Rwanda - Disequilibrium, reform, and the manufacturing sector

Country 56865 Background Papers .... .. ., o· Rwanda: Disequilibrium, Refonn and the Manufacturing Sector Catholic University of Leuven Leuven, Belgium Draft  REGIONAL PROGRAM ON ENTERPRISE DEVELOPMENT - AFRICA THE WORLD BANK Country Background Paper R WAN D A : DISEQUILIBRIUM, REFORM AND THE MANUFACfURING SECTOR Prepared by : Lodewijk Berlage, Hans Eyssen, Micheline Goedhuys, Leo Sleuwaegen, Daniel Van den Bulcke. April 1993 Tables of contents 1. Introduction 1.1 Historical background 1.2 Economic evolution and challenges 2. Main economic policies 3. Economic performance 1970-91 3.1 The evolution of the economy during the seventies 3.2 Growing economic problems, 1981-86 3.3 The economic crisis of the late eighties 4. Sectoral Composition of GDP 4.1 General overview 4.2 Manufacturing and construction 5. Public Finance 5.1 General structure of public finance 5.2 Tax receipts 5.3 Public expenditures 5.4 Public debt 6. Balance of payments 6.1 General characteristics 6.2 Composition of exports and imports of goods 7. Banking and Finance 7.1 Overview of the structure and performance of the financial sector 7.2 Structure of the credit market 7.3 Interest rates 8. Labor force, employment, wages 9. Prices and inflation 9.1 Trends in price indices 9.2 Price controls and deregulation 10. Conclusion Bibliography Annex: persons contacted List of tables Table 1 Basic data Table 2 Credit (bIn FRw, end of year) Table 3.1 Macroeconomic Statistics period Averages 1970-1990 Table 3.2 Annual rates for period 1987-90 Table 4.1 Composition of GDP at current prices (%) Table 4.2 Composition of value added in manufacturing at current prices (%) Table 5.1 Government Revenue and Expenditure (bin FRw) Table 5.2 Composition of tax revenue (% of total) Table 5.3 Composition of expenditures (% of total) Table 5.4 Sectoral composition of Public Investment Program (bin FRw and %) Table 5.5 Internal and external public debt (bill. FRw and % of GDP) Table 6.1 Balance of payments (bIn FRw) Table 6.2 Composition of exports (million FRw) Table 6.3 Composition of imports (bIn. FRw) Table 7.1 Rwanda, Interest Rates for Loans Table 7.2 Sectoral Distribution of Loans (bIn FRw and %) 1990 Table 7.3 Sectoral Distribution of Loans (bIn FRw and %) 1991 Table 8.1 Employment structure, 1989 Table 8.2 Employment in the manufacturing sector, 1985 Table 8.3 Grid of minimal wages and salaries (1992) Table 9.1 Price index of consumption goods, Kigali Table 9.2 Producer prices  1 1. INTRODUCTION. Rwanda, a small landlocked country in Central Africa with a per capita income of 320 US $ and a population of approximately 7.5 million people1, is confronted with major political and economic problems. The latent ethnic rivalry flared up after the "Front Patriotique Rwandais" in October 1990 invaded the north of the country from Uganda. The Rwandese government and the Front have been unable to work out a durable peace agreement. Renewed fighting broke out at the beginning of 1993. This resulted in a large stream of refugees, which moved southwards towards the capital, Kigali.The search for a peace agreement was complicated by the political problems arising from the democratisation process. At the economic level Rwanda is confronted with an annual growth rate of population of 3.3% resulting in a scarcity of land and the prospect of potential food shortages. Moreover during the last few years the country was hard hit by the fall of the world price of coffee, its main export crop. This paragraph gives an outline of Rwanda and of its main economic problems. 1.1 Historical background. The territory of what are now Rwanda and Burundi was originally inhabited by the Batwa, pygmoid people which now represent only about 1% of the population. The Hutus are now the most important ethnic group, with approximately 85% of the population. They settled in the area more than two thousand years ago. In the course of time they created' a number of small states. The second largest ethnic group, the Tutsis, arrived in the course of the 15th and 16th centuries. After their arrival feudal forms of social organisation developed, which were based on the ownership of cattle and in which the dominant role was played by the Tutsis. The feudal system formed the basis of the kingdoms of Rwanda and Burundi. The 1 According to the Census of August 1991 the total popUlation was 7.165 million, implying an intercensus annual growth rate of 3.3%. Before the census it had been thought that both population and the growth rate were higher (7.7 million people and an annual growth rate of 3.7%).  2 kingdom of Rwanda was insignificant until the seventeenth century. It achieved its largest gains in the nineteenth century. By then it had established "the most stratified political system to be found anywhere in Africa"2. At the end of the nineteenth century Rwanda and Burundi became German colonies. After the first world war they were given the status of mandated territories by the League of Nations and assigned to Belgium. In reality they were run from Bujumbura (Burundi) as a province of the Belgian Congo. Originally Belgium based its rule on the existing political organisation, i.e. the king (mwami) and the local authorities. But in the course of time a number of changes were introduced, e.g. a new territorial division, a new tax system and the beginning of labor legislation. The feudal relations were also maintained, though with some restrictions. Thus the Hutu farmers now had to deal with two masters, the Tutsi overlords and the Belgian colonial authorities. During the second half of the fifties a number of political parties were created, mainly on an ethnic basis. At the end of 1959 there was an insurrection of Hutus against the feudal masters, which resulted in hundreds of deaths and a first flow of refugees to neighbouring countries. The king was removed and a republic was proclaimed with the silent assent of the still ruling Belgian authorities. The country became independent on July 1, 1962. The first president was Gregoire Kayibanda, head of the main political party, MDR (Mouvement Democratique Republicain), formerly called Parmehutu. The flrst years of independence were characterised by ethnic strife, an outflow of and armed incursions by refugees and an evolution towards one party rule. On July 5, 1973 Kayibanda was removed by a military coup and replaced by Juvenal Habyarimana. The new president succeeded in alleviating the ethnic tensions. A new unique party was set up called the MRND (Mouvement Revolutionnaire National pour Ie Developpement ). Towards the end of the eighties the MRND rule and the president himself were increasingly contested. This resulted in the creation of new parties. In October 1990 an invasion from Uganda was started by the "Front Patriotique Rwandais", which was mainly made up of Tutsi refugees. This has complicated the democratisation process. The country is still in search of a new political equilibrium. In 1991 a new constitution was adopted, 2 R. Oliver (1977), The East African Interior, in R. Oliver and J.D. Fage (eds), Cambridge History of Africa, vol. 3, pp. 642-645.  3 which made made political pluralism possible. In April 1992 the so-called opposition parties formed a government with the MRND but this government is internally divided. Although there have been a number of cease fires between the national army and the Patriotic Front, the two parties have not been able to agree on a durable settlement. Renewed fighting broke out in February 1993. 1.2 Economic evolution and challenges. The structural characteristics of the economy of Rwanda can be summarized in four points: (1) Rwanda is a landlocked country, with long access routes to the ocean; (2) agriculture is by far the most important sector of the economy; there is a lack of development in other sectors; (3) population has recently been growing at an annual rate of 3.3%; land is therefore becoming scarcer; (4) exports are concentrated on a few items (mainly coffee and tea). For transport to the ocean two routes are available. One goes northwards, through Uganda, to Mombasa in Kenia. The total distance by road is 1740 km; if a combination of road and rail is used the distance is 1924 km. The other route goes southwards, through Burundi, over Lake Tanganyika and then by rail through Tanzania to the port of Dar es Salam, with a total distance of 1715 km. Transport is therefore expensive; moreover it is dependent on good relations with the neighbouring countries and on the security situation in and the relations between those countries. An alternative is to use transport by air. The high transportation costs are a natural disadvantage for the Rwandese exports. At the same time they constitute a natural barrier against imports from overseas. Trade and specialisation within the central African region could be a source of growth for the Rwandese economy. However they require co-operation between the central African countries and a stable political and economic environment in each of them. Presently there exist two frameworks for economic integration, i.e. the Communaute Economique des Pays des Grands Lacs (CEPGL). consisting of Rwanda, Burundi and Zaire, and the Zone d'Echanges Preferentiels (ZEP, in English Preferential Trade Area, PTZ), which consists of a number of eastern and southern African countries} Generally speaking the 3 Djibouti, Somalia. Ethiopia, Kenya, Uganda, Tanzania, Rwanda, Burundi, Zambia, Zimbabwe, Malawi, Swaziland, Namibia, Mauritius.  4 CEPGL has not been a success. The ZEP though has had some success : a number of internal tariffs have been reduced or abolished. Table 1 : Basic data Population (mln) 3.7 6.7 Population per square ian (1) 140.0 256.0 Share of agriculture in GDP (%) 61.6 40.1 Share of coffee in exports (%) 56.8 61.4 (1) Total area is 26,340 square kilometres. Source: World Bank; IMF, International Financial Statistics. More than in most other low income countries, agriculture is the most important sector of the Rwandese economy. It generates 40% of GDP. It provides employment for approximately 90% of the active population. As a corrollary to this urbanisation is very limited. According to the 1991 census the capital, Kigali, had a population of 237 000 people, which represents only 3.3% of the total population. The other cities are considerably smaller. Overall the degree of urbanization is approximately 10%4. Probably the most important economic problem Rwanda presently faces is the growing scarcity of land. Based on the latest two censuses population is estimated to grow at an annual rate of 3.3%. Until recently it was possible to extend the cultivated area. But such an extension is becoming increasingly problematic as already marginal land is being brought under cultivation. This raises two fundamental problems. First if ~ as in the past - agricultural productivity increases only slowly, Rwanda will probably be confronted with food shortages. A famine already occurred in some southern provinces at the end of 1989. Second other sectors than agriculture should increasingly provide employment and income 4 The figures on agricultural employment and on urbanisation are from UNDP, Human Development Report 1992. Given the well known fact that a fraction of the rural population is doing non~agricultural jobs, agricultural employment is probably overestimated.  