africa its import statistics were used in a test. The data shown below indicate the value of trade derived from the sum of all four-digit SITC imports over 1995 to 2002, and the value of total imports Uganda independently reported to UN COMTRADE. Any differences between these totals indicates inconsistencies occur in the statistics. Kenya's Reported Uganda's Imports from Kenya ($000) Energy (SITC 33) Sum of four-digit Total Reported in Year Exports to Uganda SITC products UN COMTRADE Difference 1995 47,554 206,728 213,433 6,705 1996 65,456 193,859 250,403 56,544 1997 65,494 158,865 158,865 0 1998 81,428 187,203 265,576 78,373 1999 81,795 148,280 241,878 93,598 2000 67,018 287,125 287,125 0 2001 939 126,774 281,468 154,694 2002 192,593 145,927 312,865 166,938 2003 214,550 357,327 357,327 0 Source: Totals derived from UN COMTRADE statistics. The results of these comparisons must cause unease concerning the use of Uganda's partner country data. For example, in both 2001 and 2002 the difference between Uganda's total imports as reported in COMTRADE, and the sum of four-digit data actually exceed the latter, yet in 1997 and 2000 the differences vanish. An examination of the underlying statistics suggests part of the problem is seemingly attributable to refined petroleum products classified in SITC 33. In some years, it appears Uganda's customs authorities have not been able to identify specific petroleum products at lower SITC levels. However, it is not clear whether other product groups may have also been adversely by this "adding up" problem. Two other related points should be noted. First, major discrepancies between Kenya's export statistics also occur at different levels of the SITC system. For example, at the two-digit level Kenya reported 2002 energy exports of approximately $192 million to Uganda, yet the corresponding sum of reported trade at the four-digit level was under $1 million. Second, partner country data comparisons there are major errors in Kenya's reported exports to Uganda. For example, the reported 2001 value of energy exports ($939,000 see above) is inconsistent with the data for other years and about $156 million lower than the corresponding value of Uganda's reported imports. Finally, another important aspect of these statistical problems should be noted, that is, the very inconsistent COMTRADE reporting practices of many African countries. For example, Table 2.4 lists the 13 years that COMTRADE records are available for individual African countries. As a result, if changes in Kenya's exports were being analyzed over (say) the early 1980s to the present Uganda would have to be excluded from the partner country group since it failed to report any trade statistics to COMTRADE from 1977 to 1994. If it were included it would not be possible to distinguish actual trade changes from changes due to the composition of reporting countries. On a positive note, however, the African and non- African countries that could be included as Kenya's partners accounted for about 90 percent of world trade in 2000. The major shortcomings of the approach are, as Table 2.4 shows, that missing partner country data, and serious defects in the quality of available African statistics (see the annex to this section) make it very difficult to determine what is in fact happening in Kenya's intra-African trade. Table 2.4. COMTRADE Data Availability for Selected African Countries. Country ­ Years of Data Availability Country ­ Years of Data Availability Angola ­ 1969 to 1974, 1978 to 1981, 1985, 1990, 1991 Liberia ­ 1963, 1967, 1970 to1984 Botswana ­ 2000, 2001 Madagascar ­ 1962 to 1986, 1990 to 2003 Benin ­ 1962 to 1974, 1979, 1980, 1982, 1992 to 2002 Malawi ­ 1966 to 1991, 1994 to 2001 Burkina Faso ­ 1962 to 1983, 1995 to 2002 Mali ­ 1962 to 1972, 1974 to 1980, 1982 to 1987, 1989, Burundi ­ 1965, 1974 to 1976, 1993 to 2002 1990, 1996 to 2001 Cameroon ­ 1962 to 1980, 1982, 1986, 1987, 1989 to Mauritania ­ 1962 to 1968, 1970 to 1972, 1995, 1996 1990, 1995 to 2003 Mauritius ­ 1970 to 1978, 1980 to 2003 Cape Verde ­ 1978 to 1980, 1984. 1995 to 2001 Mozambique ­ 1994 to 1997, 1999, 2001 Chad ­ 1962 to 1975, 1995 Namibia ­ 2000, 2001 Central Africa Rep. ­ 1962 to 1971, 1973 to 1980, 1989, Níger ­ 1962 to 1981, 1995 to 2003 1993 to 2001 Nigeria ­ 1970 to 1987, 1996 to 2000 Comoros ­ 1962, 1995 to 2000 Rwanda ­ 1996 to 1999, 2001 to 2003 Cote d'Ivoire ­ 1962 to 1979, 1981 to 1983, 1985, 1995 SACU* ­ 1974 to 1985, 1992 to 1999 to 2000, 2002, 2003 Sao Tome & Principe ­ 1999 to 2003 Congo ­ 1962 to 1980, 1983 to 1986, 1993 to 1995 Senegal ­ 1962 to 1975, 1977 to 1981, 1986, 1987, 1989 Dem. Rep. Congo ­ 1970, 1972 to 1978, 1985, 1986 to 2002 Djibouti ­ 1962, 1980, 1981, 1983 to 1992 Seychelles ­ 1971 to 1996, 2001, 2002 Equatorial Guinea ­ no data available Sierra Leone ­ 1963, 1964, 1972 to 1976, 1983, 1984, Eritrea ­ 2000 to 2002 2002 Ethiopia ­ 1993, 1995, 1997 to 2002 Somalia ­ 1962, 1966, 1970 to 1982 Gabon ­ 1962 to 1971, 1975 to 1983, 1993 to 1994, South Africa ­ 2000 to 2003 1996, 1997, 1999 to 2000 Sudan ­ 1963 to 1982, 1984, 1985, 1992 to 2002 Gambia ­ 1964, 1970 to 1980, 1995 to 2000 Swaziland ­ 2000 to 2002 Ghana ­ 1962 to 1984, 1992, 1996 to 2002 Tanzania ­ 1976 to 1981, 1987, 1995 to 2003 Guinea ­ 1885 to 2002 Togo ­ 1962 to 1981, 1983, 1986 to 1991, 1994 to 2002 Guinea-Bissau ­ 1970 to 1972, 1975, 1976, 1995 Uganda ­ 1976, 1994 to 2003 Kenya ­ 1976 to 1988, 1990 to 2000, 2002, 2003 Zambia ­ 1966, 1967, 1970 to 1979, 1993, 1995 to 2002 Lesotho ­ 2000 to 2002 Zimbabwe ­ 1984 to 1986, 1990, 2002 * Members of SACU began independently reporting trade statistics in 2000. Source: UN COMTRADE Statistics. 14 C. Analyzing Kenya's Export Performance: What Can Be Done? Key Point As a result of the serious analytical problems with Kenya's own COMTRADE records, its export performance will have to be analyzed using the import statistics of its trading partners. While this approach appears to work reasonably well for trade with medium and high income countries, very serious reservations must exist concerning any conclusions about Kenya's intra-African trade. A major initiative is needed to improve the accuracy and reliability of African trade statistics. As indicated, two major trade data constraints significantly restrict the analyses that can be undertaken on Kenya's export performance. The most serious problem relates to Kenya's own COMTRADE records where a failure to account for shipments from the export processing zones, as well as other large unexplained discrepancies in the available statistics, render these data all but useless for analytical purposes. Second, evidence of serious deficiencies have been found in the import statistics of many of Kenya's African country trading partners (see IMF 2000 or UN WIDER 2000). These include the failure to regularly report trade data to UN COMTRADE, evidence that intentional under-invoicing and unrecorded trade have biased available statistics, and the occurrence of major inconsistencies in available United Nations data at different levels of aggregation. For these reasons, little can be said with any degree of certainty about the magnitude and composition of Kenya's intra-African trade. This is unfortunate because many countries have devoted considerable time and effort to negotiations aimed at creating regional trade arrangements. Given the quality of the available data it is not possible to determine if these efforts were misdirected, or what the trade effects of the RTAs actually were. These observations raise a key question, that is, given the limitations of the available COMTRADE statistics what analyses can be undertaken on Kenya's trade performance. The possibilities increase if both IMF and United Nations trade data are employed for specific purposes; · International Monetary Fund Direction of Trade (DOT) statistics provide information on Kenya's aggregate (total) exports to individual countries and geographic regions. This source contains estimates for trade flows where actual COMTRADE data are missing (as is the case with Uganda in the 1980s), or where the official data appear to be highly inaccurate or incomplete (see Box 2.1 for the IMF assessment of Kenya's COMTRADE records). The IMF statistics provide useful comprehensive information for analyses of the overall level and direction of Kenya's total trade with both African and non-African countries. Their drawback is that the provide no information on the composition of trade. · Any analyses involving the composition of Kenya's exports must be based on available partner country COMTRADE records even though, as noted, these data may have serious deficiencies. One key requirement is that the countries included as partners must have consistently reported to the UN without gaps in their statistics. If gaps did existed it would be difficult to determine if observed changes in the statistics reflected a true change in Kenya's exports, or were the result of a change in the composition of reporting partner countries.8 One key point should be noted! The exclusion of non-reporting or 8 The following have consistently reported trade statistics to UN COMTRADE and will be used as the partner country group for analyzing Kenya's export performance; all original OECD members, Hong Kong, China, Republic of Korea, Singapore, Mexico, Malaysia, Brazil, Turkey, Indonesia, Israel, Hungary, Poland, Philippines, Pakistan, Argentina, Colombia, Chile, Peru, Cyprus, Ecuador, Egypt, Bangladesh, Guatemala, Malta, Macau, Mauritius, Uruguay, Romania, Trinidad and Tobago, El Salvador, Honduras, Panama, Nicaragua, Barbados, Bolivia and Greenland. 15 irregularly reporting countries ensures trade totals derived from partner country statistics will be lower than totals derived from International Monetary Fund data. · Issues relating to Kenya's intra-regional African trade are too important to be neglected entirely. Iin spite of the severe limitations of the available statistics. An effort will be made to determine what information can be gleaned from these data, although the conclusions may be highly inaccurate and misleading. Annex 2.1a. On the General Accuracy of African Trade Statistics Key Point As a result of the failure of many African countries to provide trade statistics to the United Nations, and the serious defects in the available data, little can be said with any certainty about Kenya's trade with other African countries, or about the level and composition of intra-trade in Africa. The official statistics are what the Financial Times calls "guesstimates." The possibility exists that some African countries trade policies, including Kenya's, have been seriously misdirected by the defects in available statistics. In the face of growing concerns about the reliability of UN trade statistics for Sub-Saharan countries, the World Bank previously initiated a major reconciliation study for 36 African countries' import and export data (see World Bank Economic Review, vol. 2, pp 135-156). In a feature article entitled "Enter the Twilight Zone of African Trade Statistics" the Financial Times provided its interpretation of the findings. Among the major conclusions were the following points, First, such wide differences exist in the matched partner country statistics that the data cannot be used to assess the overall level of African intra-regional trade. Among the factors responsible were smuggling, inefficient customs reporting practices, and intentional false invoicing by importers and exporters to avoid customs duties and taxes, or facilitate capital flight. Second, the COMTRADE data analyzed by the World Bank are probably equally useless for assessing the direction of intra-trade, since the countries reported by an exporter as the major destination of trade often failed to report any corresponding imports. Third, the COMTRADE data were equally deficient for determining the composition of trade, since very large discrepancies are often observed in matched partner country statistics at lower levels of detail. For example, at the four-digit SITC level differences exceeding several hundred percent were frequently found in matched data. Fourth, there often are sizeable and persistent differences in the underlying trends of African countries' partner data, as reflected in exports and matched imports, that indicates these statistics may not accurately reflect either the magnitude or the direction of trade changes. In other words, the export statistics of a specific country may indicate trade has been rising over time, while the import statistics of a trading partner may indicate the opposite (see Annex Table 2.1a). Finally, the Times acknowledged that errors of the magnitude observed in the Bank study could adversely influence government policies relating to investment, balance of payments, initiatives for the 16 liberalization of trade barriers, exchange rate policy, and a host of other factors that affect a nation's industrialization and growth. Specifically, many African countries do not have adequate and reliable statistics required to formulate appropriate trade and trade related investment policies. Annex Table 2.1a draws on East African trade data to illustrate the often contradictory nature of SSA countries' COMTRADE statistics. The table shows the value of Kenya's reported annual exports to Tanzania from 1995 to 2002 along with matched statistics on Tanzania's imports from Kenya. Is this bilateral exchange increasing or decreasing? The answer depends solely on which country's trade statistics are used for analysis. According to Kenya's trade data, exports to Tanzania fell by about 50 percent over 1995 to 2002. In contrast, the Tanzanian import statistics suggest trade with Kenya increased by about $27 million, or almost 40 percent, during this same interval. Also, Kenya's reported exports are much higher (almost $150 million in 1995) than Tanzania's matched import statistics. This is contrary to what was expected given the formers failure to include EPZ shipments in its export records. Annex Table 2.1a. Partner Country Comparisons of Kenya and Tanzania's Total Imports and Exports Value of Reported Trade ($ 000) Kenya's Reported Tanzania's Reported Percentage Year Exports to Tanzania Imports From Kenya Difference Difference (%) 1995 216,012 69,589 146,423 210.4 1996 238,687 92,589 146,098 157.8 1997 238,502 96,566 141,936 147.0 1998 228,605 96,065 132,540 138.0 1999 162,949 106,217 56,732 53.4 2000 114,495 85,805 28,690 33.4 2001 108,364 95,880 12,484 13.0 2002 106,968 96,169 10,799 11.2 2003 199,641 116,503 83,138 71.4 Note: Tanzania failed to report any trade data to the United Nations from 1988 to 1994. What were Kenya's major exports to Tanzania? Again, the answer depends on whose statistics are used for analysis. Annex Table 2.2a lists the ten largest four-digit SITC export products reported in Kenya's 1995 COMTRADE data, and also shows the matched value of Tanzania's reported imports of these items. For example, Kenya reported exports of $21 million for its second largest product (iron plates and sheets) which was about $18 million more than the corresponding value of Tanzania's reported imports. Differences of over 1,000 percent occur for six of the ten products Kenya is reporting as major exports to Tanzania. Quite clearly, these statistics cannot be used with any confidence to indicate either the level, composition or direction of Kenya and Tanzania's bilateral trade. A key point to note is that major inconsistencies in African COMTRADE records may seriously bias any conclusions concerning these countries' intra-trade. As an illustration, Annex Table 2.3a compares Uganda's reported 2002 total exports to the matched imports of 33 Sub-Saharan African partner countries while Annex Tables 2.4a and 2.5a provide similar statistics for Tanzania and Kenya. Major discrepancies are observed in all three tables. For example, Uganda's reported exports to Kenya ($61 million) are almost 8 times the corresponding value of Kenya's imports. Similarly, Uganda's exports to South Africa are about $40 million higher than the latter's matched import total. Altogether, Uganda's African exports are $51 million, or 57 percent higher, than the reported imports of the partner countries. 17 Annex Table 2.2a. Partner Country Data Comparisons for Kenya's Ten Largest 1995 Exports to Tanzania Value of Reported Trade ($ 000) Kenya's Reported Tanzania's Reported Percentage Product (SITC) Exports to Tanzania Imports From Kenya Difference Difference (%) Sugar confection (062.0) 5,834.5 73.5 5,761.0 * Beer (112.3) 24,760.5 6,242.4 18,518.1 296.7 Palm oil (424.2) 6,613.8 46.8 6,567.0 * Palm kernel oil (424.4) 6,804.2 13.7 6,790.5 * Animal & vegetable oil (431.2) 5,494.4 469.4 5,025.0 * Medicinal products (541.7) 9,167.9 3,277.9 5,890.0 179.7 Toilet preparations (553.0) 5,182.3 17.1 5,165.2 * Soaps & detergents (554.1) 15,250.6 261.8 14,988.8 * Iron bars and rods (673.2) 6,111.7 976.3 5,135.5 526.0 Iron plates and sheets (674.9) 21,066.7 2,892.3 18,174.4 628.4 ALL GOODS 216,012.0 69,589.0 146,423.0 210.4 * The percentage difference between Kenya's reported exports and Tanzania's imports for this product exceeds 1,000 percent. Source: UN COMTRADE Statistics. As a result of these and numerous other comparisons, the Financial Times concluded that "Unless African governments ­ and the donor community ­ devote more resources to data gathering and processing, most statistics from the continent will continue to be little more than highly inaccurate and unreliable guesstimates!" African countries have attached a high priority to the expansion of intra- trade. However, the available statistics appear to be of little or no utility for monitoring the level or composition of this exchange. 18 Annex Table 2.3a. Comparisons of 2002 Exports Reported by Uganda to UN COMTRADE with the Matched Imports of African Trading Partners (values in US$) Uganda's Reported Partner Country Exports to Partner Reported Imports From Reported Trade Partner Country Country Uganda Differences Angola* 0 0 0 Botswana* 15,163 86,145 -70,982 Benin 0 4,842 -4,842 Burkina Faso 0 0 0 Burundi 6,265,579 1,238,176 5,027,403 Cameroon 453,245 192,401 260,844 Cape Verde* 0 0 0 Central African Republic* 0 0 0 Cote d'Ivoire 189,920 35,122 154,798 Eritrea 1,959,751 2,535,938 -576,187 Ethiopia 217,710 65,073 152,637 Ghana 19,979 0 19,979 Guinea 0 0 0 Kenya 61,489,224 8,787,819 52,701,405 Lesotho 10,929 0 10,929 Madagascar 1,760 7,869 -6,109 Malawi* 18,674 378,368 -359,694 Mali* 21,622 14,606 7,016 Mauritius 0 85,932 -85,932 Mozambique* 74,388 0 74,388 Namibia* 12,429 11,217 1,212 Niger 5,000 0 5,000 Rwanda 12,868,615 11,055,005 1,813,610 Senegal 10,967 0 10,967 Seychelles 1,058,927 28,888 1,030,039 Sierra Leone 40,241 61,080 -20,839 South Africa 42,990,744 1,896,357 41,094,387 Sudan 5,761,131 15,109,655 -9,348,524 Swaziland 47,979 0 47,979 Tanzania 5,770,956 2,681,991 3,088,965 Togo 0 16,575 -16,575 Zambia 2,153,088 6,852,569 -4,699,481 Zimbabwe 52,623 12,124 40,499 All Above Countries 141,510,644 51,157,752 90,352,892 * These partner country comparisons utilize 2001 trade statistics since 2002 data have not yet been reported to UN COMTRADE. 19 Annex Table 2.4a. Comparisons of 2002 Exports Reported by Tanzania to UN COMTRADE with the Matched Imports of African Trading Partners (values in US$) Tanzania's Reported Partner Country Exports to Partner Reported Imports From Reported Trade Partner Country Country Tanzania Differences Angola* 503,528 0 503,528 Botswana* 224,308 406,452 -182,144 Benin 0 286,782 -286,782 Burkina Faso 0 8,577 -8,577 Burundi 7,082,692 13,446,214 -6,363,522 Cameroon 56,117 0 56,117 Cape Verde* 0 0 0 Central African Republic* 0 0 0 Cote d'Ivoire 0 229,009 -229,009 Eritrea 237,275 402,522 -165,247 Ethiopia 374,962 290,796 84,166 Ghana 3,293 0 3,293 Guinea 110,252 81,599 28,653 Kenya 35,710,092 9,991,641 25,718,451 Lesotho 0 0 0 Madagascar 696,403 826,955 -130,552 Malawi* 5,588,120 4,698,771 889,349 Mali* 56,433 0 56,433 Mauritius 293,494 649,563 -356,069 Mozambique* 1,394,796 787,060 607,736 Namibia* 34,869 49,767 -14,898 Niger 298,879 107,699 191,180 Rwanda 3,909,126 13,157,431 -9,248,305 Senegal 182,872 528,920 -346,048 Seychelles 287,373 274,666 12,707 Sierra Leone 1,236,484 24,784 1,211,700 South Africa 16,706,183 9,000,892 7,705,291 Sudan 272,256 287,556 -15,300 Swaziland 374,782 0 375,782 Togo 0 11,729 -11,729 Uganda 5,549,107 7,502,562 -1,953,455 Zambia 17,637,810 14,897,858 2,739,952 Zimbabwe 1,418,304 3,761,195 -2,342,891 All Above Countries 100,239,810 81,711,000 18,529,810 * These partner country comparisons utilize 2001 trade statistics since 2002 data have not yet been reported to UN COMTRADE. 20 Annex Table 2.5a. Comparisons of 2002 Exports Reported by Kenya to UN COMTRADE with the Matched Imports of African Trading Partners (values in US$) Kenya's Reported Partner Country Exports to Partner Reported Imports From Reported Trade Partner Country Country Kenya Differences Angola* 388,960 0 388,960 Botswana* 104,042 339,719 -235,677 Benin 67570 40,070 27,500 Burkina Faso 795401 1,592,066 -796,665 Burundi 16,368,473 15,646,315 722,158 Cameroon 187,820 112,179 75,641 Cape Verde* 0 0 0 Central African Republic* 0 74,020 -74,020 Cote d'Ivoire 0 1,175,170 -1,175,170 Eritrea 2,789,029 7,037,710 -4,248,681 Ethiopia 11,771,014 17,416,000 -5,644,986 Ghana 389,060 0 389,060 Guinea 207,883 566,829 -358,946 Lesotho 18,606 8,565 10,041 Madagascar 7,293,081 5,799,995 1,493,086 Malawi* 4,413,561 5,314,763 -901,202 Mali* 466,236 1,747,158 -1,280,922 Mauritius 0 2,522,304 -2,522,304 Mozambique* 971,093 815,847 155,246 Namibia* 22,467 140,384 -117,917 Niger 2,539 8,615 -6,076 Rwanda 39,942,540 66,702,652 -26,760,112 Senegal 577,853 499,369 78,484 Seychelles 2,218,096 1,535,145 682,951 Sierra Leone 7,813 8,313 -500 South Africa 8,818,163 10,214,928 -1,396,765 Sudan 23,508,176 32,806,110 -9,297,934 Swaziland 25,091 723,052 -697,961 Tanzania 106,969,480 96,168,440 10,801,040 Togo 15,758 5,152 10,606 Uganda 301,666,208 312,727,360 -11,061,152 Zambia 10,058,311 11,703,599 -1,645,288 Zimbabwe 2,315,141 12,648,989 -10,333,848 All Above Countries 542,379,465 606,100,818 -63,721,353 * These partner country comparisons utilize 2001 trade statistics since 2002 data have not yet been reported to UN COMTRADE. 21 III. IMPLICATIONS OF THE MARGINALIZATION OF AFRICA Key Point During the last quarter century Sub-Saharan African countries exports either declined in absolute terms, or expanded at a slower pace than world trade. Evidence suggests most African countries, including Kenya, experienced serious supply constraints that limited their ability to capitalize on opportunities in foreign markets. Inappropriate governance policies, and unfavorable internal commercial environments, were largely responsible for Africa's supply problems. However, a related problem is that many of Africa's traditional exports face unfavorable long-term global demand prospects that resulted in low (or even negative) growth and declining real prices. If Africa's marginalization is to be halted the region must diversify into new, more promising, exports and also significantly improve the internal "commercial environment" in most countries. However, there is little evidence that any of these crucial changes are occurring, For analytical purposes it would be inappropriate to attempt an analysis of Kenya's export performance and prospects without some general discussion of the longer-term problems encountered by almost all Sub-Saharan African countries. Perhaps the most striking problem has been the persistent marginalization, or diminished global importance of the region. For example, Sub-Saharan Africa's share of world trade fell dramatically over the last quarter century. As Table 3.1 shows, the region's share of global exports went from 2.3 percent in 1975 to 1.4 percent in 2003. If Africa had been able to maintain its 1975 trade share, its exports would have been about $225 billion in 2003 as opposed to their actual value of approximately $108 billion. On average, the global non-oil products market shares for the 15 largest African exporters fell by about 55 percent with Zambia and the Democratic Republic of the Congo's shares declining by over 80 percent. In contrast, East Asia was clearly the most dynamic region as its world trade share rose more than four fold to its current level of approximately 20 percent. Perhaps the most striking feature of Africa's marginalization in global trade is the major erosion of the region's ability to compete in international markets for many of its traditional exports. For example, in 1962-64 copper alloys were the regions single largest commodity export, with Sub-Saharan Africa supplying 32 percent of the original OECD countries' imports. By the early 1990s, Africa's market share dropped more than 22 percentage points to less than 10 percent. Similarly, Africa's market shares for other key commodities such as vegetable oils, palm oil, palm nuts and kernels, and groundnuts dropped between 47 to 80 percentage points below earlier levels. For the thirty largest non-oil exports combined, Africa's average share declined by more than 11 percentage points (from 20.8 percent to 9.7 percent), which implies annual trade losses of about $11 billion for these specific goods. Another major adverse factor affecting Africa's exports was the well below average growth in global demand for many of the region's major traditional products. From the early 1960s to the 1990s world trade in non-fuel goods increased at a compound annual rate of 11.8 percent; for the major African products the rate was about 4.5 percent points lower. Thus, Africa suffered a double blow ­ its apparent supply constraints led to declining global market shares for its major export products that, in turn, were of declining importance in world trade due to their low income elasticities. For example, Ng and Yeats (2002) estimate that the global income elasticity of demand for Kenya's tropical beverage crops (tea and coffee) averaged about 0.70 during the last two decades. This implies that import demand for these products should grow at a rate 30 percent below the rate of growth in world income. Elasticity estimates for hides, leather, hard fibers, and other African raw material exports were considerably lower. Three recent studies commissioned by the World Bank's Africa Region attempted to identify major problems and prospects for Sub-Saharan African countries' exports. The initial investigations focused primarily on the identification of general production and supply constraints in Africa, while the third examined longer-term demand prospects for Sub-Saharan Africa's traditional export products. The 22 major conclusions, which are summarized below, may also have a direct bearing on Kenya's trade prospects, are as follows. Table 3.1. The Relative Importance of Sub-Saharan Africa and Other Regions in World Trade Total Exports ($billion) Share of World Trade (%) Group 1975 1985 1995 2003 1975 1985 1995 2003 Australia/ New Zealand 14.1 28.3 66.6 87.1 1.8 1.5 1.3 1.2 East Asia1 39.5 194.2 896.6 1,517.4 5.2 10.5 17.8 20.2 Of which: ASEAN 22.0 72.5 322.8 481.9 2.9 3.9 6.4 6.4 European Union (15) 340.6 708.2 2,018.3 2,879.5 44.6 38.1 40.0 38.2 Other Western Europe 22.2 57.1 147.9 219.8 2.9 3.1 2.9 2.9 East Europe/Central Asia 24.2 115.9 205.4 404.8 3.2 6.2 4.1 5.4 Japan 55.7 177.2 443.0 473.9 7.3 9.5 8.8 6.3 Latin America 41.7 100.1 229.2 415.7 5.5 5.4 4.5 5.5 Middle East 45.7 99.1 152.1 285.4 6.0 5.3 3.0 3.8 NAFTA 144.6 326.0 853.2 1,160.1 19.0 17.6 16.9 15.4 North Africa 13.2 27.3 28.6 57.2 1.7 1.5 0.6 0.8 South Asia 6.5 13.6 46.0 85.8 0.9 0.7 0.9 1.1 Sub-Saharan Africa 17.7 32.4 42.0 108.4 2.3 1.7 0.8 1.4 WORLD EXPORTS 762.8 1,857.4 5,049.3 7,530.2 100.0 100.0 100.0 100.0 1East Asia is defined here as consisting of: Brunei, Cambodia, China, Republic of Korea, Hong Kong, Indonesia, Lao PDR, Malaysia, Mongolia, Philippines, Singapore, Taiwan (China), Thailand, and Vietnam. Source: International Monetary Fund Direction of Trade Statistics A. Africa's Production and Supply Problems Key Point A recent survey by the Africa Region Enterprise Development Initiative concluded internal factors like corruption, cost of finance (e.g. interest rates), crime, tax rates, economic and regulatory policy uncertainty, and macroeconomic instability (inflation and exchange rates) were having a serious negative effect on the expansion of Kenya's export production. Similar surveys by the Wall Street Journal and Heritage Foundation determined that these types of problems were endemic in Sub-Saharan Africa. As a result, Africa is at a major disadvantage in its efforts to attract foreign investment required for the expansion and diversification of its exports. Sub-Saharan African Countries continue to experience major supply constraints that limit their ability to capitalize on opportunities in foreign markets. The importance of these production problems was reflected in the following observations. · During the last three decades global Sub-Saharan African exports either declined in absolute terms or expanded at a slower pace than world trade. The displacement of exports by more efficient non- regional producers played a major role in this decline. More recently, the World Bank (2003, p. 159) estimated that Kenya's global market share losses from 1997 to 2001 reduced the growth rate of exports 23 by about 4 percent per year. The erosion of Tanzania and Uganda's market shares for their major exports generated even larger losses. · No major expansion occurred in the diversity of products exported by most Sub-Saharan African countries. Indeed the product composition of some African countries' exports became more concentrated over the last decade. Africa continues to be heavily dependent on a relatively few commodities which have been the region's traditional exports, and which appear to have relatively poor growth prospects. · Recent changes in Africa's exports indicates no general increase occurred in the number of industries in which most countries have a "revealed" comparative advantage. This is consistent with statistics showing Africa generally failed to diversify its export base and, in several countries, trade became more concentrated. That is, a fewer number of products were being exported at the end of the last decade than at the beginning. · There is little evidence that the relative importance of exports of processed domestically produced commodities increased, nor do the data indicate that the intra-industry trade between Africa and other regions grew significantly.9 Although other studies suggest that rapidly growing international trade in parts and components has been a major factor promoting interdependence and globalization, Africa generally is not significantly increasing its participation in this activity. · Surveys by the Africa Region Enterprise Development Initiative (2003) report internal factors like corruption, cost of finance (e.g. interest rates), crime, tax rates, economic and regulatory policy uncertainty, and macroeconomic instability (inflation and exchange rates) were viewed as having a serious negative effect on the expansion of export production. Related surveys by the Wall Street Journal and Heritage Foundation determined that the commercial, trade and investment environment in African countries was generally inferior to that in most other developing countries. As a result, Africa is at a major disadvantage in its efforts to attract foreign investment. B. Long-Term Demand Prospects for Traditional Exports Key Point There has been a long-term deterioration in real prices for many key commodities, like tea, coffee, and hard fibers, which are of primary importance to Kenya and other Sub-Saharan countries. Recent projections by the World Bank indicate this trend should continue through the present decade. Price and export earnings instability has been a major problem for African commodity exporting countries. The policy prescription is simple ­ diversify exports!! Former major commodity producing countries like Brazil, Thailand, Malaysia, China or Mauritius made marked shifts over the last two or three decades in the composition of their exports, moving from 9 Proponents of so called "natural resource based" industrialization strategies argue that further local processing of domestically produced agricultural products and raw materials could provide an important stimulus to industrialization and growth. The rationale is that further local processing can have important employment creation effects, it may increase trade contacts and provide benefits associated with "outward oriented" trade policies, there may be important linkage effects to other sectors of the economy, and it may generate increased export and foreign exchange earnings. However, there is little evidence to suggest a major increase in domestic processing of natural resources is occurring in Africa. See Helleiner and Welwood (1978), UNCTAD (1975), or Balassa (1968) for useful background information. 24 primary commodities to manufactures. This raises the possibility that prospects for the remaining exporters of commodities, like those in Africa, may have improved. However, the World Bank's analysis of demand prospects for Africa's traditional (commodity) exports revealed the following points; · Over the last decade, global trade in Sub-Saharan Africa's traditional commodity exports, which currently account for about three-quarters of the region's total non-oil exports, grew at a rate of 1.9 percent, or about one-third the corresponding rate for all goods. Products that the World Bank (2002, p. 12) identified as traditional exports included; tropical beverages like tea and coffee, nonferrous and ferrous metal and ores, fresh and processed sea foods, other foodstuffs like sugar, hides and leather products, minerals and products, lumber, and natural fibers. Below average growth rates for these goods is a continuation of trends observed for the 1980s. Furthermore, 1990-1999 trade growth rates for over 40 percent of Africa's traditional products were actually negative. · During the 1990s, world trade growth (5.7 percent) was roughly double that for income as measured by GDP. On average, income elasticity estimates for Africa's traditional exports were just over one-third (0.36). This implies that if world GDP expands at the recent rate of 2.5 percent, global trade in traditional products should grow by under one percent per year. As such, continued reliance on traditional exports will extend Africa's marginalization in world trade. · The recent record provides no indication that the longer-term deterioration in traditional product prices has reversed. Over 1990-1999, the World Bank estimated average real prices for all traditional products declined by about 24 percent. In a few cases, like coffee and lumber, where some modest improvement occurred, real prices still remain well below their 1980 levels. However, long-term price projections by the World Bank (2000) reinforce the basic negative outlook for traditional products and most other commodities, including two of Kenya's major exports (coffee and tea). The Bank forecasts that by 2010 real coffee prices will have fallen to about one-third their level in 1980, and that real prices for tea will have declined by about 50 percent (Ng and Yeats 2002, p. 48) · African traditional product price instability is a major problem for all exporters including Kenya. Average annual price changes for these goods generally exceeded those for the all non-oil commodity price index, while one-half of the traditional products experienced average price changes that were at least 50 percent greater. However, annual data clearly understate instability problems since traditional product prices often experience sizeable consecutive year directional changes. Over a three year period, consistent directional price shocks as high as 101 percent occurred, while changes of 150 percent were observed in four consecutive year data. These major price swings were generally associated with a "collapse" of traditional product prices as, over 80 percent of the time they were in a downward direction.10 Such a price collapse recently occurred for Kenya's coffee exports. The policy message from these findings for Africa is simple and direct. Diversify away from traditional products or continue to experience serious negative trade effects including; (i) declining or relatively low growth in global demand for these goods, (ii) falling real prices for traditional products, (iii) very unstable prices and export earnings, (iv) a continued marginalization in world trade, and (v) diminished growth and industrialization prospects. However, there is little evidence that any general 10A relatively large number of studies have analyzed the influence of commodity price instability on developing countries. Early examples include MacBean (1966), Massel (1970) and Michaely (1962). A major concern is the "boom" and "bust" price cycles of many commodities has numerous adverse financial effects on developing country exporters. Policy initiatives like the European Unions STABEX program, or UNCTAD's proposal for a Common Fund for Commodities were aimed at reducing the detrimental effects of volatile commodity price changes. 