Rehabilitation of the pastoral sector in the absence of outside intervention.
The reconstitution of the Sahelian livestock sector after the 1968-73 drought
Herders' options for herd reconstitution
The effectiveness of herders' options
Rehabilitation of the farm sector in the absence of outside intervention
Farmers options
How long does it take the farming sector to recover?
3.01 It is important to assess the capacity and speed of rehabilitation in crop and livestock production in the absence of intervention from sources external to the system or community involved for two reasons. Firstly, one must have some measure of expected rates of rehabilitation against which to evaluate the impact of various forms of intervention aimed at boosting the speed of recovery. For example, suppose that rates of herd reconstitution in the absence of intervention can be estimated at 2.5% per annum but that a programme of selective supplementation is expected to raise this to 3.5% per annum. The net gain in livestock productivity as a result of this intervention can then be compared against the cost of its implementation and its return assessed in relation to other possible policy measures.
3.02 Secondly, indigenous strategies in the post-drought period provide policy makers with guidelines as to processes of local adaptation which may be supported. Communities often display considerable resilience in the face of drought, enabling them to survive severe crises and to regain their ability to produce in the subsequent period. The policies of governments and development agencies can either re-inforce those strategies or render them less effective. For example, it is common in conditions of acute pasture shortage for herders from the Sahel to take their animals much further to the south of their usual circuits of transhumance, hoping by this means to get access to fodder in higher rainfall zones. Administrative control over the movement of herds from region to region and between nations curb the possibilities for migration and thus heighten the risks of heavy stock losses in times of drought. Similarly, temporary migration of labour from rural areas is pursued on a regular basis in many countries. For example, seasonal migration is an important source of off-farm income for many farmers in the savanna and sahelian zones of West Africa. In times of drought, this migration flow becomes crucially important, as it both reduces the number of people who must be fed by the household granary in the drought-affected area and it provides a supplementary source of income for buying food. Government policy can either aid or hamper this flow of labour from poor to better-off regions, by either minimising the paperwork and cost of movement, or conversely, by making it difficult for people to travel, cross regional or national frontiers or obtain temporary employment.
3.03 The reconstitution of herd numbers over time can be looked at, both at the level of the national herd and at the level of the individual herd-owner. The distinction must be made between the two eases of reconstitution because it can be assumed, in most cases, that imports of livestock will make little appreciable impact on the growth in total herd numbers at the national level, while much of the growth in an individual's holding may be due to acquisition of animals from various sources. In this latter case, there is no net gain in total national herd numbers, although a redistribution in livestock holdings will have occurred. The use of herd models to simulate reconstitution of livestock numbers after drought will be examined first before turning to strategies pursued by individual producers to restore their animal holdings.
3.04 Modelling the post-drought reconstitution of herds. Several writers have developed models aimed at simulating drought-induced losses and the subsequent period of herd reconstitution in the ensuing years (Dahl and Hjort, 1976; Tacher, 1975; Clark, 1984). The general format of such models is to take a herd containing a certain proportion of animals belonging to different age-classes. Usually only the female breeding herd is taken as it is the number of breeding stock available which determines current and future herd growth rates. After subjecting a herd of normal structure to drought losses of a certain order, reconstitution is stimulated by attributing given values for' fertility and mortality rates in each age-class.
3.05 The models proposed by different writers vary in terms of the different parameters chosen as typical of herds under normal conditions and in whether or not some compensatory increase in herd productivity is assumed to take place in the immediate post-drought period. It has been noted by some researchers' that herds typically experience abnormally high rates of fertility in the year or two following a drought, due to low rates of conception and birth during the drought period and to the favourable post-drought balance between livestock and pasture conditions. Thus, immediate post-drought growth rates may be very high, herds being largely composed of breeding females and many of these calving in the year after the drought. Dahl d; Hjort (1976) give examples of herds growing by as much as 25% following a drought before falling back to more normal rates of increase of between 2 and 5% per annum. However, given the uneven impact of drought losses, the numbers, age composition and output of herds will undergo considerable fluctuation over the ensuing years as the herd gradually regains its former structure.
