Tunisia - Economic development (Vol. 2 of 6) : Agriculture

c -f FILE. CO Report No. 274a-TUN T The Economic Development REPORT DESK of Tunisia WITHlN ONE WEEK Volume 1: Annex-Agriculture December 27, 1974 EMENA Region Country Programs Department II Not for Public Use Document of the International Bank for Reconstruction and Development International Development Association This report was prepared for official use only by the Bank Group. It may not be published, quoted or cited without Bank Group authorization. The Bank Group does not accept responsibility for the accuracy or completeness of the report.  CURRENCY EQUIVALENTS Currency Unit = Dinar = 1000 millimes With effect from 1955 US $1.00 = 0.42 Dinar Dinar 1.00 = US $2.381 With effect from September 28, 1964 US $1.00 - 0.52 Dinar Dinar 1.00 - US $1.90 With effect from December 20, 1971 US $1.00 - 0.48 Dinar Dinar 1.00 = US $2.08 With effect from February 1973 US $1.00 = 0.44 Dinar Dinar 1.00 = US $2.27 UNITS AND WEIGHTS AND MEASURES: METRIC British/U.S. Equivalents 1 m = 3.28 ft. 1 m ton - 0.981 g. ton 1.1 US sh. ton 1 m2 = 10.76 sq.ft. 1 kg 2.2 lb. 1 km = 0.62 mi. 1 litre = 0.22 ga. 1 km2 = 0.386 sq.mi. = 0.26 US liq. gallon 1 hectare = 2.5 acres 1 m3 = 1.31 cubic yards FISCAL E:AR January 1 - December 31  THE ECONOMIC DEVELOPMENT OF TUNISIA TABLE OF CONTENTS ANNEX I AGRICULTURE Page No. 10. THE ROLE AND PERFORMANCE OF AGRICULTURE IN THE 1960'S A. The Place of Agriculture ............................1 B. Recent Production Performance ..... ..................5 C. The Factors of Production ........................... 8 D. Imports and Exports ................................. 11 E. Agricultural Prices and Incomes ..................... 12 11. THE POTENTIAL FOR GROWTH A. Overall Demand Prospects ............................ 17 B. Overall Production Prospects ........................ 21 C. Some Individual Commodities ........... .. ............ 27 (i) Wheat and Barley ............................... 27 (ii) Olive Oil ...................................... 38 (iii) Livestock ...................................... 47 (iv) Wine ........................................... 50 (v) Other Commodities .............................. 52 (vi) Vegetables ..................................... 52 (vii) Citrus ......................................... 53 (viii) Fish ........................................... 54 D. Labor and Capital Implications ...................... 55 12. THE ROLE OF GOVERNMENT IN AGRICULTURE A. Government Institutions ............................. 60 B. Public Investment Policies . . ........ ................. 64 (i) General ........................................ 64 (ii) Irrigation ..................................... 68 C. Land Tenure ......................................... 75 (i) State-Owned Lands .............................. 76 (ii) Privately Farmed Lands ......................... 79 (iii) The Collectively-Owned Lands .... ............... Al D. Price and Mlarketing Policies ........................ 82 E. Policies for Private Investment and Credit ........ .. 85 (i) Short-Term Credit in Kind Campaign .... ......... 89 (ii) A Proposed Integrated Credit Structure .... ..... 90  TABLE OF CONTENTS (CONT'T) Page No. F. Agricultural Education, Research and Extension ..... 94 (i) Education .......... ........................... 94 (ii) Research .......... ............................ 96 (iii) Agricultural Extension ..... .................. 97 ANNEX I LIST OF MAPS Map No. 10518 Rainfall and Irrigated Areas . 10519 Soil Suitability for Crops . 10004 Location of Main Crops, I . 10005 Location of Main Crops, II . LIST OF GRAPUS Graph No. Page No. 10.1 Agricultural Output, In Value - Cereals .... ........ 15 10.2 Agricultural Output, In Value - Other Products 16 11.1 Tunisia - Areas of Bread Wheat and Durum in Hectares, 1949-1969 .31 11.2 Average Yields in Quintals Per Hectare of Bread Wheat and Durum, 1949-1969 .32 11.3 Production of Bread Wheat and Durum in 1000 Quintals, 1949-1969 ........ ............................. 33 11.4 Olive Oil Production Since 1949/1950 .... ........... 43 12.1 Organization of the Ministry of Agriculture ........ 100  10. THE ROLE AND PERFORUANCE OF AGRICULTURE IN THE 1960'S A. The'Place of Agriculture 10.1 Tunisia is less predominantly an agricultural country than some of its neighbors. In terms of GDP, employment, exports and investments, the farm sector is less important than in Morocco or Turkey, while in some aspects, it ranks below Algeria (share of employment) or Spain (share of exports). Table 10.1 shows this in some detail. The relatively minor importance of the agricultural sector is not a bad sign in itself. On the contrary, an inverse correlation can be observed between a country's per capita income and its dependence on agriculture. In the case of Tunisia however, with a per capita income roughly comparable with that of the other Maghreb countries, it is more a reflection of smaller natural endowments for agriculture and, possibly, less exploitation of these resources. 