25 diversification is occurring. Section X in this report discusses several strategies that could promote an expansion and diversification of Kenya's exports. C. Changes in the Relative Importance of East African Countries Key Point Global trade shares for Kenya, Tanzania and Uganda fell by almost fifty percent from their level in the early 1980s to the present. There is no evidence that the marginalization of East African countries differed substantially from that of Sub-Saharan Africa as a whole. However, some evidence suggests that recently adopted programs in favor of Africa, like the US African Growth and Opportunities Act (AGOA) may have halted the longer-term declines, at least in North America. Unfortunately, the recent phase-out of the Multifiber Arrangement may erode these recent gains since it will bring Kenya into more direct competition with countries like China and India.. Given the objectives of this report, an important question whether Kenya, and its neighboring countries, experienced the same general deterioration in their global trade shares as the Sub-Saharan African region. Table 3.2, which is based on UN partner country statistics, provides relevant information by showing each East African countries average share of global exports of (i) all non-petroleum products, (ii) all foodstuffs, and (iii) all manufactured goods in a 1980-85 base period. The numbers in the body of the table show each country's corresponding export shares for selected years in the 1990s and 2000s relative to the average share in the base period. For example, in 1980-85 the global export share for all Kenya's non-oil products was 0.066 percent. Its corresponding share in 1990 was 57.3 percent of that in the base period, while in 2003 its share had further fallen to about 44.6 percent of that of the 1980-1985 average. Table 3.2 confirms that the global trade shares for all three African countries declined sharply from 1980-85 to 2003, but the fall in the relative importance of both Tanzania and Uganda was greater than that for Kenya. In 2003, Tanzania's share of world non-oil exports declined to just over one-quarter of its level in the base period, while the corresponding decline was even larger for Uganda. However, a positive point is that both Kenya and Uganda's global trade shares for manufactured goods recently rose to, or above their 1980-85 base period levels. Evidence suggests that recently adopted programs, like the United States African Growth and Opportunity Act (AGOA), had a positive influence on both Kenya and Uganda's exports of manufactures (see Section V that follows). These gains were largely concentrated in the textile and clothing sectors and may be largely diluted by the removal of MFA restrictions facing the exports of countries like China and India. IV. THE GEOGRAPHIC DIRECTIONS OF KENYA'S TRADE Key Point Just as countries with highly concentrated product export structures are often advised to diversify, similar concerns apply to the direction of trade. If a relatively high share of a country's trade is with a specific geographic region it may be vulnerable to any local adverse economic effects (like the recent East Asian financial crisis). In addition, some studies have argued that a high geographic concentration of trade may lead to higher import, or lower export, prices. Previous World Bank studies suggest former African colonies that maintained an above average trade dependence with the metropolitan center generally paid higher than average import prices. 26 There are several reasons why a country should be concerned about the direction of its trade. For example, some industrial countries like the United States have expressed concerns about their dependence on the Middle East for petroleum imports, and the adverse consequences that would occur if these supplies were interrupted. Similarly, non-member countries' exports to markets where regional trade arrangements were strengthening or expanding (like the European Union or NAFTA) might experience displacement due to the exchange of country specific preferences. Several analyses stress the need forcountries to diversify the origins, and destinations, of their trade to avoid unfavorable monopoly effects associated with excessive concentration. For example, a World Bank study (Yeats 1990) found Sub- Saharan African countries pay import prices 10 to 20 percent above average due to their reliance on a limited number of suppliers. Geographic diversification has also been advised as a means of mitigating effects of unfavorable economic developments within a few major markets. For example, countries whose trade was strongly oriented toward East Asia probably were negatively affected by the region's financial crisis in the late 1990s. Finally, the last two decades witnessed a proliferation of regional trade arrangements (RTAs) among African countries whose objective was to stimulate trade between members. Is there any evidence that the RTAs are having any effect on Kenya's trade. A. The Destinations of Kenya's Exports Key Point Some recent analyses of the direction of Kenya's exports concluded there has been a positive change in the relative importance of other East African countries as markets. This assertion is open to question on two points. First, International Monetary Fund statistics indicate that East Africa's import shares are highly erratic, but now are at about the same level as they were in the mid-1970s. Second, most analyses failed to account for the importance of refined petroleum products that now constitutes approximately one-half (by value) of Kenya's exports to Uganda and Tanzania. Although the available statistics are of uncertain quality, it appears that Kenya's non-oil exports to East Africa were highly concentrated and have grown relatively slowly. Table 4.1 employs IMF Direction of Trade (DOT) statistics to examine the geographic destinations of Kenya's exports from 1975 to 2003. For comparison, the table also provides similar information on the direction of trade for all Sub-Saharan African countries combined. The data indicate East Africa and the European Union (15) are the two largest destinations for Kenya's trade with each absorbing about one-third of total exports. Second, the geographic direction of trade has been relatively stable over 1975-2003, with the exception of South Asia and NAFTA where trade shares rose by 4 percentage points or more. An examination of available COMTRADE statistics indicates the increase for South Asia was largely accounted for by growing tea exports to Pakistan (and to a lesser extent to India), while the increase for NAFTA is largely the result of increased garment exports under United States African Growth and Opportunity Act (AGOA) preferences. Table 4.1 reports that approximately one third of Kenya's current exports now go to other East African countries, but this figure is inflated by a recent surge in refined petroleum products. In 2003, Kenya exported about $250 million in refined petroleum to Uganda and Tanzania from its Mombassa distillery. If refined petroleum products are excluded, it appears that less than one-quarter of Kenya's recent exports went to East Africa. Non-metallic mineral products like cement and lime, paper and paperboard, iron and steel, and medicinal products were among Kenya's largest regional exports (see Table 9.1 that follows). 27 Table 3.2 Changes in the Relative Importance of Kenya, Tanzania and Uganda in World Trade KENYA TANZANIA UGANDA Ratio of the Current Year Share to the Ratio of the Current Year Share to the Ratio of the Current Year Share to the Average 1980-1985 Share of World Trade Average 1980-1985 Share of World Trade Average 1980-1985 Share of World Trade All Non-Oil All All All Non-Oil All All All Non-Oil All All Year Goods Foods Manufactures Goods Foods Manufactures Goods Foods Manufactures 1990 0.573 0.708 0.723 0.443 0.485 0.681 0.276 0.341 0.179 1992 0.528 0.634 0.686 0.388 0.388 0.500 0.205 0.240 0.538 1994 0.583 0.649 1.042 0.385 0.426 0.721 0.387 0.546 0.129 1996 0.606 0.712 0.996 0.435 0.516 0.540 0.465 0.632 0.639 1998 0.554 0.718 0.892 0.405 0.637 0.639 0.324 0.475 0.816 2000 0.436 0.637 0.655 0.331 0.589 0.607 0.218 0.351 0.394 2001 0.486 0.623 0.795 0.350 0.559 0.491 0.223 0.321 0.595 2002 0.499 0.600 0.893 0.346 0.522 0.602 0.245 0.341 0.986 2003 est. 0.446 0.557 0.984 0.279 0.389 0.532 0.191 0.291 1.262 Memo Item: 1980-85 Ave. 0.065 0.352 0.012 0.035 0.162 0.004 0.029 0.194 0.001 Share Source: Partner countries statistics from COMTRADE Note: Numbers in the text of the table show the share of the country's exports in any given year in the 1990s and 2000s relative to the corresponding average share in the 1980-1985 base period. For example, during 1980-85 Kenya's estimated global trade share for all non-oil exports was about six-tenths of one percent (0.065). By 2002, this share had fallen to about one-half (0.449) of its earlier value 28 Table 4.1. The Geographic Destinations of Kenya's Exports; 1975 to 2003 Share of Exports Destined For (%) Of Global Of which: Sub- which: Rest Exports East EU South Saharan East of Country Year ($ mill.) Asia China ASEAN Japan (15) NAFTA Oceania Asia Africa Africa World Kenya 1975 601.7 3.6 0.6 1.9 2.0 31.5 6.0 0.7 2.7 34.7 33.8 12.7 1985 957.5 1.4 0.4 0.9 0.8 47.7 5.8 0.3 7.1 22.0 21.3 7.9 1995 1,826.0 2.5 0.1 1.1 0.7 35.6 3.4 0.3 7.1 41.9 41.5 0.9 2003 2,535.7 3.2 0.3 1.4 0.9 32.6 10.0 0.5 8.8 32.1 31.2 1.6 All Sub 1975 17,818.4 1.5 0.6 0.6 3.8 51.9 19.0 0.3 0.4 6.0 2.3 17.0 Saharan 1985 32,454.0 2.4 0.7 0.7 1.7 58.3 18.4 0.1 0.5 5.9 1.9 12.6 Africa 1995 41,860.3 6.9 1.3 2.8 2.9 42.4 25.0 0.1 2.9 11.0 5.4 8.7 2003 109,596.1 13.2 6.3 2.3 4.7 32.9 22.3 0.9 3.8 11.8 7.4 10.5 Notes: East Asia includes Brunei, Cambodia, China, Hong Kong, Indonesia, Republic of Korea, Lao PDR, Macao, Malaysia, Mongolia, Myanmar, North Korea, Philippines, Singapore, Taiwan (China), Thailand and Vietnam. Sub-Saharan Africa includes Angola, Botswana, Burundi, Comoros, Congo Democratic Republic, Congo Republic, Eritrea, Ethiopia, Kenya, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Rwanda, Seychelles, Somalia, South Africa, Sudan, Swaziland, Tanzania, Uganda, Zambia and Zimbabwe. Rest of world countries include European countries aside from EU 15 members, the former member states of the USSR, and "bunkers" or loadings of foreign flag vessels. Source: IMF Direction of Trade Statistics. B. The Origins of Kenya's Imports Key Point According to IMF Direction of Trade statistics, Kenya sources about 12 percent of its total imports from East African countries ­ up by about 8 percentage points from the corresponding level in 1975. However, an analysis of underlying COMTRADE statistics indicates the overall change is due to a few relatively specialized products, like increased imports of fish and tobacco. On the import side, Table 4.2 shows that both East Asia and East Africa increased their import shares in Kenya's markets by about 9 percentage points, while the EU 15 experienced almost a 20 percentage point loss. In large part, the East African trade gains were largely due to a relatively few products like fish and fish preparations, which went from virtually nothing in 1995 to over $45 million in 2003. Coffee and tea was Kenya's second largest import from East Africa with trade totaling about $22 million in 2003. Given the nature of the available statistics, it cannot be determined if these imports were for domestic consumption, further processing, or for re-export. Another point is that, on both the import and export side, the direction of Kenya's trade generally confirms to that for all Sub-Saharan countries combined. There are two exceptions. Kenya has a somewhat lower share of total imports originating in NAFTA. Second, Kenya receives less than one-quarter of its total imports from EU 15 countries, which is considerably lower than the latter's share in all Sub-Saharan markets. 29 Table 4.2. The Geographic Origins of Kenya's Imports; 1975 to 2003 Share of Imports Coming From (%) Of Global Of which: Sub- which: Rest Imports East EU South Saharan East of Country Year ($ mill.) Asia China ASEAN Japan (15) NAFTA Oceania Asia Africa Africa World Kenya 1975 938.6 3.2 0.4 1.3 8.6 42.7 8.8 1.7 2.1 4.1 4.1 1.9 1985 1,436.1 8.5 0.8 6.8 10.3 39.2 5.3 1.2 1.3 1.6 1.6 1.5 1995 3,249.1 13.4 3.5 4.5 9.1 38.2 4.2 0.4 8.8 10.8 10.6 0.0 2003 4,231.9 12.8 6.3 2.8 4.8 23.2 5.4 0.5 7.4 13.2 12.8 0.9 All Sub 1975 18,791.1 4.3 2.1 0.7 7.0 55.4 10.1 0.5 1.1 6.8 4.0 14.6 Saharan 1985 26,897.8 5.1 1.4 2.0 5.7 51.2 10.6 0.4 1.2 8.5 4.1 17.5 Africa 1995 46,993.0 13.3 2.2 4.4 8.4 41.5 6.3 0.4 3.1 15.8 10.0 11.3 2003 112,489.2 15.2 6.1 4.1 4.1 37.4 8.2 1.4 2.7 12.4 7.7 18.7 Notes: East Asia includes Brunei, Cambodia, China, Hong Kong, Indonesia, Republic of Korea, Lao PDR, Macao, Malaysia, Mongolia, Myanmar, North Korea, Philippines, Singapore, Taiwan (China), Thailand and Vietnam. Sub-Saharan Africa includes Angola, Botswana, Burundi, Comoros, Congo Democratic Republic, Congo Republic, Eritrea, Ethiopia, Kenya, Lesotho, Madagascar, Malawi, Mauritius, Mozambique, Namibia, Rwanda, Seychelles, Somalia, South Africa, Sudan, Swaziland, Tanzania, Uganda, Zambia and Zimbabwe. Rest of world countries include other countries in Europe aside from the EU 15 members, the former member states of the USSR as well as "bunkers" or loadings of foreign flag vessels. Source: IMF Direction of Trade Statistics. V. THE EFFECTS OF AGOA ON KENYA'S EXPORTS Key Point United States African Growth and Opportunity Act preferences have had a positive impact on Kenya's exports in a relatively few product groups, like apparel and clothing. However, they appear to have a much more limited potential elsewhere for two reasons. Africa previously did not receive Generalized System of Preferences (GSP) treatment for its clothing exports, and United States Most Favored Nation (MFN) tariffs on these goods was quite high. In this sector, AGOA did, in fact, create important preferences for Kenya and other African countries. However, there appear to be relatively few other sectors where the same combination occurs, that is, high United States MFN tariffs and the lack of current GSP preferences. Recent US efforts to establish a Central American Free Trade Area (CAFTA) will, if successful, further dilute the trade expansion effects of AGOA. Evidence indicates the US African Growth and Opportunity Act was a major factor behind the 7 percentage point increase in the share of Kenya's exports going to NAFTA countries over 1995-2003 (Table 4.1). This was due to the fact that United States most-favored-nation (MFN) tariffs on apparel imports are high and, unlike most industrial countries, clothing was excluded from the US Generalized System of Preference (GSP) program. AGOA provides African countries with preferential access to US markets for apparel and clothing, while the Caribbean Basin Initiative (CBI) provides similar preferences for countries in the Caribbean. Given the magnitude of the spread between current MFN clothing tariffs, and the zero preferential AGOA duties, the act provided African apparel exporters with very high 30 preferential margins in some cases. However, given the current, general low level, of most United States Most Favored Nation tariffs it appears that there are relatively few sectors, aside from textiles and clothing, where AGOA can have a similar major expansionary effect on African exports of manufactured goods. An analysis of import duty "peaks" in the United States tariff schedule could provide some useful information on this point. Aggregate United States import statistics suggest AGOA has not yet had a major expansionary influence on most African countries' exports. As Table 5.1 shows, during January-June 2003, the United States imported approximately $6.6 billion in AGOA eligible products from Africa which was about $2.5 billion higher than in the same period a year earlier. However, 80 percent of the recent AGOA eligible exports consisted of petroleum products whose exports are highly concentrated in a relatively few African countries. U.S. import tariffs on oil are generally modest, except on gasoline, so it is unlikely that AGOA had a major expansionary effect here. Second, about five percent of AGOA non-oil exports consist of passenger motor vehicles. These shipments originate almost exclusively in the Republic of South Africa and are the result of the recent opening of a local BMW automobile assembly plant. Table 5.1. The Composition of Africa's AGOA Eligible Exports to the United States Export Value ($000) Share of Total (%) January-June January-June January-June January-June Product Category 2002 2003 2002 2003 Petroleum and oils 3,054,916 5,306,870 76.85 80.41 Apparel and clothing 305,303 416,530 7.68 6.30 Passenger motor vehicles* 219,741 295,611 5.53 4.48 Ferrous metals 58,621 82,134 1.47 1.24 Aluminum and alloys 32,836 35,693 0.83 0.54 Silicon 10,504 16,926 0.26 0.26 Articles of jewelry 13,098 15,038 0.33 0.23 Tobacco 16,344 7,976 0.41 0.12 Cane sugar 6,337 3,293 0.16 0.05 All other products 257,636 419,467 6.48 6.36 Total 3,975,336 6,599,538 100.00 100.00 * These shipments consist almost entirely of exports from a BMW assembly plant located in the Republic of South Africa. Source: USITC (2004), US Trade and Investment with Sub-Saharan Africa, Publication Number 3650, (Washington: USITC, December). Apparel and clothing accounted for $417 million or 6.3 percent of AGOA eligible imports in 2003 and there is little doubt that the program provided a major stimulus to trade in these goods for two reasons.11 First, current United States MFN tariffs on clothing are generally quite high ranging to 32 11AGOA provides for duty-free and quota-free access to the U.S. market without limits for apparel made in eligible Sub-Saharan African countries from U.S. fabric, yarn, and thread. Preferential treatment for apparel took effect on October 1, 2000, but beneficiary countries had to first establish effective visa systems to prevent illegal transshipment and use of counterfeit documentation, and that they have instituted required enforcement and verification procedures. The Secretary of Commerce monitors apparel imports on a monthly basis to guard against 31 percent for products like T-shirts, sweaters, and blouses, to 28 percent for men and women's trousers. Second, unlike most OECD countries, the United States previously failed to extend generalized system of preference (GSP) tariffs to textile and clothing products. As such, the preferences provided by AGOA significantly improved Africa's competitive access conditions for apparel products. However, due to the current generally low level of United States MFN tariffs, and the existing preferences under the GSP program, there appear to be few other sectors where AGOA can come close to having the trade expansionary effect it has had for textiles and clothing. Table 5.2 provides a somewhat different perspective on the effects of AGOA by showing statistics on the origins of preference receiving shipments. Over 82 percent of AGOA eligible exports originated in five primarily oil exporting African countries, with Nigeria alone being the source of about 70 percent of these shipments. In contrast, Lesotho and Kenya, the two major SSA exporters of textiles and clothing originated about 4 percent of AGOA affected products. It should be noted that major concerns have been expressed as to whether these countries can remain competitive for apparel exports once all Multifiber Arrangement restrictions are removed in 2005. Even with AGOA preferences Kenya recently experienced a significant erosion of its United States import share for apparel products like men's shirts and "other" outer garments due to a surge in imports from China, Mexico and Vietnam. Table 5.2. The Origins of Africa's AGOA Eligible Exports to the United States Export Value ($000) Share of Total (%) January-June January-June January-June January-June Exporting Country 2002 2003 2002 2003 OIL EXPORTERS Nigeria 2,340,602 4,634,726 58.88 70.23 Gabon 608,356 499,642 15.30 7.57 Congo 45,430 163,009 1.14 2.47 Guinea-Bissau 48,875 89,661 1.23 1.36 Cameroon 22,799 33,862 0.57 0.51 OTHER COUNTRIES South Africa 556,955 744,904 14.01 11.29 Lesotho 142,208 166,479 3.58 2.52 Kenya 48,875 89,661 1.23 1.36 Mauritius 55,236 64,238 1.39 0.97 Madagascar 61,070 60,870 1.54 0.92 Swaziland 28,986 53,914 0.73 0.82 Ghana 22,904 25,718 0.58 0.39 Cote d'Ivoire 15,729 25,585 0.40 0.39 Malawi 21,757 19,232 0.55 0.29 Namibia 136 9,203 0.00 0.14 Ethiopia 1,349 1,448 0.03 0.02 TOTAL 3,975,336 6,599,538 100.00 100.00 Source: USITC 2004. Annex 5.1a A Note on AGOA and Kenya's Transportation Costs for Exports to the United States Since 1974 the United States has tabulated imports, by-product-by country, on a joint free- alongside-ship (f.a.s.) and cost-insurance-freight (c.i.f.) basis. The f.a.s. valuation includes the purchase price of the product plus all charges incurred in placing it alongside the vessel at the port of exportation.12 surges. If increased imports are causing, or threatening, serious damage to the U.S. apparel industry, the President is to suspend duty-free treatment for the article(s) in question. 32 In contrast, the c.i.f. valuation measures the value of imports at the port of entry in the United States and includes all freight, insurance, and other charges (excluding import duties) incurred in bringing the merchandise from the port of exportation and placing it alongside the carrier at the port of importation.13 Since this detailed freight cost information is available for almost three decades it can be used to determine how Kenya's freight rates compare to those of its competitors, and how they changed over time.14 Another important consideration is the extent to which Kenya has been able to utilize different transport modes (that is, air versus vessel freight) for its exports. The United States customs statistics provide relevant information since they identify the type of carrier used for imports at the tariff line level. Air transport has several clear attractions including speed of transit, and the fact that it can directly deliver goods to interior markets in the United States. The further development of fast efficient air freight services clearly has a high priority if Kenya is to enhance its potential as an "off season" agricultural exporter, or to more closely integrate into international production sharing operations. Another point to note is that the United States, in particular, has taken an active role in efforts to remove bottlenecks and other cost raising inefficiencies in Kenya, and other African countries transport operations. The reasoning here is that, if Africa is paying excessive freight costs for its exports, this could negate the potential benefits associated with AGOA tariff preferences.15 Before proceeding, the two-fold objectives of this annex should be clearly noted. First, a key question that is addressed is whether, and to what extent, do Kenyan exporters encounter differentially 12An important limitation of these data is that they do not reflect the cost of inland transportation ­ which may be quite high for some African countries ­ or port charges. The importance of the latter should not be underestimated. For example, World Bank data compiled by Tyler Biggs show port charges for clearing a 20 foot container in Abidjan (Cote d'Ivoire) and Dakar (Senegal) are $1,100 and $910 respectively. In contrast, the ocean freight cost for shipping the container to the Le Havre/Hamburg area ranged from $1,350 to $1,430. These data suggest port costs may add 70 to 80 percent to the international freight costs analyzed in this annex. 13The c.i.f. to f.a.s. ratio can be used to measure nominal international freight and insurance charges for exports of product i from country j (fij) from; fij = [(Vc,ij ÷ Vf,ij) - 1.] 100 where Vc,ij represents the c.i.f. and Vf,ij is the f.a.s. value of exports of tariff line level product i from country j to the United States. For example, if the landed price of a good in the United States is $115, and the Kenyan f.a.s. export price is $100, nominal freight and insurance costs would be reported as 15 percent. Aside from the United States other countries like Australia, Brazil, Mexico and Argentina separately tabulate transport and insurance costs for their imports. 14Some commentators have suggested that Kenya and other African countries face an important export constraint associated with their "geographic remoteness" or distance to major markets. An analysis of relevant statistics does not support this contention. Kenya lies approximately 6,400 nautical miles from the nearest US East Coast port. In contrast, the distance between Thailand and the US West Coast is 6,900 nautical miles, while Malaysia and Indonesia are 7,300 and 7,500 nautical miles away, respectively. 15The United States government stressed the importance of improvements and modernization in air freight services to allow African countries to fully utilize AGOA preferences. Specifically, in a June 28, 2001 speech announcing the creation of the "Open Skies for Africa" program US Secretary of Transportation Norman Mineta observed that "without effective transportation links, national or global economies cannot continue to develop, function and expand. Without a responsive transportation system, tourism and trade cannot develop to their full potential. Now while we know it is important to consider the entire inter-modal system in transportation planning, the focus of this conference is aviation. Safe, secure and dependable air transportation is vital to the economic development and well being of the countries of Africa. Developing a safe and secure aviation infrastructure is essential if African countries want to develop strong economic and trade relationships with the United States and other nations and regions." 33 higher transport costs that negate the potentially positive effects of AGOA tariff preferences. Second, the intention is to objectively assess the general magnitude of the barriers posed by international freight costs on Kenya's exports. This question could be of major importance if future multilateral trade negotiations significantly reduce the preferential margins under AGOA. Although such an effort clearly should be encouraged, it is not the intention to analyze the potential for lowering these charges through appropriate policy actions, like measures to enhance competition for transport suppliers, or through measures to improve port efficiency. Annex Table 5.1a provides summary information on the average level of 2003 nominal freight costs, and the relative importance of air versus vessel shipments, for Kenya's major two-digit HS exports to the United States. The key points reflected in the table are as follows; · Ocean freight is by far the most important transport mode accounting for almost 78 percent (by value) of all exports to the United States. For products like furniture, sugar and confection, soaps and cleaning products, or fuels (which all have relatively high stowage factors) ocean freight is the sole mode of transport used. The opposite occurs for optical products (which are typically high-value low-bulk items) where air freight is used exclusively. · Second, nominal transport costs average about 7 percent on all products, but range to over 15 percent for 2-digit HS groups like fish, cut flowers, vegetable and fruit preparations or artificial flowers. Based on the information available, these nominal rates might be expected to roughly double if port cost were included. · Third, Kenya appears to be at a transport cost disadvantage vis-à-vis other countries as its overall nominal transport costs are more than 3 percentage points higher than those of its global competitors. To put this number in proper perspective, current United States MFN tariffs average about 4 percent. Kenya's transport cost disadvantage for apparel products is about 3 percent higher than its competitors. This adverse margin would offset some of the positive effects associated with Kenya's AGOA tariff preferences.16 Annex Table 5.2a takes a more focused look at the level and variation in Kenya's transportation costs for products exported by air. Nominal freight rates for these shipments are roughly double (14.3 percent) the average reported in Annex Table 5.1a for all products. A surprising point is how much variation occurs in nominal freight rates for six-digit products within some of these HS groups. For example, nominal freight rates ranging from 15 to 95 percent are observed within the fish and crustacean group, while freight rates for knit clothing range between 7 and 43 percent. At their upper ranges reported in Annex Table 5.2a international transport costs could have a very restrictive impact on Kenya's exports. Returning to the main point addressed in this annex, the evidence shows that differential adverse freight costs are offsetting some of the preferential tariff margins associated with AGOA. Overall, Kenya pays nominal freight costs that are about 3 percentage points higher than those of its competitors. In the United States, MFN tariffs on apparel generally range between 15 to 35 percent so the offsetting effects of the adverse transport costs are likely small. However, a somewhat different conclusion emerges for 16 An examination of the underlying six-digit HS data shows that some apparel products encounter very high nominal transport costs. For example, exports of women's cotton pajamas, women's trousers, and shawls and scarves all face freight costs in excess of 25 percent. There is some reason to believe that the average freight rates reported in Table A5.1 may understate the true importance of transportation costs as a barrier to exports. Products facing relatively high transport costs would tend to have their export volumes reduced by these charges and enter into the calculation of an overall average with relatively low trade weights. 34 products that are, or must be, shipped by air. Here Kenya's adverse freight differential is over 11 percent which indicates they have a far more "diluting" effect on United States AGOA preferences. 35 Annex Table 5.1a. 2003 Nominal Transportation Costs for Kenya's Major Two-Digit HS Product Exports to the United States; Vessel and Air Freight Shipments Combined. Kenya's Share of Nominal Freight Kenya's 2003 Exports Exports by for All Transport Transport HS Description ($000) Vessel (%) Modes (%) Margin (points) 3 Fish and crustaceans 1,516 84.8 15.0 8.5 6 Cut flowers and foliage 2,136 37.8 28.8 12.4 8 Edible fruit and nuts 6,137 97.4 2.5 -17.5 9 Coffee, tea and spices 21,538 99.2 4.8 -1.3 12 Oil seeds and fruits, seeds 169 14.0 1.8 -3.8 13 Gums and resins 6,661 89.9 0.5 -3.7 14 Vegetable plaiting material 64 96.3 14.3 4.0 15 Animal or vegetable oils 54 100.0 12.5 6.9 17 Sugars and confection 94 100.0 9.3 3.0 20 Vegetables or fruit preparations 1,443 100.0 20.0 10.2 21 Miscellaneous food preparations 4,208 95.0 4.9 0.0 22 Beverages and spirits 191 100.0 7.9 2.4 26 Tungsten concentrates 200 100.0 3.6 -10.0 27 Petroleum jelly 84 100.0 10.5 5.9 32 Tanning or dying extracts 431 42.1 11.1 7.3 33 Essential oils 371 40.2 6.9 4.5 34 Soaps and detergents 89 100.0 3.5 -2.8 39 Plastic articles 597 73.4 13.3 7.0 42 Articles of leather 59 12.2 13.5 6.2 44 Wood and wood articles 1,803 91.7 16.9 9.9 46 Basket ware and wickerwork 92 49.3 21.1 -0.1 49 Printed material 194 97.3 4.3 -0.5 59 Laminated textile fabric 106 90.8 3.9 0.1 61 Knit clothing 54,339 85.5 7.2 2.7 62 Clothing, not knit 147,389 77.2 7.5 3.0 67 Feathers or artificial flowers 90 14.3 18.4 8.9 68 Stone articles 328 95.3 9.3 -1.0 69 Ceramic products 118 75.1 13.5 -1.8 70 Glass ware 118 0.0 0.9 -6.1 71 Precious stones 1,144 36.6 5.4 4.9 73 Articles of iron or steel 56 85.3 5.7 -1.6 82 Tools or cutlery 60 21.5 7.1 2.8 84 Reactors and boilers 3,032 1.5 1.2 -1.2 85 Electrical machinery 3,547 3.9 2.3 0.2 90 Optical equipment 102 0.0 4.1 2.5 94 Furniture and bedding 108 100.0 9.1 -0.6 95 Toys and sporting goods 1,308 8.9 3.3 -3.9 96 Miscellaneous manufactures 124 78.4 13.8 8.0 97 Works of art and antiques 146 24.9 8.1 7.4 98 Special transactions 4,621 20.3 2.3 1.1 ALL GOODS 266,372 78.1 6.9 3.3 Source: USDOC, Census Bureau database. 36 Annex Table 5.2a. The Level and Variation in Nominal Transportation Costs For Kenya's Air Freight Exports. Share of Nominal Air Freight Costs (%) Kenya's Air Exports by Transport HS Description Air (%) Average Range Margin (points) 3 Fish and crustaceans 15.2 53.3 15.1 to 95.3 17.1 6 Cut flowers and foliage 62.2 27.0 16.2 to 46.8 7.0 12 Oil seeds and fruits, seeds1 86.0 1.8 -- 0.7 32 Tanning or dying extracts 57.9 13.6 -- 7.5 33 Essential oils 59.8 13.3 2.9 to 36.4 7.9 39 Plastic articles 26.6 29.7 18.1 to 42.5 3.8 42 Articles of leather 87.8 13.8 4.4 to 33.4 4.1 46 Basket ware and wickerwork 50.7 18.3 -- -16.3 61 Knit clothing 14.5 22.9 6.8 to 43.3 8.6 62 Clothing, not knit 22.8 18.1 15.9 to 22.6 3.4 67 Feathers or artificial flowers 85.7 27.1 26.7 to 27.5 16.8 69 Ceramic products2 24.9 2.0 1.7 to 2.3 -5.5 70 Glass ware3 100.0 1.1 0.1 to 1.1 -4.4 71 Precious stones 63.4 2.8 1.0 to 9.7 -0.7 73 Articles of iron or steel4 14.7 16.5 7.7 to 29.5 10.5 82 Tools or cutlery 78.5 7.6 6.4 to 8.9 2.4 84 Reactors and boilers 98.5 6.4 0.3 to 23.3 2.4 85 Electrical machinery 96.1 2.0 0.4 to 3.4 1.0 90 Optical equipment 100.0 4.6 2.1 to 6.6 1.9 95 Toys and sporting goods5 91.1 1.8 1.4 to 8.9 -6.1 97 Works of art and antiques 75.1 12.5 7.0 to 17.5 6.6 98 Special transactions 79.7 1.2 -- 0.2 ALL GOODS 21.9 14.3 11.4 1This group consists mainly of flower seeds. 2Primarily ceramic statues. 3Primarily imitation pearls. 4Primarily iron wire. 5Primarily fishing tackle line. VI. IMPLICATIONS OF THE COMPOSITION OF KENYA'S EXPORTS Key Point Countries are often advised to avoid concentrating exports in a narrow range of similar goods that may be negatively affected by new, more efficient, competitors or by factors adversely influencing the general level of demand. Compared to other Sub-Saharan African countries Kenya's exports of foodstuffs are highly concentrated, largely due to two products, namely, coffee and tea. While their recent export shares appear to be declining substantially, this is largely the result of a recent collapse in international coffee prices. There are reasons why countries should want to identify relatively important products in their trade, and also determine how the composition of these goods was changing. First, this information could help focus attention on priority sectors for liberalization of government imposed trade barriers in future trade negotiations. Second, if goods that are increasing in importance generally have common characteristics like labor, capital or natural resource intensity in production, this could help focus regional efforts to reduce any transport, or other infra-structure constraints, these types of goods encounter. Third, does the evidence indicate that a country lie Kenya's export profile is static, or are important changes 37 taking place. If the export profile is changing, how is it doing so? Fourth, it has long been recognized that primary commodities, and most other raw materials, have unfavorable long-term trade prospects (see Section IIIb). If Kenya's exports contain a relatively high and static share of these goods this could have negative implications for longer-term trade prospects.17 Table 6.1 provides summary statistics on the broad composition of Kenya's global exports from 1985 to 2003 by showing the total value of trade and the share accounted for by product groups like foodstuffs, agricultural raw materials, ores and metals, or manufactures. For comparison, the table also provides similar trade statistics for all Sub-Saharan African countries. The data for years prior to 1995 must be analyzed with some caution due to the problem (previously discussed) of missing trade statistics for partner countries like Uganda and Tanzania. Several changes in the composition of Kenya's exports warrant attention. First, the share of foodstuffs in total exports declined by 16 percentage points (to 44 percent), due primarily to a dramatic recent decline in international coffee prices, and some erosion of Kenya's European import shares by Brazil and Vietnam. Some erosion also occurred in Kenya's European Union shares for tea which carried associated export losses of about $12 million annually. Two other points are noteworthy. First, the share for Kenya's exports of refined petroleum rose almost ten fold over 1995-2003 due to an increase in production capacity at the Mombassa refinery. Second, the share of clothing more than tripled over this relatively short interval ­ as noted AGOA had a major influence on this expansion ­ with most production of apparel products originating in the export processing zones. Box 6.1 compares features of the export processing zones that have been established in Kenya with those in ten other Sub-Saharan African countries. Table 6.2 provides more detailed information on the composition of Kenya's exports in 1976 and 2003 by identifying and reporting trade values for the ten largest two-digit SITC products. The earlier year was selected on the basis that it was the first year prior to 1995 that both Kenya's major East African trading partners (Uganda and Tanzania) reported statistics to UN COMTRADE, and there also was an interest in analyzing trade changes over a somewhat longer time frame. Five products (coffee and tea, crude animal and vegetable materials, fruits and vegetables, petroleum, and non-metallic mineral manufactures) make the ten largest product lists in both years. However, the most dramatic changes relate to the remarkable increase that occurred for clothing (exports rose almost 200 fold over this period), followed by impressive gains for fish, crude agricultural materials (cut flowers), and fruits and vegetables.18 In both periods Kenya's exports were highly concentrated (although the underlying composition of goods changed) with the ten largest products accounting for approximately 80 percent of total export earnings. 