5 for the growing population. Finally Rwanda has not succeeded in diversifying its exports. In fact the concentration of exports increased after 1985, when the activities of the smelting-works for tin were halted due to the bankruptcy of the mining company SOMIRWA. In recent years the two main cash crops, coffee and tea, made up approximately 90% of total export receipts. The country is therefore very vulnerable to adverse climatic conditions and to fluctuations of prices on the world market. In 1989 e.g. a combination of unfavorable weather conditions and of falling world prices caused a drastic decrease of receipts from coffee; luckily this fall in revenue was partly compensated by increased receipts from the export of tea. Apart from exports the main other source of foreign purchasing power is development assistance in the form of gifts and soft loans. Since 1988 public transfers from abroad have exceeded export receipts in every single year. 2. MAIN ECONOMIC POLICIES. The economic program of the MRND, the unique party set up after president Habyarimana came to power in 1973, was based on the ideas of "endogenous and selfreliant" development (,'developpement endogene et autocentre") and of "planified liberalism" ("liberalisme planifie"). The former concept meant that development should be based on the use of the natural and human resources of the country itself. Yet as already indicated development assistance in the form of gifts and soft loans made a substantial contribution to the Rwandese economy. A large part of public investment was externally financed. However it should be stressed that the Rwandese government was quite cautious in attracting foreign loans. Thus in the eighties the country did not face a debt burden comparable to that of many other African countries. The meaning of "planified liberalism" has not been made quite clear. In actual fact economic policy always left room for private initiative. The government tried to maintain a favorable investment climate and a fairly liberal trade regime. At the same time the creation of public and mixed enterprises was considered necessary to satisfy the essential needs of the population in areas where private initiative was not sufficiently active. Of course the actual emphasis on the private versus the public sector fluctuated over time. Before independence, under Belgian rule the industrial development, that took place in the region, was concentrated in Bujumbura (Burundi). After independence most public  6 investment went to agriculture, transport and communications. The industrialisation, that took place, was the result of private initiative. During that period the trade and exchange rate regimes were quite liberal. There were few restriction on imports and on the repatriation of profits by foreign firms. In the second half of the seventies and in the eighties government intervention in the economy increased. The government started to create public enterprises and to participate in mixed companies. By 1990 there were 62 public firms, of which 32 were wholly-owned by the government, 12 were mixed companies with a majority government participation and the remaining 18 had majority private ownership. Public enterprises were set up because it was thought that the government had an obligation to participate in enterprises of strategic interest (coffee, tea, petroleum) and in natural monopolies (electricity). It was also thought that the national economy should be stimulated by projects that did not attract private investors (e.g. pyrethrum extraction, banana wine production). Some public enterprises were created as the result of the transfer to the public sector of projects initially set up and run by foreign aid agencies. Table 2 : Credit (bin FRw, end of year) 1986 1987 1988 1989 1990 1991 Loans to government 2.7 6.8 8.5 10.6 17.6 16.3 public firms 0.7 0.8 0.9 1.7 2.7 0.9 private firms 16.1 16.8 20.9 22.4 19.6 15.9 Total 19.6 24.3 30.3 34.6 39.9 33.1 Source: Banque Nationale du Rwanda, Bulletin, Nos 15-18.  7 In the second half of the eighties investment in public firms absorbed a substantial amount of Rwanda's public resources. Many public enterprises were dependent on subsidies for their survival. A program of reform of the sector was started up with IDA assistance. Yet public firms apparently did not crowd out the private sector from access to credit from financial institutions. Credit to public firms more than doubled between 1982 and 1983, but remained fairly stable afterwards. It increased again considerably in 1989-90, but fell back to its previous level in 1991. Larger private firms seem to have had relatively easy access to credit. Also in the eighties also government intervention in the economy was on the increase, mainly as a reaction to the deterioration of the economic situation. The foreign trade regime became less liberal. The granting of import licences ceased to be automatic. Since the end of 1989 there has been a return to more liberal trade conditions in the framework of a Structural Adjustment Program. Yet firms in the private sector keep complaining of too many and too complicated legal rules for the creation and operation of private enterprises. These rules have been studied in detail by C. Kalinijabo (1988) and by A. Nzabandora and E. Karemera (1990). Their reports contain an excellent analysis of the existing legal and administrative problems facing the private sector. 3. ECONOMIC PERFORMANCE 1970-1991. 3.1 The evolution of the economy during the seventies. During the seventies the Rwandese economy grew at an average annual rate of approximately 5%. The growth rates of agricultural and of manufacturing output were even higher and well in excess of population growth. GDP growth was especially fast after 1974. Whereas between 1970 and 1974 GDP at market prices rose at an average annual rate of less than 1%, the growth rate for the years 1974-80 was more than 7%. The explanation of the strong growth in the second half of the seventies may be found in the improvement of the terms of trade in 1976-77, due to a peak in world prices of coffee, and in a strong increase in gross fixed capital formation, which rose from less than 10% of GDP in the early seventies to 16% in 1980.  8 Table 3.1 : Macroeconomic Statistics pcriqcL,i\ veragcs 1970-1990 (1) 1970-1980 1981-1986 1987-1990 Growth of population (%) (2) 3.3 3.1 3_1 Growth of real GOP (%) (3) 5_0 3.5 0.4 A/:,rriculturc 6.3 1.2 -0.1 (a) (1) Industry 3.8 -0.9 (a) (1) Manufacturing 5.7 4.0 -0.6 (a) (1) Services 6.3 2.8 (a) (1) Growth of real GOP per capita (%) 1.7 0.4 -2.6 Composition of expenditures (% of GOP) Private consumption 81.0 81.0 81.6 Public consumption 13.2 13.0 14.6 Gross domestic investment 12.1 15.6 13.9 Gross flXed capital formation 11.7 14.8 14.1 Exports 12.9 11.5 8.7 Imports 19.2 21.1 18.7 Fiscal deficit (% of GOP) (4) 1.6 4.9 Balance of payments (% of GOP) Exports of goods and services 15.5 11.5 7.2 Imports of goods and services 24.1 22.7 18.1 Current account balance before official transfers -9.2 -11.0 -11.3 after official transfers 0.7 4.1 -5.6 Change in reserves (- = increase) -2.2 0.3 1.0 External debt (% of GOP) (5) Long term debt 6.6 16.7 27.0 Short term debt na 1.6 1.9 Terms of trade (1980 = 100) Export prices 68 95 87 (c) Import prices 56 94 89 (e) Terms of trade 124 102 98 (c) Growth of money supply 19.6 9.7 4.6 Inflation rate (yearly average) GOP deflator (6) 11.6 4.5 2.4 Consumer price index 11.6 5.2 3.0 Footnotes: (a) Period 1987-1989 (b) Period 1976-1980 (c) Period 1987-1988 Sources: (1) Up to 1986 based upon data from the World Bank, World Tables 1989-1990. From 1987 on based upon data from IMF, International Financial Statistics (Sep. 1992) (exceptions see below). (2) Population data for the period 1978-1991 have been obtained by interpolation based on the resulls of the 1978 and 1991 censuses. (3) Sectoral data at constant prices: 1987: World Bank, World Tables 1989-1990. 1988, 1989: Based upon data in current prices and scctoral deflators calculated from Miniplan, Comptcs Economiques Nationaux du Rwanda, 1989. (4) Data on fiscal deficit from 1970-1980 and from World Tablcs and for 1987-1990 from World Bank, non published data. (5) Data from 1987 onward retrieved from World Debt Tables 1991-1992. (6) GOP deflator from 1987 onward deduced from GOP dala in Banque Nationalc du Rwanda, Rapport sur I'evolution cconomiquc ct mooetairc du Rwanda, 1990.  9 Data available on public finance suggest that the government generally followed a cautious policy. With the exception of 1972 the deficit never exceeded 2% of GDP. It Was mainly financed by foreign loans. As a result foreign debt gradually increased, from the equivalent of 1% of GDP in 1972-74 to 16% in 1980. Growth was combined with moderate inflation. In most years the consumer price index rose by less than 10%. The years 1974-75 formed an exception, as the GDP deflator and the consumer price index on average went up by respectively 45% and 31 %. The external current account balance, excluding official transfers, was largely negative; on average the deficit was equivalent to 9% of GDP. But as the country received rather large amounts of foreign assistance, in most years the inclusion of official transfers resulted in a positive balance. In 1978 and 1980 though there were current account deficits equivalent to respectively 5.3 and 4.1 % of GDP. In most years capital inflows (other than transfers, but including errors and omissions), also showed a positive balance. This resulted in a gradual build-up of foreign reserves, with accelerations in 1975-76 and 1979. Simultaneously, as already stated, foreign debt gradually increased to a level of 16% in 1980. In general government policy was non-interventionist. Although there existed a system of import licenses, authorizations to import were given without problems. There were few quantitative import restrictions. But as mentioned in paragraph 2, in the second half of the seventies government increasingly participated in secondary and tertiary sector activities by creating public and mixed enterprises or participating in existing enterprises. 3.2 Growing economic problems, 1981-86. At the beginning of the eighties the Rwandese economy ran into serious problems. While according to official data GDP in 1981 rose by 10%, between 1981 and 1985 the average annual growth rate was only 1.5%. In 1986 GDP growth temporarily went up again to 7%. An important factor underlying the economic evolution were export prices, especially the world price of coffee. The export price index (1980= 100), which had reached a level of 111 in 1977, fell from 106 in 1979 to 86 in 1981. The four following years this index averaged 91. In 1986 the sudden increase in the coffee price lifted the export price index to 121 and thus contributed to the temporary increase of GDP growth just mentioned.  