10.2 Climatic conditions are such that a wide variety of rainfed crops and pastures is only possible in the northern quarter of the country, where average rainfall is over 400 mm; but even there, fluctuations around the average are extremely large, and make agriculture a hazardous enterprise, with a succession of dry years (1967 till 1969) followed by years in which crops and soils are washed away by prolonged rains (1969 and 1973). In ad- dition, part of the north is mountainous, and soils in the plains are not uniformly good. A further quarter of the country has a rainfall of 200- 400 mm which supports a rainfed agriculture combining low-density raising of sheep and goats, tree crops (predominantly olives, also almonds), and cereals under marginal conditions. Finally, in the southern half of the country, where average rainfall is below 200 mm, several oases permit the cultivation of vegetables and date palms. Outside these oases centers of activity, some extensive grazing and olive production occur in the 150-200 mm zone. Chapter 11 discusses the resource base and the further possibilities in each of the regions mentioned. 10.3 As in most Mediterranean countries, the Tunisian farmer has striven to offset the deficiency and unreliability of rainfall by the careful manage- ment of water. There is a long tradition of irrigation through private wells and, since 1950, of dam and reservoir systems. Hiere again, Tunisia has not been very lucky. Some of its water resources have a high salt content, in particular some catchment areas in the north and most of the recently explored groundwater reserves in the south; much of the irrigable land is in areas that have already a relatively high rainfall; some soils in the planned expansion of the Mejerda are of mediocre quality; low average winter temperatures and strong winds put Tunisia at a disadvantage in comparison with countries like Morocco in the highly remunerative export of irrigated off-season vegetables. These reasons alone reduce the effectiveness of most irrigation works; in addition, because of other constraints (such as land tenure problems and in- adequate extension services) the considerable investments in irrigation works during the last 10-15 years have not yet had a significant impact on output. At present the total irrigated acreage is sonme 80,000 ha, out of a total cul- tivated acreage of 4.2 million ha. The total area of Tunisia is 16.4 million ha. The problems of the irrigated areas are discussed in Chapter-12.  TABLE 10.1: AGRICULTURE'S 2 RDLE IN THE EOONOMY OF TUNISIA AND SELECTED MEDITERRANEAN COUNTRIES Tunisia Algeria Spain Morocco Turkey Average Average Average Average Average 1962-64 1970-72 1967-69 1967-69 1969-70 1969 Approximate Value of Agriculture Output (US$ Million) 360 550 6,000 820/e 2,300 Percentage of Agriculture in GDA 21 18 13 16 28 31 Agriculture Labor Force as Percentage of Total Labor Force n.a. 50 61/e 31 70 70 Agriculture Exports as Percentage of Total Exports 59 34 18 40 58 75 Food Imports as Percentage of Total Imports 18/d 23 18 17 23 5 Balance o,f Agriculture Trade, /c (US$ Million) +1O/d -16 -2 -210 +14o +365 Gross Fixed Asset Formation in Agri- culture as Percentage of Total 21 13 16/f 10 25 13 /a Including Forestry and Fishing. / Value Added at Factor Cost, in 1966 Prices / Agrialture Exports minus Food Imports. Data do not permit to show Agriculture Exports minus Imports of all Agriculture Raw Materials and Agriculture Inputs. /d Average 1965-67 /p 1969 only. _ Public Investment only; but Private Investment was recorded minimal. Sources: For Tunisia - Derived from various publications of the Ministry of Planning. For Algeria - IBRD Economic Mission Report 1971. For Morocco - IBRD Economic Mission Report 1971. For Spain - IBRD Economic Mission Report 1971. For Turkey - IBRD Economic Mission Report 1971.  10.4 During the 1960's the share of agriculture in GDP declined (Table 10.1). After stagnating from 1962 to 1970 1/, production made a jump in 1971 and again in 1972; despite damaging floods in April 1973, the 1973 crop was again above the trend. In the discussion of the performance of agricuLture in the 1960s, reference is made to analysis of trends rather than to a com- parison of output levels from year to year. Indeed, fluctuations around the trend are large, in particular because of excessive variations in yields (as much as 300 percent for grains and 1,000 percent for olive production) due, directly in the case of cereals, to variations in rainfall. The prime victim of these output fluctuations is obviously the farmer himself, as they