17 This proposition traces it origins to the writing of Raul Prebisch, former chairman of the UN Economic Commission for Latin America and Hans Singer of the United Nations. A major contributing factor to the adverse long term prospects for primary commodities is the low income elasticity of demand for many of these items. Hogendorn (1987, chapter 12) provides a useful survey of the relevant issues. Recent evidence (Ng and Yeats, 2002) shows an extensive deterioration in the terms-of-trade for commodities occurred over the last two decades, and this is projected by the World Bank to continue into the foreseeable future. 18Previous studies by the International Labor Office indicate the displacement of one agricultural worker in wealthy countries like the United States and European Union might generate as many as 50 additional jobs in an impoverished country like Kenya. The reason is that farming is very capital intensive in the former, but relatively labor intensive in low income developing countries. This observation attaches a high priority to measures aimed at further accelerating Kenya's exports of fruits and vegetables and other agricultural products. Given the fact that industrial countries' protection for some agricultural products, like tobacco and sugar, is very high the extension of AGOA type preferences to agricultural products warrants consideration by OECD members. 38 Table 6.1 The Structure of Kenya's Exports by Major Product Categories All Share in Total Exports Manufactures Sub-Group (Share in Total Exports)** Goods All Agricultural Ores & All Misc. Leather Wood & Exporter ($million) Foods Materials Fuels Metals Manufactures Goods & Rubber Paper Textiles Clothing Footwear Kenya 1985 929 77.3 7.4 1.5 0.5 11.7 1.0 2.1 0.2 0.4 0.5 0.1 1990 1,128 68.3 9.4 2.1 1.3 17.9 1.0 4.7 0.4 0.7 0.5 0.1 1995 1,797 59.6 10.6 1.2 1.8 26.0 0.9 1.7 1.3 1.0 2.6 0.2 2000 1,830 53.6 12.7 11.0 1.6 20.4 0.7 1.1 1.5 1.3 2.8 0.2 2003 2,269 43.5 15.9 11.6 1.8 26.8 0.4 0.7 1.4 0.8 9.2 0.2 All Sub- Saharan Countries 1985 48,136 17.6 5.4 47.5 13.3 11.7 4.2 0.2 0.5 0.6 0.6 0.0 1990 60,336 16.0 7.3 38.1 16.2 17.7 4.7 0.5 0.8 0.8 1.4 0.0 1995 71,553 19.2 8.0 31.4 11.8 22.7 6.9 0.7 1.1 0.9 2.0 0.1 2000 101,676 13.8 5.2 42.7 10.8 23.9 3.6 0.6 1.0 0.7 2.1 0.1 2003 109,183 15.8 5.0 40.9 10.1 24.9 3.3 0.6 0.9 0.5 2.5 0.0 Note: The product groups are classified by SITC products in Revision 2 as Foods and Feeds (0+1+22+4); Agricultural raw materials (2-22-27-28); Mineral Fuels (3); Ores and Metals (27+28+67+68); All Manufactures (5 to 8 less 67 and 68) Source: Computations based on UN COMTRADE statistics. 39 Box 6.1 Africa's Recent Experience with Export Processing Zones A recent International Labor Office (ILO 2003) survey indicates approximately 20 Sub-Saharan African countries have established export processing zones to attract foreign investment and stimulate exports. The survey estimates these zones now employ approximately 430,000 individuals with the Republic of South Africa accounting for about two-thirds of this total. In terms of the number of jobs created, Kenya is the fourth largest African zone behind South Africa, Malawi and Namibia. While most African countries established their zone in a single area, Namibia, South Africa and Kenya have zones in six or more distinct locations. Statistics prepared by Kenya's Export Processing Zone Authority indicate both the products shipped from the EPZ, and the foreign markets that are the destinations of these exports, are highly concentrated. As shown below, in 2002 exports of $177 million originated in Kenya's processing zones with about 84 percent ($148 million) of this total consisting of apparel and clothing. These shipments were destined almost exclusively for the United States, and to a far lesser extent the United Kingdom. Printing and publishing was the second largest export activity ($9.7 million) followed by horticultural products and chemicals. However, some uncertainty exists as to the accuracy of these statistics, as it does with almost all of Kenya's trade data. In the ILO survey Kenya reported the further processing of tea was one of the largest activities undertaken in the zone, yet the data provided by the EPZ Authority fail to indicate any exports of processed tea are occurring. Sector Major Destinations Exports ($000) Share (%) Clothing United States 148,266 83.8 Printing & publishing COMESA, South Africa 9,742 5.5 Horticultural products Europe 7,993 4.5 Chemicals East Africa 3,199 1.8 Edible oils Tanzania 3,172 1.8 Electronics East Africa 3,148 1.8 Pharmaceuticals Various Sub-Saharan Africa 1,411 0.8 Gemstones Thailand, United States 55 -- All Goods All Destinations 176,985 100.0 Apart from the Republic of South Africa, the composition of exports from most other African EPZs (like Kenya) appear to be in a relatively narrow range of products. For example, in Mozambique, Malawi and Kenya the zones primarily originate textile and apparel exports. As indicated below, several SSA zones established links to the rural sector by engaging in agricultural product and food processing. Country Major EPZ Products EPZ Employment No. of Zones Republic South Africa Transportation equipment, diversified 290,000 6 manufactures, and food processing. Malawi Textiles and cotton 29,000 1 Namibia Activities not specified 29,000 11 Kenya Apparel and garments, pharmaceuticals, 27,148 6 processing of tea Zimbabwe Processing of tea 22,000 7 Ghana Wood and metal products, glass fiber and 9,500 4 telecommunications Cameroon Food processing, wood manufacturing 8,000 1 Togo Shoes, computers, paints, pasta 7,213 1 Nigeria Communications, mineral processing 4,700 3 Sudan Groundnuts, sesame seeds, cotton, wood 1,033 3 products, vegetable and fruit Mozambique Textiles and agriculture n.a. 1 40 Table 6.2 Kenya's Largest Two-Digit SITC Export Products in 1976 and 2003 Export Value ($000) Share of Exports (%) SITC Product Description 1976 2003 1976 2003 Ten largest products in 1976 661,407 1,494,329 78.0 65.9 07 Coffee, tea, cocoa & spices 344,865 486,579 40.7 21.4 33 Petroleum and petroleum products 111,840 260,358 13.2 11.5 29 Crude animal and vegetable materials* 54,163 309,978 6.4 13.7 05 Fruit and vegetables** 43,008 307,892 5.1 13.6 66 Non-metallic mineral manufactures*** 26,477 47,820 3.1 2.1 26 Textile fibers, not manufactured 16,857 18,672 2.0 0.8 01 Meat and meat preparations 16,727 1,434 2.0 0.1 04 Cereals and cereal preparations 16,421 11,716 1.9 0.5 21 Hides, skins and fur skins, undressed 15,636 31,186 1.8 1.4 73 Transport equipment 15,413 18,695 1.8 0.8 Ten largest products in 2003 607,656 1,832,495 71.7 80.8 07 Coffee, tea, cocoa & spices 344,865 486,579 40.7 21.4 29 Crude animal and vegetable materials* 54,163 309,978 6.4 13.7 05 Fruit and vegetables** 43,008 307,892 5.1 13.6 33 Petroleum and petroleum products 111,840 260,358 13.2 11.5 84 Clothing 1,215 209,197 0.1 9.2 03 Fish and fish preparations 2,028 94,245 0.2 4.2 66 Non-metallic mineral manufactures*** 26,477 47,820 3.1 2.1 51 Chemical elements and compounds 12,247 45,866 1.4 2.0 89 Miscellaneous manufactured articles 7,909 38,257 0.9 1.7 67 Iron and steel 3,905 32,304 0.5 1.4 Total All Goods 847,423 2,268,595 100.0 100.0 * This product group includes cut flowers and foliage ** The largest four digit-SITC component product in this group are preserved fruit and edible nuts *** Cement is the largest component product in this group Source: UN COMTRADE Statistics. A. Does Kenya Have Dynamic Exports? Key Point Shipments of cut flowers and "off season" agricultural produce have increased at rates far above those for Kenya's total exports. Attempts to identify export opportunities for similar products deserve a high priority given their potential for reducing rural poverty. Apparel exports have also expanded at well above average growth rates due largely to the effects of United States AGOA preferences. Although they presently may not constitute a large share of Kenya's trade, there are reasons why one should attempt to identify "dynamic" (that is, fast growing) exports. Even though their current trade values may be small, if their above-average growth continues for an extended period, these items could become an important source of export earnings. Second, if the dynamic products have common production characteristics this could be of interest. For example, if they are (say) highly labor- or resource-intensive, both the reasons for their growth, and whether similar export opportunities exist in 41 related goods, could have important implications. Third, there is an obvious interest in identifying dynamic products to focus attention on the removal of any trade barriers they face in regional or global markets. Fourth, correlations show a relatively strong relationship often exists between the growth rates for specific product exports in the 1980s and 1990s. As such, the identification of dynamic products may provide useful information concerning their future growth prospects. Conversely, it may also be important to identify exports that have been declining in relative importance. Is their decline due to the increased use of superior substitute products (such as the substitution of synthetic fibers for Kenya's jute and sisal exports), or to the erosion of Kenya's import shares. Alternatively, declining exports may also be the result of changes in market access barriers, like the imposition of new trade restrictions, or to the unilateral reduction of trade barriers facing competitive suppliers. Situation where this may occur include the proposed extension of US preferences to central American countries (CAFTA), or the planned extension of the European Union.19 Table 6.3 lists the fastest-growing four-digit SITC (Rev. 2) products in Kenya's exports to the European Union (15) over 1995-2003, while Table 6.4 provides similar data for the United States. Both tables show the value of exports in each year, each item's share of total exports, and the change in the value of exports. The lower portion of the tables provide similar information for those products that experienced the largest declines in exports. Four important points emerge from these statistics; · Coffee appears among the products with the largest export earnings decline in the EU 15 as trade fell by about $206 million. In large part, the European Union results reflect a collapse in coffee prices as the total (global) value of EU imports fell by over 50 percent from their $8.5 billion level in 1995. However, Kenya's export losses in European markets were worsened further by erosion of its market share which fell from 3.3 to 1.8 percent, a decline that carried associated export earning losses of about $38 million (see Table 7.1 for further evidence concerning this point).20 · Although, the value of EU imports was essentially flat, Kenya's tea exports fell by almost $20 million. The primary reason was a 2.4 percentage point decline (to 28 percent) in Kenya's market share (see Box 7.1). Supply problems associated with a prolonged drought are cited as a factor behind the erosion of Kenya's market share for tea. · Exports of cut flowers and fresh vegetables to the European Union over 1995 to 2003 expanded at annual rates close to 10 percent, which was far better than the corresponding rate for all exports. An important point is that four of the fastest growing products in EU trade (combined 1995 to 2003 exports grew by over $250 million) were all agricultural products that may have benefited from Kenya's position as an off season agricultural producer. An important question is whether Kenya can further 19A simple hypothetical example can help clarify this line of reasoning. Assume that both Kenya and a given Central American country currently export a specific good to the United States. Further assume that the central American country pays a 10 percent MFN tariff, which would be reduced to zero under CAFTA, while Kenya pays no duty on the product due to AGOA. These developments would lower tariffs paid by the Central American country relative to those facing Kenya and could result in some displacement of the latter's exports. 20 According to the International Coffee Organization, Vietnam increased its global exports of coffee more than threefold from 1995 to the early 2000s, and now originates more than 13 percent of world exports (as opposed to 5 percent in 1995). Partly as a result of these new sources of supply, the ICO estimates the price received by Kenyan coffee exporters fell from about 177 US cents per pound in January 1995, to about 60 cents in December 2002. Preliminary estimates show the decline continuing with Kenyan coffee exporters receiving about 36 cents per pound in December 2003. 42 Table 6.3. Dynamic and Declining Products in Kenya's Recent Exports to the European Union (15) Exports ($000) Share of Total Exports (%) Export Change ($000) SITC Product 1995 2000 2003 1995 2000 2003 2000-2003 1995-2003 0 to 9 Total exports 852,383 736,300 912,518 100.0 100.0 100.0 176,219 60,135 FAST GROWING PRODUCTS 292.7 Cut flowers and foliage 99,451 142,755 229,521 11.7 19.4 25.2 86,766 130,071 054.5 Other fresh or chilled vegetables 66,340 120,783 153,066 7.8 16.4 16.8 32,282 86,726 292.6 Bulbs and tubers 5,138 16,390 33,138 0.6 2.2 3.6 16,748 28,000 057.9 Fresh or dried fruit. 18,295 16,599 35,991 2.1 2.3 3.9 19,392 17,697 778.8 Other electrical machinery 120 142 10,878 0.0 0.0 1.2 10,737 10,758 037.1 Prepared or preserved fish. 10 13,539 10,642 0.0 1.8 1.2 -2,897 10,632 121.2 Tobacco, stripped 1,028 11,860 11,410 0.1 1.6 1.3 -449 10,383 661.2 Portland cement 0 8,666 8,830 0.0 1.2 1.0 164 8,830 292.4 Plants used in perfumery 993 1,155 9,196 0.1 0.2 1.0 8,040 8,202 034.3 Fish fillets, fresh or chilled 9,251 212 17,397 1.1 0.0 1.9 17,186 8,146 061.1 Sugar 0 1,044 6,683 0.0 0.1 0.7 5,639 6,683 121.1 Tobacco, not stripped 898 2,445 5,271 0.1 0.3 0.6 2,827 4,373 892.8 Printed matter. 804 2,245 3,965 0.1 0.3 0.4 1,720 3,161 894.2 Children s toys and games. 123 1,144 3,258 0.0 0.2 0.4 2,113 3,134 697.4 Products of art 134 1,651 2,919 0.0 0.2 0.3 1,268 2,785 278.5 Quartz and mica 3,922 6,460 6,336 0.5 0.9 0.7 -124 2,415 515.6 Hetero-cyclic compounds 0 0 1,712 0.0 0.0 0.2 1,712 1,712 121.3 Tobacco refuse 28 668 1,515 0.0 0.1 0.2 848 1,487 541.7 Medicaments 22 11 1,433 0.0 0.0 0.2 1,422 1,412 764.9 Parts of telecom apparatus 384 1,730 1,769 0.0 0.2 0.2 39 1,384 MAJOR DECLINING PRODUCTS 667.3 Precious & semi-precious stones 2,845 1,988 341 0.3 0.3 0.0 -1,647 -2,504 288.2 Other non-ferrous base metal waste 2,986 600 39 0.4 0.1 0.0 -562 -2,947 058.9 Fruit otherwise prepared 52,012 43,820 47,699 6.1 6.0 5.2 3,879 -4,312 881.3 Photographic apparatus 4,359 14 6 0.5 0.0 0.0 -8 -4,353 611.4 Leather of other bovine cattle 4,722 2,700 116 0.6 0.4 0.0 -2,584 -4,606 265.4 Sisal & other fibers 9,306 3,748 4,186 1.1 0.5 0.5 438 -5,120 611.6 Leather of other hides or skins 10,758 4,280 4,401 1.3 0.6 0.5 121 -6,357 034.4 Fish fillets frozen 16,150 92 5,858 1.9 0.0 0.6 5,766 -10,293 074.1 Tea 151,775 120,683 130,930 17.8 16.4 14.3 10,248 -20,845 071.1 Green or roasted coffee 282,781 127,775 76,658 33.2 17.4 8.4 -51,117 -206,123 Source: Based on EU imports from UN COMTRADE Statistics. 43 Table 6.4. Dynamic and Declining Products in Kenya's Recent Exports to the United States Exports (000) Share of Total Exports (%) Export Change ($000) SITC Product 1995 2000 2003 1995 2003 2003 2000-2003 1995-2003 0 to 9 Total Exports 108,190 115,261 266,248 100.0 100.0 100.0 150,987 158,058 FAST GROWING PRODUCTS 843.9 Other textile outer garments 2,094 21,592 90,684 1.9 18.7 34.1 69,092 88,590 842.3 Trousers 9,482 18,032 42,904 8.8 15.6 16.1 24,872 33,422 845.1 Jerseys and pullovers 96 65 26,352 0.1 0.1 9.9 26,287 26,255 845.9 Other knit outer garments 82 55 20,290 0.1 0.0 7.6 20,235 20,208 846.2 Knit under garments of cotton 348 44 66,45 0.3 0.0 2.5 6,601 6,296 057.7 Edible nuts 3,352 243 5,969 3.1 0.2 2.2 5,726 2,617 843.5 Blouses of textile fabrics 977 24 3,180 0.9 0.0 1.2 3,156 2,203 752.5 Peripheral electronic units 0 373 1,763 0.0 0.3 0.7 1,390 1,763 098.0 Edible products n.e.s. 2,525 1,324 4,208 2.3 1.1 1.6 2,884 1,683 292.7 Cut flowers and foliage 104 197 1,787 0.1 0.2 0.7 1,590 1,682 764.3 Radiotelephonic equipment 173 96 1,705 0.2 0.1 0.6 1,609 1,533 843.4 Women's skirts 19 49 1,284 0.0 0.0 0.5 1,235 1,266 058.9 Fruit otherwise prepared or preserved 146 0 1,385 0.1 0.0 0.5 1,385 1,239 843.3 Women's dresses 203 7 1,296 0.2 0.0 0.5 1,289 1,094 763.8 Other sound recorders 0 93 1,062 0.0 0.1 0.4 969 1,062 844.2 Under garments excluding shirts 45 0 1,095 0.0 0.0 0.4 1,095 1,050 844.3 Women's under garments 0 0 910 0.0 0.0 0.3 910 910 074.1 Tea 4,521 10,503 5,401 4.2 9.1 2.0 -5,102 880 759.9 Parts of office equipment 234 173 798 0.2 0.2 0.3 625 564 846.3 Under garments knitted 89 0 626 0.1 0.0 0.2 626 537 MAJOR DECLINING PRODUCTS 532.2 Tanning extracts 1,141 511 431 1.1 0.4 0.2 -80 -710 657.5 Twine and cordage 733 0 2 0.7 0.0 0.0 2 -731 663.3 Manufactures of mineral materials 924 457 173 0.9 0.4 0.1 -284 -751 842.9 Other textile outer garments 1,305 269 150 1.2 0.2 0.1 -119 -1,155 667.3 Precious & semi-precious stones 1,848 1,053 359 1.7 0.9 0.1 -694 -1,489 292.5 Seeds of fruit & spores 1,908 158 145 1.8 0.1 0.1 -13 -1,763 658.4 Bed and table linen 1,841 3 3 1.7 0.0 0.0 0 -1,838 752.4 Digital central storage units 1,960 233 56 1.8 0.2 0.0 -177 -1,904 292.9 Other materials of vegetable origin 19,104 7,809 6,688 17.7 6.8 2.5 -1,120 -12,416 844.1 Men's shirts 22,802 6,399 5,455 21.1 5.6 2.0 -944 -17,347 Source: Based on United States imports from UN COMTRADE Statistics. 44 capitalize on its geographic location to expand other types of agricultural exports, or exports of agro- industrial products. · Although Kenya's US exports grew by about $160 million, these gains were highly concentrated in a relatively few clothing products. Exports of one four-digit SITC product, namely, textile outer garments grew by $88 million (roughly 55 percent of the total increase), while trousers added an additional $33 million. Altogether, the fastest growing Kenyan exports were all clothing products whose shipments grew by $168 million. This figure implies that the net change in Kenya's other exports to the United States would have been negative without these four apparel products. B. Have Kenya's Exports Become More "Complementary"? Key Point A general problem often encountered in efforts to establish regional trade arrangements is that the types of goods African countries export are quite different than those other SSA countries import. If the profile of Kenya's exports is changing in ways that more closely match other African countries imports, this would indicate a growing potential for expanded intra-regional trade. An important related question is whether Kenya's exports are becoming more similar to OECD members imports. However, a "trade complementarity" index shows that such positive changes have not been occurring. Furthermore, the values of the index are currently so low that few opportunities exist for a meaningful expansion of intra- or inter-regional trade. An important question is whether the structure of Kenya's exports has been changing in ways that more closely match the structure of its major partner countries imports. If such changes have occurred this has positive implications since it would increase the value and quantity of goods that could be traded, and augment the interdependence between Kenya and its trading partners. However, previous empirical analyses, based on statistics for the mid-1990s, indicated there were major dissimilarities between the types of goods most Sub-Saharan African countries (including Kenya) exported, and the types of goods the region imports (Yeats 1998). An important related question is whether Kenya's export profile has recently become more similar to the import profiles of its major non-African trading partners. The so called "trade complementarity index" can provide useful relevant information. Furthermore, changes in the index over time can help determine whether Kenya's trade profile is becoming more, or less, compatible. The index of trade complementarity between two countries k and j (Ckj) is defined as, Cij = 100 - (mik - xij ÷ 2) where xij is the share of good i in the exports of country j, and mik is the share of good i in the imports of country k. The index is zero when no good exported by one country is imported by the other, and 100 when the export-import shares exactly match. As such, it is assumed that higher index values indicate more favorable prospects for broader trade contacts (in terms of the number of products that could be exchanges) between countries.21 21Several caveats should be noted. The index has also been used to assess the potential for trade between countries where actual trade contacts are zero or minimal (say between Kenya and Mongolia). Use of the index to assess the potential for new trade contacts presumes that a country can expand production and trade of its exports on a relatively constant cost basis. Second, considerations must be given to relevant factors like economic geography. High complementarity indices may be misleading if the countries are geographically distant, or have other natural 45 Table 6.5 reports 1976, 1995 and 2003 complementarity indices for Kenya's exports to selected trading partners. To help put these statistics in proper perspective, the notes to the tables shows similar information for trade among some East Asian countries. Three points are important. First, there is no indication that Kenya's export profile is changing in ways that are more complementary with the imports of its major trading partners. Kenya's 2003 complementarity index for global exports (27.1) is, in fact, slightly lower than it was in 1976. Second, Kenya's indices are quite low compared to the East Asian countries, which indicates major dissimilarities exist between the goods Kenya exports, and the goods its partners import. Third, Kenya's indices are influenced by one item ­ refined petroleum products. In 2003, Kenya exported over $200 million worth of refined petroleum, mainly to countries in East Africa. The relatively high share of petroleum in both Kenya's exports and global imports elevates the complementarity index, and overstates the similarities between non-oil exports and partners imports. Table 6.5 Complementarity Indices for Kenya's Trade with Selected Countries Total Exports ($000) Trade Complementarity Index* Kenya's Trading Partner 1976 1995 2003 1976 1995 2003 High Income Countries European Union (15) 122,088 662,384 966,256 39.0 26.8 27.3 United States 113,730 617,628 709,174 26.2 24.1 24.2 Middle & Low Income Burundi 54 168 42 35.1 26.0 32.6 Egypt 1,364 5,991 7,533 24.5 23.6 22.2 India 5,269 31,430 57,906 17.6 19.4 18.5 Pakistan 1,096 7,761 9,696 28.3 23.1 23.0 South Africa 6,010 21,306 33,991 19.9 22.8 21.9 Sudan 555 541 2,582 21.5 27.0 27.0 Tanzania 507 693 799 29.9 26.4 27.6 Uganda 343 637 438 31.8 28.1 35.8 WORLD 896,827 4,874,735 7,160,831 28.6 27.6 27.1 *For comparison the following reports some recent trade complementarity indices for other countries; China and East Asia (47.0), Hong Kong and East Asia (54.9), Korea and East Asia (67.7), Malaysia and East Asia, Singapore and East Asia (61.3), Taiwan and East Asia (68.5) Notes: The trade complementarity index between Kenya j and countries k is defined as: Cij = 100 - (mik - xij ÷ 2) where xij is the share of good i in the exports of Kenya j and mik is the share of good i in the imports of country k. The index ranges between zero when no good exported by one country is imported by the other, and 100 when export-import shares exactly match. Higher index values purportedly indicate more favorable prospects for a successful trade integration between countries. Source: Computations based on partners data from UN COMTRADE Statistics. barriers that make trade unprofitable. Third, relative size differences can be very important. If exporter i can only supply a very small share of country j's import needs this would be a negative factor, even if their trade complementarity indices were quite high. Finally, the index assumes that countries assign equal priorities for trade expansion to all goods. If there are different priorities for (say) manufactures as opposed to raw materials this complicates the use of the index. 46 C. Has Kenya Diversified Its Exports? Key Point The diversification of Kenya's exports away from traditional products has a very high priority. However, several empirical measures of export concentration suggest little progress has been made in this direction over the last two decades. Statistics show a high export concentration in products with very poor growth and real price change prospects is a major problem for almost all Sub-Saharan Africa countries. Does the available evidence suggest Kenya has been in any way successful in diversifying its exports. A second related issue is whether regional trade became more diversified due to the proliferation of African regional trade arrangements? Proponents of African RTAs often argue regional markets can be an initial base for the launch of new exports, which subsequently are traded elsewhere once positive scale and other "learning" effects were achieved.22 The question is important since it is generally assumed that serious negative effects are generally associated with a high concentration of exports exist. In previous analyses of the magnitude and effects of trade concentration, three empirical indices often have been employed. These include, · A count of the number of products exported. Two related problems are how to distinguish between established and marginal exports, and at what level of aggregation should products be defined. UNCTAD utilized an approach that seems sensible and differentiates goods at the four-digit level of the SITC. To be included in the count, a product had to account for at least one-half of one percent of total exports. · A second index is the share of a country's total exports accounted for by the largest four-digit SITC products. Previous studies based this measure on the largest, or three largest, products. The higher the shares of these products the higher is the level of export concentration. · The so called "Hirschmann" index has been used as a measure of trade concentration (see UNCTAD, Handbook of International Trade and Development Statistics, various issues). This index ranges between 0 and 1, with lower values indicating less concentrated trade structures. The Hirschmann index is defined as; Hj = ((xi/X)2 where xi is the value of exports of commodity i (normally defined at the four-digit SITC level) and X is the total value of country j's exports. Table 6.6 presents 1976, 1995 and 2002 results when these concentration indices were calculated for Kenya's exports to individual low-medium and high income countries. Although the results are mixed, there is some indication of a slight diversification occurring over 1995 to 2003, although the changes clearly were not of major significance. The share of the three largest products in exports to the European 22This line of reasoning is frequently advanced in support of regional trade arrangements (RTAs) among developing countries. The argument goes that, within the preferential tariffs of the RTA, member countries may be able to develop new export products that it previously could not trade competitively. At some point the experience within the protected markets increases competitiveness to a point that the country may begin to export these new products to non-regional markets. See UNCTAD (1976) for an example of this line of reasoning. 47 Union (15) fell from 63 to 56 percent, although the actual number of products exported remained virtually constant back to 1976. Over 1995-2003, the share of the three largest products exported to the United States actually rose by 14 percentage points (to 83 percent), due largely to the expansionary influence AGOA preferences had on apparel products. Similarly, over this period, the share of the three largest exports to Uganda almost doubled. In this case, the significant increase in the three product share was largely due to refined petroleum products. These data strongly suggest that, as is the case for almost all Sub-Saharan African countries, an effective Kenyan diversification strategy is badly needed (see Box 6.2). Table 6.6 Concentration Indices for Kenya's Exports Number of Products Top Three Product Export Concentration Exported a/ Share (%) Index b/ Trading Partner 1976 1995 2003 1976 1995 2003 1976 1995 2003 High Income Countries European Union (15) 263 252 260 68.5 63.0 56.4 0.54 0.41 0.36 United States 84 123 133 88.1 69.1 83.4 0.61 0.43 0.61 Middle & Low Income Burundi /c 118 107 151 35.0 35.1 28.1 0.24 0.25 0.21 Egypt, Arab Rep. 27 13 17 96.5 99.0 99.2 0.75 0.96 0.98 India 26 48 87 88.4 53.4 50.5 0.57 0.37 0.35 Pakistan 18 39 42 92.9 97.3 95.0 0.74 0.95 0.92 South Africa .. 146 196 .. 83.8 33.3 .. 0.63 0.23 Sudan /c 221 39 100 51.8 85.6 90.6 0.32 0.71 0.87 Tanzania 402 358 424 21.7 40.4 40.8 0.18 0.29 0.33 Uganda /d 389 468 495 47.6 31.4 59.6 0.30 0.21 0.51 Memo Item East African Community/e 439 499 529 29.0 26.8 54.1 0.20 0.19 0.47 Notes: /a Number of products exported at SITC 4-digit level in Revision. 1 /b The concentration index (Hirschmann index) is defined as: Hj = ((xi/X)2 where xi is the value of export products at SITC 4-digit level and X is the total exports in country's j. The value of the index ranges between 0 to 1, with lower values indicating less concentrated exports. /c Due to the missing data, 2002 statistics are substituted for 2003. /d Due to incomplete data, energy products are excluded from the 1995 statistics /e Tanzania and Uganda combined. Source: Computations based on partners import data from UN COMTRADE Statistics. South Africa failed to report trade statistics to the United Nations in 1976. 48 Box 6.2 Dimensions of the Export Concentration Problem in Africa Various empirical and policy studies show African countries exports are more heavily concentrated than those from any other region. In addition, the products that form the "core" of most SSA exports are normally primary commodities that have major unfavorable characteristics, including below average growth prospects, long-term declines in real prices, competition from synthetics, and high levels of demand and price instability. As indicated in the statistics shown below, the problem of concentrated exports is pervasive across almost all African countries. In addition, the export profiles of some countries are even more highly concentrated than those of Kenya. Seven countries, namely, Angola, Chad, Niger, Gabon, Equatorial Guinea, Burundi and Nigeria actually receive more than ninety percent of their total export earnings from only three products. All but five countries get more than 50 percent of their total export revenues from just three four-digit SITC products. Country Three Largest Exports Three Product Share (%) Angola Petroleum, diamonds, fuel oil 98.2 Chad Raw cotton, shellac, resin 97.9 Niger Petroleum, radio-isotopes, fuel oil 93.3 Gabon Petroleum, saw logs, manganese ore 92.6 Equatorial Guinea Petroleum, saw logs, frozen fish 92.3 Burundi Coffee, non-monetary gold, tea 92.1 Nigeria Petroleum, lubricating oils, fuel oil 90.8 Gambia Diamonds, crustaceans, vegetables 88.7 Benin Raw cotton, cotton seed, edible nuts 87.5 Republic of Congo Petroleum, fuel oil, saw logs 86.1 Guinea Aluminum ore, diamonds, petroleum 84.4 Rwanda Coffee, tea, non-ferrous ores 84.3 Uganda Coffee beans, fish fillets, raw cotton 79.8 Burkina Faso Raw cotton, raw sugar, sesame seeds 76.3 Ethiopia Coffee, sesame seeds, sheep skins 75.9 Malawi Tobacco stripped, tobacco not stripped, tea 74.6 Togo Raw cotton, natural phosphates, coffee 70.1 Zambia Refined copper, beryllium, non-ferrous ore 69.2 Sudan Petroleum, sesame seeds, raw cotton 66.6 Cameroon Petroleum, saw logs, non-conifer wood 59.9 Ghana Cocoa, diamonds, unwrought aluminum 53.8 Kenya Tea, coffee beans, cut flowers 53.2 Mauritius Raw sugar, knit undergarments, pullovers 49.4 Tanzania Edible nuts, coffee beans, tobacco stripped 48.3 Zimbabwe Tobacco stripped, ferro-alloys 44.1 Madagascar Crustaceans, pullovers, spices 35.2 South Africa Non-monetary gold, platinum, coal 29.3 While most development economists acknowledge that the need for a diversification of exports is one of Africa's most pressing economic problems, there is little general evidence that any broadening of SSA countries' trade base is occurring. In fact, after extensive analysis of available trade statistics, a World Bank study (Ng and Yeats, 2002) concluded the opposite was occurring and that some African countries exports became more highly concentrated over the last two decades. Source: Ng and Yeats (2002). 49 D. Implications of An "Export Prospects" Index Key Point Assuming that historical global trade growth rates for specific products continue, a key question is how favorable, or unfavorable, are the overall export prospects for specific African countries. Relevant information is available from an index that establishes a concordance between the share of a product in a country's total exports, and the rate of growth of that product in world trade. This "export prospects" index suggests that Kenya should expect its exports to grow at about one-half the rate of growth in global trade. Very similar results are projected for all Sub-Saharan African countries. These results indicate that unless Kenya diversifies into new, more promising product lines, or significantly reduces production costs for traditional exports, it will continue to be marginalized in world trade. A key question relating to rational expectations for future exports is how well a country's current trade profile positions it among relatively high growth products. The previous analysis showed Kenya and other Sub-Saharan African countries exports were highly concentrated in a relatively few products. If these products generally have lower than average growth prospects, African countries must diversify into new product lines, or continue to be "marginalized" in global trade. The so called "growth prospects" index can help assess how favorable or unfavorable is the outlook for the "basket" of goods a country exports. The index first establishes a concordance between the share of each four-digit SITC product (i) in a country's (j) exports, and the recent world trade growth rate for that product. That is, if sij is the current share of product i in country j's total exports, and Ri is the rate of growth of the product in world trade, the export prospects index (Pj) is defined as, Pj = [sij×Ri]/Rw Where Rw is the rate of growth of world trade in all products.23 It should be noted that the index only reflects the influence of global demand growth and assumes that no changes occur in a country's competitive position. Countries with an index above unity may be thought of as having above average export growth prospects, while the reverse is true for those with indices below unity. Furthermore, because the index is measured relative to the growth rate for world trade it shows how relatively favorable, or unfavorable, are a country's export prospects compared to world trade. For example, an index value of 0.50 suggests a country's exports should be grow at one-half the rate of world trade. An index value of 1.50 indicates its exports should grow 50 percent faster. Table 6.7 shows the results when the growth prospects index was calculated for 26 Sub-Saharan African countries. The average index (0.58) suggests that Sub-Saharan countries' exports should expand at a rate just over one-half that for world trade. However, there is considerable variation in the indices 23 Several caveats should be noted. First, use of this measure is justified by correlations showing longer-term growth rates for trade in most products are relatively stable. A basic assumption of the index is the continuation of past trends. As an example, rank correlations between the 1980 to 1989 and 1990 to 1999 export growth rates for all four-digit SITC products were statistically significant at the 99 percent confidence level. Second, information provided by the index differs from a measure based on the growth rate of a country's total exports over a given period. The latter may be biased by changes in the relative importance of products during the interval. Third, the measure can provide information on the likely influence of a policy induced change in the composition of exports on trade growth rates, but it provides no information concerning the costs, or feasibility, of implementing the trade changes. Such, changes may take the form of government incentives to diversify exports into new product lines. 