3.06 Table 3.1 gives the estimated number of years required for cattle herd reconstitution after losses varying from 10 to 90% and is taken from Tacher's model of Sahelian cattle herds. Diagram 3.1 presents this data in another form. There is a progressive flattening in the slope of the lines showing the time-path for reconstitution. This indicates that as mortality rates rise, the reconstitution period increases more than proportionately. For example, while losses of 20% will take 3 years to make good, losses of double this figure (i.e. 40%) will take far longer than double the time to reconstitute. Diagram 3.2 contrasts the speed of reconstitution for goats and cattle following different orders of loss. The rapid reproduction rates found among small stock allows them to reconstitute their number after drought relatively quickly. It can be seen from the diagram that after losses of 70% (or a survival rate of 30%), goat populations will be back to their former level in about 8 years. The slower growth of cattle herds means that they take much longer to reconstitute after losses of lesser significance.
Table 3.1. Estimated Time for Cattle Herd Reconstitution after Varying Levels of Drought Loss.
|
Percentage losses among herd |
Number of years taken for herd reconstitution |
|
20% |
3 |
|
30% |
10 |
|
40% |
12 |
|
50% |
21 |
|
60% |
80 |
|
70% |
43 |
|
80% |
61 |
|
90% |
85 |
Source: Tacher (1975).
3.07 The higher survival rates of small stock in times of drought as compared with cattle and the faster rate of growth among the former explains changes in the composition of livestock populations in the post-drought period and in particular the dominance of goats as reconstitution takes place. This shift in the species composition of herds is seen in the following section
3.08 The general conclusion of the herd models discussed above and the derived rates of growth may be compared with the evolution of animal numbers in the Sahel, following the losses of the prolonged drought period of 1968-73. As may be seen from Table 3.2 most Sahelian states had by 1983 seen a restoration of the national herd, although a shift had taken place in the relative importance of different species and in their geographic distribution. The evolution of livestock numbers since 1973 provides support for the conclusion that overall drought losses were around 30% for cattle in most countries, with somewhat lower rates among sheep and goats, although the incidence of loss varied considerably between regions within any single country. Only in the more northerly zones did cattle losses exceed 25 to 30%, so that cattle numbers were approximately restored over the ten-year period, as would be predicted by the model described above. Very high losses among cattle of 80 to 100% for herds in northeast Mali (Marty, 1975), while of critical importance to the herders concerned, were probably of low national or international significance. These losses were composed not only of livestock deaths but also of distress sales of stock, some of which were subsequently slaughtered for meat while some were bought by other herd-owners.
Table 3.2. Reconstitution of Livestock Numbers in the Sahel after the Drought of 1968-73 (millions of head).
|
Country |
1972 |
1974 |
1982/83 |
|
Burkina Faso | |||
|
Cattle |
2.40 |
1.60 |
2.95 |
|
Small stock |
4.05 |
3.00 |
4.50 |
|
Mali | |||
|
Cattle |
5.40 |
3.70 |
5.40 |
|
Small stock |
11.18 |
7,70 |
3.95 |
|
Senegal | |||
|
Cattle |
2.51 |
2.32 |
2.33 |
|
Small stock |
2.70 |
2.53 |
3.36 |
Source: FAO (1985 a, b, c).
In the latter case, while the sales of an animal to another represents a net loss to the individual producer, at the national level this transaction represents a redistribution of available livestock rather than a net loss of herd capital. In addition, some animals which had been thought lost to the national herd by their migration to states further south are likely to have returned to the Sahelian countries with the better rainfall and pasture conditions of the mid-70s.
3.09 Earlier, reference was made to several herd models used to estimate the speed of herd reconstitution in the absence of outside intervention, after losses of varying intensity. It was seen for example that for losses of 30%, it will take about 10 years for a herd to regain its former numbers. Individual herders, however, can pursue a number of strategies in the event of loss in order to reduce the length of time taken for their own herd numbers to re-build on the basis of natural reproduction alone. These include: supplementing livestock output with income earned elsewhere, concentration of resources on rapid reproducers, like sheep and goats, the acquisition of stock from other households or attempts to raise the productivity of remaining livestock holdings. Most of these strategies involve the re-distribution of existing livestock and thus do not exert an overall impact on the overall rate of growth of the national herd. If however stock are available from elsewhere (for example if Malian herders can buy breeding stock in Burkina Faso), individual herders' strategies for herd-rebuilding can add to total number of animals owned at the national level.