are likely to affect his real income to the extent that opposite changes in farm- gate prices do not occur, or are not relevant. Cases in point are: (a) Farmers producing for subsistence rather than for the market. (b) Producers of crops of which the price is not determined by domestic demand and supply, but set by the authorities (wheat) or dependent on export markets (olive oil). (c) For a livestock farmer, a year of low rainfall means no weight increase of his animals, and a reduction of his production po- tential of coming years because of a reduction in the herd (high mortality, forced selling at rock bottom prices, as in 1967). (d) Modern farmers use more purchased inputs (fuel, fertilizer, pesticides) than traditional farmers 2/. Because the amount spent on these tends to be stable in the short run, the fluc- tuations in net value of output (their income) are larger than those in gross value of output (their production). 10.6 The effects for the rest of the economy are also serious. Since half of the population lives on the land and contributes less than one fifth to the national product, the average per capita agricultural income is far below that in the rest of the economy (about US$80 versus about US$400). Al- though the average farmer thus spends little on the products of the other sectors, a 25 percent change in agricultural GDP may mean for the other sec- tors a change of at least 5 percent in demand for their products, without counting the indirect effects. The effects of these sudden changes in buying power tend to be concentrated in certain regions, such as Sfax, where olive production is predominant, and may hit particular sectors, such as transporta- tion, suppliers of agricultural equipment or producers of current consumer goods. 1/ 1962 is chosen, rather than a prior date, because (a) statistics are reportedly more accurate as from that date, (b) prior to 1962 a large part of agricultural production was produced by foreigners, and (c) the period since 1962 corresponds with the one studied by the Tunisian authorities, in the "Retrospectives 1962-1972." 2/ On the average, purchased inputs as a proportion of gross value of output does probably not exceed 20%, but for the large cereal producers it may be as much as 40-50%.  - 4 - Table 10.2: ANNUAL CHANGES IN AGRICULTURE OUTPUT (1962-1972) IN CONSTANT 1966 PRICES 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 - +39.4 -1h.0 +14.4 -10.8 -10.5 +16.3 - 9.1 + 6.2 +24.7 +31.6 Source: Derived from Ministry of Planning figures, as revised in April 1974.  - 5- B. Recent Production Performance 10.6 On the basis of the latest revised figures 1/ average real growth in the value of aggregate output for the period 1962-1970 was close to 1.5 percent (using the least square method). Just how much the year-to-year changes have fluctuated around this average is shown in Table 10.2.. This table also shows how much the trend can change by adding or subtracting a year or two at the beginning or at the end of the period considered, even when applying the least squares method. 10.7 In 1962 the planning authorities published 1971 production targets for agriculture. At the time, these appeared plausible. It is now clear that they were too optimistic. The following table compares some production targets and actual results in 1971, a year of better than average weather. Table 10.3: TARGETS FOR AGRICULTURE IN THE 1960's AND ACTUAL PERFORMANCE Target Actual 1971 1971 Cereal Production ('000 tons) 1,116 8.50 - of which Durum Wheat ('000 tons) 665 460 Pulses Production ('000 tons) 134 130 Vegetables Production ('000 tons) 777 714 Mutton Production ('000 tons, live weight) 35 40 Beef Production ('000 tons, live weight) 33 28 Sheep Herd ('000 female units) 2,320 1,850 Target Actual 1961-71 1961-71 Olive Trees Planted ('000 ha) 335 300 Other Fruit Trees Planted ('000 ha) 105 80 Sources: Targets - "Perspectives Decennales de Developpement 1962-1971," Secretariat d'Etat au Plan et aux Finances. Actual - Ministere du Plan. The discrepancy between targets and actual performance is even more st,riking if expressed in planned and actual increases, rather than absolute levels. For example, the production of Durum wheat was planned to increase from 363,000 t. (average 1955-59) to 665,000 t (in 1971); the actual increase was only 100,000 t. It would be wrong to attach too much significance to the absolute level of output or growth rates for several reasons. The agricultural sector has an important subsistence element, the production of which is diffi- cult to ascertain, although considerable progress has been made in estimating 1/ Statistical Appendix, Table 2.7.  - 6 - cereal harvests, which has subsequently invalidated some previous data. The arbitrary method of estirmating livestock production by estimating slaught- erings but ignoring herd changes imposes further caution. 