50 that ranges from 0.20 for Mali to 1.18 for Mauritius. As such, the latter is the only African country whose exports seemingly have the potential to grow at a rate roughly equivalent to that for world trade.24 Kenya's export growth prospects index (0.55) places it quite close to the average for the Sub- Saharan African countries. In part, Kenya's index is influenced by a relatively high share of coffee and tea in total exports. Ng and Yeats (2002) estimate the income elasticity for tea in major OECD markets is very low so rising income levels should not have a significant positive effect on the growth of exports of this product. Table 6.7 The Export Prospects Index for Sub-Saharan African Countries Growth Prospects Growth Prospects Country Index Country Index Angola 0.62 Mauritius 1.18 Benin 0.47 Mauritania 0.56 Cameroon 0.46 Mozambique 0.57 Congo, Democratic Rep. 0.70 Nigeria 0.62 Congo, Republic 0.57 SACU 0.81 Cote d'Ivoire 0.46 Senegal 0.56 Ethiopia 0.40 Sudan 0.50 Gabon 0.46 Tanzania 0.57 Ghana 0.62 Uganda 0.42 Guinea 0.39 Zambia 0.64 KENYA 0.55 Zimbabwe 0.58 Liberia 0.50 Madagascar 0.90 MEMO ITEM Malawi 0.74 Average 0.58 Mali 0.20 Range 0.20 to 1.18 Source: Computations based on UN COMTRADE Statistics. There is a clear and important message in Table 6.7 for both Kenya and the other Sub-Saharan African countries. Diversify exports!25 The available statistics indicate that a continued reliance on Africa's current exports will result in continued below average trade growth rates and the continued global marginalization of Africa. 24This observation is subject to qualifications. Mauritius' relatively high export growth prospects index is due to a high share (53 percent in 2003) of clothing in total exports. In 2005, Multifiber Arrangement quotas on clothing exports are scheduled to be phased out and Mauritius' exports will subject to heightened competitive pressures from countries like China, India, Bangladesh and Vietnam. This may cause the share of clothing in Mauritius' exports to decline which, in turn, would lower the country's export growth rate. Lesotho and Kenya are other African countries with relatively high shares of clothing in total exports. 25Section X in this report entitled "Prospects for Export Diversification" offers several suggestions as to how this might be accomplished including procedures for the identification of labor intensive manufactures in which Kenya should have a comparative advantage. This section also argues that the import and local assembly for export of foreign produced components holds considerable promise as a means of diversifying, 51 VII. COMPETITIVE FACTORS AND THE CHANGE IN KENYA'S EXPORTS Key Point Any aggregate analysis of changes in Kenya's exports to industrial countries could be misleading if the influence of underlying factors influencing individual products are not properly accounted for. Kenya's 1995- 2003 export expansion of $123 million to the United States is largely the result of AGOA preferences for apparel products which improved Kenya's competitive position. Major developments in the European Union included a collapse in coffee prices and demand, as well as some erosion of Kenya's import market share by countries like Brazil and Vietnam. On the positive side, Kenya's exports of cut flowers to the EU rose by $130 million, largely as a result of an improvement in its import market share. However, when the results for all products are examined the data show a general erosion occurred in Kenya's international competitive position. The previous analyses showed Kenya's exports to industrial markets expanded at a relatively modest pace since the mid-1990s. Factors relating to both supply and demand influenced these trade changes, and an important question concerns the extent to which an improvement in Kenya's overall competitiveness may have been a factor. This point can be addressed by decomposing Kenya's export growth into three factors, two of which relate to demand changes and changes in the ability to compete.26 The influence of demand for a specific product can be measured by the change in the total value of imports in a given country or regional market. In calculating the influence of this factor, one assumes that Kenya maintains its import market share for the commodity. Specifically, if Do,j and Di,j represent (say) EU total imports of product j, at time period o and t respectively, the change in Kenya's exports that would be attributed solely to demand Ed,i is: Ed,i = (so,j) × (Dt,j - Do,j) where so,j is Kenya's share of country i imports of product j (defined at the four-digit level of the Revision 2 SITC) from all countries in some initial period o, and the summation is over all goods traded. Therefore, the above computation will show the change in country i's (that is, Kenya's) exports that would have occurred if only changes in demand took place. Second, the change in the competitive position of a country like Kenya can be measured by the difference between the exports that would have occurred if its market share had not changed, and the exports that were in fact realized. This competitive factor change (Ec,i) is: Ec,i = (st,j - so,j)(Dt,j) where st,j is Kenya's share of imports of the product in a specific country in period t, and the summation is over all goods imported. Any differences between changes in a country's total exports and the sum of these two "demand" and "competitive" factors are due to product diversification.27 For Kenya and other 26A detailed description and early application of this procedure can be found in GATT (1966). The analysis also constituted a large part of Irving Kravis' (1970) classic analysis of the influence of trade on the 20th century growth of developing countries, that is, "Trade as a Handmaiden of Growth: Similarities Between the Nineteenth and Twentieth Centuries." 27An illustrative example may help explain this approach. Assume country i exports one product j to the EU and has a 20 percent import share with exports $20 million in 1995, and a 25 percent share with exports of $37.5 million 52 African countries the magnitude of this latter term is of major importance since the diversification of exports is a high priority policy issue. Using this analytical approach Table 7.1 examines the underlying causes of 1995-2003 change in Kenya's major exports to the EU 15. Aside from total trade, the products shown here were previously identified as this country's most rapidly increasing, and decreasing exports (see Table 6.3). The focus is on these specific products in an attempt to identify the sources of the major trade changes that occurred. The statistics reported for "total exports" summarizes the influence of competitive and demand changes on all four-digit SITC products that are exported by Kenya. Kenya's EU exports rose by about $60 million, but this overall change was heavily influenced by developments within two four-digit SITC groups. Specifically, Kenya's coffee exports to the European Union, which totaled over $280 million in 1995 fell by more than 50 percent due to a collapse of international coffee prices, i.e., a demand factor. Kenya also experienced some erosion of its EU market share for coffee, due largely to inroads by Brazil and Vietnam, that resulted in export losses of about $37 million. See Box 7.1 for addition information concerning recent developments in European Union markets for tropical beverage products. Offsetting changes occurred for exports of cut flowers where an improvement in market share generated trade gains of over $100 million. A further noteworthy point is that EU demand was virtually flat for the "basket" of goods that Kenya exports as this factor generated an increase in trade of only $342,000. Table 7.2 provides similar statistics for Kenya's similar statistics for Kenya's major exports to the United States. As noted, all of the trade gains over 1995-2003 are due to five four-digit SITC apparel products, and the source of the increased exports was an increase in Kenya's import market shares. Exclusive of these items, which all receive significant trade preferences under AGOA, Kenya's exports to the United States would have declined by approximately $18 million. If one omits cut flowers in Europe and apparel exports to the United States (it could be argued that both reflect unrepresentative products) what message emerges from the statistics in Tables 7.1 and 7.2. Without these relatively few items, Kenya's overall competitive change factor would have been negative with losses of approximately $80 million. There is no general evidence of an improved ability to compete, rather any general conclusions drawn from these statistics point in the opposite direction. in 2003. During this period regional demand for j rose from $100 to $150 million. The change in i's exports attributed solely to changes in demand would be: Ed,i = .20($150 - $100) = $10 million while the change due to the competitive factor is: Ec,i = (.25 - .20) × $150 = $7.5 million This example assumes that the country experiences no diversification in its exports. That is, the export change attributable to supply and demand fully account for the total trade change that occurred. 53 Table 7.1 The Influence of Demand and Competitive Changes on Kenya's Exports to the EU (15) Exports ($000) Source of Export Change ($000) Demand Competitive Net SITC Product (Rev. 2) 1995 2003 Factor Factor Change 0 to 9 Total exports 852,383 912,518 342 61,674 62,016 FAST GROWING PRODUCTS 292.7 Cut flowers and foliage 99,451 229,521 27,638 102,433 130,071 054.5 Other fresh or chilled vegetables 66,340 153,066 56,118 30,608 86,726 292.6 Bulbs and tubers 5,138 33,138 4,498 23,501 28,000 057.9 Fresh or dried fruit. 18,295 35,991 14,133 3,564 17,697 778.8 Other electrical machinery 120 10,878 14 10,744 10,758 037.1 Prepared or preserved fish. 10 10,642 3 10,629 10,632 121.2 Tobacco, stripped 1,028 11,410 27 10,356 10,383 661.2 Portland cement 0 8,830 0 8,830 8,830 292.4 Plants used in perfumery 993 9,196 210 7,993 8,202 034.3 Fish fillets, fresh or chilled 9,251 17,397 19,099 -10,953 8,146 061.1 Sugar 0 6,683 0 6,683 6,683 121.1 Tobacco, not stripped 898 5,271 -156 4,530 4,373 892.8 Printed matter. 804 3,965 538 2,623 3,161 894.2 Children s toys and games. 123 3,258 83 3,052 3,134 697.4 Products used for domestic purposes 134 2,919 67 2,718 2,785 278.5 Quartz and mica 3,922 6,336 2,892 -477 2,415 515.6 Hetero-cyclic compounds 0 1,712 0 1,712 1,712 121.3 Tobacco refuse 28 1,515 -1 1,488 1,487 541.7 Medicaments 22 1,433 54 1,358 1,412 764.9 Parts of telecom apparatus 384 1,769 232 1,152 1,384 MAJOR DECLINING PRODUCTS 667.3 Precious & semi-precious stones 2,845 341 -435 -2,068 -2,504 288.2 Other non-ferrous base metal waste 2,986 39 -610 -2,337 -2,947 058.9 Fruit otherwise prepared 52,012 47,699 16,027 -20,339 -4,312 881.3 Photographic apparatus 4,359 6 468 -4,820 -4,353 611.4 Leather of other bovine cattle 4,722 116 1,489 -6,095 -4,606 265.4 Sisal & other fibers 9,306 4,186 -3,445 -1,675 -5,120 611.6 Leather of other hides or skins 10,758 4,401 -1,167 -5,190 -6,357 034.4 Fish fillets frozen 16,150 5,858 9,444 -19,736 -10,293 074.1 Tea 151,775 130,930 -8,219 -12,626 -20,845 071.1 Green or roasted coffee 282,781 76,658 -168,471 -37,652 -206,123 Source: Computations based on EU imports from UN COMTRADE Statistics. Given their un-typical success as an export, Box 7.1 examines some of the developments that led to the impressive success of cut flowers as a Kenyan export. This sector clearly has important 54 implications for the type of commercial environment needed for the development of other export ventures. One final point. As previously noted, the difference between the total 1995-2003 change in Kenya's exports to the EU and United States, and that attributable to demand and competitive supply factors, must be due to export diversification. The statistics in Table 7.1 show these differences were actually negative (by almost $1.9 million) which indicates Kenya's exports actually became more concentrated. Similar, conclusions emerge from an analysis of changes in the United States market where new exports may have added about $1 million to Kenya's exports in 2003. In short, there is little evidence that suggests a diversification of Kenya's exports to the EU and United States occurred. A. How Important is Kenya as a Supplier of US Garment Imports? Key Point While apparel exports to the United States are of major importance to Kenya, US import statistics show that Kenya is a very marginal supplier compared to China, Mexico, Bangladesh, Vietnam and a number of Central American countries. Furthermore, Kenya's apparel exports are highly concentrated in a very limited number of products. As a result, Kenya's clothing exports are very vulnerable to post-MFA competitive changes that might occur in a relatively few product lines. In both 2002 and 2003 experienced serious erosion of its US import shares for several products even with Multifiber Arrangement restrictions in place. Although clothing exports to the United States are clearly of major and growing importance to Kenya (shipments of these goods totaled $202 million in 2003), a related question is how important is Kenya as a US supplier of these products. This point is of interest given the phase out of all remaining MFA restrictions in 2005, and the possible impact this may have on Kenya. Has Kenya become an important established supplier for the specific product lines it exports, or is it operating purely at the margin? If Kenya is operating at the margin it might be more adversely affected by the more competitive post-MFA environment. A second question is which countries are Kenya's major competitors for the products it exports, and how have their pre-MFA phase out import shares been changing? Table 7.3 draws on United States COMTRADE import statistics for answers. The table identifies Kenya's major four-digit SITC garment export products to the United States in 1995 and 2003.28 It also shows similar statistics for three major competitive suppliers for each item whose 1995-2003 market shares recorded the largest gains. To help put this information in perspective, the table also shows similar statistics for all clothing products combined (that is, SITC group 84). Three important points are evident in these statistics; 28 Two products were excluded from these tabulations, namely, men's shirts (SITC 844.1) and "other" outer garments (SITC 842.9). Table 7.3 has been designed to provide some indication as to how well Kenya will likely be able to compete after the full MFA phase out. The previous analysis (Table 7.2) answered this question for the two excluded products. Kenya was experiencing serious erosion of its market share for the excluded items as early as 2003. 55 Box 7.1 The Development of Kenya's Exports of Cut Flowers Kenya previously attempted to diversify away from traditional commodities like tea and coffee to processed agricultural products such as preserved fruit, to new types of niche products such as "off-season" fresh vegetables" and cut flowers, or to manufactures such as apparel and leather products. The results have been decidedly mixed. While Kenya is now the largest African cut-flower grower and one of the biggest exporters of fresh horticultural produce, the country has been less successful in manufacturing. Notwithstanding initial positive achievements, the provision of incentives to export-oriented manufacturing firms failed to sustain export growth, with the notable exception of garments. In the late 1960s, Kenya emerged as a supplier of "off-season" fruits and vegetables to the United Kingdom, and then to other European markets. Besides the impressive trade gains in fresh horticultural produce, Kenya started to develop cut-flower exports. This industry underwent a major transformation as a result of foreign investment by a Danish company that was granted tax and other financial incentives. The company brought in capital and expertise to generate spin-offs that had positive linkages to horticultural service industries. Several expatriate professionals left the company and started up their own small flower businesses ­ which further increased the value and volume of Kenya's cut flower exports. In the 1970s, the Horticultural Crops Development Authority managed an experimental program to train smallholding farmers in flower cultivation and to organize their harvest for export. The great expansion of the sector in the 1980s increased the demand for technical assistance, which gave rise to a technical support cluster of specialized service providers. Cut flower exports took-off in the 1990s in conjunction with significant reforms in import procedures, foreign exchange and air freight services, improvements in infrastructure and active investment promotion. Historically dependent on foreign capital and expertise, the industry has increasingly seen the emergence of Kenyan entrepreneurs with significant levels of expertise, to the point that the country is now largely self- sufficient in in-house knowledge required production and international marketing expertise. The statistics shown below document Kenya's success in exporting cut flowers to the European Union, which is its single largest foreign market. Although the Netherlands is by far the largest provider of cut flowers, Kenya is now in second place with an import market share of almost 6 percent. Kenya's competitive share gains in the EU from 1995 to 2003 resulted in increased exports of almost $110 million. Import Share (%) Value of Imports ($000) Exporter 1995 2000 2003 1995 2000 2003 World 100.0 100.0 100.0 3,016,232 3,059,606 3,974,293 Netherlands 63.9 60.5 64.7 2,018,776 1,851,991 2,569,474 Kenya 3.1 4.7 5.8 99,451 142,695 229,521 Israel 4.9 5.0 3.4 155,414 153,273 135,786 Italy 4.1 2.7 2.5 129,695 83,110 99,595 Colombia 3.5 3.0 2.5 111,239 92,799 98,400 Spain 1.9 2.8 2.1 60,312 84,330 85,271 United States 3.0 2.9 1.9 94,665 90,010 77,067 Ecuador 0.9 2.4 1.9 28,064 73,523 73,764 Zimbabwe 1.5 2.0 1.6 47,233 61,307 63,069 If there is one negative factor affecting this sector it is the perceived longer term supply surplus relative to demand. In a recent industry analysis the US ITC (Industry and Trade Summary: Cut Flowers, Report No. 3580) noted that "due to the increased global supply of fresh cut flowers, especially roses, since the early 1990s, fresh cut flower import prices have fallen significantly." The ITC indicated it expected these trends would continue in the future as new foreign suppliers became operative. 56 Table 7.2 The Influence of Demand and Competitive Changes on Kenya's Exports to the US Exports ($000) Source of Export Change ($000) Demand Competitive Net SITC Product (Rev. 2) 1995 2003 Factor Factor Change 0 to 9 Total Exports 108,190 266,248 33,092 123,901 156,993 FAST GROWING PRODUCTS 843.9 Other textile outer garments 2,094 90,684 2,029 86,560 88,590 842.3 Trousers 9,482 42,904 8,525 24,898 33,422 845.1 Jerseys and pullovers 96 26,352 116 26,140 26,255 845.9 Other knit outer garments 82 20,290 102 20,106 20,208 846.2 Knit under garments of cotton 348 6,645 711 5,585 6,296 057.7 Edible nuts 3,352 5,969 1,848 769 2,617 843.5 Blouses of textile fabrics 977 3,180 371 1,833 2,203 752.5 Peripheral electronic units 0 1,763 0 1,763 1,763 098.0 Edible products n.e.s. 2,525 4,208 4,253 -2,570 1,683 292.7 Cut flowers and foliage 104 1,787 28 1,654 1,682 764.3 Radiotelephonic equipment 173 1,705 467 1,065 1,533 843.4 Women's skirts 19 1,284 11 1,254 1,266 058.9 Fruit otherwise prepared or preserved 146 1,385 119 1,120 1,239 843.3 Women's dresses 203 1,296 8 1,086 1,094 763.8 Other sound recorders 0 1,062 0 1,062 1,062 844.2 Under garments excluding shirts 45 1,095 21 1,030 1,050 844.3 Women's under garments 0 910 0 910 910 074.1 Tea 4,521 5,401 1,790 -910 880 759.9 Parts of office equipment 234 798 13 551 564 846.3 Under garments knitted 89 626 -61 598 537 MAJOR DECLINING PRODUCTS 532.2 Tanning extracts 1,141 431 270 -979 -710 657.5 Twine and cordage 733 2 301 -1,032 -731 663.3 Manufactures of mineral materials 924 173 1485 -2,235 -751 842.9 Other outer garments 1,305 150 31 -1,186 -1,155 667.3 Precious & semi-precious stones 1,848 359 212 -1,701 -1,489 292.5 Seeds of fruit & spores 1,908 145 664 -2,427 -1,763 658.4 Bed and table linen 1,841 3 3,841 -5,679 -1,838 752.4 Digital central storage units 1,960 56 -383 -1,521 -1,904 292.9 Other materials of vegetable origin 19,104 6,688 6,100 -18,516 -12,416 844.1 Men's shirts 22,802 5,455 525 -17,872 -17,347 Source: Computations based on United States imports from UN COMTRADE Statistics. 57 Box 7.2. Recent Developments in EU 15 Markets for Tropical Beverage Products The previous analysis showed Kenya experienced losses exceeding $250 million on tropical beverage exports (tea and coffee) to the European Union (15) over 1995 to 2003, and that both erosion of market share and unfavorable demand conditions contributed to this decline. Given the importance of these products in total exports further analysis is warranted as to who was responsible for the erosion of Kenya's market shares. The following tabulations draw on EU 15 COMTRADE import statistics and show changes in the relative importance of individual countries in European markets for tropical beverages. Aside from Kenya, imports are shown for countries whose EU market shares increased, or decreased, by more than 3 percentage points. EU 15 Import Share (%) EU Imports ($000) 1995-2003 Product /SITC/Exporter 1995 Change 1995 2003 COFFEE (074.2) -ALL COUNTRIES 100.0 -- 7,255,232 2,932,937 Brazil 17.2 12.0 1,245,456 856,473 Vietnam 3.5 6.0 255,645 279,492 Peru 1.6 2.3 115,307 115,324 Switzerland 0.1 2.1 7,379 63,215 India 1.7 1.8 125,837 103,731 Honduras 3.5 1.4 253,054 142,969 Mexico 0.4 1.2 32,363 48,889 Cameroon 2.4 -1.1 173,544 37,166 Ecuador 1.5 -1.3 108,787 7,205 Kenya 3.9 -1.3 282,780 76,658 Costa Rica 3.9 -1.4 280,324 72,111 Uganda 5.8 -2.5 421,365 96,391 Congo, Dem. Rep. 2.8 -2.7 205,220 4,812 El Salvador 4.8 -3.0 348,203 53,314 Cote d'Ivoire 4.3 -3.0 312,063 36,890 Colombia 19.9 -7.6 1,440,998 358,993 TEA (074.1) - ALL COUNTRIES 100.0 -- 545,377 515,856 China 13.1 2.0 71,537 77,747 South Africa 1.6 1.7 8,651 17,085 Malawi 2.5 1.6 13,495 21,146 Vietnam 0.4 1.2 2,296 8,281 United Arab Emirates 0.0 1.0 0 5,137 Mauritius 2.3 -2.2 12,387 329 Kenya 27.8 -2.4 151,770 130,930 India 21.7 -3.0 118,461 96,406 Sri Lanka 14.1 -3.6 76,952 54,063 Three key points emerge from these statistics. First, Kenya's coffee exports to the EU experienced a major decline associated with a plunge in international prices that caused the total value of EU imports to fall by more than 50 percent. Second, Kenya and other countries import market shares were eroded by the 18 percentage point increase in the shares of Brazil and Vietnam. In this relatively short period, Brazil's EU import share for coffee rose by about 12 percentage points and is now close to thirty percent. In spite of these impressive competitive gains, falling prices caused the value of Brazil's coffee exports to decline by about $400 million. Third, although the level of EU demand for tea remained relatively stable, Kenya did experience export losses of over $10 million due to competitive gains by countries like China, South Africa, Malawi and Vietnam. 58 · First, Kenya's apparel product exports are highly concentrated in terms of their diversity. Kenya only exports six different four-digit SITC products to the United States, and one single item (Other Textile Outer Garments ­ SITC 843.9) accounted for 45 percent of total clothing exports. At the four- digit SITC level the Revision 2 classification system identifies 28 individual clothing (SITC 84) products. Kenya's clothing exports are vulnerable to competitive changes that might occur in a relatively few product lines. · Kenya is a very marginal supplier for all the clothing product lines it exports. In no case does its import share exceed one percent. For all apparel products combined, Kenya's 2003 exports were roughly eight percent of those originating in Vietnam, and less than three percent of those from Mexico. As indicated, China is the dominant supplier for the US market with 2003 clothing exports of about $12 billion, or roughly 60 times the current value of Kenya's total US apparel exports. · In spite of the recent introduction of United States AGOA preferences, Kenya's trade shares for its product lines remained quite stable. This contrasts with the 4 to 5 percentage point (or more) market share gains made by countries like Vietnam, Guatemala and Honduras. Overall, there is very little in these statistics that suggests Kenya is in a relatively strong competitive position vis-à-vis other clothing exporters. In fact, one development is particularly troubling. Table 7.2 shows that, in spite of the extension of AGOA preferences, Kenya's exports of men's shirts (SITC 844.1) fell by about $17 million. An analysis of the underlying statistics shows that Kenya's exports were largely displaced by four more competitive suppliers, namely, China, Bangladesh, Mexico and Vietnam. Market share losses also occurred in a second line (other outer garments) that had associated export earnings losses of over $1 million. · There appears to be a general view among apparel executives, and even among some Kenyan government officials themselves, that the 2005 phase out of MFA restrictions will have major adverse long-term consequences for clothing exports. In the absence of quotas on third countries there seems to be a generally held view that production will shift from Kenya and other African countries to low cost producers like China and India (see Box 7.3). 59 Box 7.3 Long-Term Effects of the Removal of MFA Restrictions on African Exporters. Although Kenya's apparel exports are not directly affected by current Multifiber Arrangement restrictions, there are concerns that their removal from other suppliers may have serious negative effects on Sub-Saharan African exporters. The reasoning here is that, even with AGOA preferences in the United States, and Everything But Arms Preferences (GSP) in Europe, Kenya may not be fully competitive with countries like China or India once the quotas on their exports are removed in 2005. This view holds that the MFA restrictions have had a significant positive effect on African countries' exports since they provided protection against more efficient third country producers. One related point should be noted. It is unlikely that the full effects of the phase out will occur in the short-term. Textile producers may have important "sunk" costs in capital and equipment, or longer term contractual obligations with raw material suppliers or distributors. To help anticipate the effects of the removal of MFA restrictions, the United States International Trade Commission (2004) conducted interviews in the United States and abroad with buying managers of major U.S. importers of apparel and home textiles--namely, the large apparel and home textile companies and retailers-- regarding their current sourcing strategies, likely changes to these strategies following quota elimination in 2005, and reasons for the expected changes. USITC staff conducted interviews with representatives of East Asian firms that produce or purchase textiles and apparel worldwide, and also are major sources of investment in the production of such goods in third countries. ITC staff also interviewed representatives of textile and apparel producers in India, which is considered by many U.S. apparel companies and retailers as the major low-cost alternative to China, and representatives of textile and apparel producers in sub-Saharan Africa, Mexico, and Central America. The analytical framework and competitive assessment was based on information obtained from a wide range of sources, including a review of the literature and testimony presented to the Commission at the hearing and in written statements. The conclusions of the ITC survey can be summarized as follows. China is expected to become the "supplier of choice" for most U.S. importers (the large apparel companies and retailers) because of its ability to make almost any type of textile and apparel product, at any quality level at a competitive price. However, the extent to which China continues to expand its shipments following quota elimination will be constrained by the possible use by the United States, and other major importers, of the textile-specific safeguard provisions in China's WTO protocol of accession. To reduce the risk of sourcing from only one country, U.S. importers indicate they plan to expand trade relationships with other low-cost countries, such as India, as alternatives to China. The ITC survey suggests one or two other low-cost countries, like Bangladesh or Pakistan, are expected to emerge as major suppliers for a narrower but still significant range of goods, such as mass-produced basic knit cotton tops and woven cotton shirts and pants (Bangladesh) or men's and boys' cotton apparel (Pakistan). Some firms indicated they would consider Caribbean Basin Economic Recovery Act (CBERA) beneficiary countries, particularly those located in Central America, as a major source of supply if a Central American or hemispheric free- trade agreement is negotiated that permits the use of regional fabrics or third-country (e.g., Asian) fabrics. Among the members of the Association of South East Asian Nations (ASEAN), the only countries considered competitive as major alternate suppliers to China or India, are Vietnam and, to a lesser extent, Indonesia. Although many countries may see their U.S. market share decline, exceptions may occur at the firm level, reflecting the importance of longstanding relationships between U.S. apparel companies and retailers and their foreign suppliers, as well as the efficiency, flexibility, and experience of foreign suppliers in producing certain articles. A number of countries likely will become major "second tier" suppliers to U.S. apparel companies and retailers for niche goods or services. As U.S. firms seek to balance cost, flexibility, speed, and risk in their sourcing strategies, they likely will look to the second-tier suppliers to meet those needs that are not met by the first-tier suppliers. For example, production of certain goods likely will remain in Mexico and the CBERA region to service U.S. buyers' quick turnaround or mid-season order requirements, particularly for replenishment of basic items offered in a wide range of different sizes, such as men's dress shirts and pants. Quick-turn orders are sometimes needed for fashion goods. Turkey and Colombia also are considered capable suppliers for quick-turn business. 60 Box 7.3 Continued According to industry sources, Sub-Saharan Africa (SSA) is not a particularly low-cost area for production of textiles and apparel, given efficiency adjusted labor costs, low productivity, long lead times, and high cost of other inputs compared with those in Asia. Most companies located their production in SSA because of quotas on other suppliers. These quotas, combined with duty-free, quota-free access to the EU and, since October 2000, to the U.S. market, has led to increasing exports of mainly apparel items from Africa. Most companies interviewed indicated that because of the importance of quotas, it will be difficult for SSA to compete in a quota-free world. They indicated that EU and AGOA preferences will not be enough to keep the industry competitive except in the area of manmade-fiber and wool apparel, where SSA is competitive and U.S. duties are high. Some SSA companies reported they are already losing sales in the EU market to countries such as Bangladesh, even with EU quotas in place. A statement submitted by its embassy to the ITC hearings, the Government of Kenya indicates it shares this pessimistic view of likely post MFA phase out developments. The Embassy states that "AGOA has enabled Kenya to redevelop its textile and apparel sector. AGOA's implementation created jobs, introduced new technologies, increased exports to the United States, and created foreign investment in the apparel industry. All of these benefits are expected to disappear with the elimination of quotas in 2005. The quota elimination will expose Kenya to competition with the world's leading textile and apparel manufacturers, such as China. The implementation of AGOA did not allow enough time for Kenya's textile and apparel sector to become competitive with such countries." (Ms. Lina Ochine, Commercial Attaché of the Kenyan Embassy, written submission to the Commission, Jan. 24, 2003. 61 Table 7.3. The Relative Importance of Kenya as a United States Clothing Supplier US Import Share US Imports ($000) Product (SITC Rev. 2) Exporter 1995 1995-03 Change 1995 2003 Other outer garments (843.9) Vietnam -- 4.9 767 521,917 Mexico 12.2 2.9 653,353 1,602,412 Cambodia -- 2.8 7 307,733 Kenya -- 0..9 2,094 90,684 World 100.0 -- 5,333,049 10,655,768 Trousers & breeches (842.3) Mexico 18.4 7.5 660,298 1,768,473 Vietnam -- 4.6 135 312,065 Nicaragua 0.8 1.5 29,452 155,470 Kenya 0.3 0.3 9,482 42,904 World 100.0 -- 3,589,834 6,817,394 Jerseys & pullovers (845.1) Guatemala 0.5 4.1 30,547 572,595 Honduras 2.2 2.6 126,751 606,664 Mexico 5.2 1.5 294,402 836,422 Kenya -- 0.2 96 26,352 World 100.0 -- 5,704,625 12,482,946 Knit outer garments (845.9) China 5.6 9.0 196,136 1,160,481 Vietnam -- 5.0 433 396,835 Mexico 6.6 2.9 230,960 757,849 Kenya -- 0.3 82 20,290 World 100.0 -- 3,510,403 7,947,058 Knit under garments (846.2) Honduras 5.5 10.3 188,059 1,039,602 El Salvador 3.8 6.1 130,132 651,452 Pakistan 1.2 2.6 42,967 251,224 Kenya -- 0.1 348 6,645 World 100.0 -- 3,440,198 6,575,542 Textile Blouses (843.5) Indonesia 6.6 4.1 137,420 307,119 India 13.9 2.6 288,981 473,962 Philippines 1.4 1.9 28,862 93,531 Kenya -- 0.1 977 3,180 World 100.0 -- 2,078,619 2,867,537 ALL CLOTHING (84) Vietnam -- 3.5 18,317 2,562,336 Mexico 7.0 3.2 2,904,892 7,257,515 China 14.9 2.0 6,203,008 12,078,555 Kenya -- 0.2 37,848 201,820 World 100.0 -- 41,618,374 71,511,802 Source: Computations based on UN COMTRADE Statistics. 62 VIII. INTRA-INDUSTRY TRADE AND PRODUCTION SHARING Key Point International production sharing, which often involves the importation and further assembly of parts and components in developing countries, can significantly broaden the range of new products a country can successfully export. In the absence of production sharing, a developing country like Kenya would have to master entire production processes for a good in order to become a viable competitor in world markets. Production sharing provided a major stimulus to trade and growth in East Asia, the Caribbean, and even between industrial countries themselves. However, evidence suggests that Kenya and other African countries have failed to fully capitalized on opportunities that exist in this activity. Some earlier analyses of factors promoting trade and growth concluded a growing level of inter- industry trade (IIT) can play an important positive role (see Grubel and Lloyd 1975, Greenway et. al. 1994, Feenstra 1998, or Hoekman and Djankov 1996). There are several reasons for this assumption. Intra-industry exchange produces gains from international trade, over and above those from comparative advantage, because it allows countries to take advantage of larger markets. By engaging in IIT, a country can often simultaneously reduce the number of similar goods it produces, while the variety of goods available to consumers is increased. By manufacturing fewer varieties, a country may be able to produce each on a larger scale, often with higher productivity gains and at a lower cost. More recently it has been acknowledged that there is another aspect of intra-industry trade that may convey even more benefits than the exchange of similar final products. Examples of the latter would involve (say) exports of Toyota automobiles from Japan to Germany, and the shipment of BMWs or Volkswagens in the opposite direction. Intra-industry trade increasingly consists of the exchange of parts and components of a specific product (say television picture tubes) in one direction, while trade in the final product (television sets) occurs in the reverse direction between two countries. Several summary measures provide information on the extent intra-industry trade is taking place, or is changing. One such measure is the IIT ratio. This index ranges between zero and one, with larger values indicating greater trade between firms in the same industry. Higher IIT ratios suggest that gains from specialization in differentiated products are being exploited, and that the participating countries are increasing their interdependence.29 Alternatively, the IIT ratio may indicate the extent to which intra- trade in similar products, at somewhat different stages of fabrication, is taking place. Table 8.1 presents intra-industry trade ratios for Kenya and selected African countries in 1995 and 2003. For comparison, the table also shows similar statistics for Brazil, Republic of Korea and Taiwan (China) that are at a more advanced stage of industrialization. The key point evident from these 29The IIT index is defined as, IIT = 1 - [|Xijk - Mijk| ÷ (Xijk + Mijk)] Where Xijk represents exports from industry i by country j to country k, and Mijk represents corresponding import values. Industries are defined at the three-digit level of the SITC system and the analysis is confined to manufactured goods, that is, items classified in SITC groups 5 through 8 less nonferrous metals (SITC 68). It should be noted that, at this level of detail the SITC groups will contain both fully fabricated equipment as well as parts of equipment needing further assembly. 