3.10 The pastoralist is faced with a major problem in the post-drought period, that of having reduced herd numbers producing a lower income in the form of milk supplies and saleable stock. Herd size may have fallen below a level capable of supporting the human group formerly dependent on it. As a result, the herder has several options:
3.11 Firstly, to increase the rate of herd exploitation by increasing milk offtake or selling more stock. This policy will lead to the disappearance of the herder's capital sooner or later.
3.12 Secondly, to supplement income from the herd with that obtained elsewhere, for example by the outmigration of some household members or by the pursuit of a number of supplementary activities carried out in association with care of the herd. The success of such a strategy depends on the extent to which herd management patterns change as a consequence. As discussed earlier, the outflow of adult male labour from the pastoral household can have adverse effects on the way the herd is managed. Where herders must combine farming with the care of their animals, the heavy labour demands of cultivation will reduce the possibilities for efficient herd management. Transhumance over long distances during the rainy season may have to be abandoned because labour is tied Up in farming. However, in some eases, the combination of farming with herding may be relatively easy. For example, several households can group their animals together and send one or two of their members with the Joint herd' thereby minimising the labour demands on individual households. Alternatively, agricultural production can be fitted into transhumance circuits in such a way as to minimise disruption in grazing patterns (as described by Cunnison (1966) for Baggara nomads of the Sudan).
3.13 The third option is to convert remaining holdings into small stock, whose rates of reproduction are considerably higher than those for cattle and which can supply a more even supply of milk throughout the year. Having built Up a large holding of sheep and goats over a period of years, the animals can then be sold in order to buy- Cattle. The viability of this option depends on the relative price ratios of small to large stock in the period immediately following the drought and in subsequent years. The success of this strategy is hampered by the tendency for small stock prices to be fairly high after a drought' due to strong demand among herders for these species as many producers try to reconstitute their holdings in this way. In addition, as noted by Dahl and Hjort (1979), this strategy is very risky, as small stock tend to be more susceptible to disease.
3.14 Fourthly, herders can acquire animals from other herd-owners, through legal or illicit means. The pastoral literature presents many examples of livestock "loans" between different producers and details the variety of terms and contexts within which these exchanges occur (see Toulmin 1983b, pp 12-18, for a summary of these animal loans). Some of these loans are re-distributive in function and motive, involving the transfer of rights to offspring of loaned animals to the borrower, enabling the latter to build up his own livestock capital over time. An example of this is exhibited by the habbanaae system of loans among the pastoral WodaaBe of the Sahel, whereby a cow is loaned to another herder, the latter receiving the first three calves born before returning the cow to her owner (Dupire, 1962). Other livestock transfers do not offer the same prospect for gaining an independent herd, the temporary allocation of animals to another being made on less favourable terms, such as granting access to a quantity of milk, or payment in grain or cash. The latter examples are essentially herding contracts, involving the exchange of a herder's labour for a wage denominated in cash or in kind. It is generally held that the frequency of loans will decline in the period following a drought, given the fall in total numbers of livestock available and the lack of surplus animals at the level of the individual household. However, Marty (1975) finds that they remained the most important means by which herders gained access to animals after the 1973 drought for the group of pastoralists surveyed in north-east Mali.
3.15 Theft or raiding of stock is an alternative means by which an individual herder or group of pastoralists can make good their losses relatively quickly. In the past, this was probably a much more frequent means by which people could expand their herd numbers, but is of much less importance today' except perhaps in remote border areas where control by the government is weak.
3.16 The fifth option open to herders is to try to raise productivity levels of existing stock, so that herd numbers increase more rapidly and/or a higher level of offtake can be maintained without running down herd capital. For example, herders can acquire access to inputs - medicines, supplementary fodder with which to reduce mortality rates, increase milk production or raise fertility. However' these inputs are often expensive and in very short supply. Herders who have recently suffered heavy stock losses are unlikely to be able to finance the purchase of such inputs on a significant scale. Alternatively, herders can change management practices to raise herd performance, for example by reducing milk offtake to raise calf growth and survival rates, by adapting circuits of transhumance and by devoting more time and attention to their livestock. However, as noted above, herders face pressures in the opposite direction following a drought. With few breeding cows left, the herder may have to raise milk offtake to feed his family. With the outflow of labour, a shortage of herding labour acts as a constraint on the pursuit of more efficient herd management practices, such as herd division and transhumance to salt deposits.