10.8 The principal reasons why output has stagnated are the following: (a) Climatic conditions. During the 1960's spring and fall rains were on average 30 to 40 percent less than during the 1950's. While the effect of rainfall is most obvious for cereals and olives, it is probably even more significant for livestock and for the groundwater level, which is important for well irrigation (Chapter 11, Cereals and Livestock). (b) The departure of foreign farm-operators. The departure of tech- nical and management expertise resulted in temporary disorganiza- tion of, in particular, grain and wine production. (c) The effort to bring farms into cooperatives. The growth and termination of the cooperative movement are discussed in Chapter 2. Its disruptive effects were mainly felt in the livestock sector. (d) Lower than expected returns from investments. Since independ- ence, Tunisia has made heavy public investments in agriculture, in particular in reforestation and erosion control, and in irri- gation. By their nature, the former have not yet had a signif- icant impact on output; other constraints, such as tenure prob- lems and inadequate extension services, have kept down irrigated production. Olive trees planted at the beginning of the dec- ade will only come into full production in the coming years. 10.9 The above factors have not in the same way and to the same degree affected the output of individual commodities. The relative importance of cereals and animal products has changed somewhat, more as a result of these factors than because of changes in consumption patterns. Table 10.4: COMPOSITION OF AGRICULTURAL PRODUCTION (in % of gross output, valued at 1966 prices) Average Average 1962-63 1969-70 Cereals 28 20 Olives 7 7 Other tree crops 18 16 Vegetables 14 17 Meat and eggs 22 29 Other 11 11 Total 100 100  - 7 - The graphs at the end of this Chapter show the actual behavior of the output of each commodity group. Some salient features are briefly indicated in the following paragraphs. Chapter 11 discusses the potential for further growth of each major commodity. 10.10 The acreage of cereals has not changed much, and the recent gains in output are due to better yields. These in turn are mainly due to favor- able weather, and to a less extent to the introduction, since 1968, of HYV. A study for the period 1945-1969 shows a high correlation between timely and sufficient rainfall and grain yields. This explains why the trend in average grain output (but probably also the performance of livestock and tree crop subsectors) was disappointing. The extensive methods of cultivation of the large farms, which produce the bulk of output, are the basic reason for the low average yield. Moreover, much of the land under crops would be better for livestock production. Low-density farming of good land should be penalized by, for example, a tax on production potential. This could raise average yields considerably and discourage owners from leaving land fallow. If it were not for the protected prices, 1/ the acreage under cereals would be much smaller. This would also be the case if the small farmers could be assured of a regular supply of grains at low prices. Most of them prefer not to run risks and to grow cereals for their own needs under marginal conditions. 10.11 The acreage under fruit trees increased by some 30 percent during the 1960's. Olives remain predominant both in terms of acreage and value; they also remain the most important export crop. Fluctuations in output are very big. The largest crop in history occurred in 1971/72, while the11972/73 crop was also above average. The acreage under almond trees is increasing fast, the market outlook is quite favorable and net value added per hectare of almond trees is on average double or triple that of olive trees. Citrus is the second most important tree crop. Production in Cap Bon has gone down in recent years because of a fall in available irrigation water following a fall in the groundwater table. It is not certain yet whether this fall is due to the lower rainfall of the 1960's and therefore temporary, or whether it is a result of over-exploitation of groundwater resources. Date produc- tion, which is for agronomic reasons concentrated into relatively small areas on either side of the Chott, has remained stable. New plantations have been started, but are not yet in production. The market for the Tunisian date ex- ports is very favorable. Production is, however, very sensitive to uncontrol- lable factors; rain in September can destroy the entire crop. Wine production has also decreased sharply over the decade, because of a reduction of|outlets and lack of know-how. Lately the short- and long-term export prospects have improved considerably, and production is being resumed. 