63 Table 8.1 Intra-Industry Trade Ratios for Kenya and Other African Countries Machinery & Other All Country Year Transport Manufactures Manufactures Kenya 1995 0.127 0.428 0.265 2003 0.160 0.326 0.266 AFRICAN COMPARATORS Angola 1995 0.021 0.013 0.016 2003 0.017 0.044 0.026 Congo Dem. Republic 1995 0.024 0.024 0.025 2003 0.049 0.012 0.023 Cote d'Ivoire 1995 0.074 0.234 0.163 2003 0.096 0.177 0.133 Ghana 1995 0.037 0.063 0.046 2003 0.024 0.066 0.043 Madagascar 1995 0.010 0.161 0.106 2003 0.032 0.142 0.107 Mauritius 1995 0.093 0.170 0.152 2003 0.152 0.210 0.198 South Africa 1995 0.269 0.429 0.361 2003 0.459 0.434 0.454 Tanzania 1995 0.057 0.162 0.095 2003 0.061 0.128 0.089 Uganda 1995 0.017 0.014 0.016 2003 0.122 0.118 0.105 Zimbabwe 1995 0.082 0.272 0.180 2003 0.119 0.275 0.198 OTHER COMPARATORS Brazil 1995 0.482 0.423 0.480 2003 0.644 0.356 0.527 Korea 1995 0.485 0.536 0.514 2003 0.533 0.601 0.555 Taiwan, China 1995 0.605 0.440 0.543 2003 0.589 0.437 0.549 Note: The category of Machinery and Transport Equipment is classified as SITC 7; Other Manufactures as SITC 6+8-68; All Manufactures as SITC 5+6+7+8-68. The IIT ratios are computed in the SITC 3-digit products in Revision 2. Source: Computations based on partners' data from UN COMTRADE Statistics. 64 statistics is that a relatively small intra-industry trade base exists in Kenya and most of the other African countries (South Africa is an exception). The African IIT ratios are generally under 0.25 (which is less one-half those for Brazil, Korea and Taiwan (China). Also, the 1995-2003 changes that occurred in the ratios were, again with the exception of South Africa, minimal. Furthermore, the ratios of several of the African countries like Angola, the Democratic Republic of Congo and Ghana are very low (under 0.05) and static. While growing levels of intra-industry trade are normally viewed as indicators of growing global linkages, there is little indication that this activity is expanding in most African countries. One further point should be noted. Kenya's intra-industry trade ratio for all manufactured goods is relatively high compared to several other African countries. An analysis of the underlying statistics indicates this is due primarily to the import of components, and export of assembled products, rather than intra-trade of finished assembled products like office machinery, or "ready to use" telecommunications equipment. An example of the latter would be trade in BMWs and Toyota cars (finished goods) between Germany and Japan. In Kenya's case this exchange involves the importation of unfinished parts (like parts of textile clothing) and the exports of similar closely related assembled products like trousers or shirts. Section X of this report argues the further expansion of these types of operations may be a very effective way for Kenya to diversify its export base. A. How Big is Production Sharing in Kenya? Key Point With the exception of the Republic of South Africa, empirical evidence suggests Kenya and other African countries are not effectively capitalizing on opportunities that exist in international production sharing. Inter-regional comparisons of the composition of trade in components indicates Kenya may potentially have unrealized opportunities in several sectors like office machinery and electronics. This scenario would have Kenya importing parts of office machinery (possibly from Europe or Asia) for further assembly, and then exporting the finished products to third countries. Countries like China actively encouraged the import of office machinery and electronic components in order to break into markets for high tech products in which it otherwise would have been uncompetitive. It is generally recognized that a remarkable increase occurred in international production sharing as reflected in trade in components, or partially assembled manufactured goods.30 Production sharing involves the development of specialized (often) labor intensive operations within vertically integrated international manufacturing activities. As an example, parts of electronic components and office equipment are now assembled for multinational firms in places like Indonesia, Malaysia or the Philippines. Parts of wearing apparel and leather goods are assembled in Jamaica and the Dominical Republic for re-export to the US market, or to third countries. One estimate by the World Bank puts the value of assembly exports from the Caribbean at over $3 billion. Among the many industries where 30Production sharing is the internationalization of a manufacturing process in which several countries participate in different stages of a specific good's fabrication. The process is of considerable economic importance since it allows stages of production to be located where they can be undertaken most efficiently. For example, assembly operations for many types of goods are generally labor intensive and tend to gravitate toward developing countries. Ng and Yeats (1999, p. 13) show East Asian intra-trade in goods normally used in production sharing grew at an annual rate of 21 percent from the mid-1980s to mid-1990s. This was approximately 7 percentage points higher than East Asia's intra-trade in all goods. 65 major parts of a production process have been internationalized include automobiles and transport equipment, television and radio receivers, sewing machines, calculators, office equipment, electrical machinery, power and machine tools, typewriters, cameras and watches (USITC 1994). The increase in empirical information on production sharing was the result of modifications to the Standard International Trade Classification (SITC) system. In its original form, the SITC did a less than adequate job of distinguishing between trade in final goods, as opposed to parts and components. At its lowest (five-digit) level the SITC Revision 1 identified about 800 individual products ­ only 10 of which consisted of "parts" of manufactured goods that normally would undergo further assembly. However, in the late 1970s and early 1980s, many countries adopted the more detailed SITC Revision 2 system that expanded the number of product groups composed solely of parts and components. The coverage of these groups was most extensive within the machinery and transport equipment sector (SITC 7) where about 60 three, four, and five-digit product classifications were established that consist solely of components of manufactured equipment intended for further assembly.31 This data source greatly facilitated empirical analyses relating to production sharing. The Harmonized System (HS) classification scheme further expanded the statistics on trade in parts and components, although the coverage of these data, in terms of countries and years available, are still somewhat limited. Table 8.2 utilizes the UN SITC Revision 2 trade statistics to examine the composition and relative importance of components in African trade.32 The table identifies each product by SITC number and name, and indicates the value and share of each item in all African components trade. For example, parts of telecommunications equipment accounts for about 24 percent of all African imports of components, and about 11 percent of the region's exports. The table also shows the share of components in all manufactures (exclusive of chemicals) imports and exports. Several interesting points emerge from Table 8.2. First, as expected, African countries collectively had a major trade deficit in these goods as the 2003 value of imports ($11.2 billion) was more than five times the value of exports. World Bank studies indicate developing countries often tend to have a comparative advantage in assembly operations that are often relatively labor intensive. Second, African trade in components is concentrated in a relatively few items. Over 50 percent of all component imports occur in just three product groups, namely, telecommunications equipment, motor vehicles, and switchgear. This pattern differs somewhat from that in East Asia where parts of office machinery and equipment account for about 20 percent of all imports, and about 26 percent of all exports of these goods. 31 This section is based exclusively on product groups defined by the UN as consisting solely of parts and components. This almost certainly causes estimates of African production sharing to be downward biased. Some other traded SITC 7 products, like television picture tubes, or threads and fibers in SITC 6, likely experience further assembly operations, but it is not possible to accurately determine what is their true end use. A qualification should be noted for one group, namely, telecommunications equipment. The official UN description for this item is telecommunications equipment and parts, which indicates that it has not been possible to determine whether some of the items included are components or are products that have legitimate end uses. Separate statistical tests suggests that most of the items included in this group are in fact components. 32According to United Nations trade data the United States and Germany are the two largest exporters and importers of components, but Singapore, Hong Kong, Japan, Malaysia, and China ranked among the 10 largest countries participating in this exchange. In the late 1990s, East Asian (including Japan) global trade in parts and components ($165 billion) fell between North America ($152 billion) and OECD Europe ($239 billion). 66 Table 8.2 Sub-Saharan Africa 2003 Trade in Parts and Components Imports ($000) Exports ($000) SITC Component Description* Value ($000) Share (%) Value ($000) Share (%) 711.9 Steam boilers 24,740 0.22 972 0.05 713.19 Aircraft combustion engines 21,250 0.19 1,254 0.06 713.31 Engines for outboard motors 31,157 0.28 274 0.01 713.32 Other engines, nes 41,890 0.37 430 0.02 713.9 Internal combustion engines 417,497 3.72 110,573 5.65 714.9 Engines & motors, nes 255,138 2.27 55,036 2.81 716.9 Rotating electric motors 91,135 0.81 8,560 0.44 718.89 Water turbines 20,214 0.18 206 0.01 721.19 Cultivating equipment 27,504 0.25 1,193 0.06 721.29 Harvesting machinery 16,519 0.15 873 0.04 721.39 Dairy machinery 4,653 0.04 122 0.01 721.98 Wine making machinery 2,350 0.02 288 0.01 721.99 Agricultural machinery, nes 10,020 0.09 1,069 0.05 723.9 Construction machinery 913,767 8.14 37,123 1.90 724.49 Spinning machinery 40,177 0.36 2,102 0.11 724.69 Knitting machinery 30,770 0.27 879 0.04 724.79 Textile machinery 14,201 0.13 741 0.04 725.9 Paper making machinery 46,228 0.41 2,281 0.12 726.89 Bookbinding machinery 2,333 0.02 43 0.00 726.9 Printing & typesetting machinery 47,367 0.42 2,621 0.13 727.19 Grain milling machinery 18,267 0.16 583 0.03 727.29 Food processing machinery 62,627 0.56 4,845 0.25 728.19 Machine tools for special industries 24,041 0.21 1,126 0.06 728.39 Mineral working machinery 124,583 1.11 72,729 3.72 728.49 Special industry machines 227,590 2.03 22,768 1.16 736.9 Metal working machines 54,701 0.49 4,942 0.25 7371.9 Foundry equipment 8,251 0.07 113 0.01 737.29 Rolling mills 33,925 0.30 6,257 0.32 741.49 Refrigerating equipment 57,099 0.51 4,096 0.21 742.9 Pumps 136,089 1.21 17,636 0.90 743.9 Centrifuges & filters 183,627 1.64 71,290 3.64 744.19 Fork lift trucks 7,201 0.06 894 0.05 744.9 Loading machinery 418,088 3.72 62,781 3.21 745.19 Power hand tools 35,591 0.32 1,709 0.09 745.23 Packing machinery 90,655 0.81 2,509 0.13 749.99 Non-electric machinery, nes 112,941 1.01 12,721 0.65 759 Office machinery 789,495 7.03 81,805 4.18 764 Telecommunications equipment 2,674,522 23.83 210,458 10.75 771.29 Electric power machinery 31,583 0.28 5,224 0.27 772 Switchgear 854,498 7.61 82,052 4.19 775.79 Domestic electrical equipment 4,760 0.04 3,113 0.16 775.89 Electro-thermal appliances 10,945 0.10 859 0.04 776.89 Electronic components, nes 10,586 0.09 8,481 0.43 778.19 Electric accumulators 9,487 0.08 1,878 0.10 778.29 Electric lamps & bulbs 1,991 0.02 188 0.01 778.89 Electric machinery, nes 14,811 0.13 598 0.03 784 Motor vehicles 2,408,076 21.45 510,112 26.06 785.39 Cycles 192,269 1.71 4,524 0.23 786.89 Non-motor vehicles 33,849 0.30 11,785 0.60 791.99 Railroad equipment 55,326 0.49 17,550 0.90 67 Table 8.2 Continued Imports Exports SITC Component Description Value ($000) Share (%) Value ($000) Share (%) 792.9 Aircraft 308,888 2.75 30,719 1.57 821.19 Chairs and parts of furniture 68,422 0.61 424,864 21.70 821.99 Other furniture 45,118 0.40 21,159 1.08 874.29 Measuring machines 23,082 0.21 15,883 0.81 881.19 Still cameras, nes 4,950 0.04 277 0.01 881.21 Cameras under 16mm 4,069 0.04 1,272 0.06 881.29 Other apparatus 3,994 0.04 1,870 0.10 884.11 Unmounted optical elements 15,226 0.14 1,732 0.09 885.29 Clocks and watches 3,821 0.03 6,121 0.31 899.49 Umbrellas and canes 943 0.01 1,485 0.08 All above parts and components 11,224,898 100.00 1,957,645 100.00 As % of manufactures, excl. chemicals 19.58 8.01 * The official UN description for these products begins with the term "Parts of ....." This has been omitted below in the interest of brevity. Source: Based on partners data from UN COMTRADE Statistics, On the export side there are some surprises including the fact that over $400 million, or 22 percent of African exports of components, consist of parts of chairs and furniture. The underlying statistics show this exchange almost entirely originates in the Republic of South Africa and Europe (primarily Germany) is the main destination of these shipments. This point is important since it shows that the most geographically remote country in Africa was still able to effectively participate in international production sharing operations.33 It is not possible to easily determine what portion African components imports are locally processed, and then re-exported. However, evidence from East Asia indicates the domestic assembly of imported components for re-export is of major importance. Second, it must be acknowledged that trade in components can provide African countries with an opportunity to export products that they otherwise would not be able to do so. A concrete example is the BMW assembly plant in South Africa, and the RSAs rapidly increasing exports of automobiles and transportation equipment. A priority issue is how effective has Kenya been in identifying, and capitalizing on, new export opportunities based on the assembly of imported components. Concerning this latter point, comparisons of the composition and growth of African trade in parts and components with that in other geographic regions could be useful. The idea here is to identify items in which production sharing is of major importance outside Africa, and then attempt to identify related constraints to similar activities within Africa. Section X, which follows, argues that foreign direct investment will be of major importance to the establishment of production sharing operations in Kenya 33The pattern observed here runs counter to that between developed and developing countries, that is, components are being produced and shipped from South Africa to Europe for assembly. This is die to the fact that "stowage" factors for assembled furniture are generally far higher than they are for furniture parts. Relative transport costs explain why components are being exported from a developing to a developed country. 68 and other African countries. Unfortunately, African governance and self imposed commercial policies make the region relatively unattractive to foreign investors. Table 8.3 provides a different perspective on this exchange by showing the value of individual African countries imports and exports of components in 1990, 1995 and 2003. Perhaps the most surprising point is the extent to which the Republic of South Africa dominates African trade in components. With $1.7 billion in 2003 exports, the RSA accounts for about 86 percent of Sub-Saharan Africa's shipments of components while, on the import side, its share is just under 50 percent. Kenya is Africa's third largest importer of components, with 2003 trade amounting to about one quarter billion dollars. B. Implications of Production Sharing Outside Africa Key Point Production sharing clearly has unrealized potential for accelerating Kenya's trade, industrialization and growth. An analysis of how countries in the Far East, Europe, or Latin America are exploiting production sharing may provide useful information as to how Kenya might further capitalize on this activity. For example, production sharing between North America and the Caribbean initially developed through the export of clothing and footwear parts from North America for further assembly. These activities then broadened into other areas like the assembly of consumer electronics and office equipment. As far as related developments in other regions are concerned, comparisons between the African and East Asian experience are inevitable. In East Asia, production sharing was a catalyst for the superior trade and economic performance of these countries (an event often referred to as the East Asian miracle). World Bank statistics show Asian trade in parts and components expanded more than five-fold from the mid-1980s to the mid-1990s.34 However, aside from South Africa, Table 8.3 indicates production sharing has not increased at anything close to a similar rate in Africa. Over 1990 to 2003, the value of imports of components actually declined in the case of Burundi, Democratic Republic of the Congo, and in other cases appear to have had little or no influence on assembly related exports. An analysis of Kenya's imports of components reveals a broadly similar pattern to that of other Sub-Saharan countries (see Table 8.4). Parts of telecommunications equipment (SITC 764) constitute about one-third of Kenya's imports of components, followed by parts of motor vehicles. An unusual feature of Kenya's trade is that imports of components appear to have declined by about $80 million over 34It is sometimes argued that Africa's "remoteness" may act as a constraint to industrialization and growth. In this context it should be noted that production sharing between Japan and Indonesia is relatively high and has been growing rapidly. In terms of "great circle" miles the distance between Tokyo and Jakarta is 3,122 miles, while the distance between Nairobi and Rome is 2,900 miles. The distance between Nairobi and Barcelona is about 200 miles more. These comparisons suggest that the importance of Africa's geographic isolation may have been overstated. 69 Table 8.3 Changes in Major African Countries Trade in Parts and Components from 1990 to 2003 Component Trade Values ($000) Share of Parts in Total Trade (%) Trade Flow/Country 1990 1995 2003 1990 1995 2003 African Countries' Exports Angola 935 3,784 8,079 0.02 0.02 0.09 Burundi 183 67 132 0.16 0.16 0.32 Congo, Dem. Rep. 1,089 875 1,271 0.06 0.06 0.12 Congo, Rep. 726 1,329 5,296 0.06 0.06 0.25 Cote d'Ivoire 2,065 8,231 10,150 0.07 0.07 0.21 Ghana 3,513 4,617 2,980 0.29 0.29 0.17 Kenya 9,176 15,372 19,291 0.81 0.81 0.85 Madagascar 14,242 339 2,101 1.63 1.63 0.18 Mauritius 9,205 17,371 39,888 0.74 0.74 2.31 Nigeria 7,785 16,343 45,771 0.06 0.06 0.19 South Africa 247,615 855,254 1,668,244 1.52 1.52 4.61 Sudan 934 1,679 11,821 0.18 0.18 0.46 Tanzania 3,875 6,008 4,511 0.85 0.85 0.56 Uganda 666 1,369 4,342 0.28 0.28 0.99 Zambia 764 1,254 2,069 0.06 0.06 0.30 Zimbabwe 6,128 11,783 5,204 0.48 0.48 0.34 African Countries' Imports Angola 193,772 174,445 589,122 12.42 12.42 13.74 Burundi 21,568 18,626 9,475 13.21 13.21 7.49 Congo, Dem. Rep. 215,273 80,243 72,311 17.29 17.29 10.56 Congo, Rep. 74,904 101,766 133,069 12.08 12.08 13.28 Cote d'Ivoire 115,414 204,166 161,991 7.46 7.46 7.66 Ghana 116,417 227,045 244,958 10.39 10.39 8.50 Kenya 326,277 327,202 248,222 18.08 18.08 9.45 Madagascar 83,105 60,072 70,509 12.55 12.55 6.78 Mauritius 89,380 117,885 167,863 8.80 8.80 9.23 Nigeria 845,724 668,632 1,695,592 16.36 16.36 13.57 South Africa 2,918,284 4,684,771 5,104,523 23.70 23.70 18.46 Sudan 99,846 95,301 244,412 11.04 11.04 11.00 Tanzania 154,364 198,982 192,735 16.26 16.26 11.25 Uganda 57,056 57,911 84,722 11.37 11.37 8.06 Zambia 127,540 147,154 134,601 19.15 19.15 15.29 Zimbabwe 142,253 261,147 137,071 17.50 17.50 10.94 Source: Based on partners data from UN COMTRADE Statistics. 70 Table 8.4 The 2003 Composition of Kenyan and Other Sub-Saharan African Countries Imports of Parts and Components; (values in $000) Kenya Other SSA Countries SITC Component Description Value Share (%) Value Share (%) 711.9 Steam boilers 2,592 1.04 22,148 0.20 713.19 Aircraft combustion engines 124 0.05 21,126 0.19 713.31 Engines for outboard motors 1,540 0.62 29,616 0.27 713.32 Other engines, nes 170 0.07 41,720 0.38 713.9 Internal combustion engines 9,623 3.88 407,874 3.72 714.9 Engines & motors, nes 3,807 1.53 251,332 2.29 716.9 Rotating electric motors 951 0.38 90,185 0.82 718.89 Water turbines 1,476 0.59 18,738 0.17 721.19 Cultivating equipment 906 0.37 26,598 0.24 721.29 Harvesting machinery 421 0.17 16,098 0.15 721.39 Dairy machinery 121 0.05 4,532 0.04 721.98 Wine making machinery 148 0.06 2,202 0.02 721.99 Agricultural machinery, nes 527 0.21 9,493 0.09 723.9 Construction machinery 1,290 0.52 912,477 8.31 724.49 Spinning machinery 1,111 0.45 39,066 0.36 724.69 Knitting machinery 1,309 0.53 29,461 0.27 724.79 Textile machinery 1,820 0.73 12,381 0.11 725.9 Paper making machinery 1,389 0.56 44,838 0.41 726.89 Bookbinding machinery 68 0.03 2,265 0.02 726.9 Printing & typesetting machinery 1,653 0.67 45,714 0.42 727.19 Grain milling machinery 745 0.30 17,522 0.16 727.29 Food processing machinery 3,647 1.47 58,980 0.54 728.19 Machine tools for special industries 425 0.17 23,616 0.22 728.39 Mineral working machinery 2,629 1.06 121,954 1.11 728.49 Special industry machines 8,157 3.29 219,433 2.00 736.9 Metal working machines 2,039 0.82 52,663 0.48 737.19 Foundry equipment 56 0.02 8,195 0.07 737.29 Rolling mills 1,035 0.42 32,890 0.30 741.49 Refrigerating equipment 2,028 0.82 55,071 0.50 742.9 Pumps 2,681 1.08 133,409 1.22 743.9 Centrifuges & filters 3,041 1.23 180,586 1.65 744.19 Fork lift trucks 283 0.11 6,918 0.06 744.9 Loading machinery 4,909 1.98 413,179 3.76 745.19 Power hand tools 694 0.28 34,897 0.32 745.23 Packing machinery 5,187 2.09 85,468 0.78 749.99 Non-electric machinery, nes 2,100 0.85 110,842 1.01 759 Office machinery 15,465 6.23 774,030 7.05 764 Telecommunications equipment 82,592 33.27 2,591,930 23.61 771.29 Electric power machinery 595 0.24 30,989 0.28 772 Switchgear 17,464 7.04 837,034 7.63 775.79 Domestic electrical equipment 85 0.03 4,676 0.04 775.89 Electro-thermal appliances 377 0.15 10,568 0.10 776.89 Electronic components, nes 298 0.12 10,288 0.09 778.19 Electric accumulators 667 0.27 8,820 0.08 778.29 Electric lamps & bulbs 41 0.02 1,950 0.02 778.89 Electric machinery, nes 1,307 0.53 13,504 0.12 784 Motor vehicles 34,378 13.85 2,373,698 21.62 785.39 Cycles 5,797 2.34 186,471 1.70 786.89 Non-motor vehicles 5,071 2.04 28,778 0.26 791.99 Railroad equipment 543 0.22 54,782 0.50 71 792.9 Aircraft 9,255 3.73 299,632 2.73 821.19 Chairs and seats 348 0.14 68,074 0.62 821.99 Other furniture 1,659 0.67 43,459 0.40 874.29 Measuring machines 578 0.23 22,503 0.21 881.19 Still cameras, nes 151 0.06 4,799 0.04 881.21 Cameras under 16mm 105 0.04 3,964 0.04 881.29 Other scientific apparatus 57 0.02 3,937 0.04 884.11 Unmounted optical elements 597 0.24 14,628 0.13 885.29 Clocks and watches 39 0.02 3,782 0.03 899.49 Umbrellas and canes 51 0.02 892 0.01 All Above Products 248,222 100.00 10,976,676 100.00 As a Percent of Mfgs. (excl. Chemicals) 15.41 19.70 As a Percentage of all Goods 9.45 13.03 Source: Based on partners data from UN COMTRADE Statistics. 1995-2003, which implies a reduction in this activity. In other regions, the exchange of these goods generally outpaced general trade growth rates. Focused World Bank analyses have revealed the diversity of production sharing in different regions. For example, In the Far East, electronic components, like micro-chips and transistors, are produced in Japan (this operation is capital intensive) and then sent to relatively poorer countries like Thailand, the Philippines, Indonesia, or China for further assembly. The latter process may involve wiring the components into metal boards, or the final assembly of office equipment, computers, or electronic machines. These finished goods may then be re-exported to Japan or to third countries. The World Bank estimates that in 1996 the value of East Asian intra-trade in parts and components intended for further assembly totaled $165 billion. If serious problems with the internal commercial environment can be corrected (see Section X) these types of assembly operations appear suitable for Kenya. In the United States, synthetic yarns and fabric are produced and exported to Caribbean countries, like Jamaica and the Dominican Republic, where they are sewn into semi-finished, or finished clothing. This activity is encouraged by special provisions in the United States tariff schedules that apply duties only to the value added component of the product assembled abroad, and not on the actual value of the finished good. Estimates indicate the value of Caribbean exports of manufactures produced using imported components rose to approximately $5 billion in the mid-1990s. Statistics provided by Kenya's EPZ authorities are very imprecise, but an effort should be made to determine if further opportunities exist for the local assembly of foreign produced parts of apparel and clothing. In the European Union, metal parts and sheets are produced and exported to Eastern Europe for cutting, shaping and further assembly. EU tariffs also encourage these production sharing activities by applying duties only to the value added component of the foreign assembled good. The value of OECD intra-trade in parts and components has been estimated at over one-quarter trillion dollars in the mid- 1990s. Kenya might have a comparative advantage in these types of operations, particularly if other African countries were the eventual destination for the assembled metal products. Intra-developed and developing country production sharing is of major global importance. In the automotive industry, for example, car parts are produced in Japan and then sent to the United States or Europe for final assembly. Kenya is importing crude petroleum from various sources, refining these imports, and then exporting processed petroleum products to other East African countries. Do similar 72 opportunities exist for the further processing other primary commodities like foodstuffs, wood and paper products, or metal ores?35 Further consideration is clearly warranted as to what types of production sharing activities that have been used successfully in other geographic regions might also be exploited by Kenya. IX. DEVELOPMENTS IN REGIONAL MARKETS Key Point The statistics in this section that analyze Kenya's regional trade must be treated with considerable caution due to concerns about data quality. One major finding, which appears to be well documented, is that Kenya's refined petroleum products have been steadily growing in relative importance and their share of total exports to Uganda and Tanzania is approaching 50 percent. Some "optimistic" assessments of the growing importance of East African intra-regional trade failed to account for the major importance of refined petroleum products. Available statistics do not support the proposition that any long-term product diversification in Kenya's regional trade occurred, indeed it seemingly became more concentrated. Kenya's non-oil regional exports grew at an annual rate of just over 3 percent since the mid-1970s which was well below world trade growth rates. While both the IMF Direction of Trade Statistics, as well as data reported by Kenya, Uganda and Tanzania to UN COMTRADE, concur that Kenya's East African exports are relatively important, there are major uncertainties concerning the magnitude and composition of this exchange. For example, comparisons of partner country statistics for Kenya and Tanzania indicate that total intra-trade of the two countries has been changing in different directions (Table 2.5). According to Tanzania's import statistics, trade with Kenya has been rising sharply since the mid-1990s, while Kenya's export statistics indicate trade with Tanzania has been declining. Furthermore, major differences occur in matched import (Tanzania) and export (Kenya) statistics for products at lower levels of detail. These differences often ranged from several hundred to over one thousand percent (Table 2.6). Perceptions concerning what is happening in East African intra-trade will vary depending on whose statistics you believe, that is Kenya, Tanzania, or Uganda's data ­ or none of the three. A further problem is that a large volume of the exchange between East African countries is "unauthorized" trade that fails to get recorded in official customs statistics. In the preceding sections, a decision was made to use partner country import statistics for analyses of Kenya's export performance. This decision was based on the fact that Kenya's COMTRADE export records have major analytical shortcomings in that shipments from the export processing zones are not included, and data on a relatively high value of exports appear to be otherwise unaccounted for, or "lost" (see Box 2.1). However, for analyses of East African intra-trade there are important uncertainties associated with this approach. First, several international organizations, including the IMF, maintain that under invoicing of imports (to avoid taxes, import duties, or quantitative trade control measures) is a major problem throughout East Africa. This raises the possibility that the biases in Uganda and Tanzania's import statistics may approach, or exceed, those in Kenya's export data. Second, this report demonstrated that Uganda's import statistics are inconsistent at different levels of aggregation. That is, trade totals derived by summing reported three or four-digit SITC import values may be considerably lower than the independently reported trade totals (see Box 2.2). As a result, at relatively low levels of product detail, Uganda's import statistics may provide inaccurate information on Kenya's export 35 There are important reasons why initiatives to encourage food processing, either for export or domestic consumption, should be assigned a high priority. Some estimates suggest that 25 percent or more of local harvests in some developing countries may be lost to spoilage. However, food processing activities like canning, freezing, drying or juicing generally extends the shelf life of agricultural products and reduces the problem of spoilage. 