3.17 It is debatable how effective are the strategies pursued by herders in the post-drought period in significantly increasing the speed at which the individual herder can regain his former herd numbers. Individual case-studies are presented by a number of researchers but little can be concluded from these except at the anecdotal level. However, a comparison of two studies among the pastoral Fulani of Niger in the 1980s and twenty years earlier does suggest certain factors which have reduced the effectiveness of herders' strategies over the past few decades.
3.18 White's study (1984) notes that WodaaBe herders have faced severe constraints since the drought of 1973 because their herd size has fallen below the minimum for household viability. As a result of this, herders are forced to sell many of their animals at a very young age and must sometimes sell breeding stock in order to finance grain purchases. Herders also take on entrusted cattle from sedentary livestock-owners to supplement their incomes. However' the presence of these animals in their herd constrains the pattern of transhumance they can follow, as the owners of entrusted stock do not want their animals to be taken far from their control and -supervision. This constrained transhumance limits herd productivity.
In addition to these factors, the outflow of migrant labour in the dry season further reduces the efficiency of herd management techniques and contributes to the persistence of the household in a vicious circle, in which it is unable to reestablish a herd of sufficient size to become fully self-supporting,
3.19 Dupire's work (1872) among the Fulani of Niger in the 1950 and 60s presents a picture of greater; flexibility, with many households following a cycle of loss and subsequent rehabilitation' by their temporary settlement in agriculture and the gradual re-building of herds using small stock. The difference between the situation described by the two writers is probably attributable to several factors: (i) the relatively favourable climatic conditions over much of West Africa in the 1950s and 60s, in contrast to the period after 1970; (ii) the exceptional severity of the 1968-73 drought in terms of livestock losses in contrast to less devastating droughts of previous decades; (iii) the growth in human and livestock populations over the past twenty years which has put heavy pressure on available pasture resources; and (iv) the growing involvement of non-pastoral producers in the livestock sector which has led to rising livestock prices and increased competition for the use of grazing and water resources.
3.20 Post-drought recovery by herders will be accelerated where the fall in livestock numbers allows a favourable balance to be restored between available grazing and animal fodder needs. However, while very heavy animal losses provide good conditions for those with some surviving stock, they make rehabilitation more difficult for those who have lost their entire herd. Herders in the latter situation will face high livestock prices, especially for breeding stock, relative to the other commodities available to them, such as their labour power, grain, etc. The pursuit of supplementary incomes by pastoral household with some surviving stock can have both positive and negative effects on the speed of herd reconstitution. On the one hand, diversion of labour from managing the herd reduces the level of herd productivity. On the other hand, the receipt of income from elsewhere reduces the pressure of demand from the pastoral household on available milk supplies and on animals to be sold to finance cereal purchases. With calves receiving more milk, calf mortality will be lower and calf growth to maturity faster, leading to a higher overall level of herd growth and productivity.
3.21 An earlier section summarised the likely effects of drought on livestock productivity and holdings within farming areas. It focussed on the consequences of draft animal losses for crop production in the years following drought, due to the weakness or death of some animals and to the distress sales of others by farmers needing to get cash to buy grain. Diagram 3.3 presents a schematic picture of the different lengths of time required for rehabilitation of farming areas, according to the level of draft animal loss experienced. The modelling of the farming sector's rehabilitation after drought is of greater complexity than that of herd development over time. This is because the significance of draft losses will depend on the nature of the farming system and the effectiveness of different options pursued by farmers in order to maintain crop output. In order to assess the likely fall in crop output due to work oxen losses, questions such as the following must be answered:
(i) What proportion of land is normally ploughed and weeded by oxen?(ii) how much time is available for land preparation before sowing?
(iii) what effect is there on yields from late sowing, from sowing on unploughed land or from weeding by hand rather than by plough?
(iv) can land be prepared by hand and, if so, what is the area that can be dealt with?
(v) what proportion of households have the necessary animals and equipment for their own ploughteam and thus what slack is there in the system for ploughing using borrowed or shared animals?
3.3 Draft Animal Population: Ethiopia, Botswana and Mali. (thousands of head).
|
Country |
Work Oxen |
Horses and Donkeys |
Total Cattle |
|
Ethiopia |
5,670 |
5,465 |
26,300 |
|
Botswana |
420 |
167 |
2,818 |
|
Mali |
400 |
575 |
5,400 |
Sources: Ethiopia (FAO Production Yearbook, 1983)
Botswana (FAO, 1984a)
Mali (FAO, 1985b)
3.22 The three countries looked at in this paper - Ethiopia, Botswana and Mali - all depend to a greater or lesser extent on work oxen for their draft power requirements. The draft animal populations of each country are shown below in Table 3.3. Ethiopia has an ancient heritage of plough cultivation going back for more than 1,000 years; in the case of Botswana, plough technology was brought in during the late nineteenth century, for Mali, ploughs have only become of importance in the last 20 to 30 years, the number of farmers using this technology increasing by an estimated 9 to 10% per annum over the last decade (Delgado, 1980).