10 The growth of GDP in 1980-81 was partly the result of an increase in government expenditure, especially of salaries, a belated response to the increase in government receipts from coffee during the coffee boom of 1977-79. Falling receipts from coffee exports and rising expenditures resulted in an increase in the public sector deficitS. In 1983, as a reaction to these tendencies, the government adopted more restrictive budgetary and monetary policies. By the middle of the eighties, the financial situation of the government had improved again. Falling export receipts and rising expenditures on imports, due to the second oil shock, resulted in a deterioration of the current account of the balance of payments. The current account balance, including official transfers, turned negative in 1980 and remained so throughout the period under consideration (and indeed throughout the eighties). The deficit was maintained through a combination of an increase of foreign debt and a loss of foreign reserves. Foreign debt rose from 16% of GDP in 1980 to 23% in 1986. Foreign exchange reserves fell from $ 186 mIn in 1980 to $ 113 mIn in 1985. In 1986 they increased again as a result of a higher world price of coffee. To improve the external current account the government imposed restrictive conditions on the allocation of import licences, raised import duties on luxuries and prohibited imports of certain goods, which competed with domestic production. This policy was to some extent succesful. The current account deficit contracted in 1983 and 1984 and rose moderately in 1985-86. As already noted GDP increased in the early eighties, but over the years 1983-86 it stagnated. More specifically between 1979 and 1986 value added in agriculture hardly increased. While the growing population probably hit the ceiling of arable land, the lack of improvements in productivity left few possibilities to raise output. Finally it should be stressed that prudent monetary policies kept the inflation rate at a relatively low annual average of approximately 5%. 5 This is a conjecture as data comparable to those of the preceding period are not available.  11 3.3 The economic crisis of the late eighties and the Structural Adjustment Program. After a temporary improvement in 1986 the economic situation of Rwanda further deteriorated. One underlying determinant of the crisis was the falling world price. of coffee. Furthermore unfavorable weather conditions caused a drastic fall of the volume of coffee exports in 1988-89. Another unfavorable shock was the invasion by rebels at the end of 1990 and the ensuing border war in the North of the country. GDP fell in 1987 and 1988. Earlier data showed another drastic fall of GDP in 1989, but revised data suggest an increase of 4% in that year. According to the same data GDP stagnated in 1990 and decreased again in 1991. The public sector deficit reached a record level of 8% of GDP in 1987 (13.1 % if grants are excluded). One major factor underlying this deficit was the increasing subsidy to the coffee sector, which resulted from a growing gap between the domestic producer price and the world price. After 1987 the deficit decreased. It rose to a new record in 1990. This has been attributed to rising defence expenditure and falling tax receipts, both resulting from the war. While the full effect of both evolutions could have been expected in 1991, the fiscal deficit for that year actually was lower. The deficits were covered by internal and external lending and by accumulating payment arrears. The external current account deficit, including official transfers, reached record levels of 6.2% and 6.3% of GDP in 1987 and 1988 respectively. Excluding official transfers the deficit remained higher than 10% of GDP. This unfavorable evolution was the result on the one hand of a sharp decrease in the value of exports in 1987 and further decreases in the following years and on the other hand of imports, which remained at their historical 1986 record level of in the two following years. As already stated the falling value of exports resulted from the unfavorable evolution of world markets after 1986 and from the decreasing volume and quality of Rwandese coffee exports in 1987-88. This coincided with a decline in other traditional primary exports and an appreciation of the Rwandese franc. After 1988 the tightening of import restrictions contributed to a reduction of imports. The deficit of the current account was covered by concessionallending and by drawing on foreign currency reserves. The latter decreased from 64 mIn US$ at the end of 1987 to 70 mIn and 44.5 mIn US$ at the end of respectively 1989 and 1990. Faced with a foreign exchange crisis Rwanda called on the IMF and the World Bank. An agreement was reached in September 1990. The invasion by rebels in October 1990 necessitated a revision of this agreement, which was finalised in February 1991. The  12 Table 3.2 : Annual rales for period 1987-90 (1) 1987 1988 1989 1990 Growth of real GOP (%) (2) -0.8 -1.6 4.0 0.1 Agriculture -2.0 -5.2 7.4 na Industry 1.8 -0.7 -3.8 na Manufacturing 1.8 -2.7 -1.0 na Services -0.9 6.0 3.3 na Growth of real GOP per capita (%) -3.7 -4.6 0.8 -3.0 Composition of expenditures (% of GOP) Private consumption 80.2 81.3 82.1 82.8 Public con~umption 13.4 14.1 13.6 17.2 Gross domestic investment 15.6 15.0 13.3 11.6 Gross fixed capital formation 15.6 14.4 13.7 12.6 Exports 9.8 9.1 8.1 7.7 Imports 19.1 19.5 17.2 19.2 Fiscal deficit (% of GOP) (3) 8.0 5.3 3.3 8.1 Balance of payments (% of GOP) Exports of goods and services 8.2 7.6 6.6 6.4 Imports of goods and services 20.3 19.2 16.6 16.2 Current account balance before official transfers -11.7 -12.3 -10.6 -10.6 after official transfers -6.2 -6.3 -5.1 -4.7 Change in reserves (- = increase) 0.6 1.2 2.1 0.1 External debt (% of GOP) (4) Long term debt 26.5 26.7 25.3 29.6 Short term debt 1.8 1.8 1.9 2.1 Terms of trade (1980 == 100) Export prices 79.6 94.7 na na Import prices 90.2 88.0 na na Terms of trade 88.2 107.6 na na Growth of money supply 10.3 7.3 -4.1 5.6 Inflation rate (yearly average) GOP deflator (5) 1.3 3.2 3.9 Consumer price index 4.1 3.0 1.0 4.2 Sources: (1) From 1987 on based upon data from IMF, International Financial Statistics (Sep. 1992) (exceptions see below). 1986 data based upon data from the World Bank, World tables 1989-1990. (2) Sectoral data at constant prices: 1987 : World Bank, World Tables 1989-1990 1988, 1989 : Based upon data in current prices and sectoral deflators calculated from Miniplan, Comptes Economiques Nationaux du Rwanda, 1989. (3) Fiscal deficit data are from World Bank, non-published. (4) Data from 1987 onward retrieved from World Debt Tables 1991-1992. (5) GOP daflalor from 1987 onward deduced from GDP data in Banque Nationale du Rwanda, Rapport sur "evolution economique et monetaire du Rwanda, 1990.  13 revised agreement covered the period 1991-93. It aimed at an increased reliance on market forces and private initiative and at a more export oriented appoach. The agreement specified measures and orientations in the fields of trade and foreign finance, budgetary policy, public enterprise, credit policy and price policy. More specifically the following measures were either actually taken or scheduled to be taken. (1) Trade and international finance: the Rwandese franc was devalued by 45% on November 9, 1990. The understanding was that its value would be adjusted when necessary. In June 1992 it was again devalued, this time by 15%. Imports were to be liberalised; import prohibitions have been suppressed in 1991 and the import tariff structure has been simplified. Export promotion measures will be considered. (2) In the field of budgetary policy the revenue administration was to be reinforced and the fiscal system was to be improved. On the expenditure side arrears would be eliminated and subsidies would be reduced. The subsidy to the coffee sector has been reduced in 1991 but it was not abolished. A revolving three year public investment program has been establised. (3) Efficiency in public firms sector would be improved by rehabilitaton of viable firms, privatisation where possible and liquidation of non-viable firms. (4) Credit to the public sector would be restricted. Credit ceilings for individual banks would be replaced by indirect controls. (5) In the field of price policy, with the exception of monopolies, the control of profit margins has been eliminated in November 1990. The number of products subjected to price controls has been reduced. Economic activity in 1991 and 1992 remained affected by the military situation. Apart from its direct costs, trade was hampered by military road controls. Above all the war resulted in a climate of political and economic uncertainty, which had a serious negative impact on the level of economic activity. A positive aspect of the economic evolution over the period 1986-90 was the mild level of inflation. Measured by the consumer price index the average annual inflation rate was 3%. However it is sometimes suggested that official data underestimate the real price increases. In 1988-89 prices might have risen by 10-20%6. In 1991, as a result of the devaluation, the official inlation rate rose to 20%; in 1992 it was down again to less than 10%. 6 World Bank, Rwanda (1991), Financial Sector Review, pp. 10-11.  14 4. SECTORAL COMPOSITION OF GDP. 4.1 General overview. Table 4.1 shows the relative importance, in terms of value added, of agriculture, industry and services for selected years in the eighties. The share of agriculture in GDP has fallen drastically. According to a World Bank estimate, it was 80% in 1960, as compared with only 49% in 1980 and 41% in 1989. An overwhelmingly large share of value added in agriculture is realised in the production of food crops. The share of export crops has recently declined. Table 4.1: Composition ofGDP at current prices (%) 1980 1985 1988 1989 Agriculture 47.8 43.2 38.6 41.0 Export crops 4.7 3.6 3.0 1.8 Other products 43.1 39.5 35.6 39.3 Industry 22.3 24.0 22.7 21.5 Mining 1.8 0.3 0.2 0.4 Manufacturing (1) 15.9 14.7 15.2 14.0 Construction 4.6 9.0 7.2 7.0 Services 29.9 32.8 38.7 37.5 (1) including utilities. Source: Ministere du Plan, Comptes Economiques de la Nation, 1981 and 1989. The share in GDP of industry (including utilities and construction) rose from an estimated 6% in 1960 to 22% in 1980. In the eighties this percentage fluctuated without showing a clear trend. It should be noticed that a large share of industrial production consists of value added in the informal or traditional sector. One particular subsector, traditional beverages, which consists mainly of the production of banana beer, accounted for 17% of value added in industry and 26% of value added in manufacturing. The production of traditional beverages is a semi-agricultural activity; its output is almost entirely locally consumed.  15 Table 4.2: Composition of value added in manufacturing at current prices C%) 1980 1985 1988 1989 Food industry 73.2 65.0 62.4 Traditional beverages 41.3 25.1 26.7 Industrial beverages 18.9 23.9 25.8 Butchers, slaughterhouses ) 1.0 1.3 1.4 Dairy ) 0.2 0.1 Coffee, tea 5.2 7.0 0.6 Rice mills 0.5 1.3 0.3 Oil, sugar 3.0 0.9 0.6 Sigarettes, jam 3.0 4.6 Bakeries, flour mills 3.4 2.3 2.2 Textile 4.3 5.3 6.2 Textile, leather 0.9 2.5 3.2 Confection 3.4 2.8 3.0 Basketry 3.3 3.4 3.6 Wood works 3,4 4,4 4.0 Paper 1.2 1.2 1.2 Chemicals 2.2 4.1 5.0 Non metal minerals 53 9.8 10,4 Metals (1) 7.2 6.7 7.1 Total (bIn FRw) 16.5 24.6 24.9 (1) including garages Source: see table 4.1. The share of mining in GDP has fallen, especially after the mining company SOMIRWA went bankrupt in 1985. The share of services in GDP has increased from an estimated 14% in 1960 to 30% in 1980 and 37.5% in 1989 4.2 Manufacturing. The manufacturing sector is dominated by the food subsector, which accounted for 73% of value added in manufacturing in 1980 and for 62% in 1989 (see table 4.2). While  16 beverages in 1980 represented 60% of value added in manufacturing, by 1989 this share had fallen to 52%. The decrease was entirely the result of the smaller share of traditional beverages. The relative importance of the formal beverages subsector had risen, thanks to the increasing production of the modern breweries BRALIRWA and OVIBAR. The share of industrial activities in the coffee and tea sectors, which used to be higher than 5% of value added in manufacturing, fell drastically in 1989. This was the result of falling output value, combined with a slightly increased value of the raw materials, the latter reflecting the price support for coffee and tea producers. Among the remaining manufacturing subsectors, the most important were textiles, non- metal mineral products and metal works. The most notable increases occurred in the textile, chemical and mineral non-metal subsectors. The share of the metal subsector in value added in manufacturing was approximately 7% and that of wood works 4%. 