73 performance. In short, serious biases may exist in all available data on the level and composition of East African regional trade. For this reason, conclusions drawn from these statistics need to be interpreted with considerable caution. Given these qualifications, Table 9.1 presents statistics on Kenya's ten largest exports to Uganda and Tanzania in 1976 and 2003. The choice of years was largely determined by the major gaps that occur in Tanzania and Uganda's COMTRADE records from the late 1970s to the mid-1990s (see Table 2.3). Second, the statistics are presented at a relatively aggregate two-digit SITC level due to the inconsistencies in Uganda's trade data below this level, that is, statistics on trade in petroleum products may not be available below this level. Nevertheless, several important points are evident, · First, most previous analyses of East African intra-trade failed to recognize the importance of refined petroleum products in this exchange. In part, this may have been due to the use of IMF Direction of Trade data which reports overall trade levels, but tells nothing about product composition. According to the statistics in Table 9.1 refined petroleum products currently account for $219 million, or about 46 percent of Kenya's exports to Tanzania and Uganda. This is an increase of about 7 percentage points from their share in 1976. Table 9.1 The Largest Products in Kenya's Exports to Uganda and Tanzania in 1976 and 2003 Value of Exports ($000) Share of Exports (%) Product (SITC) 1976 2003 1976 2003 TOTAL EXPORTS 158,976 473,433 100.0 100.0 TOTAL NON-OIL EXPORTS 95,839 252,538 60.3 53.3 PETROLEUM EXPORTS 63,137 220,895 39.7 46.7 Ten Largest Products in 2003 109,202 359,809 68.7 76.0 Petroleum Products (33) 62,784 219,381 39.5 46.3 Non-Metallic Mineral Manufactures (66) 3,350 31,540 2.1 6.7 Paper and Paperboard (64) 4,451 18,235 2.8 3.8 Iron and Steel (67) 3,401 18,227 2.1 3.8 Medicinal Products (54) 6,375 17,203 4.0 3.6 Miscellaneous Manufactures (89) 4,418 14,836 2.8 3.1 Other Metal Manufactures (69) 8,307 10,582 5.2 2.2 Non-Electrical Machinery (71) 10,480 10,137 6.6 2.1 Textile Yarn and Fabrics (65) 3,947 9,991 2.5 2.1 Perfume Materials (55) 1,688 9,676 1.1 2.0 Ten Largest Products in 1976 120,139 315,863 75.6 66.7 Petroleum Products (33) 62,784 219,381 39.5 46.3 Non-Electrical Machinery (71) 10,480 10,137 6.6 2.1 Other Manufactures of Metal (69) 8,307 10,582 5.2 2.2 Other Chemical Materials (59) 6,699 6,532 4.2 1.4 Transport Equipment (73) 6,419 9,102 4.0 1.9 Medicinal Products (54) 6,375 17,203 4.0 3.6 Electrical Machinery (72) 6,085 4,624 3.8 1.0 Paper and Paperboard (64) 4,451 18,235 2.8 3.8 Miscellaneous Manufactures (89) 4,418 14,836 2.8 3.1 Miscellaneous Food Preparations (09) 4,121 5,229 2.6 1.1 Source: Kenya's exports as tabulated from the import statistics of Uganda and Tanzania. The choice of years in this table was necessitated by the fact that Tanzania and Uganda failed to report trade statistics to the United Nations from 1977 to the mid-1990s. 74 · Second, previous studies which compared trade totals often noted that Tanzania and Uganda have a major deficit in their trade with Kenya ­ this came to about $372 million in 2003. A point that has generally been missed is the importance of petroleum in this respect. For all non-oil goods Uganda and Tanzania's deficit with Kenya is about 60 percent lower, or approximately $151 million. Box 9.1 Characteristics of Kenya's 2003 Trade with the Republic of South Africa There are reasons why Sub-Saharan African countries may have a special interest in their trade with the Republic of South Africa. The RSA is the most industrialized economy in the region and appears to be in the process of becoming an increasingly important supplier of some types of capital equipment SSA countries import. South Africa's longer-term prospects as a exporter of transport equipment was enhanced by a recent decision by the BMW corporation to establish an auto assembly plant there. Second, due to its relatively high industrial base, the RSA may become an increasingly important market for many of the raw materials and foodstuffs other African countries export. Third, differences in the level of industrialization and labor costs between South Africa and other African countries could have positive implications for the development of regional production sharing similar to that in East Asia. This scenario has the RSA (like Japan in East Asia) increasingly emerging as an exporter of parts and components for assembly in regional low-wage, less-industrialized, African countries. The statistics shown below report 2003 trade between Kenya and South Africa in terms of major product groups. At present, South Africa is only of very minor importance as a export market from Kenya. RSA imports from Kenya are less than $14 million which is about 4 percent of Kenya's exports to Uganda. Underlying statistics show two products dominate Kenya's exports in the "other manufactures" group, as printed matter and cotton yarn account for approximately two-thirds of all trade in this group. Another point is that trade is highly out of balance, as Kenyan imports from South Africa ($287.8 million) are more than 20 times the corresponding value ($13.8 million) of Kenya's exports. Exports to South Africa* Imports from South Africa** Product Group (SITC) Value ($000) Share (%) Value ($000) Share (%) ALL PRODUCTS (0 to 9) 13,846 100.0 287,760 100.0 Foods and feeds (0+1+22+4) 2,652 19.1 30,887 10.7 Agricultural materials (2-22-27-28) 1,380 10.0 4,376 1.5 Mineral fuels (3) 0 0.0 5,938 2.1 Ores and metals (27+28+67+68) 2,826 20.4 98,158 34.1 Chemicals (5) 2,557 18.5 55,567 19.3 Transport and machinery (7) 1,127 8.1 45,465 15.8 Other manufactures (6+8-67-68) 3,289 23.8 47,354 16.5 Miscellaneous products (9) 16 0.1 15 0.0 An analysis of the underlying statistics shows that trade in parts and components between South Africa and Kenya has been developing rapidly along expected lines. In 2003, Kenya imported $10.4 million worth of components from South Africa, with parts of refrigerating equipment, telecommunications equipment and motor vehicles accounting for more than one-half of the total. These developments should be monitored closely. If Africa, in connection with South Africa, can in some small way, replicate the East Asian pattern of production sharing it could become an important source of growth for the region. * Kenya's exports to South Africa as reported in RSA import statistics. * Kenya's imports from South Africa as reported in RSA export statistics. · Third, the available evidence does not support the proposition that Kenya's regional trade somehow facilitated a de-concentration of trade. In 1976, Kenya's three largest products in Kenya's regional exports (petroleum products, non-electrical machinery, and metal manufactures) accounted for 51 percent of total export earnings. The corresponding share for the three largest products was 7 percentage points higher in 2003. Refined petroleum products were the largest single export in both 75 1976 and 2003. Over this interval their share of total trade rose 7 percentage points to just under 50 percent. · Fourth, there is little evidence that the regional trade facilitated an important expansion of exports of new product lines. Non-metallic mineral manufactures rose almost ten-fold over 1976 to 2003, but the underlying statistics indicates this was largely attributable to the expansion of trade in established products like cement and lime. This point is also reflected in the fact that exports of each of the ten largest products in 2003 totaled at least $1.6 million in 1976. Box 9.1 provides some related information on Kenya's trade with the most industrialized African country, that is, the Republic of South Africa. Table 9.2 addresses a related question, what is Kenya importing from the region and how has the composition of this trade changed. These statistics indicate Kenya's imports from Tanzania and Uganda were, and still are, highly concentrated, but there have been some sizeable underlying changes. The Table 9.2 The Largest Products in Kenya's Imports from Tanzania and Uganda in 1976 and 2003 Value of Imports ($ 000) Share of Total Imports (%) Product (SITC) 1976 2003 1976 2003 TOTAL IMPORTS 31,555 101,081 100.0 100.0 Ten Largest Products in 2003 17,305 89,265 58.5 88.3 Fish and preparations (03) 386 45,094 1.3 44.6 Coffee, tea and spices (07) 981 22,652 3.3 22.4 Textile fibers (26) 2,796 5,823 9.4 5.8 Cereals and preparations (04) 16 4,107 0.1 4.1 Textile yarn and fabrics (65) 1,732 2,630 5.9 2.6 Perfume materials (55) 224 1,997 0.8 2.0 Beverages (11) 56 1,848 0.2 1.8 Fruit and vegetables (05) 1,644 1,763 5.6 1.7 Tobacco and manufactures (12) 8,865 1,678 30.0 1.7 Miscellaneous manufactures (89) 606 1,672 2.0 1.7 Ten Largest Products in 1976 24,749 14,859 83.6 14.7 Tobacco and manufactures (12) 8,865 1,678 30.0 1.7 Textile fibers (26) 2,796 5,823 9.4 5.8 Non-ferrous metals (68) 2,280 91 7.7 0.1 Electrical machinery (72) 2,058 665 7.0 0.7 Textile yarn and fabrics (65) 1,732 2,630 5.9 2.6 Fruit and vegetables (05) 1,644 1,763 5.6 1.7 Rubber manufactures (62) 1,534 213 5.2 0.2 Iron and steel (67) 1,460 696 4.9 0.7 Clothing (84) 1,364 1,300 4.6 1.3 Electric energy (35) 1,016 0 3.4 0.0 Source: Kenya's imports as tabulated from the export statistics of Uganda and Tanzania. relative importance of Kenya's imports of fish and fish preparations rose dramatically (an increase of about 43 percentage points) from 1976 to 2003. The shipments originate entirely in Tanzania and are highly concentrated in two four-digit SITC product groups; fresh or chilled fish fillets (SITC 034.3), where Kenya's imports totaled $27.5 million, and frozen fish fillets with imports of $18.8 million. Overall, the total 2003 value of Uganda and Tanzania's exports to Kenya were only about $101 million, or less than 8 percent of their combined global exports. 76 As was the case with Kenya's regional exports, Table 9.2 does not indicate there has been any general diversification of Kenya's imports from East Africa. In fact, the share of the three largest products imported in 2003 (fish and preparations; coffee, tea and spices; textile fibers) was roughly 25 percent higher than the corresponding three product trade share (47.1 percent) in 1976. Fish and preparations accounted for 45 percent of Kenya's total regional imports in 2003, which is about 15 percentage points higher than the share of the largest import product (tobacco) in 1976. While it is acknowledged that there may be increasingly important trade data quality problems as one moves to lower levels of aggregation, Table 9.3 identifies the four-digit SITC products that experienced the largest 1995-2003 increases in Kenya's exports to Uganda.36 Two important points emerge from these statistics. First, since the mid-1990s Kenya's total non-oil exports appear to have (at best) been stable in value, but within the limits of the accuracy of the underlying data they probably declined in value. According to these statistics, Kenya's 1995-2003 exports of several products like iron plates, wheat flour, medical furniture and trucks declined by between $6 to $16 million. Second, the underlying data provides little indication that regional trade was a catalyst for new products, as the major positive export changes were for products already exported in the mid-1990s. The few exceptions include; unmilled wheat, malt beer, and cigarettes. Similar conclusions emerge from an analysis of Kenya's 1995-2003 exports to Tanzania. This exchange was basically stagnant, but did register an overall increase of about $10 million, largely due to a relatively few products like; boilers and radiators, medical products, and articles used in the packing of goods. A few of the items in the table had very low or nil exports in 1995 (like sugar confection, boilers and radiators, or glassware, but there is no evidence of a significant expansion in the number and types of product lines Kenya exports to Tanzania. A key question is just what should a country like Kenya expected from regional trade. The evidence examined in this section suggests there has been no significant increase in the number of new products traded among East African countries, longer term growth rates for this exchange have been below those for the expansion of world trade. One of the major problems is that East African countries' non-oil exports are highly concentrated in a very few products ­ none of which are of major importance in regional imports. In all likelihood, the problem of non-complimentarity can only be resolved over a fairly long interval. While this point clearly reduces the attraction of many proposals for African regional trade arrangements there are other potential problems, one of which relates to the very small size of many of these countries import markets. For example, the 2003 combined imports of the three East African Community countries (about $7 billion) were less than those of either Peru or Costa Rica, and about one-third those of Viet Nam. Clearly the very small size of markets created by many African RTAs 36If tariff rates for similar products differ substantially there may be an important incentive to misclassified goods in tariff lines where import duties are relatively low. To the extent that such intentional misclassifications are occurring it would tend to reduce the reliability of lower level trade statistics. For example, if tariffs are lower on (say) cane sugar than they are on beet sugar, importers would have an incentive to incorrectly invoice beet as cane sugar. For these, and other reasons, it is often recommended that countries establish a relatively similar tariff bands for their imports. For example, all imports might be taxed at one of three set tariff rates (say 5, 10 or 15 percent). 77 Table 9.3. Dynamic and Declining Products in Kenya's Recent Non-Oil Exports to Uganda Export Value ($000) Percent of Total (%) Export Change ($000) SITC Export Product (SITC No.) 1995 2000 2003 1995 2000 2003 2000-2003 1995-2003 Total non-oil exports 203,057 127,610 173,848 100.0 100.0 100.0 46,239 -29,208 MAJOR PRODUCT GAINERS 223.9 Flours of oil seeds 85 0 8,125 0.0 0.0 4.7 8,125 8,040 674.6 Iron sheets 735 9,726 3,334 0.4 7.6 1.9 -6,392 2,599 642.1 Boxes & packing containers 1,731 2,506 4,321 0.9 2.0 2.5 1,814 2,590 541.7 Medicaments 6,509 5,416 9,031 3.2 4.2 5.2 3,615 2,521 658.1 Sacks and bags 321 120 2,705 0.2 0.1 1.6 2,585 2,385 048.2 Malt 499 1,745 2,825 0.2 1.4 1.6 1,080 2,326 641.5 Paper and paperboard 1,957 3,084 3,778 1.0 2.4 2.2 693 1,820 895.2 Pens and pencils 124 138 1,887 0.1 0.1 1.1 1,749 1,763 553.0 Perfumery and cosmetics 995 775 2,743 0.5 0.6 1.6 1,968 1,748 554.2 Organic surface-active agents 291 2,309 2,015 0.1 1.8 1.2 -294 1,724 562.9 Fertilizers 175 822 1,877 0.1 0.6 1.1 1,054 1,701 641.3 Kraft paper 1,422 2,497 2,927 0.7 2.0 1.7 430 1,504 041.2 Other wheat not milled 0 0 1,362 0.0 0.0 0.8 1,362 1,362 112.3 Beer made from malt 7 915 1,159 0.0 0.7 0.7 244 1,153 641.7 Corrugated paper 477 1,778 1,564 0.2 1.4 0.9 -213 1,087 728.4 Machines for special industries 303 94 1,268 0.1 0.1 0.7 1,174 965 523.2 Metallic salts 379 827 1,290 0.2 0.6 0.7 463 911 651.5 Synthetic yarn 64 409 923 0.0 0.3 0.5 514 859 073.0 Chocolate 17 476 791 0.0 0.4 0.5 314 774 122.2 Cigarettes 1 836 765 0.0 0.7 0.4 -71 764 MAJOR DECLINING PRODUCTS 061.1 Sugars 2,366 702 210 1.2 0.5 0.1 -491 -2,156 673.3 Iron shapes 2,782 1,143 574 1.4 0.9 0.3 -569 -2,209 684.2 Aluminum and alloys 4,295 1,561 1,844 2.1 1.2 1.1 283 -2,451 554.1 Soaps and detergents 4,336 787 1,757 2.1 0.6 1.0 970 -2,580 278.3 Common salt 9,153 4,846 6,569 4.5 3.8 3.8 1,723 -2,584 431.2 Vegetable oils & fats 3,174 349 348 1.6 0.3 0.2 -1 -2,826 782.1 Trucks 4,564 1,714 1,055 2.2 1.3 0.6 -659 -3,509 661.2 Portland cement 28,952 12,395 22,997 14.3 9.7 13.2 10,602 -5,955 821.2 Medical furniture 6,189 39 41 3.0 0.0 0.0 2 -6,148 674.9 Iron sheets 19,425 1,454 6,853 9.6 1.1 3.9 5,398 -12,572 046.0 Flour of wheat 15,513 2,510 4 7.6 2.0 0.0 -2,505 -15,508 Source: Kenya's exports as tabulated from Uganda's import statistics from UN COMTRADE Statistics. 78 Table 9.4. Dynamic and Declining Products in Kenya's Recent Non-Oil Exports to Tanzania Export Value ($000) Percent of Total (%) Export Change ($000) SITC Export Product (SITC No.) 1995 2000 2003 1995 2000 2003 2000-2003 1995-2003 Total non-oil exports 68,829 62,864 78,890 100.0 100.0 100.0 1,6026 10,061 MAJOR PRODUCT GAINERS 812.1 Boilers & radiators for residential buildings 4 104 3,976 0.0 0.0 5.0 3,872 3,972 541.7 Medicaments 3,278 6,016 6,980 4.8 4.8 8.8 964 3,702 893.1 Articles used for packing 576 2,817 3,562 0.8 0.8 4.5 745 2,986 684.2 Aluminum and alloys 573 1,978 2,891 0.8 0.8 3.7 914 2,318 091.4 Margarine and lard 411 1,089 1,972 0.6 0.6 2.5 883 1,561 592.1 Starches and wheat gluten 13 1,572 1,416 0.0 0.0 1.8 -156 1,403 723.4 Construction and mining machinery 37 129 1,254 0.1 0.1 1.6 1,124 1,217 665.2 Glassware 37 20 1,234 0.1 0.1 1.6 1,214 1,197 661.1 Quicklime and slaked lime 2 295 1,163 0.0 0.0 1.5 868 1,162 697.4 Domestic articles 33 1,311 1,167 0.0 0.0 1.5 -144 1,134 044.0 Maize 0 173 1,075 0.0 0.0 1.4 902 1,075 674.6 Sheets & plates of iron 59 83 962 0.1 0.1 1.2 879 903 786.8 Other vehicles not mechanically propelled 30 82 909 0.0 0.0 1.2 826 878 553.0 Perfumery and cosmetics 19 837 876 0.0 0.0 1.1 39 857 523.2 Metallic salts 573 874 1,395 0.8 0.8 1.8 521 822 062.0 Sugar confectionery 74 508 816 0.1 0.1 1.0 308 743 658.3 Travel rugs and blankets 3 32 679 0.0 0.0 0.9 648 677 772.1 Switches & relays 56 122 708 0.1 0.1 0.9 586 652 598.9 Chemical products and preparations 119 299 744 0.2 0.2 0.9 445 626 554.2 Organic surface-active agents 49 316 653 0.1 0.1 0.8 338 604 MAJOR DECLINING PRODUCTS 741.6 Simple laboratory equipment 753 32 73 1.1 1.1 0.1 42 -679 673.2 Bars & rods iron 976 115 196 1.4 1.4 0.2 80 -781 699.6 Miscellaneous articles of base metal 1,116 222 177 1.6 1.6 0.2 -45 -939 642.1 Boxes & packing containers 2,959 1,940 1,905 4.3 4.3 2.4 -34 -1,054 783.1 Busses and public transport vehicles 1,244 393 101 1.8 1.8 0.1 -292 -1,143 784.1 Chassis fitted with engines 1,287 47 0 1.9 1.9 0.0 -47 -1,287 665.1 Containers of glass 2,872 1,114 455 4.2 4.2 0.6 -659 -2,417 892.8 Printed matter n.e.s. 4,325 307 137 6.3 6.3 0.2 -171 -4,188 112.3 Beer made from malt 6,242 155 51 9.1 9.1 0.1 -104 -6,191 782.1 Trucks and vehicles for the transport of goods 17,557 2,881 1,029 25.5 25.5 1.3 -1,852 -16,528 Source: Kenya's exports as tabulated from Tanzania's import statistics from UN COMTRADE Statistics. 79 are insufficient to attract significant increases in trade related investments, or to facilitate the achievement of economies of scale that might be achieved through more open export oriented trade policies. Proposals for African regional trade arrangements can be traced back to at least the 1960s and 1970s. A point that should be recognized is that some of the reasons for the early proposals are clearly of greatly weakened validity. One of the major reasons for the interest in regional trade arrangements was to help African countries overcome any constraints associated with the small size of their own domestic markets, and the importance of government imposed trade barriers in foreign markets. However, the Kennedy, Tokyo and Uruguay Rounds of multilateral trade negotiations, along with the Generalized System of Preferences, the Lome convention, and the implementation of AGOA reduced tariff barriers against most African exports to zero or insignificant levels.37 In short, the international environment has become so "open" that African countries can now easily use international markets to overcome any constraints associated with the small size of their domestic economies. It should be noted that the so called Asian "newly industrialized countries (NICs) were able to capitalize on opportunities in the 1960s and 1970s ­ when trade barriers were considerably higher than at present ­ to accelerate their own industrialization and growth (see Box 9.2). The Asian countries success is attributable to the fact that they created an appropriate internal environment that was favorable to trade and commercial activity (see Section X which follows). Although the issue has received relatively little attention, there are reasons why the implementation of East African regional trade preferences should be treated with some caution, since they have the potential to be detrimental to industrialization and growth. East African countries do not have a comparative advantage in the types of machinery and capital goods that are vital for regional development, and are of paramount importance in regional imports. If trade preferences, within Africa's relatively high tariffs and nontariff barriers, did stimulate some intra- trade in these sectors, it could negatively influence the region's competitiveness in other international markets.38 For example, if other developing countries (say those in East Asia) source imports of manufacturing inputs, like textiles and fibers, or production equipment, from global low cost producers, while preferences cause East African countries to turn to higher-cost less-efficient sources in the region (whose products might also be less reliable and of lower quality), this could reduce Kenya and other East African countries ability to compete in non- regional markets. Globally, East Africa would continue to be marginalized! The discussion to this point raises a policy issue of major importance, that is, what priorities should be assigned to a trade liberalization through the exchange of regional preferences, as opposed to a general liberalization on a most-favored-nation (MFN) basis. 37Aside from tariffs, international transportation costs can also pose an important barrier to international trade. A point that has not received adequate recognition is that developing countries like Kenya may have more options for reducing their international freight costs than is often realized. An annex to this report examines the level and structure of Kenya's nominal transport costs for exports to the United States and the possible implications of the US Open Skies for Africa program. 38This is precisely what seems to have happened in the case of Mercosur (a regional trade arrangement initially between Argentina, Brazil, Paraguay and Uruguay. High MFN tariffs against non-member countries, coupled with tariff preferences for members, stimulated production and internal consumption of capital intensive machinery and transportation in which the Mercosur countries did not have a comparative advantage in production. The internal use of this "high cost" capital equipment was one factor contributing to the balance of payments and other financial problems in the late 1990s. 80 Clearly, there are important considerations that make the former far less attractive. These include the "non-complementarity of most African countries exports and imports, the very small size of most markets created by African regional trade arrangements, and the potential for stimulating intra-trade in high cost products which are not globally competitive. In addition, a series of multilateral trade negotiations like the Tokyo and Uruguay Rounds, as well as the adoption of special preferences like AGOA, have greatly expanded export opportunities for countries like Kenya in major foreign markets. A general "efficiency enhancing" reduction of trade barriers on a most-favored-nation basis would best allow Kenya to capitalize on these opportunities.39 X. PROSPECTS FOR EXPORT DIVERSIFICATION Key Point In terms of potential new export ventures, there is some evidence that Kenya may have more options for diversification than is generally thought. In this section consideration is given to two related points; how might potential new export products be identified, and to what extent internal conditions in Kenya act as constraints to the development of new exports. For example, evidence from the Africa competitiveness report suggests that Kenya is at a disadvantage vis-à-vis more than one half the other African countries surveyed. The preceding analysis argued Kenya and other African countries export prospects are unfavorable due to their high concentration in products that are of diminishing relative importance in world trade. If Africa continues to rely on these goods the region will experience negative economic effects such as a continual declining share of world trade and below average growth and industrialization prospects. Long-term real prices for Kenya's two major commodity exports (coffee and tea) have been on a secular decline, while nominal prices for these products often record major year-to-year fluctuations which make development planning difficult. The policy prescription is clear. Diversify away from traditional products! Unfortunately, recent statistics show Kenya and other African countries made little progress in this direction. In fact, Ng and Yeats (2000, pp. 22-23) found that "no major changes in the diversification of African exports occurred in the 1990s, indeed several statistical indices suggest some African countries' exports became more concentrated". African anti-competitive, anti-open, domestic policies were cited as a major reason for this failure (Ng and Yeats 1997).40 39In a recent IMF study Yang and Gupta (2005) effectively argue the liberalization of trade barriers on an MFN basis holds far more promise for African countries than does a further exchange of regional preferences. In part, this conclusion is based on an analysis of regional trade statistics which indicate African RTAs have had no discernable effect in increasing the relative importance of intra-trade, nor have they increased the number of products traded, nor have they enhanced Africa's competitiveness in global markets. The authors also argue the major disparities between the types of goods African countries export and import greatly reduces the potential for increased intra-trade. 40 Could World Bank policies do more to promote diversification? Brownbridge and Harrigan (1996) examined changes in the export profiles of African countries involved in World Bank structural adjustment programs (SAPs). They observe (p. 419) that, over a decade, sixty percent of the recipients exported a smaller number of products at the end of the period than at the beginning. The authors concluded that "The emphasis placed by structural adjustment programs on price reforms to boost exports is likely to be of greater benefit to the traditional primary commodity exports than to non-traditional products, since the latter often face other severe constraints such as inadequate technological, managerial, technical and marketing skills, infrastructure, and lack of finance. Remedying these constraints has not been addressed in most SAPs in Sub-Saharan Africa." 81 Box 9.2 Trade Barriers Facing East Asian Exporters at the Start of their Industrialization Drive One way to assess the importance of foreign trade barriers now facing Kenya and other African countries is to compare them with those that faced the "newly industrialized countries" (NICs) when the latter began their successful export oriented industrialization drive in the mid-1960s. Since the NICs experience showed that the external barriers they faced could be overcome, such an inter-temporal comparison could help indicate whether Africa's export problems are largely attributable to external protectionism, or to African internal policies and conditions bearing on the export sector. The statistics shown below report the average level of tariffs countries like Republic of Korea, Hong Kong, Singapore and Taiwan (China) faced in major OECD markets in the late 1960s. Industrial countries import duties facing their exports averaged about 17 percent and reached a high of 19.5 percent in the United Kingdom. The statistics also indicate that developed countries' tariffs then incorporated a high degree of discrimination against developing countries. This is reflected in the considerably higher than average tariffs on developing countries exports than on all shipments. Two other points should be noted. First, in the 1960s GSP schemes had not yet been adopted so the Asian NICs had to compete with other suppliers on an equal MFN basis. Second, the degree of escalation in OECD tariffs was far greater than it is today. Mid-1960 Tariff Averages on Mid-1960 Tariff Averages on Imports of Import Market Total Imports of Manufactures Manufactures from Developing Countries United States 11.5 17.9 United Kingdom 15.2 19.5 European Community 11.0 14.3 Sweden 6.6 9.8 Japan 16.1 18.0 All Developed Countries 10.9 17.1 Source: UNCTAD, The Kennedy Round Estimated Effects on Tariff Barriers, (TD/6/Rev. 1), (New York: United Nations, 1968). In contrast to the situation the Asian NICs faced (and overcame), one estimate placed the average level of OECD tariffs on Africa at well under one percent. This was before the introduction of recent tariff programs in favor of Africa (like AGOA). The evidence clearly indicates that Africa's sub-par export performance cannot be attributed to OECD tariffs. Could it be that nontariff measures are now more restrictive against African exports than they were against the Asian NICs. Available evidence strongly suggests the opposite. For example, the Short Term Textile Arrangement was initiated in 1961 and followed by the Long Term Arrangement in 1962. Both placed important restrictions on the NICs exports of textile and clothing products. "Voluntary" export restraints (VERs) were extensively applied to NIC exports in the 1970s and 1980s, yet no such restraint was ever applied to Africa's exports. Two important conclusions emerge from these statistics. First, with the possible exception of some agricultural products Africa's trade problems are not the result of an unfavorable external environment. As such, the focus should be on Africa's own internal commercial environment. Second, these observations generate reservations about the potential adverse effects of preferential regional trade arrangements. If the RTAs promote the exchange of relatively high cost products, for which member countries do not have a comparative advantage in production, this may have serious negative effects on industrialization, trade, and growth. 82 A recent OECD (2003) report stressed the importance of creating an appropriate internal commercial environment for encouraging export diversification, and also accented the need for a comprehensive strategy for trade expansion. As Box 10.1 indicates these strategies may differ depending on a country's objectives, its natural resource base and other endowments, or its established trading contacts. While acknowledging the potential importance of external market Box 10.1 Examples of Successful National Diversification Strategies Chile: the successful transformation of the Chilean economy rested upon the adoption of a coherent policy package that the private sector helped adopt. Three major lessons emerge from this experience. First, domestic trade liberalization was implemented in conjunction with appropriate fiscal and monetary polices to encourage investment and restructuring, as well as with industrial policies correcting for market failures and weaknesses in the private sector (that is, dissemination of market information and financial support for research and development). Second, export diversification was incorporated in an overall development strategy to promote several strategic sectors such as forestry and wood products, to attract foreign direct investment in export-oriented or science-based sectors, to promote regional trade integration, and to strengthen linkages between the resource-based sectors and the rest of the economy. Finally, private-public partnership was promoted to ensure the success of policy reform. The Chilean case was cited as an example of how natural resource based sectors can sustain national growth for long periods, while at the same time favoring vertical diversification and creating the knowledge base for new exports. Costa Rica: the gradual transition to an open trade regime minimized recessions and large- scale unemployment, while bolstering export growth (with an annual average rate of 14 per cent over the 1961-82 period). The active participation of private sector organizations to this reform process dramatically increased its effectiveness. Initially, emphasis was placed on competitive and stable exchange rates and direct subsidies to compensate for the anti-export bias, to promote traditional exports (defined as coffee, banana, beef and cocoa). As export diversification (from the mid-1980s onward) gathered momentum, tariffs were reduced and other distortions removed. Regional trade integration helped to address limitations associated with the small size of the internal market. Investment in science, technology and human resource development helped this process. Equally important was an active foreign direct investment policy to encourage multilateral activities in certain targeted sectors such as electronics. Malaysia and Thailand stand out as successful examples of both vertical and horizontal diversification. Both governments adopted a dual strategy to upgrade natural resource-based industries (such as palm oil and rubber products in Malaysia, and agricultural and fish products in Thailand) and to encourage labor-intensive manufactured exports, most notably clothing and electronics. Agriculture played a key role in the industrialization process. The development of traditional (e.g. rice and rubber) and high-value, export-orientated agricultural crops stimulated the growth of agro-industry. In the case of palm oil and rubber, Malaysia set up specialized agencies to promote production and upgrading, and used the proceeds of production and export taxes to finance research and development investments. Both countries established export processing zones and licensed bonded warehouses as a means of stimulating manufactured exports and attracting foreign investment. FDI came mostly from neighboring Asian countries (Japan and Asian NIEs). The development of natural resource-based sectors helped both countries to cope with the economic downturn after the mid-1990s, which affected manufactures exports most severely. Source: adapted from OECD 2003 83 access conditions, the OECD urges far more attention be paid to correcting domestic constraints to export development. A first step in building the capacity to trade competitively would be to identify key domestic barriers to international business development, and to take measures to improve the local "commercial environment." The latter include measures such government policy constraints (e.g. the anti-export bias associated with exchange rate misalignment and barriers to international trade), financial market constraints (that is. limited provision of export credit and insurance), poor infrastructure and administrative constraints, such as high transport costs and governmental red tape, and limited knowledge of trading practices (e.g. lack of information on foreign market structure, a lack of contacts with potential importers, and a lack of marketing expertise). Clearly foreign investment will be of major importance if Kenya is to diversify exports into new lines, and the ability to attract investment is crucial to the success of these efforts. Kenya may be disadvantaged by several physical characteristics, like its relatively small internal market size. While government policy makers have little control over such variables, they clearly can influence key trade, monetary, fiscal, foreign investment, or tax regulations that have a major influence on the internal commercial environment, and the country's ability to attract foreign investment. An increased importance has been attached to the fact that Kenya and other African countries often pursue economic policies that make them relatively unattractive to potential investors. For example, in an address to the UN Economic Commission for Africa, World Bank President James Wolfenson graphically highlighted the general importance, and negative influence, of these commercial and governance problems throughout Sub-Saharan Africa, "But what do we see when we look at Africa? We see that Africa is missing out. Of $300 billion in total foreign private capital flows, Sub-Saharan Africa received about $12 billion. And of that, only $2.6 billion in direct investment ­ a trivial number in relation to the size and potential of the continent. But we also have to face facts. It is not just because the private sector is myopic that less than 1 percent of direct investment comes to Africa. Africa needs to set itself up to attract private investment and that means a clean regulatory environment, it means a judicial system that works, it means property rights, corporate law, predictability in taxes and, in relations to governments, it means capacity building, health care, and the infra-structure to go along with it. And it means that corruption must be stamped out. Without these, private investors simply will not invest." A. Evaluating Kenya's Commercial Environment Key Point Policy makers in specific countries previously had difficulty in determining how their domestic commercial environment compared with those implemented elsewhere. Several recent efforts to compile comprehensive cross-country indices of the quality of governance and commercial policies now provide relevant information. These statistics suggest that domestic commercial policies make Kenya relatively less attractive for foreign investment compared to many other developing countries. Less than 35 percent of all Latin American countries have a domestic commercial environment inferior to that in Kenya, while two-thirds of the East Asian countries have a better environment. A recently released Wall Street Journal-Heritage Foundation survey concludes Kenya's internal commercial environment is having a major retardation effect on trade, investment and growth. How does the commercial environment in Kenya compare to that in other developing and developed countries? A practical problem relating to this point is that policy makers in a specific African country previously had difficulty in determining how their own domestic commercial 84 environment compared to those implemented elsewhere, or even if their economic and governance policies were relatively favorable or repressive. Specifically, Kenya must compete with other countries for foreign investment and a key question is whether its own internal policies make it relatively attractive, or unattractive, to foreign finance. If Kenya is relatively unattractive, which specific policies are primarily responsible for the country's negative image. While it previously would have been difficult to address questions like these, several recent initiatives to compile comprehensive cross-country indices of the quality of governance and the commercial environment can provide much useful relevant information. One such initiative involved an effort by Transparency International (TI) to construct numeric indices of the extent of corruption in over 100 countries by using detailed interviews with government officials and local and foreign businessmen. The indices ranged, theoretically, from a value of zero for a country perceived to be totally corrupt, to a value of 10 for a country perceived to be totally clean.41 More recently, the Wall Street Journal and Heritage Foundation (WSJ-Heritage 1997) compiled, and annually updated, an index that measures the general commercial policy environment in over 150 developed and developing countries. This initiative produced similar indices on government policies relating to ten specific business factors, namely, · Corruption in the judiciary, customs service, and government bureaucracy; · Non-tariff barriers to trade, such as import bans and quotas as well as strict labeling and licensing requirements; · The fiscal burden of government, which encompasses income tax rates, corporate tax rates, and trends in government expenditures as a percent of output; · The rule of law, efficiency within the judiciary, and the ability to enforce contracts; · Regulatory burdens on business, including health, safety, and environmental regulation; · Restrictions on banks regarding financial services, such as selling securities and insurance; · Labor market regulations, such as established work weeks and mandatory separation pay; · Informal market activities, including corruption, smuggling, piracy of intellectual property rights, and the underground provision of labor and other services The WSJ-Heritage Indices for each of these ten variables are assigned on the basis of objective criteria, and can take a value of one to five with lower values indicating an environment more conducive to economic growth.42 Since the index values are assigned on the basis of 41 Kenya's most recent TI index was 1.9 which signifies a very high level of official corruption. Only 10 of the 133 countries for which the index was calculated had a lower corruption index. These were; Angola, Azerbaijan, Cameroon and Georgia (all with an index of 1.8), Myanmar and Paraguay (with indices of 1.6), Haiti (1.5), Nigeria (1.4), and Bangladesh (1.3). See Transparency International (2004) for a discussion of the methodology used in the construction of this index. 