10.12 The acreage under vegetables has increased considerably in the last few years, but for most of the 1960's it remained virtually unchanged! despite the considerable effort made in irrigation. The main vegetables are peppers, potatoes, tomatoes, and watermelons. Only a small share is exported as early vegetables. Although Tunisia has only limited possibilities in the exlport 1/ Until the recent price increase, following the massive grain import by the USSR, Tunisian prices were consistently above world market prices.  - 8 - market for early vegetables, it appears that inadequate use is being made of these limited possibilities. Recent trials with strawberries, asparagus and flowers seem encouraging. The main problem seems to be the absence of an effective market organization that would emphasize quality and centralize marketing efforts abroad, and the lack of adequate extension services. 10.13 The performance of the livestockc subsector during the last 10 years is not quite clear, but seems disappointing. During the years of enforced cooperativization which coincided with a series of droughts, slaughterings increased considerably, as did uncontrolled exports of herds to Algeria and Libya. It is reported that the sheep herd declined from 1.6 million in 1962 to 1.2 million in 1968, but recovered to 1.85 million in 1972. The same thing happened with cattle, althoughi recovery has not been as quick. The increased demand for livestock products has sharply increased prices, which are now two or three tir,mes as highi as in 1969-70. Two factors seem to be responsible: (a) prices in 1969 were abnormally low because of an increase in the slaughtering rate; (b) demand has increased rapidly because of the per capita income gains of the urban population and the growth of tourism. C. The Factors of Production 10.14 The degradation of cultivable land, due to heavy rains, occasional sandstorms, and the disappearance of the vegetation cover in a number of places, has been slightly slowed down by reforestation and erosion control. Actually, althouglh the total acreage of agricultural land has remained un- changed, its production potential has increased considerably because of a doubling of the irrigable area which permits high value crops and double cropping, and important plantings of olive and other fruit trees (over 400,000 ha). Because of constraints discussed later, the investment in irrigation and the newly planted trees (olive trees start producing significantly only after 10 to 15 years, and almond trees and citrus after 5 to 10 years) have not so far had much impact on output. During the 1962-71 period, some D 190 million, representing two-thirds of the total investment in agriculture, was spent to increase or maintain soil productivity. With heavy migration from the rural areas, the agricultural, labor force has probably fallen over the decade, but what is not clear is the trend in under-employment and unemploy- ment. A recent shortage of farm laborers due to exceptionally high production levels has increased their bargaining position, and led to an increase in wages above the minimum level of 0.6 Dinar per day, improving the relative income position of a large part of the agricultural population. 10.15 The capital base increased considerably during the 1960's, far more through public than private investment initiative. Table 10.5 summarizes the investment effort and shows that irrigation has absorbed some 40 percent of public investment (Government Budget and Foreign Aid) and forestation and erosion control 35 percent. Bank credit has mainly taken the form of lending to the cooperatives; only D 12 million of it was private borrowing. Over half of self-financed investment was in machinery and a further quarter in tree  -9- Table 10.5: GRDSS INVESTMENT IN AGRICULTURE, 1962-1971 (in millions of Dinars) DESTINATIDN Governmnit Foreignav Bank Self Budget Aid Credit Financing Total Forestation and Erosion 52 16 - - 69 Control Irrigatibn 63 16 1 1 81 livestock and Fodder 3 3 2 3 11 Production Machinery and Equipment 2 2 8 31 43 Planting Olive and Other 15 3 8 16 42 Fruit Trees Research and Extension 12 10 - - 23 Fisheries 1 - - 4 5 Other - - 3 1 4 Total 1h8 51 23 55 278 a/ Outside Government Budget. Source: "Les Investissements dans l'Agriculture," Ministry of Agriculture, March 1972.  - 10 - planting. The figures given in the table are gross and therefore do not re- flect depreciation; net investments must have been considerably smaller. For example, in 1961 there were 2,700 tractors less than 6 years old, for which replacement needs can be estimated at 400 units a year; imports during 1962-71 averaged 900 units per year. Investment or disinvestment due to changes in herd size is not included in the statistics. 