5. PUBLIC FINANCE The classification of Rwandese public finance data is not at all transparent 7. The administrative set-up contains four budgets: the ordinary budget, the development budget, the 'budget pour ordre' and the accounts outside of the budget. Each budget has its own revenues and expenditures. Revenues include receipts from loans. Different sources use different methods to classify the expenditure data of the four budgets as current or capital expenditures. They also give different figures for the fiscal deficit. The public finance data used in this paper are from Republique Rwandaise, Programme d'Ajustement Structure!, Document Cadre de Politique Economique et Financiere a Moyen Terme (September 1990) and from the World Bank (unpublished documents). 5.1 General structure of public finance. General data of the structure of government revenue and expenditure, the deficit and its financing are presented in table 5.1. Revenue consists of revenue proper and grants. The share of taxes in total revenue, including grants, fluctuated around a level of 70%. In 1991 it fell to 56%. The share of grants in total revenue fluctuated between 15% and 21 %. In 1991 it rose to 35%. 7 The World Bank (World Tables) and the IMF (International Financial Statistics) do not publish data on public finance for Rwanda.  17 Table 5.1 : Government Revenue and Expenditure (bIn FI3c\.VJ. 1985 1986 1987 1988 1989 1990 1991 1992 (prov.) REVENUE Tax revenue 18.8 21.5 20.2 20.0 20.9 18.8 21.8 24.0 Non-tax revenue and adjustment 2.4 2.1 3.0 3.3 3.4 2.8 3.2 3.1 Total revenue 21.2 23.5 23.1 23.3 24.4 21.6 25.0 27.1 Grants 4.9 4.5 4.1 6.2 5.3 5.9 13.7 (a) Total revenue and grants 26.0 28.0 27.2 29.5 29.7 27.5 38.7 27.1 EXPENDITURES (commitments) CUrent expenditure 17.3 20.2 24.0 23.8 23.5 30.2 35.9 44.8 Capital expenditure 14.9 15.4 16.8 14.7 12.2 13.2 17.1 7.1 Total expenditures 32.3 35.6 4O.S 38.5 35.7 43.4 53.0 51.S Total revenue-current expenditure 3.9 3.4 -0.8 -0.5 0.9 -S.6 -10.9 -17.7 Lending minus repayment 0.2 0.5 0.2 0.2 0.2 0.1 0.2 0.0 SURPLUS (commitments) -6.4 -S.l -13.8 -9.3 6.2 -16.1 -14.5 -24.S Change in arrears (- = decrease) -0.4 0.0 0.9 1.0 -2.2 3.5 6.2 0.0 SURPLUS (cash) -6.7 -8.1 -12.9 -8.3 -8.4 -12.6 -8.3 -24.8 Discrepancy 0.0 0.0 0.0 0.5 -0.2 0.3 0.0 FINANCING 6.7 8.1 12.9 7.8 8.6 12.3 8.3 15.3 Foreign 6.0 6.0 7.8 5.0 4.1 5.4 6.5 5,7 (b) Domestic 0.7 2.0 5.0 2.7 4.4 6.9 1.8 9.6 As%ofGDP Total revenue 12.2 13.9 13.6 13.2 12.8 10.9 11.8 Grants 2.8 2.7 2.4 3.5 2.8 3.0 6.5 Current expenditure 10.0 12.0 13.9 13.5 12.4 15.2 17.0 Capital expenditure 8.6 9.1 9.8 8.3 6.4 6.7 8.1 Deficit (commitments) 3.7 4.8 8.0 5.3 3.3 8.1 6.9 (a) data not yet available (b) including 1.1 bin FRw"aid to the budget" Source : R6publique Rwanda (1990), Programme d'Ajustemcnl StrucLUrel, Document Cadre de Politique Economiquc ct Financicrc it Moyen Tcrmc (OCl. 1990 - Sept. 1993) and World Bank.  18 Over the years 1986-89 total revenue stagnated in nominal terms; as a percentage of GDP it fell from almost 14% in 1986 to less than 13% in 1989. In 1990 there was a noticeable decrease in tax and in non-tax revenues. In 1991 they rose again, without reaching, in terms of GDP, their 1989 level. In most years the amount of grants was equivalent to approximately 3% of GDP; in 1991 it rose to 6.5% of GDP. In the second half of the eighties, total expenditure was equal to approximately 20% of GDP.1t was higher in 1986-88, but lower in 1989. In 1991 it reached the exceptionally high level of 25% of GDP. The share of current expenditure in the total rose from somewhat more than one half in 1985 to more than two thirds in 1990-91. During the years 1985-89 total revenue (excluding grants) more or less covered current expenditure. The rise of the latter in 1990 and 1991 was not covered by a rise of revenue nor by a decrease of capital expenditure. It resulted in a fiscal deficit (on a commitment basis) of8.1% ofGDP in 1990 and of 6.5% in 1991. The deficit on a cash basis differs from the one based on commitments by the change in arrears. The latter increased in 1987-88, and especially in 1990-91. A large fraction of the deficit was financed by foreign loans, but domestic fmancing has gradually become more important. The structural adjustment program adopted in 1990 provided for a continuation of the restrictive budgetary policy. Tax receipts were to be raised by an improvement in the tax administration system. Moreover the global tax system was to be critically examined. In the short run receipts was to be be raised both as a result of the devaluation and of an increase of the turnover tax rate. In addition a moratorium was established on new exemptions in the framework of the investment code of 1987; a revision of this code would be undertaken. As for expenditures, the control of the growth of wages, which was earlier established, was maintained. An evaluation of the public administration was made with the purpose of rationalising its structure. Subsidies to public firms which provide private goods and services, were to be eliminated. The subsidy for the coffee sector, which was to be abolished, has been reduced, but not removed. As for capital expenditures a triannual gliding Program of Public Investments (PIP) has been established. in collaboration with the World Bank.  19 The invasion by rebels and the political turmoil have delayed the execution of the program. Moreover as already stated these events have resulted in a drastic rise of current expenditure and of the fiscal deficit. 5.2 Tax receipts. The structure of tax receipts is presented in table 5.2. Approximately three quarters of the tax revenue come from indirect taxes. More than half of the indirect taxes used to come from import and export duties. However in recent years their share has somewhat fallen; but in 1991 it increased again. Revenue from import duties was in most years equal to 25 to 30% of total tax receipts. As revenue from export duties consists almost entirely of the export tax on coffee, it fluctuates with the value of coffee exports. Table 5.2 : Composition of tax revenue C% of total) 1985 1986 1987 1988 1989 1990 1991 1992 (est) (prov) Tax on income and profits 21.8 19.8 23.0 25.9 25.1 26.0 22.3 24.4 Corporate 7.3 5.8 7.6 6.0 504 5.7 3.2 6.8 Business licenses 2.7 2.9 3.1 3.4 3.5 3.9 3.5 3.5 Individual 9.0 8.8 95 lOA 10.3 lOA 9.7 8.0 Other 2.8 2.3 2.8 6.1 5.8 6.1 5.9 6.2 Taxes on property 2.1 2.0 2.2 2.7 2.8 3.1 2.1 204 Domestic taxes on goods and services 28.0 24.8 31.0 32.8 35.7 35.9 36.8 32.9 Excises 28.0 22.1 24.3 26.0 28.8 29.2 23.9 21.2 Sales tax 2.7 6.6 6.8 6.9 6.7 12.8 11.7 Road Fund 1.3 4.6 3.2 3.3 3.3 3.5 4.7 Taxes on international trade 48.0 52.1 39.1 35.4 33.1 31.7 35.4 35.6 Import Duties 30.7 27.8 25.7 27.4 25.1 22.4 28.6 29.3 Export duties 15.9 23.2 12.1 6.8 6.9 8.0 5.8 5.4 Coffee 13.4 22.9 12.1 6.3 6.6 7.8 5.8 5.4 Other 25 0.3 0.0 0.4 0.3 0.2 0.0 0.1 Other taxes 1.4 1.1 1.3 1.2 1.1 1.2 1.1 0.8 Source: see table 5.1.  20 The receipts from domestic indirect taxes has gradually increased. This was mainly the result of rising revenue from the turnover tax. As in most low income countries direct taxes in Rwanda produce a much lower share of total tax receipts than indirect taxes. In the years 1985-89 receipts from taxes on firms (corporate income tax and business licenses) and from the personal income tax each represent approximately 10% of total tax receipts. In 1990 their share in the total was considerably lower, but it rose again in 1991. 5.3 Public expenditures The structure of current expenditures in the years 1985-91 is shown in table 5.3; the structure of public investment expenditure in the years 1990-91 is presented in table 5.4. Table 5.3 : Composition of current expenditures (% of total) 1985 1986 1987 1988 1989 1990 1991 1992 (prov) Wages and salaries 50.5 47.0 42.2 44.9 46.8 38.6 37.0 32.1 Purchases of goods and services 27.8 29.3 24.2 25.4 23.7 33.0 43.4 45.8 Interest payments 7.1 6.9 7.6 8.5 8.6 7.0 12.0 10.2 on domestic debt 4.9 4.8 5.1 5.7 5.1 4.9 9.1 5.2 on foreign debt 2.2 2.1 2.4 2.9 3.1 2.1 2.8 5.0 Subsidies and transfers 13.5 13.9 23.2 18.0 20.9 21.4 7.6 11.9 OCIR coffee 0.4 2.1 13.3 6.8 8.8 13.3 2.4 5.4 Other 13.1 11.8 9.9 11.2 12.1 8.1 5.2 6.4 Unclassified 1.1 2.8 2.9 3.1 0.8 Source: see table 5.1.  21 Table 5.4: Sectoral composition of Public Investment Program (bin FRw and %) 1990 1991 Share of lotal (%) Agriculture 4.6 5.6 25 Industry 0.4 1.2 4 Infrastructure 6.5 9.2 38 Transport 0.1 o Energy 0.7 1.0 4 Roads 2.9 5.1 19 Water/sewers 2.5 2.0 11 Post & Telecom 0.3 0.6 2 Human resources 5.3 5.9 27 Education 2.1 2.7 12 Health 3.1 3.0 15 Population/Social action 0.2 0.1 1 Other 0.7 2.2 7 Total 17.5 24.0 100 Source: World Bank. In 1985-89 expenditures on wages and salaries made up almost half of the current expenditures. After 1989 their share fell quite drastically, partly as a result of restrictions on wages in the public sector. Until 1989 the share of outlays on goods and services in total current expenditures was approximately one quarter. Afterwards it rose considerably, probably due to purchases of military material. Although gradually increasing, the share of interest payments on the public debt remained relatively smalL This was the result of a fairly low level of internal debt and of the fact that external debt is mainly on concessional terms. Finally the relative importance of subsidies and transfers fluctated with the subsidies to  22 the coffee sector. It rose strongly in 1987, reaching a level of more than 20% of current expenditures. This was mainly the result of the fall in the world price of coffee and the . maintanance of the domestic price at the high level of 1986. As already mentioned the coffee subsidy was to be completely eliminated, but this has not yet been realised. Other subsidies and transfers, financed on the ordinary budget, on average made up 12% of total current expenditures.Their importance decreased after 1989. The functional distribution of current expenditures on the ordinary budget is published by the National Bank. As expected there has recently been a strong increase of expenditure on defence. As shown in table 5.1 capital expenditure increased in 1986-87, but subsequently decreased, probably as a result of the policy to reduce the fiscal deficit. Capital expenditure was raised in 1991. But in 1992 it seems to have fallen to an all time low. Table 5.4 shows data on the public investment program for the years 1990-91. The main component of public investment was expenditure on infrastructure. This represented 38% of the total. Half of this expenditure was on roads. Agriculture and human resources each got approximately one quarter of the total, while 4% was for investment in industry. 