42For example, in the construction of the taxation index five gradations of income tax levels were set. These ranged from no taxes on income, or a flat tax of 10 percent or less ­ a situation assigned a value of 1.0 ­ as opposed to an environment where a top rate of over 50 percent was applied, along with a tax on average income between 20 and 25 percent. This latter "worst case" environment was assigned an index of 85 clearly specified empirical standards, the potentially corruptive influence of subjective judgments is greatly reduced.43 Finally, an overall index of the quality of the national economic environment was derived for each country by averaging the ten WSJ-Heritage policy indices. This overall index is viewed as a measure of the attractiveness of a country for foreign investment and its ability to sustain industrialization and economic growth.44 The four broad categories of economic freedom in the Index are: · Free -- countries with an average overall score of 1.99 or less; · Mostly Free -- countries with an average overall score of 2.00 to 2.99; · Mostly Not Free -- countries with an average overall score of 3.00 to 3.99; and · Repressed -- countries with an average overall score of 4.00 or higher. Box 10.2 provides empirical information on the general relationship between index levels and GDP per capita as well as changes in the index for Sub-Saharan African countries. The message from these statistics is clear. Countries that implemented governmental policies that produced more favorable commercial environments have been able to achieve relatively higher national income levels According to the WSJ-Heritage indices, Kenya is ranked 67th out of 126 developing countries, and its overall index of 3.26 places it in the "mostly not free" country group -- which indicates a generally unfavorable commercial environment. Kenya's worst scores (4 or more) occur for the relative height of trade barriers, government regulation of industry, and the size of the informal business sector. Its best scores (2.0) are for monetary policy and a relative absence of wage and price controls. Table 10.1 reports Kenya's indices for each of the 10 commercial policy measures along with WSJ-Heritage comments as to why these scores were assigned. 5. Similar gradations were made for intermediate corporate tax rates and index values assigned accordingly. The corporate and income tax indices were then averaged to arrive at an overall "tax environment" index for the country. 43As an illustration, a trade policy index of 1 was assigned to a country with average tariffs of less than or equal to 4 percent, while an index of 2 was assigned to countries whose tariffs were in the 4 to 9 percent range. The worse rating (an index of 5) was assigned to countries with average tariffs exceeding 19 percent. WSJ-Heritage also examined various published studies by the IMF, WTO, World Bank and UNCTAD to determine the extent that nontariff protection was used to supplement that from tariffs. Where NTBs were extensively applied the tariff derived index was increased by one point. 44Ng and Yeats (1999) employ the WSJ-Heritage index to show conclusively that countries with relatively attractive commercial environments achieved significantly higher levels of GDP per capita, they experienced higher growth rates for exports, imports and GDP, and were more successful in integrating into the global economy. Regression results indicated the WSJ-Heritage commercial indices explain over 60 percent of some measures of economic performance, which implies that a country's own national policies are a major determinant of its rate of growth. 86 Box 10.2 Income Levels and the General Commercial Environment in Africa. Empirical analyses document the fact that countries maintaining a favorable internal commercial environment generally experience superior economic performance and growth. Countries with a "free" commercial environment have achieved GDP per capita levels that are roughly 30 times higher than those classified in "repressed" or "mostly not free" groups. When examined independently, a similar pattern emerges for the Sub-Saharan African countries, that is, those with a superior "mostly free" commercial environment achieved GDP per capita levels that were three to four times higher than those that pursued less attractive commercial policies. These observations raise an important question. Have recent changes in Sub-Saharan African countries generally been positive or negative? GDP Per Capita (US$) Commercial Environment All Countries Sub-Saharan Africa Free 30,144 -- Mostly Free 9,674 2,304 Mostly Not Free 1,177 745 Repressed 1,093 539 Overall, the commercial environment improved recently in sub-Saharan Africa, with 21 countries' WSJ-Heritage indices moving in a positive direction. However, the majority of these countries - 30 out of 42 ­ remain in the "mostly not free" category. Of the 10 index components, government intervention registered the greatest net improvement, with direct government involvement declining in 18 countries. The largest negative changes occurred for fiscal burden factor as marginal tax rates rose in 21 countries. Five countries in this region (Rwanda, Ethiopia, Cape Verde, Senegal, and Mauritania) are among the world's 10 most improved in terms of their overall business climate. Rwanda experienced the single greatest degree of improvement overall: an impressive feat considering it previously was one of the countries with the greatest decline in economic freedom. Rwanda has improved its trade policy, government intervention, monetary policy, and regulation scores. However, four SSA countries, namely, Namibia, Madagascar, Lesotho, and Gabon are among the 10 countries whose scores worsened by the world's widest margins. In addition, Zimbabwe continues to have the worst commercial environment of any Sub-Saharan country. As a result, unemployment stands at about 80 percent, inflation is over 200 percent per annum. On the other end of the spectrum, the Index suggests Botswana continues to have the most favorable commercial environment in the region despite a worse score this year. Both its trade policy and fiscal burden of government indices recorded negative changes. Uganda, which privatized 74 businesses over the past decade and is targeting 85 more, has the second best commercial environment. In Kenya's case, the Index suggests the commercial environment has essentially been static over the last eight years. During 1997 to 2000, the overall WSJ-Heritage index averaged 3.12 as opposed to 3.25 for the 2001 to 2004 interval. Currently, Kenya records its worst indices for trade policy, government regulation, and the functioning of the informal market. 87 An important question is whether each of the ten commercial policy measures reported in Table 10.1 should, or should not, be considered of equal importance. While the normal practice is to assume each variable is of equal importance, regression analyses by Ng and Yeats (1999) indicate the trade policy variable has a major, above average, influence on the speed with which individual developing countries are becoming more fully integrated into the global economy.45 However, in a recent survey of the commercial environment in Kenya, the Africa Regional Program on Enterprise Development (2003) reported that crime and corruption was frequently reported to be the most serious constraint on commercial activity in Kenya. Table 10.1 Kenya's Scores for Major Policy Variables Influencing the Commercial Environment (Scores: 1 best possible, 5 worst possible) Trade Policy: Score 5 (Recent change ­ worse: High level of protectionism) Comment: The World Bank reports that Kenya's weighted average recently rose to 15.5 percent from 12.4 percent. Non-tariff barriers include the required use a Kenya appointed inspection firm for imports, special packaging and labeling requirements, and burdensome licensing requirements. Fiscal Burden of Government: Score 3.6 (Recent change ­ stable) Comment: Kenya's top income tax rate is 30 percent. The top corporate income tax rate is 30 percent. Government expenditures as a share of GDP increased by the same amount (0.6 percentage point to 26.3 percent) in 2001 as they did in 2000. Government Intervention in the Economy: Score 2.5 (Recent change ­ improvement) Comment: The government consumed 16.8 percent of GDP in 2001, down from the 18 percent reported for the previous year. In the July 2001­June 2002 fiscal year, World Bank reported Kenya received 7.4 percent of its total revenues from state-owned enterprises and government ownership of property. Monetary Policy: Score 2.0 (Stable: Relatively low level of inflation) Comment: From 1993 to 2002, Kenya's weighted average annual rate of inflation was 3.80 percent. Capital Flows and Foreign Investment: Score 3.0 (Stable: Moderate barriers) Comment: Kenya's government has relaxed its screening standards and is developing a one-stop shop for investment approval. According to the U.S. Department of State, foreign and domestic investment is restricted in those sectors in which the state has a monopoly. These include the power, telecommunications, and ports sectors. The government often discriminates in favor of domestic bids. 45 These regressions utilized a World Bank (1996, Appendix 2) "speed of integration index" which incorporated the rate of change in four underlying changes, namely, (i) changes in real trade as a share of GDP, (ii) changes in a country's institutional investor ratings, (iii) foreign direct investment (FDI) as a share of GDP, and (iv) the share of manufactures in total exports. The index is an average of the scores for these four variables after they were standardized to have a mean of 0 and a standard deviation of 1. The index ranges between values of ± 3.0 with higher positive index values identifying countries that were more successful in integrating globally. Negative values indicate the links were either deteriorating or developing at a below average rate. 88 Banking and Finance: Score 3.0 (Stable: Moderate level of restrictions) Comment: Kenya's banking system is troubled. As of mid-2002, an estimated 41percent of loans were non-performing, with most of these loans held by state-controlled banks. According to the U.S. Department of State, "The banking problems in Kenya are the result of poor bank management, inadequate government supervision, political pressure to make loans that are rarely paid, and current economic conditions." Wages and Prices: Score 2.0 (Improvement: Moderate level of intervention) Comment: According to the Economist Intelligence Unit, "Price controls were abolished some years ago, but the legal mechanism to control prices exists in part IV of the Restrictive Trade Practices, Monopolies and Price Controls Act." The government intervenes in agriculture markets to various degrees to support farmers. Security of Property Rights: Score 3.0 Comment: Expropriation of property is unlikely in Kenya. However, the Economist Intelligence Unit reports, "Although... arrangements [are] more secure than in many other African countries, abuses and disputes are common. The country's judicial system is widely regarded as overloaded, inefficient and often corrupt. There is little confidence in the lower courts." Government Regulation of Industry: Score 4.0 Comment: Kenya's bureaucracy remains significantly burdensome. The Economist Intelligence Unit reports "investors should be aware that the official register is in a deplorable state; it has never been computerized or properly updated." The government gives local authorities "discretion to choose the appropriate schedule of fees to charge, depending on the size and level of development of the local authority concerned"; but businesses complain that, because of this discretion, they sometimes have to "pay more for a single business permit than they have paid before for many trading licenses." Informal Market: Score 4.5 Comment: Evidence indicates a substantial volume of commercial activity in Kenya occurs through informal channels and goes unreported. In part this is due to a cumbersome bureaucracy and red tape. Source: Wall Street Journal-Heritage Foundation (2004). Table 10.2 provides a somewhat different perspective on the question of how attractive is Kenya for foreign investors relative to other countries.46 Here, Kenya's overall commercial 46 It must be noted that preconceived notions about social and other domestic conditions will certainly influence transnational corporations decisions about investing in specific countries. These are not accounted for in the Wall Street Journal-Heritage Foundation approach. However, reports such as the following news article that appeared recently in the New York Times (January 5, 2005, p. 5) clearly will have a serious negative impact on investment decisions "Workers at Kenya's main market killed 6000 rats, trucked away 750 tons of garbage and sucked 70 tons of human waste out of the latrines in three days of the first major cleanup of the market in 30 years, a government minister said. The Wakulima Market, which supplies fresh food to most of Nairobi's three million was a public health hazard, the rubbish piling up seven feet deep in some places, said local government Minister Musikari Kombo, who ordered the 89 policy index is compared to those for other regional groups of developing countries ­ like those in Latin America or East Asia. The table shows the average and median index for each regional group, and also indicates Kenya's position relative to the other countries. For example, Kenya's index (3.26) is slightly below (better) than the average for all other Sub-Saharan African countries, while the commercial environment appears to be more attractive in 35 percent of the other African countries. These conclusions are in broad agreement with the findings of a recent Africa Region Program for Enterprise Development (2003) survey of constraints facing commercial enterprises operating in Kenya.47 These statistics suggest that domestic commercial policies make Kenya relatively less attractive for domestic or foreign investment compared to many other developing countries. Less than 35 percent of all Latin American countries have a domestic commercial environment inferior to that in Kenya, while two-thirds of the East Asian countries have a better environment. The East Asian average for this group is pulled down by countries like North Korea, China, Burma and Vietnam. Overall, over 50 percent of all developing countries appear to have implemented commercial policies that make them, at least equally, or more attractive to foreign investment and commercial activity. Table 10.2. The Relative Position of Kenya's WSJ-Heritage Overall Commercial Environment Index Compared to Averages for Major Developing Country Groups. Developing Kenya's Position in Group Country Group (% of Countries with a Better Kenya's Index or Worse Business Environment Commercial Worse Better Developing Country Group Index Average Median Environment Environment Sub-Saharan Africa 3.26 3.36 3.43 64.4 35.1 East Asia 3.26 3.22 3.05 33.3 66.7 South Asia 3.26 3.44 3.53 80.0 20.0 North Africa & Middle East 3.26 3.12 2.94 35.3 64.7 Central Asia 3.26 3.72 3.70 85.7 14.3 Eastern Europe 3.26 2.92 3.08 26.3 73.7 Latin America 3.26 3.04 3.02 34.6 65.4 ALL ABOVE COUNTRIES 3.26 3.20 3.18 47.6 52.4 Source: Wall Street Journal-Heritage Foundation (2004). closing of the market for cleaning last week. We were lucky to be spared a major outbreak of disease, he said. City workers used more than 42,000 gallons of water in the cleanup operation." 47 In summarizing its findings the ARPED noted corruption was rated as a severe or major obstacle by about three-quarters of the firms surveyed which was much higher than in neighboring countries. Inadequate infrastructure was cited as the second biggest problem as unreliable electricity and telecommunications were major problems for exporters. Crime was also cited as a major problem as almost four percent of sales were spent on the private provision of security. Finally, the survey finds that domestic wage levels, relative to productivity, is not globally competitive. In part, this is due to the fact that the capital stock of Kenya's enterprises is aging with one-third of all equipment being between 11 and 20 years old. 90 A somewhat different, but closely related question is "Can Kenya compete?" Stated differently, if foreign and domestic enterprises based in Kenya attempted to launch a new national or international commercial venture would it likely be a success of failure. According to the World Economic Forum (2004) the latter is the most likely outcome. Kenya's internal environment and policies make the it generally uncompetitive ­ even relative to other African countries (see Box 10.3). The importance of these findings should not be understated. Foreign investment is crucial to any successful diversification effort and Kenya's domestic commercial environment of the amount of FDI that will be forthcoming. However, as Table 10.2 and Box 10.3 shows, there is little or nothing about Kenya's commercial environment or international competitiveness that distinguishes it from the average developing country. Box 10.3 Can Kenya Compete? Implications of the Africa Competitiveness Report. The conceptual approaches employed both by WSJ-Heritage and the World Economic Forum are similar in nature. Both establish objective guidelines for assigning scores to specific policy variables deemed to be important. However, the two have slightly different objectives. WSJ-Heritage attempts to determine how conducive a country's internal environment is to commercial activity. This point also reflects on its ability to attract foreign investment. In contrast, the Forum attempts to determine how well a country is situated to compete internationally with other countries. Both indices conclude that there is little that distinguishes Kenya from most other Sub-Saharan African countries. Kenya's Rank Within 25 African Within 102 Comparator Policy Variable Countries Countries in Total Business cost of crime 25 97 Impact of AIDs on foreign direct investment 22 97 Reliability of police services 22 89 Dispersion of public funds 21 93 Telephone infra-structure quality 19 93 Postal efficiency 19 89 Overall infra-structure quality 18 88 Property rights 13 67 Quality of electrical supply 11 82 Extent of bureaucratic red tape 12 75 Availability of scientists and engineers 5 54 Control of inflation 4 28 OVERALL COMPETITIVE INDEX 15 83 The above statistics compare Kenya's scores for several policy variables influencing competitiveness with those for 25 other African countries and 101 comparator countries. For the African group Kenya ranks 25th in the business cost of crime which indicates it has the worst environment for this variable than any other African country. Other variables with an important negative influence on Kenya's competitiveness are the impact of AIDs on foreign investment, the reliability of police services, telephone infra-structure and postal efficiency. Kenya's overall competitiveness score of 15 indicates it is below average compared to the other African countries Source: African Competitiveness Report (2004). 91 B. Identifying Products for a "Successful" Diversification Key Point A crucial question is what procedures are available for identifying potentially new export ventures for Kenya. In Ghana, the World Bank utilized surveys of individuals and commercial enterprises engaged in various export activities to compile a list of possible new export products. A similar initiative should be considered for Kenya. Since it is generally assumed that a country like Kenya should have a comparative advantage in the production of goods that are intensive in the use of low-skill low-wage labor inputs, available census of manufactures could be used to identify such products. However, the success of any new export venture will require major changes in Kenya's internal commercial environment. Aside from issues relating to the commercial environment, a second key policy question is how can Kenya identify and encourage potentially successful new export ventures. There are several approaches that might provide useful information. The first, which the World Bank recently tested in Ghana, involves surveys of individuals and commercial enterprises engaged in various export activities to identify new products that might be traded successfully. The second involves a "case study" approach. UNCTAD calculates and publishes aggregate indices of export concentration for developed and developing countries, on an annual basis, from the present back to the early 1970s. Where these statistics show major diversifications occurred, data on the underlying composition of exports could identify the specific products that produced these positive results. This information could be helpful for countries, like Kenya, that are attempting to diversify. The reasoning here is that a "ladder" of products exists, ranging from low-skill labor intensive goods to high-technology manufactures, that countries export at different stages of their industrialization. Analyses of past changes in the exports of countries that successfully diversified could help identify the types of products comprising different stages of these ladders. In other words, this information could, on the basis of other countries' experience, help Kenya differentiate between potentially successful, and probably unsuccessful new export ventures. Third, a focused analysis on Africa's own export performance might convey useful information if it identified specific products that had both relatively high growth rates in world trade and stable or rising African import shares. The latter would indicate that Africa was able to retain competitiveness in these high growth products.48 1. The Survey Approach Economic surveys are used for gathering information on likely future developments relating to activities such as manufacturing output, prices, or general economic conditions. For example, the United States survey of consumer confidence is conducted to help determine how 48Several trade related technical assistance support programs maintained by the United Nations Conference on Trade and Development (UNCTAD) might provide useful information for formulation of a diversification strategy. UNCTAD's Trade Point System has extensive global information on firms who are buyers of specific products, which has been useful information for assessing prospects for items that are potential candidates for export diversification. The United Nations International Trade Center (ITC) also has an established program to help developing countries identify and promote new export ventures. 92 consumers foresee changes in the national economy, while the survey of manufacturing purchasing agents is conducted to help predict prices for capital goods and equipment. The underlying theory is that a survey of individuals closely involved in an activity should have specialized knowledge that is useful for anticipating probable changes. As noted earlier, the USITC (2004) recently employed a survey of industry executives to help anticipate the likely effects of the 2005 MFA phase out. A survey of local domestic and foreign enterprises involved in different aspects of Kenya's foreign commerce, including financial, production, and logistical operations, could help identify new potentially successful export ventures. Due to their current involvement in Kenya's trade, those surveyed may posses useful knowledge relating to possible export opportunities and constraints. A World Bank (2001) trail of the survey approach in Ghana produced useful information for formulating an export diversification strategy. Ideally, two stages would be used. The first would attempt to produce a relatively short list of potential export products, while the second stage would focus on identification of any special related production, financial, marketing or distribution problems. United Nations COMTRADE records could be used to generate important supplemental information on market demand and supply conditions for these products. Unlike the situation in Ghana, many required institutional arrangements are already in place in Kenya. Specifically, Africa Regional Program on Enterprise Development (ARPED) has undertaken several surveys relating to the commercial environment in Kenya and other East African countries, and has established potentially useful contacts with many individuals and enterprises. There may be advantages in conducting such a survey in association with the UN International Trade Center or UNIDO since both organizations have already compiled much relevant information on demand, supply, and market access conditions in major international markets. Available evidence indicates that Kenya's export processing zones have had an important influence on generating some new exports, particularly in sectors like garments and pharmaceuticals. A relevant question is what new export products could be competitively produced in Kenya's zone. The International Labor Office (2003) recently compiled a detailed list of products exported from all African and non-African export processing zones (see Box 6.2). The proposed ARPED survey might specifically address the question of which products identified in the ILO report might be manufactured in Kenya's zones if appropriate incentives were provided. A final issue that might be addressed would be an attempt to identify possible products for diversification utilizing the assembly of foreign produced parts and components. As previously indicated, production sharing might allow Kenya to diversify into product lines (like medium tech equipment) that it otherwise would not be able to do so. Aside from the direct survey, selected comparator countries COMTRADE records could provide useful information on specific countries where international production sharing was successfully used to diversify into new export lines.49 49 For example, some governments have actively encouraged the importation and assembly of specific electronic components reasoning this could be an initial first step leading to the development of a full comparative advantage in "high-tech" industries. See World Bank (1994a) for a discussion of China's policies in this respect. The use of financial and other government incentives to attract offshore assembly operations in the Caribbean is well documented by the World Bank (1994b) 93 2. Implications of Low Income Countries' Comparative Advantage Efforts to formulate a diversification strategy for Kenya should empirically attempt to identify potential new export ventures in which the country may have an actual comparative advantage. Economic theory indicates these types of goods should consist primarily of items manufactured using relatively low-skill highly labor intensive products. The World Bank and United States National Bureau of Economic Research previously utilized information compiled in national surveys of manufactures to identify such products. An important finding was that countries like Kenya may have more options for export diversification than is generally recognized, and that the range of potential new export ventures extends well beyond footwear, textiles, and apparel products (see Box 10.4). Analyses of historical changes of some developing countries' export profiles clearly show that specialization in labor intensive exports played a major role in their transformation from raw materials and commodity exporters to manufactured goods. For example, the structure of exports from most of the so called "newly industrialized countries" experienced a remarkable two stage transformation that began in the 1960s and early 1970s. The first stage witnessed a major rapid expansion of goods produced using labor intensive manufacturing processes, that is, items like textiles, clothing, footwear, leather and rubber goods. Most analyses concur that a key factor serving as a catalyst for this expansion was the establishment of favorable "export oriented" commercial policies within these countries themselves. The second stage of this transformation saw these countries becoming increasingly important producers of more sophisticated "high tech" products like consumer electronics, office equipment and machinery. The shift to this stage was facilitated by the experience gained as exporters of labor intensive manufactures, and the participation of these countries in global production sharing operations. The statistics presented in Table 10.3 illustrate the nature of the transformation that occurred in the structure of ten "newly industrialized country" exports during their labor intensive good production and export stage. As an example, over 1965 to 1985 the value of labor exports from Taiwan (China) increased by over $22 billion, the share of these goods in total exports rose by about 40 percentage points, and the share of clothing in all labor intensive exports rose to about 16 percent. A point to note is that the US did not include apparel products in its GSP scheme so the impressive share increases that occurred for countries like Korea, China, and India were in the face of relatively high pre-Uruguay Round tariffs. Far more dramatic changes occurred in the export profiles of countries like Thailand, the Philippines, and Turkey where the share of labor intensive products in total exports rose more than five-fold to approximately 50 percent. At this point, the share of these goods became relatively static, or even declined somewhat, as exports shifted to more lucrative, and profitable, high-tech and semi-capital intensive goods. 94 Box 10.4 Implications of Industry Level Labor Intensity Indices The trade statistics analyzed in this report indicated Kenya has a highly concentrated structure of exports in products that generally have relatively low growth prospects in international trade. The policy prescription called for a diversification into new, more promising, product limes. Specific proposals for product diversification should recognize two points. First, new export ventures should be highly intensive in a factor Kenya has in abundance, that is, low cost relatively unskilled labor. Second, the proposals need not be confined to full production processes for a given product, but might also assembly operations for foreign produced components if these operations are also labor intensive. In order to assist countries like Kenya evaluate prospects for potential new exports, the World Bank initiated a study aimed at identifying goods that were normally produced using labor intensive manufacturing processes. Detailed surveys of industrial production statistics were undertaken and various measures of labor intensity were computed for over 600 individual industry sectors. A key statistic was the value added per employee in the industry to the average value added in all manufacturing sectors. Two specific points should be noted regarding this approach. First, detailed studies by the US National Bureau of Economic Research showed that goods manufactured by relatively labor intensive processes in a given country like the United States were generally manufactured by similar processes in other countries, irrespective of their average income levels. Second, it should be noted that there is an inverse relation between the numeric value of the industry index and the level of labor intensity for the product. That is, the lower the numeric value of the index the higher is the labor intensity of the product. It also follows that products with very high index values (that is, well above 100) are capital intensive in production. A surprising result of the survey was that manufacturing processes in a relatively high number of industries were dependent on highly labor intensive production methods and, therefore, suitable for manufacture in developing countries like Kenya. While textile and clothing products were often identified as highly labor intensive. The following table lists a few of the other products that also utilize very high labor inputs. Example of Industrial Products Normally Using Labor Intensive Manufacturing Processes Industry (Labor Intensity Index) Industry (Labor Intensity Index) Industry (Labor Intensity Index) Poultry and dressing plants (41.7) Wood TV and radio cabinets (50.3) Women's handbags (48.2) Fresh and frozen fish (59.7) Wood kitchen cabinets (67.5) Leather glovers (43.0) Curtains and draperies (42.5) Wood office furniture (72.7) Leather belts (60.5) House furnishings (59.7) Upholstered furniture (57.3) Leather tanning (68.9) Textile bags (50.5) Bookbinding (53.2) Luggage (62.7) Canvas products (57.6) Watches and clocks (66.7) Earthenware food utensils (47.6) Hardwood flooring (45.1) Musical instruments (64.4) Pottery (57.7) Hardwood plywood (57.4) Dolls (66.5) Artificial flowers (59.6) Softwood plywood (55.2) Costume jewelry (64.7) Buttons (57.5) Nailed wood boxes (49.9) Fabricated rubber products (83.3) Woven carpets (62.0) Folding paper boxes (72.8) Rubber & plastic footwear (47.4) Cordage and twine (63.7) Envelopes (81.1) Boot and shoe stock (52.7) Wood household utensils (65.7) Wood containers (45.2) House slippers (46.2) Handbags (67.2) Wood household furniture (50.1) Men's footwear (54.9) Sewing machines (71.1) Brooms and brushes (71.1) Women's footwear (53.0) Residential light fixtures (73.5) Note: The lower the level of the intensity index the higher the labor intensity of the product. An index value of 50.0 indicates the product employed twice the average labor inputs for all sectors of the economy. A final important point was that the Bank's analysis suggested the potential for developing countries in these types of activities was indeed promising. In global trade, many of these products had relatively high growth rates. In addition, while developing countries were increasing their international market shares for many of these products, there was clearly scope for a further major expansion. 95 Table 10.3. The Importance of Labor Intensive Exports as a Catalyst for Growth in Newly Industrialized Developing Countries Share of Labor Share of Textiles and Value of Labor Intensive Intensive Manufactures Clothing in Labor Manufactures Exports ($ in Total Exports (%) Intensive Exports (%) million) Exporting Country 1965 1985 1965 1985 1965 1985 Taiwan, China 45.5 85.6 13.4 15.9 143 22,608 Korea, Republic of 51.1 78.7 19.3 27.5 60 14,599 China 19.2 57.7 25.1 43.9 114 6,007 Singapore 31.5 74.5 2.4 8.8 13 5,056 Brazil 6.2 26.7 1.9 7.1 82 5,020 India 44.3 62.3 10.1 27.3 526 3,369 Malaysia 12.5 47.1 7.4 10.0 81 3,334 Thailand 8.7 49.1 1.3 25.4 27 3,071 Philippines 10.9 54.4 28.5 23.8 88 2,528 Turkey 8.7 51.5 1.5 38.9 35 1,519 Source: Yeats (1988) Does the evidence suggest that Africa is moving along an export oriented growth path similar to that leading to the NICs industrialization and growth.50 Unfortunately, there is little to indicate this is in fact happening. As Table 10.4 shows, labor intensive manufactures account for under 6 percent of Sub-Saharan Africa's exports (exclusive of the Republic of South Africa), which is slightly lower than their corresponding share in the mid-1990s. One four-digit SITC product (jerseys and pullovers) accounts for 22 percent of total exports of these goods, while the six textile and apparel products listed in Table 10.4 account for 80 percent of all labor intensive manufactures exported from the region. Two related points should be noted regarding these observations, · First, surveys of executives of major apparel producing companies indicate their decisions to locate in Africa were heavily influenced by MFA restrictions against low cost countries like China and India. The removal of these restrictions in 2005 is expected to result in a sizeable erosion of Africa's import shares for these goods in major markets. As Table 10.4 indicates, such a development would have a major negative impact on Africa's existing (highly concentrated) export base for labor intensive manufactures. · Second, evidence suggests that differences in the NICs and Africa's ability to capitalize on export opportunities for labor intensive manufactures may largely be attributable to internal 50In the 1970s and 1980s there was considerable enthusiasm in some quarters for the use of the use of so called "natural resource based" industrialization strategies as opposed to the specialization in labor intensive manufactures (UNCTAD 1972, 1975). This approach suggested the further processing of locally produced primary commodities could upgrade the quality of raw material producing developing countries' exports and also provide a stimulus to industrialization. However, a recent analyses of exports consisting of different stages of commodity processing chains shows Africa has made little progress in this direction (Ng and Yeats 2000). Part of the problem is that some commodity processing functions, particularly those for ferrous and nonferrous ores, are very capital intensive and unsuitable for location in developing countries. 