10.16 On average, the actual value of investment represented 80 percent of the initial objective. In physical terms, the shortfall was greater. Moreover, the actual distribution of investment, by headings, was not the same as the planned distribution (see Table 10.6). Table 10.6: PLANNED AND ACTUAL INVESTMENTS IN AGRICULTURE, 1962-71 (in millions of Dinars) Planned Actual Actual as % 1962-71 1962-71 of Planned Forestation, Erosion Control 55 69 126% Irrigation 113 81 71% Livestock & Fodder Production 48 11 22% Machinery and Equipment 63 43 69% Planting Olive and Other Fruit Trees 32 42 130% Research and Extension 28 23 83% Fisheries 5 5 108% Other 7 4 65% Total 351 278 80% Source: "Les Investissements dans l'Agriculture", Ministry of Agriculture, March 1972. 10.17 The use of current inputs hardly increased during the 1960's. Ac- cording to the Ministry of Plan, the total cost of inputs (in constant 1966 prices) fluctuated between D 17 and D 19 million during 1962-70, i.e. around 15 percent of the value of total farm output. Until 1970 expenditure on seeds and fuel remained virtually constant, while there was no clear upward trend in the use of fertilizer or of feed concentrates. Since then, input use seems to have intensified. 1/ Fertilizer use increased sharply in 1971 and 1972, and is now at a level comparable withi that in Morocco and Algeria. Comparing figures with Spain, a country with roughly similar natural endowments, would indicate that possibilities of increased use are very good. 1/ Statistical Appendix, Table 7.6.  - 11 - Table 10.7: FERTILIZER USE Iii 1970/71 N P 0 K 0 Total Arable land /a 2 5 2 (------in thousand tons------- (million ha) kg/ha Tunisia 15.5 28.0 5.0 48.5 4.5 11 Morocco 38.0 44.5 15.5 88.0 7.9 11 Algeria 40.0 52.6 26.0 118.6 6.8 17 Spain 540.1 428.6 206.0 1174.7 20.6 57 /a Including land under permanent crops. Source: Compiled from FAO Production Yearbook 1971. D. Imports and Exports 10.18 While agricultural exports represent a large share of total exports (34 percent, average 1964-71), they are only a small proportion of total farm output, possibly of the order of 15-20 percent. 1/ For some individuatl com- modities the proportion is much higher--some 60 percent of wine production, 60 percent of olive oil, nearly 40 percent of citrus, 2/ and virtually all esparto grass after transformation. The proportion of the output of most other agricultural commodities, including most vegetables, exported is; on the average less than 10 percent. The major part of food consumption is locally produced. The share of imported food in total consumption is current- ly of the order of 20-25 percent. Some of the imports cannot be prodtLced under Tunisian conditions in any case (coffee, tea, spices, etc.), but: for certain other products, net imports are substantial 3/ -- wheat 45 percent, sugar over 90 percent, meat 15 percent, dairy products 50-60 percent, edible oil 5-10 percent, 4/ tobacco 50 percent. 10.19 A major consequence of the very moderate growth and fluctuations in output has been the reversal of the agricultural trade balance in recent years. Until 1967, exports exceeded imports by some D 8-15 million annually.1 Since then, net imports of the same order of magnitude have been common. The ex- ceptional olive crop of 1971 hardly brought exports up to imports. 1/ A comparison between the value of exports and the value of grossloutput shows a proportien of 30-35%. But output is valued at farm gate prices and exports at FOB prices, so that comparison is not valid. 2/ Averages over the period 1966-71. 3/ Estimates based on 1969-71 production and trade figures. 4/ Based on the following average 1969-71 figures: olive oil production 65,000 tons, of which exported 40,000 tons; imports of other edible oils 45,000 tons.  - 12 - 10.20 The value of exports in current prices remained roughly the same during the 1960's. In real terms it declined somewhat. The composition has substantially changed, however: cereals have all but vanished and wine dti minished considerably; on the other hand exports of fruits, fish and animal products have increased; despite the extreme fluctuations in output, olive oil continued to be by far the main export. Imports increased sharply--from D 20-25 million in the early 1960's to around D 40 million at the end of the decade. Cereals and edible oil accounted for most of this rise, followed more recently by dairy products, meat and sugar. Details of exports and -im- ports of agricultural commodities are given in Statistical Annex Tables 7.4 and 7.5 and changes in composition of agricultural trade, in the table below. Table 10.8: COMPOSITION OF AGRICULTURAL TRADE Exports Imports Average Average Average Average 1960-62 1969-71 1960-62 1969-71 m0Do % m.D. % m.D. % m.Do % Livestock products 0.9 3 2,1 6 1.6 7 5.6 13 Fish 0.3 1 0.9 3 - - 001