5.4 Public debt. During the decade of the eighties total public debt has gradually increased, from the equivalent of 18% of GDP in 1980 to 61 % in 1990. In relative terms the highest growth was registered for the internal debt. It was equivalent to 4% of GDP in 1980 and to 19% in 1990. Until 1988 approximately one quarter of the internal claims were held by the National Bank. In 1989 and 1990 its share increased to 45%. The structural adjustment program stipulated a gradual reduction of bank credits to the public sector. The increase of defense expenditures related to the military situation, has made this objective difficult to attain. From 1980 to 1990 external debt rose from 14.5% to 43% of GDP. The devaluation of the Rwandese franc resulted in a sudden increase of the foreign debt expressed in the national currency. Half of the foreign debt was owed to the World Bank-IDA, 20% to the African Development Bank/Fund and somewhat more than one quarter to bilateral donors. The debt service-export ratio, though increasing, remained at a sustainable level of less than 20%.  23 Table 5.5: Internal and external public debt (bill. FRw and % of GDP) 1980 1985 1986 1987 1988 1989 1990 Internal debt 4.0 14.3 16.5 21.1 24.3 28.8 36.2 of which BNR 1.0 3.2 4.0 7.0 6.2 9.6 16.6 External debt 15.7 32.6 37.0 45.1 50.6 50.0 82.5 Multilateral creditors 9.8 22.9 26.2 29.2 34.0 38.0 63.3 Bilateral creditors 4.9 8.4 9.7 13.1 14.4 10.0 17.1 Supplier credit 1.0 1.4 1.1 2.8 2.2 2.1 2.2 As % ofGDP Internal debt 3.7 8.3 9.7 12.3 13.8 15.2 18.7 External debt 14.5 18.8 21.7 26.3 28.8 26.3 42.7 Source: Banque Nationale du Rwanda. 6. BAlANCE OF PAYMENTS. 6.1 General characteristics. Detailed information on the balance of payments is presented in table 6.1. As already stated in paragraph 3 the current account has been in deficit in every single year since 1980. Both the goods and the services accounts have a negative balance. Exports of goods cover less than half of the imports. For services the imbalance is even more important. In relative terms the strongest disequilibrium concerns the transport of goods. As Rwanda is a landlocked country and far removed from the oceans, transport costs are relatively high and payments for transport services are far superior to receipts. Interest payments on the public debt amount to only five percent of payments for services. Transfers are not sufficient to compensate for the deficit on the goods and services accounts. Most of the transfers have a public character and consist of development assistance in the form of gifts.  24 Table 6.1 : Balance of payments (bln FRw) 1986 1987 1988 1989 1990 1991 Current account Balance on goods -6.6 -11.6 -12.3 -11.8 -10.3 -16.6 Exports 16.1 9.6 9.0 7.S 8.5 12.0 Imports 22.7 21.3 21.3 19.6 18.8 28.5 Balance on services -10.4 -9.2 -8.3 -7.2 -8.7 -10.3 Exports 4.6 4.5 4.3 4.2 3.9 5.8 Imports 15.0 13.6 12.6 11.4 12.6 16.1 Balance on goods and services -16.9 -20.7 -20.6 -19.1 -19.0 -26.8 Transfers 10.8 10.0 11.5 11.0 11.9 22.6 of which: public 10.3 9.4 10.6 10.3 11.4 20.0 Balance -6.1 -10.7 -9.1 -8.1 -7.1 -4.2 Capital account Direct investment 1.5 1.4 1.6 1.2 0.6 0.6 Other long term capital 5.8 7.2 4.9 3.5 3.2 10.9 of which net public lending 6.0 7.2 5.0 3.6 3.2 11.0 loans received 7.0 8.8 6.2 5.2 4.6 13.0 repayments 1.0 -16. -1.2 -1.6 -1.4 -2.0 Short term capital 1.6 0.9 -0.2 0.5 -1.6 4.0 Balance 8.9 9.5 6.3 5.2 2.2 15.4 Change in reserves -2.6 1.3 3.0 3.1 3.4 -10.9 (- = increase) Source: Banque Nationale du Rwanda, Bulletin, Nos 15, 16 and IS.  25 The deficit on current account after transfers is more or less covered by a surplus on the capital account. Most of the receipts on the capital account consist of net public lending. In the years preceding 1991 both net receipts from loans and direct investment had considerably fallen. As a result of the structural adjustment program, in 1991 net receipts from loans increased considerably, while direct investment in real terms stagnated at a low level. The overall balance was negative from 1987 to 1990, but positive in 1991. Table 6.2: Composition of exports (million FRw) Animal products 288 273 271 359 216 223 Hides 288 273 271 359 214 223 Vegetable products 14608 8581 7884 6513 7419 10401 Coffee 12569 7324 6548 4691 5425 7210 Tea 1445 971 1082 1557 1737 2797 Pyrethrum 160 121 109 151 160 279 Quinquina 192 106 61 61 30 24 Plants and live roots 44 59 32 36 55 67 Minerals 263 70 30 529 505 815 Cassiterite 94 47 381 295 320 Tin 130 Colombo-tantalium 7 110 171 381 Wolfram 34 13 38 38 115 Gold 10 23 Manufactures 32 4 43 120 151 107 Beer 1 12 7 PVC tubes and pipes 2 28 26 47 38 Hoes 1 10 9 Traditional sculptures 4 4 4 2 7 13 Corrugated iron and steel 10 64 60 Cement 1 35 Subtotal 15191 9075 8229 7521 8291 11546 Reexports, personal goods, 947 600 782 856 187 425 illegal exports Total 16138 9675 9011 8377 8478 11971 Source: Banque Nationale du Rwanda, Bulletin, Nos 15-18.  26 6.2 Composition of exports and imports of goods. The composition of exports of goods is shown in table 6.2 and that of imports in table 6.3. The bulk of Rwandese exports consists of agricultural goods. They account for 90% of total export receips (excluding receipts from re-exports, personal goods and illegal exports). In recent years more than 80% of export receipts were generated by coffee and tea, the former making up more than 60% and the latter more than 20% of the total. The main other agricultural export items are hides and pyrethrum (a natural raw material for insecticides ). Minerals account for seven to eight percent of export receipts and manufactures for less than two percent. In the first half of the eighties exports of tin had some importance. But tin production was halted in the mid 1980s, when Rwanda's large mining company, SOMIRWA (Societe Miniere du Rwanda) was liquidated. Rwanda now continues to export tin ore (cassiterite). Recently there was an increase of exports of colombo-tantalite. Table 6.3 : Composition of imports (bin. FRw) Consumption goods 12.5 8.4 7.6 7.6 6.5 10.8 Equipment 9.0 7.6 7.6 6.9 4.8 6.7 Inputs 4.0 8.0 9.1 8.3 8.0 16.0 Oil and energy 5.0 4.1 4.0 3.9 3.7 4.9 Total 1 30.6 28.2 28.3 26.7 23.1 38.5 (1) These figures include transport and other costs; they are therefore higher than the imports in the balance of payments. Source: Banque Nationale du Rwanda, Bulletin, Nos 16 and 18. The share of manufactures in exports is quite volatile. This is even more so for exports of individual manufactured goods, which can suddenly surge or fall. In 1991 e.g. exports of corrugated iron and steel plates practically ceased, while those of cement surged. Apparently Rwandese exports of manufactures are the result on the one hand of surpluses on the domestic market, and on the other hand of shortages in neighbouring countries.  27 These include countries of the PTZ (Preferential Trade Zone, in French ZEP, Zone d'Echanges Preferentiels), especially Burundi, Uganda and Kenya, as well as Zaire. The distribution of imports is shown in table 6.3. The most important class of import goods are intermediate inputs, primarily for the manufacturing sector. Their share in total imports rose from 25% in 1988 to 34% in 1990 and an exceptionnally high 41% in 1991. Concurrently shares of other categories of import goods declined. Imports of equipment have declined between 1986 and 1990. To the extent that the structural adjustment program had an effect on imports, it seems to have induced a shift towards industrial inputs, not towards investment goods. The share of oil and other energy imports was lower than 15% in all the years since 1987. 7. BANKING AND FINANCE8 7.1 Overview of the structure and performance of the financial sector. The forma! fmancia! sector in Rwanda is structured as follows. 1. The central bank, called the Banque Nationale du Rwanda, is the emission institute and is in principle charged with the supervision of the financial sector. 2. There are three commercial banks, the Banque Commerciale du Rwanda (BCR), the Banque de Kigali (BK) and the Banque Continentale Africaine du Rwanda (BACAR). Each of these banks was set up by European financial institutions, but the government has a participation in all three. However their day-to-day management is largely the responsibility of the foreign shareholders. 3. There are two savings institutions, a public one, the Caisse d'Epargne du Rwanda (CER) and a private one, the Union des Banques Populaires (UBP), which is the umbrella organization for the local"banques populaires". 4. There is one public institution for development finance, the Banque Rwandaise de Developpement (BRD). 5. There is one housing finance institution, owned by the government, the Caisse Hypothecaire du Rwanda (CHR). Its activities are financed by its equity capital and by loans. 6. In addition there are three insurance companies. One of them, the Societe 8 This section is to a large extent based on World Bank (1991), Rwanda, Financial Sector Review.  28 Nationale d'Assurances du Rwanda (SONARWA) is government controlled, while the other two, the Societe Rwandaise d'Assurances (SORAS) and the recently created Ingoboka, are privately owned. Information on the informal credit sector is scarce. It is known that there exist numerous informal savings and credit associations, but there no data are available on their importance. As in most low income countries credit between relatives, friends and people of the same area is likely to be significant, but again there is information on such credit. In comparison with other countries with a similar per capita income the formal financial system in Rwanda is rather rudimentary. The M2/GDP ratio recently fluctuated around 17% as compared with an average of 22.3% for Africa as a whole. Only the "banques populaires" and the CER have established a branch network in the rural areas. The bulk of the commercial bank loans are extended to the private sector. There is a relatively high concentration on a limited number of borrowers. As a result of the economic downturn banks are recently confronted with a deterioration of the quality of their portfolios. Because of this their main problems are now undercapitalization and insufficient provisions. The Caisse d'Epargne du Rwanda, which is government owned, was created with the objective of promoting private savings. The CER has been virtually bankrupt for more than a decade and is only kept afloat by increased deposits of the Government and public enterprises at below market interest rates. The Union des Banques Populaires was established with the support of the Swiss Government. It consists of local cooperative banks and of the Caisse Centrale in Kigali. The latter manages deposits of member banks. Recently the portfolio of the UBP also deteriorated. The UBP continues to receive assistance from the Swiss government in the form of grants to promote the creation of new banks and to cover part of the operating costs of the Caisse Centrale. The Banque Rwandaise de Developpement is the main source of investment credit. The Rwandese Government is the majority shareholder. The remaining shares are owned by foreign aid agencies and by local private investors. The deterioration of the economic situation has also affected the BRD's portfolio. But its financial situation is considered to be relatively solid. Prior to 1987 the financial system was segmented. Commercial banks could provide only  29 short and medium term credit (up to five years) while the BRD was only permitted to provide long term credit. In 1987 commercial banks got permission to make long term loans up to a specified percentage of total credit, and the BRD was allowed to provide short term credit. As the government itself is one of the beneficiaries it has encouraged banks to extend long term loans even beyond the ceiling. The BRD has not yet started to provide short term credit. 