96 factors like the quality of governance, the nature of the local commercial environment, and the quality of available infrastructure needed to support export oriented production.51 Table 10.4. The Ten Fastest Growing Labor Intensive Manufactures in All Sub-Saharan African Countries' (Exclusive of Republic of South Africa) Exports. Share in all SSA Labor Intensive Exports (%) Value of Exports ($000)** SITC Description* 1995 2003 1995 2003 Change 845.1 Jerseys and pullovers 9.40 22.01 251,850 589,647 337,797 843.9 Other outer garments 3.08 14.71 82,479 393,895 311,416 846.2 Knit cotton under garments 7.64 16.15 204,615 432,694 228,078 845.9 Other knit clothing 2.58 9.99 69,144 267,447 198,303 842.3 Trousers of textile fabrics 5.37 11.84 143,951 317,150 173,199 634.1 Sawn wood 7.22 10.43 193,300 279,436 86,135 652.2 Woven cotton fabrics 2.33 4.81 62,312 128,887 66,575 611.5 Sheep and lamb leather 2.82 4.26 75,652 114,079 38,427 893.9 Plastic articles 0.69 1.74 18,502 46,495 27,993 634.2 Plywood 1.64 2.64 43,821 70,656 26,936 All Labor Intensive Manufactures 100.0 100.0 2,695,621 3,343,399 647,778 *Labor intensive manufactures are defined at the four-digit SITC level and constitute over 200 individual items classified in SITC 6 and 8. **The Republic of South Africa is excluded from these tabulations since its comparative advantage appears to differ considerably from that of other Sub-Saharan African Countries. Source: Based on partners data from UN COMTRADE Statistics. 3. Fast Growing Exports Some useful information for formulating a diversification strategy for Kenya might come from analyses which attempted to identify "dynamic" or relatively fast growing African exports. The reasoning here is that, if other African countries have been able to successfully export these products, they may also be suitable for Kenya. Fast growing products are of special interest given the relatively low growth prospects for most of Kenya's and other regional countries' traditional exports. For an assessment, the following three-step procedure was employed. First, a list was compiled of all four-digit SITC Rev. 2 products where 2003 global imports from Africa where at least $30 million. This "cut-off" was established to distinguish between situations where regional trade might be the result of special "irregular" circumstances, as opposed to where a permanent trade base was established. In order to provide further supportive evidence on this last point, we 51 In the case of Kenya, Siggel et. al. (2003) note "Several recent studies, including the present one, show that inadequate and poorly maintained infrastructure has emerged as a major impediment to Kenya's industrial growth and other economic activities. A study, which covered more than 200 industrial enterprises in the country's main industrial towns, observes that the poor state of infrastructure continues to be a serious impediment to business activity and manufacturing (ARPED 2003). According to the firms interviewed during the study, the state of the country's infrastructure, especially with regard to electricity, water, freight transport, port-handling facilities, telephones, waste disposal, and security, had deteriorated substantially compared with a few years earlier. 97 next eliminated all products where SSA exports failed to exceed $1 million in each year from 1995 to 2003. This procedure produced a list of almost 200 products for further analysis. Next, 1990-2003 global import growth rates were calculated for these goods from both Africa and all exporting countries. The products were then ranked in descending order of the global trade growth rates. All products with above average global trade growth rates were retained for further analysis. This procedure made it possible to differentiate between four different groups of dynamic products. · Competitive-dynamic products. World trade in these items grew at above average rates while the rate of growth of imports from Africa actually exceeded the global averages. The 48 four-digit products listed in Table 10.5 fall in this category and accounted for African exports of $13 billion in 2003. With the exception of thirteen four-digit SITC products, all are manufactured goods. Passenger motor vehicles are the single largest product on the list, with 2003 African exports (primarily from South Africa) of over $2 billion. Two petroleum derivative products are also among the "fast growth" products with 2003 exports exceeding $1 billion. However, the general impression that emerges from these statistics is that there is a relatively broad range of products for which world trade grew at above average rates, and which African countries were able to maintain or improve their international competitive position. Three other groups of products were identified which warrant further scrutiny. · Dynamic Products. African exports of these goods grew faster than world trade, but slower than global trade in the item. Although African import shares for these goods declined over the decade, their above average growth rates worked against Africa's longer-term marginalization in world trade. Among the products in this group were several textile and apparel products whose trade may have been strongly affected by AGOA. · Static dynamic products. Global trade in these products grew at an above average rate, yet imports from Africa grew less rapidly than world trade. Platinum dominates this group, but other important static-dynamic products include printing paper, miscellaneous base metals, and toys and games. · Declining dynamic products. Global trade in these goods grew at an above average pace, yet the annual rates of growth of imports from Africa were negative. Products classified in the this group include palm oil, vegetable tanning extracts, and safety glass among others. A point of interest concerns the African origins of the competitive-dynamic product exports. Most of these goods originated in the southern cone of Africa. SACU accounted for 71 percent of all exports, followed by Mauritius (16 percent), and Madagascar and Zimbabwe (3 percent each). However, all SSA countries with the exception of Reunion and Rwanda recorded some competitive-dynamic product exports. How Africa was able to establish an export base for such items is a question of importance. 98 Table 10.5 Fast Growing Exports for Which African Countries Increased World Market Shares 2003 Global Imports Annual Import ($ million) 2003 SSA Growth 1990-03 (%) From Market From SITC Export Product (Revision 2) Africa World Share (%) Africa World 072.2 Cocoa powder 122 1,287 9.5 46.7 12.8 034.3 Fish fillets 176 2,184 8.1 44.1 12.1 743.6 Filtering machinery 1,097 14,818 7.4 40.1 8.8 764.3 Radiotelephonic goods 76 78,562 0.1 36.9 20.3 781.0 Passenger motor cars 2,333 379,276 0.6 36.7 6.7 762.1 Radio receivers 38 8,735 0.4 34.3 5.8 341.3 Petroleum gases 1,445 47,920 3.0 31.4 7.8 776.3 Diodes and transistors 33 29,247 0.1 31.1 11.7 713.2 Combustion engines 104 38,788 0.3 29.8 7.7 893.2 Toilet articles 34 1,881 1.8 29.5 7.2 523.3 Salts of metallic acids. 112 2,208 5.1 29.2 5.9 551.4 Air fresheners 55 9,320 0.6 28.7 10.7 511.1 Acyclic hydrocarbons 149 7,402 2.0 28.0 5.0 112.1 Wine of grapes 526 17,690 3.0 26.8 6.0 334.1 Motor spirits 1,807 152,489 1.2 25.7 15.3 122.2 Cigarettes 52 12,227 0.4 24.9 5.1 741.5 Air conditioning units 39 16,503 0.2 24.2 9.4 625.1 Pneumatic tires 100 15,415 0.6 24.0 6.4 553.0 Perfumery and cosmetics 107 30,313 0.4 22.8 10.3 112.4 Spirits and liqueurs 39 13,103 0.3 22.5 4.3 642.8 Articles of paper 44 13,411 0.3 21.0 7.8 277.2 Natural abrasives 44 710 6.2 20.8 -0.5 672.7 Iron or steel coils 643 21,568 3.0 20.6 5.3 821.1 Chairs and other seats 460 33,923 1.4 20.5 9.5 533.1 Other coloring matter 33 6,766 0.5 20.4 5.8 892.8 Printed matter. 38 11,574 0.3 20.3 5.2 282.0 Scrap metal 163 13,971 1.2 19.7 6.3 761.1 Television receivers 35 34,021 0.1 19.7 6.0 625.2 Tires n.e.s. 69 8,797 0.8 19.5 5.8 098.0 Edible preparations 156 21,560 0.7 19.2 8.8 773.1 Insulated wire 119 39,609 0.3 18.9 9.2 512.1 Acyclic alcohols 285 16,436 1.7 18.6 7.7 081.4 Flours of meat or fish 48 2,368 2.0 18.5 2.2 776.4 Electronic microcircuits 37 215,600 0.0 18.4 13.3 246.0 Pulpwood 384 2,657 14.4 18.0 2.2 872.0 Medical instruments 35 34,978 0.1 17.7 10.4 233.1 Synthetic rubber 34 8,434 0.4 17.6 4.5 782.1 Trucks 177 64,341 0.3 17.4 5.9 764.9 Parts of telecommunications 76 77,861 0.1 17.2 9.5 99 Table 10.5 Continued 2003 Global Imports Annual Import ($ million) 2003 SSA Growth 1990-03 (%) From Market From SITC Export Product (Revision 2) Africa World Share (%) Africa World 598.9 Chemical products n.e.s. 145 45,784 0.3 16.6 7.5 843.9 Other outer garments 471 34,949 1.3 16.3 7.4 778.8 Other electrical machinery 79 54,801 0.1 16.2 8.8 541.9 Pharmaceutical goods 34 8,063 0.4 15.8 9.6 743.9 Parts of centrifuges 71 11,501 0.6 15.8 7.8 714.9 Parts of engines & motors 55 29,199 0.2 15.8 6.7 784.9 Other motor car parts 497 170,260 0.3 15.7 6.9 778.3 Electrical motor equipment 43 17,919 0.2 15.6 7.2 845.9 Other knitted outer garments 319 23,639 1.4 15.4 8.1 ALL ABOVE PRODUCTS 13,039 1,904,069 0.7 24.0 8.5 0 to 9 ALL GOODS 109,183 7,189,836 1.5 4.7 6.1 Note: The fastest African growth products are drawn from those having exports above $30 million in 2003 and average annual growth rates of at least 15 percent during 1990-2003 (excluding armored vehicles (SITC 951.0) Source: Computations based on world imports from UN COMTRADE Statistics. 4. The Economic Implications This section's intention was not to identify specific products that might be included in a diversification strategy for Kenya, but to examine several procedures that might be useful for this effort. One important point appears evident. That is, the range of potential new exports may be considerably broader than is generally recognized (see Box 10.4 and Table 10.3). A key question, however, is whether Kenya can make the required improvements in its commercial environment needed for such new export ventures to be successful. Foreign investment will have to play a key role in the expansion and diversification of Kenya's exports.52 However, until the Kenyan authorities adopt bold direct measures to combat corruption, serious crime problems, widespread bribery, social issues relating to major health care problems and training of the local labor force, and an inferior and deteriorating infrastructure, it must be admitted that there are few incentives to invest in Kenya. This point is of fundamental importance in the design of any trade expansion and diversification strategy. 52See, for example, Box 7.2 for an analysis of the key role that Danish foreign investment played in the development of Kenya's cut flower industry. 100 XI. PRIORITY POLICY ISSUES Key Point Two major problems must be addressed before any rationale work can begin on the formulation of a trade related industrialization strategy. The first relates to the quality of Kenya's national trade statistics which are totally inadequate for empirical analysis of trade performance or policy formulation. The international donor community can help Kenya upgrade the quality and utility of its trade statistics. The second priority problem relates to the nature of the commercial environment in Kenya. This problem will have to be addressed directly by the local authorities themselves and not by outsiders. While the underlying issues are clearly complex, this report identified two high priority initiatives that must be directly addressed. The first involves a pressing need for improving the quality of Kenya's official trade statistics, and a need to significantly improve the general commercial environment. A. Trade Data Issues This report concluded Kenya's official export statistics are seriously flawed and had little practical utility for policy formulation or analyses of export performance. Kenya has unilaterally departed from United Nations guidelines and fails to include shipments from its export processing zones in its UN COMTRADE records.53 However, the statistical problems appear to go considerably beyond this omission as IMF data indicate almost $1 billion in exports have not been recorded in Kenya's 2002 trade statistics, and about $1.5 billion has been "lost" from the combined 2002-2003 data. This report has been unable to resolve a question that has major policy implications. That is, are the discrepancies in Kenya's trade statistics largely due to errors made in the compilation and tabulation of the data, or do they reflect the effects of incentives to intentionally falsify customs invoices. From a policy perspective, this report strongly recommends that a concerted effort be quickly made to improve the quality and reliability of Kenya's trade statistics. This will likely require a thorough review of customs procedures to determine why such large scale underreporting of Kenya's exports has occurred. Such a review should also investigate whether the adoption of modernized electronic procedures could significantly improve customs operations and the compilation of trade statistics. Specifically, over 80 developing and transition countries now use UNCTAD's ASYCUDA computerized customs clearance software and Kenya should consider adopting such a system. Second an immediate effort is needed to improve the reporting of exports from the export processing zones, and to incorporate this information in Kenya's UN COMTRADE statistics. Very "rough" numbers tabulated by the EPZ authorities have little or no analytical value since they are not compiled in terms of any accepted international trade classification system like the SITC or the Harmonized System. 53There is relatively strong indication that the Kenyan authorities have just lost track of what is going into, or coming out of the export processing zones. For example, in a recent ILO study Kenya reported that the manufacture of apparel and medications, and the processing of tea, were major activities in the zone (see Box 6.2. However, data provided by the EPZ authorities contain no evidence of tea processing. Important inconsistencies also exist between statistics on exports provided by the authorities and the reported imports of Kenya's trading partners. 101 Third, there is an obvious need for a regional trade data reconciliation study. Export statistics reported by (say) Kenya often differ markedly from the corresponding import statistics of other East African countries. As a result very little can be said with any degree of certainty about the level, composition, and trends in intra-regional trade. B. The Implications of Kenya's Commercial Environment Ample evidence exists that Kenya's trade and growth problems are tied directly to problems in the internal commercial environment. Recent statistics published by Transparency International, for example, ranks Kenya among the World's most corrupt countries. The 2004 Africa Competitiveness Report indicates Kenya is less competitive than average for a group of 28 other Sub-Saharan African countries. When comparisons are extended to non-African countries Kenya's competitiveness places it in the lowest 17th percentile. Surveys conducted Africa Regional Program on Enterprise Development indicate Kenya places well below other African comparator countries when it comes to the cost of crime and the quality of infrastructure. Finally, Wall Street Journal-Heritage Foundation analyses report that the commercial environment in Kenya is substantially less favorable than in other developing countries that are competitors for foreign investment. The implications of this point should be clearly noted. The international donor community can assist Kenya with many other development problems. However, problems associated with crime, bribery and corruption, and various other negative features of the commercial environment will have to be directly addressed by the Kenyan's themselves. World Bank President James Wolfensohn stressed this point in a recent (August 2003) address in Nairobi; "For nearly two decades after its independence in 1964, Kenya was headed to be an African success story--GDP growth rates averaged 6.5 percent a year, the economy was buoyant, investor confidence was high, the international community was generous in its support, and agriculture and trade thrived. Roots of decline could be traced to the late seventies attributable to public mismanagement and deep-rooted, "institutionalized" corruption. Indeed Kenya's experience serves to confirm the assertion that corruption is not just a political issue, but an economic one. As global poverty rates declined by the 1990s, Kenya's increased from 48 percent at the start of the 1990s to 56 percent by the end of 2002. By 2002, Kenya's economy was growing by only 1.1 percent. This decline has already turned up in the country's human development indicators: Primary education enrolment rates fell to 82% by 1995 after peaking to 91% in 1989. In an address to an Anti-Corruption Conference in Nairobi, Wolfensohn stressed the role of Kenyans in bringing about the much-needed reform. He noted "It is the Kenyans themselves-- government, private sector, and citizens alike--who now want the Kenya that was, the Kenya that stood for integrity, the leadership, the Kenyan values which were values of hard work and were values of education and commitment to equity." Economic recovery in Kenya going forward will be driven in large part by the government's ability to win back the confidence of donor and investment communities. Investment is not possible without good governance and an environment free of corruption. Good governance is not just going to engage investors, it will also have a positive impact on trade. 102 REFERENCES Africa Region Enterprise Development Initiative (1996). Report on Round 3 of the Kenya Survey, (Nairobi and Gotenborg: Department of Economics University of Nairobi and Department of Economics University of Gotenborg). Africa Region Enterprise Development Initiative (2003). 2003 Kenya Investment Climate Assessment, (Washington: World Bank). Ajayi, S. and M. Khan (2000). External Debt and Capital Flight in Sub-Saharan Africa, (Washington: International Monetary Fund). Amjadi, Azita, Ulrich Reinke and Alexander Yeats (1996). "Did External Barriers Cause the Marginalization of Sub-Saharan Africa in World Trade," World Bank Discussion Paper Number. 348, (Washington: World Bank). Balassa, Bela (1965). "The Structure of Protection in Industrial Countries and its Effects on the Exports of Processed Goods from Developing Countries," in UNCTAD, The Kennedy Round Estimated Effects on Tariff Barriers, (TD/6/Rev. 1) (New York: United Nations). Brownbridge and Harrigan (1996). "Positive Terms of Trade Shocks and Structural Adjustment in Sub-Saharan Africa," Development Policy Review, vol. 14. Collier, Paul (1995). "The Marginalization of Africa," International Labour Review, 134 (4-5): 541-57. Dollar, David (1992). "Outward Oriented Developing Economies Really Tend to Grow More Rapidly," Economic Development and Cultural Change, (vol. 4, no. 3). Easterly, William and R. Levine (1994). Africa's Growth Tragedy, mimeo, (Washington: World Bank). Edwards, Sebastian (1993). "Openness, Trade Liberalization and Growth in Developing Countries," Journal of Economic Literature, (vol. 31, no. 3). Feenstra, Robert (1998). Trade and Disintegration of Production in the Global Economy, Journal of Economic Perspectives, (Fall). GATT (1966). International Trade, 1965, (Geneva: GATT). Grubel and Lloyd (1975). Intra-Industry Trade: The Theory and Measurement of International Trade in Differentiated Products, (New York: John Wiley). Helleiner, G.K. and Douglas Welwood (1978). Raw Material Processing in Developing Countries and Reductions in the Canadian Tariff, (Ottawa: Economic Council of Canada, April) Hoekman , Bernard and S. Djanken (1996). "Intra-Industry Trade, Foreign Direct Investment, and the Re-Orientation of Eastern European Exports," unpublished manuscript, (Washington: World Bank). ILO (2003). ILO Database on Export Processing Zones, (Geneva: ILO, Sector Activities Department). International Monetary Fund (2000). External Debt and Capital Flight in Sub-Saharan Africa, edited by S. Ajaya and M. Khan, (Washington: International Monetary Fund. 103 Keesing, D. (1967). "Outward Looking Policies and Economic Development," Economic Journal, June. Kravis, Irving (1970). "Trade as a Handmaiden of Growth: Similarities Between the Nineteenth and Twentieth Centuries," The Economic Journal, December. MacBean, A. I. (1966). Export Instability and Economic Development, (Cambridge: Harvard University Press). Massel, B. F. (1970). "Export Instability and Economic Structure," American Economic Review, 60 (September 1970). Michaely, M. (1962). Concentration in International Trade, (Amsterdam: North Holland). Ng, Francis and Alexander Yeats (1997). "Open Economies Work Better! Did Africa's Protectionist Policies Cause its Marginalization in World Trade?," World Development, July. Ng, Francis and Alexander Yeats (1999), Good Governance and Trade Policy: Are They the Keys to Africa's Global Integration and Growth?, World Bank Policy Research Working Paper Number 2038, (Washington: The World Bank). Ng and Yeats (2002). "On the Recent Trade Performance of Sub-Saharan African Countries: Cause for Hope or More of the Same," World Bank Africa Region Working Paper No. 7, (Washington: World Bank). Ng, Francis and Alexander Yeats (2003). "Major Trade Trends in East Asia: What are Their Implications for Regional Cooperation and Growth," Policy Research Working Paper No. 3084, (Washington: World Bank). OECD (2003). Export Diversification in Low Income Countries: An International Challenge After Doha, (OECD Development Center Technical Paper No. 209, June) Riedel, J. (1984)."Trade as an Engine of Growth Revisited," Economic Journal, (March). Siggel, Eckhard et. al. (2003). "Policy Reforms, Competitiveness and Prospects of Kenya's Manufacturing Industries, 1984-1997, and Comparisons with Uganda, in Mwangi Kimenyi, John Mbaku and Ngure Mwaniki (editors), Restarting and Sustaining Economic Growth and Development in Africa, (United Kingdom: Ashgate Publishers). Transparency International (2004). The Global Corruption Report 2004, (Berlin: Transparency International, 2004). UNCTAD (1972). Commodity Problems and Policies, (Geneva: United Nations). UNCTAD (1975). Measures to Expand Processing of Primary Commodities in Developing Countries, (Geneva: December). UNCTAD (1976). Economic Cooperation Among Developing Countries, (Nairobi: United Nations, March). UN WIDER (2000). Is Africa a Net Creditor? New Estimates of Capital Flight From Severely Indebted Sub-Saharan African Countries, by J. Boyce and L. Ndikumana, (Helsinki: UN WIDER) US International Trade Commission (1994). Production Sharing: Use of U.S. Components and Materials in Foreign Assembly Operations, 1991-1994, (Washington: USITC, May). 104 US International Trade Commission (2003). Industry and Trade Summary: Cut Flowers, (Washington: USITC Publication 3580) US International Trade Commission Investigation No. 332-448 (2004). Textiles and Apparel: Assessment of the Competitiveness of Certain Foreign Suppliers to the U.S. Market under section 332(g) of the Tariff Act of 1930 (19 U.S.C. 1332(g), (Washington, USITC). World Bank Economic Review (vol. 4, no 2). "On the Accuracy of Economic Observations: Do Sub-Saharan African Trade Statistics Mean Anything," World Bank Economic Review, May. World Bank (1994a). China: Foreign Trade Reform, (Washington: World Bank). World Bank (1994b). Coping with Changes in the External Environment, (Washington: World Bank Latin America and Caribbean Division, May). World Bank (1996). World Development Report, 1996, (Washington: World Bank). World Bank (1997). World Development Report, 1997, (Washington: World Bank). World Bank (2000). Global Commodity Markets: A Comprehensive Review and Price Forecasts, (Washington: World Bank). World Bank (2001). Ghana: International Competitiveness ­ Opportunities and Challenges Facing Nontraditional Products, (World Bank: Macroeconomics 4 Africa Region, January 31). World Bank (2003). Kenya: A Policy Agenda to Restore Growth, Report No. 25840-KE, (Washington, World Bank Africa Region, August 18). World Economic Forum (2004). The African Competitiveness Report 2004, (Geneva: World Economic Forum). Yang, Yongzheng and Sanjev Gupta (2005). Regional Trade Arrangements in Africa: Past Efforts and the Way Forward, IMF Working Paper WP/05/36, (Washington: IMF, February). Yeats, Alexander (1981). Trade and Development Policies, Leading Issues for the 1980s, (London: Macmillan Press). Yeats, Alexander (1987). "The Escalation of Trade Barriers," in J. Michael Finger and Andrzej Olechowski eds., The Uruguay Round: A Handbook on the Multilateral Trade Negotiations, (Washington: World Bank). Yeats (1990). "Do African Countries Pay More For Imports?, World Bank Economic Review, (May). Yeats, Alexander (1991). Do Natural Resource Based Industrialization Strategies Convey Important Price Benefits for Commodity Exporting Developing Countries? World Bank Policy Research Paper Number 580, (Washington: World Bank). Yeats, Alexander (1996). Did External Barriers Cause the Marginalization of Sub-Saharan Africa in World Trade?, World Bank Discussion Paper Number 348, (Washington: The World Bank). Yeats, Alexander (1998a). Just How Big is Global Production Sharing? World Bank Policy Research Working Paper Number 1871, (Washington: The World Bank). Yeats, Alexander (1998b). What Can be Expected from African Regional Trade Arrangements? World Bank Policy Research Working Paper Number 2004, (Washington: World Bank). 105 Africa Region Working Paper Series Series # Title Date Author ARWPS 1 Progress in Public Expenditure Management in January 1999 C. Kostopoulos Africa: Evidence from World Bank Surveys ARWPS 2 Toward Inclusive and Sustainable Development in March 1999 Markus Kostner the Democratic Republic of the Congo ARWPS 3 Business Taxation in a Low-Revenue Economy: A June 1999 Ritva Reinikka Study on Uganda in Comparison with Neighboring Duanjie Chen Countries ARWPS 4 Pensions and Social Security in Sub-Saharan October 1999 Luca Barbone Africa: Issues and Options Luis-A. Sanchez B. ARWPS 5 Forest Taxes, Government Revenues and the January 2000 Luca Barbone Sustainable Exploitation of Tropical Forests Juan Zalduendo ARWPS 6 The Cost of Doing Business: Firms' Experience June 2000 Jacob Svensson with Corruption in Uganda ARWPS 7 On the Recent Trade Performance of Sub-Saharan August 2000 Francis Ng and African Countries: Cause for Hope or More of the Alexander J. Yeats Same ARWPS 8 Foreign Direct Investment in Africa: Old Tales November 2000 Miria Pigato and New Evidence ARWPS 9 The Macro Implications of HIV/AIDS in South November 2000 Channing Arndt Africa: A Preliminary Assessment Jeffrey D. Lewis ARWPS 10 Revisiting Growth and Convergence: Is Africa December 2000 C. G. Tsangarides Catching Up? ARWPS 11 Spending on Safety Nets for the Poor: How Much, January 2001 William J. Smith for How Many? The Case of Malawi ARWPS 12 Tourism in Africa February 2001 Iain T. Christie D. E. Crompton ARWPS 13 Conflict Diamonds February 2001 Louis Goreux ARWPS 14 Reform and Opportunity: The Changing Role and March 2001 Jeffrey D. Lewis Patterns of Trade in South Africa and SADC 106 Africa Region Working Paper Series Series # Title Date Author ARWPS 15 The Foreign Direct Investment Environment in March 2001 Miria Pigato Africa ARWPS 16 Choice of Exchange Rate Regimes for Developing April 2001 Fahrettin Yagci Countries ARWPS 18 Rural Infrastructure in Africa: Policy Directions June 2001 Robert Fishbein ARWPS 19 Changes in Poverty in Madagascar: 1993-1999 July 2001 S. Paternostro J. Razafindravonona David Stifel ARWPS 20 Information and Communication Technology, August 2001 Miria Pigato Poverty, and Development in sub-Saharan Africa and South Asia ARWPS 21 Handling Hierarchy in Decentralized Settings: September 2001 Navin Girishankar A. Governance Underpinnings of School Alemayehu Performance in Tikur Inchini, West Shewa Zone, Yusuf Ahmad Oromia Region ARWPS 22 Child Malnutrition in Ethiopia: Can Maternal October 2001 Luc Christiaensen Knowledge Augment The Role of Income? Harold Alderman ARWPS 23 Child Soldiers: Preventing, Demobilizing and November 2001 Beth Verhey Reintegrating ARWPS 24 The Budget and Medium-Term Expenditure December 2001 David L. Bevan Framework in Uganda ARWPS 25 Design and Implementation of Financial January 2002 Guenter Heidenhof H. Management Systems: An African Perspective Grandvoinnet Daryoush Kianpour B. Rezaian ARWPS 26 What Can Africa Expect From Its Traditional February 2002 Francis Ng Exports? Alexander Yeats ARWPS 27 Free Trade Agreements and the SADC Economies February 2002 Jeffrey D. Lewis Sherman Robinson Karen Thierfelder ARWPS 28 Medium Term Expenditure Frameworks: From February 2002 P. Le Houerou Robert Concept to Practice. Preliminary Lessons from Taliercio 107 Africa Region Working Paper Series Series # Title Date Author Africa ARWPS 29 The Changing Distribution of Public Education February 2002 Samer Al-Samarrai Expenditure in Malawi Hassan Zaman ARWPS 30 Post-Conflict Recovery in Africa: An Agenda for April 2002 Serge Michailof the Africa Region Markus Kostner Xavier Devictor ARWPS 31 Efficiency of Public Expenditure Distribution and May 2002 Xiao Ye Beyond: A report on Ghana's 2000 Public S. Canagaraja Expenditure Tracking Survey in the Sectors of Primary Health and Education ARWPS 33 Addressing Gender Issues in Demobilization and August 2002 N. de Watteville Reintegration Programs ARWPS 34 Putting Welfare on the Map in Madagascar August 2002 Johan A. Mistiaen Berk Soler T. Razafimanantena J. Razafindravonona ARWPS 35 A Review of the Rural Firewood Market Strategy August 2002 Gerald Foley in West Africa Paul Kerkhof Djibrilla Madougou ARWPS 36 Patterns of Governance in Africa September 2002 Brian D. Levy ARWPS 37 Obstacles and Opportunities for Senegal's September 2002 Stephen Golub International Competitiveness: Case Studies of the Ahmadou Aly Mbaye Peanut Oil, Fishing and Textile Industries ARWPS 38 A Macroeconomic Framework for Poverty October 2002 S. Devarajan Reduction Strategy Papers : With an Application Delfin S. Go to Zambia ARWPS 39 The Impact of Cash Budgets on Poverty Reduction November 2002 Hinh T. Dinh in Zambia: A Case Study of the Conflict between Abebe Adugna Well Intentioned Macroeconomic Policy and Bernard Myers Service Delivery to the Poor ARWPS 40 Decentralization in Africa: A Stocktaking Survey November 2002 Stephen N. Ndegwa ARWPS 41 An Industry Level Analysis of Manufacturing December 2002 Professor A. Mbaye Productivity in Senegal ARWPS 42 Tanzania's Cotton Sector: Constraints and December 2002 John Baffes 108 Africa Region Working Paper Series Series # Title Date Author Challenges in a Global Environment ARWPS 43 Analyzing Financial and Private Sector Linkages January 2003 Abayomi Alawode in Africa ARWPS 44 Modernizing Africa's Agro-Food System: February 2003 Steven Jaffee Analytical Framework and Implications for Ron Kopicki Operations Patrick Labaste Iain Christie ARWPS 45 Public Expenditure Performance in Rwanda March 2003 Hippolyte Fofack C. Obidegwu Robert Ngong ARWPS 46 Senegal Tourism Sector Study March 2003 Elizabeth Crompton Iain T. Christie ARWPS 47 Reforming the Cotton Sector in SSA March 2003 Louis Goreux John Macrae ARWPS 48 HIV/AIDS, Human Capital, and Economic April 2003 Channing Arndt Growth Prospects for Mozambique ARWPS 49 Rural and Micro Finance Regulation in Ghana: June 2003 William F. Steel Implications for Development and Performance of David O. Andah the Industry ARWPS 50 Microfinance Regulation in Benin: Implications of June 2003 K. Ouattara the PARMEC LAW for Development and Performance of the Industry ARWPS 51 Microfinance Regulation in Tanzania: June 2003 Bikki Randhawa Implications for Development and Performance of Joselito Gallardo the Industry ARWPS 52 Regional Integration in Central Africa: Key Issues June 2003 Ali Zafar Keiko Kubota ARWPS 53 Evaluating Banking Supervision in Africa June 2003 Abayomi Alawode ARWPS 54 Microfinance Institutions' Response in Conflict June 2003 Marilyn S. Manalo Environments: Eritrea- Savings and Micro Credit Program; West Bank and Gaza ­ Palestine for Credit and Development; Haiti ­ Micro Credit National, S.A. 109 Africa Region Working Paper Series Series # Title Date Author AWPS 55 Malawi's Tobacco Sector: Standing on One Strong June 2003 Steven Jaffee leg is Better than on None AWPS 56 Tanzania's Coffee Sector: Constraints and June 2003 John Baffes Challenges in a Global Environment AWPS 57 The New Southern AfricanCustoms Union June 2003 Robert Kirk Agreement Matthew Stern AWPS 58a How Far Did Africa's First Generation Trade June 2003 Lawrence Hinkle Reforms Go? An Intermediate Methodology for A. Herrou-Aragon Comparative Analysis of Trade Policies Keiko Kubota AWPS 58b How Far Did Africa's First Generation Trade June 2003 Lawrence Hinkle Reforms Go? An Intermediate Methodology for A. Herrou-Aragon Comparative Analysis of Trade Policies Keiko Kubota AWPS 59 Rwanda: The Search for Post-Conflict Socio- October 2003 C. Obidegwu Economic Change, 1995-2001 AWPS 60 Linking Farmers to Markets: Exporting Malian October 2003 Morgane Danielou Mangoes to Europe Patrick Labaste J-M. Voisard AWPS 61 Evolution of Poverty and Welfare in Ghana in the October 2003 S. Canagarajah 1990s: Achievements and Challenges Claus C. Pörtner AWPS 62 Reforming The Cotton Sector in Sub-Saharan November 2003 Louis Goreux Africa: SECOND EDITION AWPS 63 Republic of Madagascar: Tourism Sector Study November 2003 Iain T. Christie (E) D. E. Crompton AWPS 63 République de Madagascar: Etude du Secteur November 2003 Iain T. Christie (F) Tourisme D. E. Crompton AWPS 64 Migrant Labor Remittances in Africa: Reducing Novembre 2003 Cerstin Sander Obstacles to Development Contributions Samuel M. Maimbo AWPS 65 Government Revenues and Expenditures in January 2004 Francisco G. Carneiro Guinea-Bissau: Casualty and Cointegration Joao R. Faria Boubacar S. Barry 110 Africa Region Working Paper Series Series # Title Date Author AWPS 66 How will we know Development Results when we June 2004 Jody Zall Kusek see them? Building a Results-Based Monitoring Ray C. Rist and Evaluation System to Give us the Answer Elizabeth M. White AWPS 67 An Analysis of the Trade Regime in Senegal June 2004 Alberto Herrou-Arago (2001) and UEMOA's Common External Trade Keiko Kubota Policies AWPS 68 Bottom-Up Administrative Reform: Designing June 2004 Talib Esmail Indicators for a Local Governance Scorecard in Nick Manning Nigeria Jana Orac Galia Schechter AWPS 69 Tanzania's Tea Sector: Constraints and June 2004 John Baffes Challenges AWPS 70 Tanzania's Cashew Sector: Constraints and June 2004 Donald Mitchell Challenges in a Global Environment AWPS 71 An Analysis of Chile's Trade Regime in 1998 and July 2004 Francesca Castellani 2001: A Good Practice Trade Policy Benchmark A. Herrou-Arago Lawrence E. Hinkle AWPS 72 Regional Trade Integration inEast Africa: Trade August 2004 Lucio Castro and Revenue Impacts of the Planned East African Christiane Kraus Community Customs Union Manuel de la Rocha AWPS 73 Post-Conflict Peace Building in Africa: The August 2004 Chukwuma Obidegwu Challenges of Socio-Economic Recovery and Development AWPS 74 An Analysis of the Trade Regime in Bolivia August 2004 Francesca Castellani in2001: A Trade Policy Benchmark for low Alberto Herrou-Aragon Income Countries Lawrence E. Hinkle AWPS 75 Remittances to Comoros- Volumes, Trends, October 2004 Vincent da Cruz Impact and Implications Wolfgang Fendler Adam Schwartzman AWPS 76 Salient Features of Trade Performance in Eastern October 2004 Fahrettin Yagci and Southern Africa Enrique Aldaz-Carroll AWPS 77 November 2004 Alan Gelb Implementing Performance-Based Aid in Africa Brian Ngo Xiao Ye AWPS 78 Poverty Reduction Strategy Papers: Do they December 2004 Rene Bonnel matter for children and Young people made Miriam Temin vulnerable by HIV/AIDS? Faith Tempest AWPS 79 Experience in Scaling up Support to Local December 2004 Jean Delion Response in Multi-Country Aids Programs (map) Pia Peeters in Africa Ann Klofkorn Bloome 111 Africa Region Working Paper Series Series # Title Date Author AWPS 80 What makes FDI work? A Panel Analysis of the February 2005 Kevin N. Lumbila Growth Effect of FDI in Africa AWPS 81 Earnings Differences between Men and Women in February 2005 Kene Ezemenari Rwanda Rui Wu AWPS 82 The Medium-Term Expenditure Framework The Challenge of Budget Integration in SSA April 2005 Chukwuma Obidegwu countries AWPS 83 Rules of Origin and SADC: The Case for change Paul Brenton in the Mid Term Review of the Trade Protocol June 2005 Frank Flatters Paul Kalenga AWPS 84 Chukwuemeka Sexual Minorities, July 2005 Anyamele Violence and AIDS in Africa Ronald Lwabaayi Tuu-Van Nguyen, and Hans Binswanger AWPS 85 Poverty Reducing Potential of Smallholder Agriculture in Zambia: July 2005 Paul B. Siegel Opportunities and Constraints Jeffrey Alwang AWPS 86 Infrastructure, Productivity and Urban Dynamics in Côte d'Ivoire July 2005 Zeljko Bogetic An empirical analysis and policy implications Issa Sanogo AWPS 87 Poverty in Mozambique: Louise Fox Unraveling Changes and Determinants August 2005 Elena Bardasi, Katleen Van den Broeck AWPS 88 Operational Challenges: Community Home Based Care (CHBC) August 2005 Nadeem Mohammad forPLWHA in Multi-Country HIV/AIDS Programs Juliet Gikonyo (MAP) forSub-Saharan Africa AWPS 89 Framework for Forest Resource Management in August 2005 Giuseppe Topa Sub-Saharan Africa AWPS 90 Kenya Exports Prospects and Problems October 2005 Francis Ng Alexander Yeats