7.2 Stmcture of the credit market. Before 1987 the credit market was formally segmented along two lines. On the one hand the activity of financial institutions was limited to specific forms of credit: the commercial banks could only provide short and medium term credit, while the BRD and the CHR were only permitted to make long term loans. On the other hand a ceiling on credit was established for each individual financial institution. But this ceiling applied only to credit to non-priority sectors. Credit to priority sectors could be provided ouside of the ceiling (flhors plafond"). The priority sectors were defined quite broadly. The list of priority sectors consisted of eight categories, including among others: production, transformation, stocking and marketing of goods of the agricultural, livestock and crafts sectors; exports; productive investments generating employment, with a minimum value added of 20 percent, especially those involving the transformation of local national resources. Priority sectors could obtain loans at favorable interest rates. Both of these forms of formal segmentation have been abolished. In 1987 commercial banks were allowed to give medium and long term credit up to a certain percentage of total credit. As a matter of fact they were encouraged to extend such credit. No measures were taken when they went beyond the maximum allowed because the government itself was a main beneficiary of the banks' long term loans. Institutions for long term credit were also allowed to give short term loans. But only in 1991 the articles of association of the BRD were changed to allow it to operate in the commercial credit sector. In 1989 the central bank restricted the list of priority sectors to four categories. These included the first two mentioned above, but not the third one. On August 30, 1990 the  30 distinction between priority and non-priority sectors was abolished and with it the system of preferential interest rates. All credits had to be within the ceiling. However a non-formal segmentation continues to exist as small enterprises have problems of access to formal sector credit. 7.3 Interest rates. Before the adoption of the Structural Adjustment Program the intrest rates on loans provided and deposits held by financial institutions were fixed by the BNR. Moreover priority sectors were charged a lower rate than other sectors (see table 7.1). On August 30, 1990, even before the adoption of the Structural Adjustment Program, the system of preferential rates was abolished. Henceforth the BNR determined a maximum interest rate for loans, a minimum rate for deposits and a rediscount rate. The maximum interest rate on loans was fIXed at 12%. It was raised to 19% in November 1990 and reduced to 15% in July 1992. No information is available on interest rates on informal credits. Table 7.1; Rwanda. Interest Rates for Loans (1) (percent per annum) November July April August November July 1967 1979 1987 1988 1990 1990 1992 Priority Sectors ~9.00 9.00 ~9.oo ) ) Non Priority sector ~9.00 13.0-17.0 ~ 12.00 ) ~12 ~19 5-year Development Bond 5.0 9.00 9.00 7.00 6-month Treasury Bill 2.0 6.25 7.50 5.50 (1) Excluding coffee and tea exports, which are determined separately. (2) The distinction between priority and non-priority sectors was abandoned on August 31, 1990. Source: BNR. Tabe17.2 Sectoral Distribution of Loans (bIn FRw and %) 1990 Short Medium long Total Under Outside ·term term term ceiling ceiling Total (bIn FRw) 10.1 4.7 5.5 20.3 8.1 12.2 Sha res ( %l Agriculture 0.3 2.5 6.0 2.3 0.5 3.5 Mining 0.0 0.0 0.3 0.1 0.0 0.1 Manufacturing 14.6 19.6 50.1 25.4 12.5 34.1 1. Food Products 3.6 2.4 B.O 4.5 0.4 7.2 Coffee 1.4 0.1 0.0 0.7 0.0 1.2 Tea 1.9 0.0 1.7 1.4 0.0 2.3 Other 0.4 2.3 6.3 2.4 0.4 3.7 2. Beverages and tobacco 0.9 1.3 7.2 2.7 1.8 3.3 3. Textile and leather 1.7 9.9 6.3 4.8 0.4 7.8 3.1 Textile 1.6 9.9 6.3 4.8 0.3 7.8 3.2 Leather 0.1 0.0 0.0 0.0 0.1 0.0 4. Nood 0.3 0.2 0.9 0.4 0.4 0.5 5. Paper, printing 0.7 0.9 2.5 1.3 1.1 1.3 6. Chemical 4.7 3.5 14.7 7.1 5.7 8.1 7. Mineral non-metallic 0.2 0.4 5.0 1.6 0.2 2.4 8. Metal 2.3 0.6 4.6 2.5 2.2 2.8 9. Other 0.2 0.5 0.8 0.4 0.2 0.6 Energy and water 0.4 0.0 0.0 0.2 0.5 0.0 Construction 9.1 21.8 28.0 17.2 29.5 9.0 Transport and communication 2.8 22.1 0.1 6.5 11. 9 2.9 Services 69.4 29.6 15.2 45.4 39.3 49.5 of which commerce 67.9 23.5 8.1 41. 4 37.5 43.9 of which financial, etc 0.3 3.1 1.8 1.4 0.4 2.0 of which public 1.1 2.9 5.3 2.7 1.4 3.6 of which foreign trade 42.0 4.9 0.1 22.0 17.0 25.4 w Non-classified 3.4 4.4 0.4 2.8 5.8 0.9 Total 100.0 100.0 100.0 100.0 100.0 100.0 Source BNR Tabel 7.3 Sectoral Distribution of Loans (bln FRw and %) 1991 Short Medium long Total Under Outside term term term ceiling ceiling Total (bln FRw) B.4 3.6 5.1 17.2 9.7 7.5 Shares ( %) Agriculture 0.6 3.3 7.3 3.2 1.3 5.6 Mining 0.1 0.0 0.3 0.2 0.1 0.2 Manufacturing 19.0 21.1 46.9 27.7 14.4 44.9 1. food Products 1.5 2.7 7.9 3.6 1.3 6.7 Coffee 0.9 0.1 0.0 0.5 0.7 0.2 Tea 0.0 0.0 1.5 0.5 0.0 1.0 Other 0.6 2.6 6.3 2.7 0.6 5.5 2. Beverages and tobacco 0.0 0.0 6.0 1.8 0.0 4.1 3. Textile and Leather 3.1 10.8 4.7 5.2 1.2 10.4 3.1 Textile 3.1 10.8 4.7 5.2 1.2 10.4 3.2 Leather 0.0 0.0 0.0 0.0 0.0 0.0 4. Wood 0.6 0.3 0.9 0.6 0.6 0.7 5. Paper, printing 2.6 1.0 2.5 2.2 2.4 2.0 6. Chemical 6.6 4.1 14.1 8.3 5.5 11.8 7. Mineral non-metallic 0.4 0.8 5.6 2.0 0.4 4.1 8. Metal 4.1 0.7 4.3 3.4 2.8 4.3 9. Other 0.2 0.8 0.9 0.5 0.2 0.9 Energy and water 0.1 0.0 0.1 0.0 0.1 0.0 Construction 15.4 24.9 29.0 21.4 29.1 11. 6 Transport and communication 3.5 19.3 0.0 5.8 8.0 3.0 Services 56.2 28.0 16.1 38.3 41.5 34.3 of which commerce 52.9 21.5 7.9 32.9 38.2 26.1 of which financial, etc 0.4 3.5 1.8 1.5 0.4 2.9 of which public 2.9 3.1 6.4 4.0 3.0 5.3 of which foreign trade 27.3 6.9 0.0 14.9 19.9 8.5 Non-classified 5.1 3.4 0.3 3.3 5.5 0.5 w N Total 100.0 100.0 100.0 100.0 99.9 100.0 Source: BNR  33 8. LABOR MARKET. 8.1 Labor force and employment: Formal and informal sector. According to the most recent data, provided by the "Enqu~te Nationale sur l'Emploi" and published by the "Ministere du Plan et de la Cooperation" in 1992, the population at working age (i.e. between 14 and 74 years old) was estimated to reach 3,541,100 in 1989 or 52,5 % of the total population. Of this total 3,143,000 persons were actually employed, while 25,500 were registered as searching for work ("population disponible a la recherche d'un emploi"). More than 372,000 people were considered to be inactive. Almost 95% of the so-called active population of 3,168,500 was to be found in the rural areas. Table 8.1 shows that 94.5 % of total employment was in the informal and only 5,5 % in the formal sector. About 90% of the population of 14 years and older was active in agriculture. Employment in agriculture represented 1,8% of formal and 88,3 % of informal employment. The secondary sector (manufacturing, utilities and construction), with 86,000 persons, represented 2,7% of total employment, while manufacturing taken separately only provided 45,000 jobs or 1.4% of total employment. Less than 12,000 jobs, or 26% of manufacturing employment, were in the modem or formal sector. Manufacturing accounted for 6.8% of employment in the formal sector and 1.1 % in the informal sector. It has been estimated (Roy, 1992) that employees in the formal sector could be classified in the following legal categories. About 41,000 persons, or 23% of the total were to be found in the central administration, 15,000 (8%) in the state owned enterprises and about 7,000 (4%) in mixed companies. This leaves about two thirds or 110,000 persons in the private sectors. From these figures one can deduct that only about 3.5% of the labor force of more than 14 years old, was engaged in a broadly defined private formal sector. Almost half (46%) of the Rwandese, who were active in the formal sector, had followed only primary education, as compared with one quarter who had not received any formal education. 15.5% had a degree of vocational or technical education and 9% a degree of general secondary education. Some 3% held a higher education degree. For details of employment in the manufacturing subsectors one has to go back to data published by the "Direction Generale de la Politique Economique" of the Ministry of Finance for 1985 (table 8.2.). Total employment in manufacturing was put at 10,200 for the formal and at 116,600 for the informal sector. While the former figure is comparable Table 8.1 : Employment structure, 1989 Formal Informal Total 1000 %of %of 1000 %of %of 1000 %of total formal total informal total Agriculture 57.8 1.8 33.4 2774.7 88.3 93.4 2832.6 90.1 Mining 2.6 0.1 1.5 2.1 0.1 0.1 4.7 0.1 Manufacturing 11.8 0.4 6.8 33.3 1.1 1.1 45.1 1.4 Utilities 2.1 0.1 1.2 0.5 0.0 0.0 2.6 0.1 Construction 19.5 0.6 11.3 18.7 0.6 0.6 38.2 1.2 Trade, hotel, rest. 10.9 OJ 6.3 69.1 2.2 2.3 80.0 2.5 Transport and communication 2.9 0.1 1.7 4.5 0.1 0.2 7J 0.2 Banking, insurance 2.5 0.1 1.4 0.7 0.0 0.0 3.1 0.1 Other services 62.9 2.0 36.3 57.1 1.8 1.9 120.0 4.1 Non declared 8.8 0.3 0.3 8.8 OJ Total 173.1 5.5 100.0 2969.5 94.5 100.0 3142.5 100.0 Source: Ministere du Plan, Enqu~te Nationale sur l'Emploi 1989 (1992). W J:'-  35 to the formal employment as estimated for 1989, the figure for the informal sector is much higher and clearly not consistent with the estimate based on the 1989 survey. Data for the informal sector are therefore not to be regarded as very reliable. Nevertheless certain conclusions can be derived from table 8.2. Table 8.2 : Employment in the manufacturing sector. 1985 (number of employees) Formal Informal Total Food,beverages 1,857 69,354 71,211 Textiles, clothing 597 13,497 14,094 Woodworking 479 6,060 6,539 Metalworking 3,414 2,158 5,572 Other Manufacturing 3,867 25,552 29,419 Total manufacturing 10,214 116,621 126,835 Source: Ministere des Finances. The subsector food and beverages had the highest overall employment and represented 56% of total manufacturing employment. This high percentage was the result of its share of about 60% in the informal segment of the market. Textiles and clothing ranked second in terms of total employment with 11 %, again largely because of the influence of the informal sector. In terms of formal employment the metalworking sector was only slightly less important than "other manufacturing". It was also the only sector where formal employment was more important than informal. In woodworking and other manufacturing, the informal market segment again dominated the formal segment. 8.2 Unemployment and underemployment. The number of unemployed, as tallied in the national employment survey, amounted to only 25,470 people, 58% of whom were men and 41% women. The survey defined as unemployed those persons who were actively looking for work and who had not carried out any work for remuneration during the survey period. More than half of the unemployed (54%) had only received primary education and one fifth had received no education at all. Sixty percent of the unemployed was situated in the rural areas. Information on underemployment is limited to so-called visible and involuntary underemployment. Underemployment is calculated with reference to the number of normal working hours (35 in the survey) during a week. According to this definition, 1.970.500 persons or 62,7% of the labor force were underemployed.  36 One out of four underemployed persons worked less than 15 hours per week, while slightly more (27%) had a job for 15 to 25 hours per week and almost half (47%) worked between 25 and 35 hours. These proportions were very much affected by factors such as: the rural areas which harboured 93% of the underemployment; the level of education as 56% of the underemployed had no formal education or training and 40% had only received primary education; and the agricultural sector which accounted for 94% of the underemployment. This indicates that underemployment is very much a structural problem. 8.3 Wage level, labor unions and labor legislation. 8.3.1 Wage level. The number of wage earners outside of traditional agriculture was 274,500. 55% of them were employed in the private sector, 32% in the public sector and 10% in the so-called parapublic sector. More than four out of five wage earners were men. About 60% of the employees received a monthly wage that was lower than 5,000 FRw9. This proportion reached 69% in the rural areas. In the public sector 40% of the wage earners had a monthly wage of less than 5,000 FRw, while 36% earned between 5,000 and 15,000 FRw. In the parastatal sector the proportion which earned less than 5,000 FRw. amounted to 75% as compared with 16% in the category between 5,000 and 15,000 Frw. These latter proportions were very similar to those in the private sector where they were respectively 74% and 18%. In principle there is a structure of minimal salaries, divided in 5 categories and based on education and experience. A distinction is made between non skilled workers (with a wage range between 2,5000 and 5,000 FRw.), apprentices (between 6,750 and 9,750 FRw.), skilled or qualified workers (between 12.000 and 19.360), technicians (between 16,650 and 23,805 FRw.) and higher technicians (between 22,080 and 31,625 FRw.) These official categories are presented in table 8.3. The application of the minimum wage rate for beginning workers is not compulsory for the moment. Yet the categories themselves are applicable, except for the non qualified workers. It has been suggested that these categories should be eliminated, except for the non skilled workers meaning that negotiations should determine the wage rate (Roy, 1992). Others have proposed that the 9 79.98 FRw = 1 US $ in 1989.  37 minimum wage rates should be maintained but that the statutory automatic annual increases of 4% should be stopped. Table 8.3 Grid of minimal wages and salaries (1992) Categories Required Minimum wages Education Category A Non qualified workers Grade 1 : beginning non skilled person 2,500 Grade 2: specialised non skilled person none 3,750 Grade 3 : regular non skilled person 5,250 Category 2 Apprentices Grade 1 : beginning apprentice 6,750 Grade 2 : specialised apprentice none 8,260 Grade 3: confirmed specialised apprentice 9,750 Category 3 Skilled (or qualified) workers Grade 1 : normal workman lower 12,000 Grade 2 : qualified worker secondary 14,400 Grade 3 : specialized worker education 16,560 Grade 4 : confirmed specialized workers technical 19,320 Category 4 Technicians higher Grade 1 : beginning technician secondary 16,560 Grade 2 : specialized technician education 20,640 Grade 3 : specialised confirmed technician technical 23,805 Category 5 Higher technicians Grade 1 : beginning higher technician higher 22,080 Grade 2 : Specialized higher technician education 25,300 Grade 3 : confirmed specialized higher technician University 31,625 education Category 6 Middle Management Category 7 Top Management  38 8.3.2 Trade unions. When the "Code du Travail" was adopted in 1967, and even after changes to this code were made in 1977 and 1978, trade unions were more a scheme to transmit party directives to the workers than real representative organisations. Although the history of trade unions in Rwanda goes back more than fifty years (Roy, 1992), the establishment of a central trade union, i.e. the "Centrale des Syndicats des Travailleurs du Rwanda" (CESTRAR) is relatively recent. Its first conference was held in 1985 and the election of the first representatives took place in 1987. The real start occurred in 1989 with the establishment of the basic unions and the election of the members of the National Bureau. CESTRAR was separated from the unique political party in 1991. It has about 72,000 members and regroups 11 different branches. The other trade unions, of which only three have published their statutes in the "Journal Officiel", are smaller and do not list membership, e.g. UMURIMO (Association des Syndicats Chretiens) and STIR (Syndicat Interprofessionnel des Travailleurs du Rwanda), which existed before CESTRAR but was not really active. While in the past the demands of CESTRAR were mainly related to the wage level, they are now negotiating for a reduction of the number of working hours per week from 45 to 40. It may be mentioned that on the side of the employers, a more or less similar evolution occurred. The "Association des Employeurs du Rwanda", the traditional organisation, is considered to be too close to the former unique party. As the "Chambre du Commerce et de l'Industrie du Rwanda" (CCIR) was also considered to have close links with the unique party and to be mostly concerned with commercial issues, about twenty industrialists founded the "Association des Industriels du Rwanda" (AIR) at the time when the structural adjustment programme was launched. At the end of 1992 AIR had about 50 member companies, including most of the larger ones, and was drafting proposals on how to improve the situation of the industrial sector in Rwanda. 8.3.3 Labor legislation. The labor legislation was developed when the trade unions were mainly concerned with the application of collective conventions. The "Code du travail" regulated the working conditions in detail, rather than creating a legal framework within which the employers' and workers' representatives could come to negotiated settlements. A number of studies have stated that the labor legislation in Rwanda has become too rigid and inflexible (Kalinijabo and Nsengimana, 1992 and Roy, 1992).  39 The 'Code du Travail' stipulates that the maximum working week should not exceed 45 hours. The Minister of Employment interpreted this as a strict principle. The trade unions are requesting a reduction to 40 hours per week. Labor legislation in Rwanda is also rather strict with respect to the employment of women and children. With regard to women it has been stated that the laws are overprotective and that changes should be envisaged. Roy (1992) has proposed a special commission to study this issue. He also suggests a reduction of the number of official holidays from 15 to 8 and the elimination of the "conges extraordinaires". 8.3.4 National Social Action Program. The Structural Adjustment Programme, has been accompanied by a "Programme National d'Actions Sociales" (PNAS) to alleviate the negative effects of the Structural Adjustment Program for special vulnerable population groups and to selectively accelerate the growth programs. The PNAS targets the population groups which are likely to suffer intensely from the initial effects of SAP, e.g. persons whose health and financial situation is so bad that they can not help themselves without special aid. The deplaced persons who need reconversion, also come into this special category. With a total indicative funding of 100 million US$ for six years, the PNAS consists of several central parts, i.e. the program for security of basic food deliveries (PSA Programme de Securite Alimentaire), the Program of development of initiatives at the basis (PAIB) and programs with a high employment component (HIMO - Haute Intensite de Main-d'oeuvre). 9. PRICES AND INFLATION 9.1 Trends in price indices. As has been observed in paragraph 3, inflation was not a major problem in Rwanda before 1990. This is illustrated by the consumer price indices for Kigali shown in table 9.1. Over the period 1982-90 consumer prices in Kigali rose on average by 3.3% per year. Other consumer price indices show more or less the same evolution. The rise in food prices was higher than average, but still limited to an annual rate of 4.3%. The evolution of the producer prices of food crops, as shown in table 9.2, is not consistent with that of consumer prices. Similarly the evolution of the general index of producer prices is not consistent with that of the GDP deflator. According to the data shown in table 9.2 producer prices for manufactured goods remained more or less stable over the period 1982-89, but increased by 14% in the latter year. The devaluation of November 1990 resulted in considerable price increases, especially in 1991 when the consumer price index for Kigali went up by almost 20%. As could be expected, price increases were highest for imported goods (34%) and lowest for local goods (9.6%); consumer prices of food rose by 13.6%. In the same year producer prices of food crops and of manufactured goods rose respectively by 10% and 21 %. In 1992 the rise of consumer prices slowed down. Table 9.1 : Price index of consumption goods, Kigali (March-June 1980 = 100) 1986 1987 1988 1989 1990 1991 1992 Imported goods 119.2 122.4 128.0 129.4 135.9 182.5 205.3 Mixed goods 116.7 112.7 113.8 114.2 126.2 142.0 152.9 Local goods 113.3 120.7 123.7 125.1 126.5 138.6 147.9 All goods 115.2 119.9 123.6 124.7 130.0 155.5 170.2 Source: Banque Nationale du Rwanda, Rapport sur l'Evolution Economique et Monetaire du Rwanda 1990 ; Bulletin, No 18. Table 9.2 : Producer prices (1982 = 100) 1986 1987 1988 1989 1990 1991 Food crops 130.2 134.5 142.0 179.4 198.1 218.6 Export crops 100.9 102.5 102.5 102.7 85.6 90.0 Manufactured and traditional goods 109.0 107.4 99.5 114.0 130.0 157.4 General index 123.9 127.0 131.1 161.3 176.89 197.0 Source: see table 9.1  41 9.2 Price controls and deregulation. Before 1990 there existed a system of price controls based on the law of 5th July 1967. This law provided for three systems of price controls (Kalinijabo, 1988). (1) Fixed prices : this system was applicable to export crops (producer prices), beer and lemonade, fuel and some services like transport and insurance. (2) Minimum prices: applicable to food crops. (3) Maximum prices : applicable to imported goods and to locally manufactured goods. The system was based on maximum profit margins. Importers and domestic producers had to declare the cost price to the Minister who was competent for commerce. There was a ceiling on the profit margins that could be added to the cost price. The maximal margins were 15% for producers and wholesalers and 25 % for retailers. The system was criticised for a number of reasons. The administration did not have the means to control all prices; there existed parallel markets with higher prices. Inherent to the system were arguments over the cost price between the economic agents and the administration. The system of minimum prices for food crops was not workable : actual prices frequently were lower than the minima (Nzabandora and Karemera, 1991). The price control law of 1967 was abolished in July 1990. In November of that year the controls on profit margins were removed except for natural monopolies, i.e. utilities, for which prices are fIXed. At the same time the system of fIXed prices was also abolished, except for a limited number of goods and services, including retail prices of petroleum products and of pharmaceutical products and public health service consultation rates. Minimum producer prices were maintained for coffee, tea, wheat, paddy, cane sugar and pyrethrum. 10. CONCLUSION. Rwanda is a landlocked, densely populated, mainly agricultural country with a low rate of urbanization. Due to a population growth of 3.1 % per year land is becoming scarce. Growth of productivity in agriculture is hardly sufficient to cover the nutrition needs of the growing population. In the future agriculture will probably not be able to provide  42 employment for the growing labor force. After independence in 1962 there was an increase of investment in the modern manufacturing sector. But in the eighties the share of manufacturing in GDP stagnated at a level of 15%. Most manufacturing activity is in the informal sector. According to a survey in 1989 out of a total employment of 45,000 in manufacturing only 12,000 was in the formal and 33,000 in the informal sector. An earlier survey gave a much higher estimate of employment in informal manufacturing. Food and beverages was the most important subsector. The government traditionnally followed a prudent policy. Rwanda did not face a debt problem comparable to that of other African countries. After 1973 the government had a tendency to expand its role in the form of public or mixed ownership of enterprises. It also intervened in other areas, e.g. in the credit sector and in price setting. But the foreign trade regime was quite liberal. In the eighties the economy came under increasing strains, resulting from the fall of coffee prices and from an increase of government expenditure, a belated response to the coffee boom of 1977-79. Moreover in the middle of the eighties tin production was halted, resulting in an increasing concentration of exports on coffee and tea. The government reacted by restricting imports. Import licences were not any more granted quasi- automatically, as used to be the case in the seventies. The coffee boom of 1986 provided temporary relief. But after this boom the renewed fall of coffee prices intensified the foreign exchange and fiscal problems. In 1990 the government worked out a structural adjustment program with the IMF and the World Bank. The Rwandese franc was devalued. The program provided among others for a more liberal regime in the fields of foreign trade, credit and prices. The invasion by rebels in the north of the country, which started at the end of 1990, and internal political strife had a negative impact on economic activity and interfered with the realisation of the structural adjustment program. The situation became especially critical after renewed fighting at the beginning of 1993, which caused new disruptions and resulted in a stream of refugees.  43 BIBLIOGRAPHY Association des Industriels du Rwanda, Memorandum sur les principaux probH~mes eprouves par les industriels, Kigali, janvier 1992, 17p. BARSACQ, Jean Louis et KANEYE, Hassane, Un cadre incitatif pour les micro-entreprises rurales non agricoles au Rwanda, PNUD-BIT, Kigali, fevrier 1992, 39 p. BEZY, Fernand, Rwanda 1962-1989, Bilan Socio-Economique d'un Regime, Institut d'Etudes du Developpement, Louvain-la-Neuve, 1990. BITjPNUDjCNFPP. Donnees de base en rapport avec la formation professionnelle au Rwanda, Kigali, mai 1992, 19p. KALINIJABO Charles et NSENGlMANA Marc, Etude sur Ie Marche de Travail et la Legislation du Travail au Rwanda, Rapport Intermediaire, Banque Mondiale, Kigali, mai 1992, 62p. (confidentiel). KALINIJABO, Charles, Obstacles au developpement du Secteur prive au Rwanda, Ministere des Finances et de l'Economie, Groupe Consultatif sur Ie developpement du secteur prive au Rwanda, septembre 1988, Kigali, 170p. NGlRABA1WARE Augustin, MUREMBYA Leonidas et MEAD Donald, Les moyennes et grandes entreprises rwandaises du secteur manufacturier prive: Etude diagnostique et impact des politiques economiques, Document de Travail No 9, Programme de Soutien it. la reforme de la politique industrielle et de l'emploi, Ministere des Finances et de l'Economie, Kigali, aoftt 1988, 177p. NGUYEN-HUU Khiem, Analyse de l'influence du Programme d'Ajustement Structure1 sur le developpement du Secteur manufacturier au Rwanda, P.A.S., Minitere du Plan, Kigali, avril 1991, 41 p. NTAGARA, Aloys, Entreprises Rwandaise, Trente ans apres, Kigali 1992, 60p. Nzabandora, Andre et Karemera Eduard, Rapport du Consultant charge de conseiller Ie gouvernement sur la lancinante question de procedures, formalites et tracasseries administratives, Kigali, janvier 1991, 106 p. Republique Rwandaise, ONUDI, INTERG, Elaboration d'un programme de developpement industrie1 et artisanal de reference. Analyse diagnostic des activites industrielles et artisanales existantes, Rapport de liere phase, volume 1, septembre 1990, volume 2, octobre 1990. Identification des Activites et Idees de Projet, Rapport de 2ieme phase, janvier 1992, 197 p. Republique Rwandaise. Banque Nationale du rwanda, Rapport sur l'Evaluation economique et monetaire du Rwanda 1990, Kigali 1992, 169 p. Repuhlique Rwandaise. Document Relatif a la preparation de 1a Table Ronde, No.8, Strategie Sectorielle de l'Industrie Artisanat et Mines, SFC (Sedes-Cegos), Boulogne-Billancourt, 23 p. + annexes.  44 Republique Rwandaise. Forum de concertation sur les problemes souleves pour les operateurs des secteurs de l'industrie et du transport routier international a la suite de la mise en oeuvre des mesures d'ajustement structurel, Rapport de synthese, Gisenyi 16-18 3vril1992, Kigali 1992, 30p. Republique Rwandaise. GASORE, Fran<;:ois e.a. Etude des Differentes Interventions d'Assistance technique pour la Promotion des PMI et Artisanat au Rwanda. Ministere de l'Industrie et de I'Artisanat, Kigali, juin 1990,223 p. Republique Rwandaise. Ministere de l'Industrie et de l'Artisanat, Etude Strategique Sectorielle de l'Agro Industrie, Volume 1, Synthese des travaux de la premiere phase de l'Etude Diagnostic de l'existant. Kigali, decembre 1990,459 p. Republique Rwandaise - ONUDI. Etude strategique de l'Agro Industrie Diagnostic approfonie des unites industrielles, Rapport de 2ieme phase, volume 2, fevrier 1991. Republique Rwandaise. Ministere de l'Industrie et de l'Artisanat, Diagnostic de la situation du secteur industriel, minier et artisanal, Rapport No 1 de la Cellule de Planification, Kigali, fevrier 1990, 150p. Republique Rwandaise. Ministere des Finances et de l'Economie, Direction Generale de la Politique economique, Recueil des textes iegisiatifs et reglementaires, relatifs a la loi de 5 aoO:t 1987 portant code des investissements, Kigali 1987, 134 p. Republique Rwandaise. Ministere du Plan et de la Cooperation, Direction Generale de la Statistique, Enquete Nationale sur l'Emploi, Volume 2, Resultats Synthetiques, janvier 1991, Kigali. Republique Rwandaise. Ministere du Plan. Analyse de la Situation Economique et Mesures de Redressement, Document du Comite Central, Kigali, decembre 1989. Republique Rwandaise. Ministere du Plan. Bulletin de Statistique, Direction Generale de 1a Statistique, Supplement AnnueL no. 17, janvier 1990. Republique Rwandaise. Ministere du Plan. Evaluation de la mise en oeuvre du P.A.S. (novembre 1990 - juin 1991), Kigali,juillet 1991. Republique Rwandaise. Ministere du Plan. Groupe Technique du Travail, Etude sur l'Integration Regionale en Afrique Orientale et Australe, volume 2. Rapport analytique, Kigali, 71 p. Republique Rwandaise. Ministere du Plan. La politique socio-economique du Rwanda 1992-1994. Conference de Table Ronde des Partenaires au Developpement Economique et Social de la Republique Rwandaise. Geneve, 2-3 juillet 1992, Kigali. Republique Rwandaise. Ministere du Plan. Programme de Soutien a la reforme de la politique industrielle et de I'empioi, Rapport d'activite, 1986-1989, Kigali, mars 1990, 95p. Republique Rwandaise. PNUD, Programme d'appui au developpement des micro-entreprises rurales non agricoles et aux initiatives de base (PAIB), Document operationnel, avril 1992, 39 p. annexes.  45 Republique Rwandaise. Programme d'Ajustement Structurel, Document Cadre de Politique Economique et Financiere it Moyen Terme (octobre 1990, septembre 1993), Kigali, septembre 1990. Republique Rwandaise. Ministere du Plan, Comptes Economiques Nationaux du Rwanda 1989, Direction Generale de la Statistique, Direction de la Comptabilite Nationale, no.13, janvier 1992, Kigali. ROY Paul-Martel, La legislation du travail et Ie marche de travail au Rwanda, Banque Mondiale, Montreal, juin 1992, 99p. (Confidentieel - dient vrijgegeven door Min. du Travail). SPRAUL, Marianne, Impact de l'ajustement par la devaluation sur les entreprises industrielles privees rwandaises, Universite de Clermont Ferrand, mai 1992, SIp. annexes.(FK). Technoservice Rwanda, Comment Demarrer une entreprise au Rwanda, Guide pour l'Entrepreneur, Kigali 1991, Printer Set, 122 p.  46 ANNEX: PERSONS CONTACTED 1. National administration and organisations AOUNI, Abdelkadir, Directeur du Departement d'Economie, Universite Nationale du Rwanda. FRANCOIS, Andre, Banque Rwandaise de Developpement. GASORE RUKARA, Pierre, Coordinateur Programme PRIME-MINIPLAN. HABINEZA, Jean de Dieu, Secretaire permanent, Association des Industriels du Rwanda (A.LR.). HAKIZIMANA, Deogratias, Division de Promotion des Projets Industriels, MICOMART. KALIMUNDA, Joseph, Directeur Division Emploi et Statistique du Travail. KIGUFI, Thomas, Secretaire General, Chambre de Commerce et d'Industrie du Rwanda (CCIR). KUBWIMANA, Bernardin, Chef de Service Production SO RAS. MAKUZA, Cyprien, Directeur du Departement de Management, Universite Nationale du Rwanda. NAHIMANA, Alphonse, Chef de Service, Statistiques de Credit, Departement de la Banque Nationale du Rwanda. NSABlMANA, Dismas, Directeur General de Ia Planification, Ministere du Plan et de la Cooperation. NTONDO, Apollinaire, Adjoint du Secretaire General CCIR. NYALlHAMA, Jean Marie Vianney, Directeur general de l'Industrie, MICOMART. NYAMWIGENDAHO, Charles, Directeur de l'Assistance aux Entreprises Industrielles, MICOMART. NYARWA YA, Jean Baptiste, Direction de la Statistique. NZIRORERA, Innocent, Affaires Sociales, A.LR. RUClRA, Gerard, Directeur des Etudes et Statistiques, Banque Nationale du Rwanda. TWAGlRAYEZU, Alexis, Directeur de la Statistique, Miniplan. UWIMANA, Antoine, Direction de la Politique Economique MINIPLAN. 2. Multilateral and bilateral donors BISA-SAMALI, Octavien, Economiste, PNUD. BOCKSTAL, Christine, Experte-Coordinatrice, PAlB, Bureau Internationale du Travail. DEBAR Johan, Economiste, Ambassade de Belgique. DIJCKERMAN, Dirk, Programme Officer, US-AID. DURLET, Claude, Directeur Section Economique, Ambassade de Belgique. FRANZETII, Christophe, Charge de Programme PNUD j ONUDI. JORGENSON, Bernadette, US-AID. RAJONSON, Etienne, Conseiller Technique Principal PNUDjBIT.