3.23 Soils vary considerably between the three countries, the Sahelian and savanna soils of Mali being far lighter than those of highland Ethiopia and of Botswana, where heavier clays and vertisols are found. In the case of Ethiopia, soils must often be worked 4 to 6 times to achieve a fine enough seedbed for the traditional crop of teff, in contrast to the rapid ridging of sand soils done by many farmers in Mali. The total volume and timing of rainfall also vary very considerably within and between each country. For highland areas of Ethiopia the short rains of February to April provide the opportunity for much land preparation to take place before fields are sown in June when the main rains begin. In arable areas of Botswana and in the more northerly farming zones of Mali, time available for land preparation before sowing is much more limited and working of the soil is often minimal as a result. Part of the millet crop may even be sown on unploughed land in order to maximise the plant's growing period, although the subsequent weeding problems are severe.
3.24 Given the above factors and their varying importance between farming regions, one would expect the consequences of work oxen losses to vary considerably according to the characteristics of the farming system involved.
3.25 There are a number of options which farmers can pursue to restore levels of crop production and holdings of draft animals: the share or loan of draft animals between households, use of other stock for pulling the plough, hand cultivation of soils, hire of tractor services, changes in crop composition, purchases of fertiliser, supplementation of remaining stock, turning to income earned elsewhere or waiting for livestock holdings to re-grow. Each of these is discussed below.
3.26 Sharing or loans of work oxen from other households. Where oxen losses have been relatively slight, farmers may be able to borrow a single animal or plough team from another household. Such loans are common in many farming systems, loans being repaid on a variety of terms often involving the exchange of several days hand labour against each Slough team day borrowed (see Vierich and Sheppard, 1980 for Botswana; Fulton and Toulmin, 1982 for Mali). Alternatively, two households with A single ox each can arrange to take turns in using the oxen pair, as described by Gryseels and Anderson (1983) for Ethiopia. With very heavy work oxen losses in a particular region, loans are likely to be less easy to arrange for those having lost their own animals. For the ease of Botswana, a recent survey notes that there may be considerable reserve capacity in the availability of draft animal power in normal times, so that minor losses, of say 20%, have little effect on the overall area cultivated. More extensive sharing of draft power takes place and animals can be worked more intensively in order to maintain total output at expected levels.
3.27 Use of other animals for draft. If losses of oxen have been high, a farmer can continue to cultivate his land using other animals for the ploughteam, such as milk cows, horses, donkeys or camels. These animals will have a lower productivity in work but will partially compensate for the loss of trained oxen. In extreme cases, even human labour has been known to be used for pulling the plough for example in the period following the great rinderpest epidemic in Ethiopia in the 1890s when an estimated 90% of the country's draft oxen were lost (Wolde Mariam, 1984). However, if work oxen holdings have been badly affected by drought, it is likely that other stock will also have either suffered high mortality or have been sold to purchase food grains.
3.28 Hand cultivation. In the absence of draft power, farmers may be able to cultivate part of their land by means of hand tools. However, this will be at the cost of lower crop output' due to the smaller area cultivated and the less effective tillage that can be achieved by hand as against plough techniques. Estimates of the land area which can be cultivated by hand vary from 10 to 50% of that which can be managed by a plough team, depending on the nature of soils and time available for land preparation. Where weeding is usually done with a plough, resort to hand techniques will lead to lower yields from the less optimal timing of this operation.
3.29 Hire tractor services. This option is open only to those farmers with access to this service at a reasonable cost. Hiring a tractor tends to be more expensive than hire of a ploughteam for the same work and usually requires a cash outlay rather than repayment in labour or other services (Vierich and Sheppard, 1980). For this reason, farmers who find themselves without work oxen may also be without the funds to finance ploughing by tractor. Data from Botswana would suggest however that a considerable number of farmers resort to hire of tractors both in normal years and in times of drought, a survey done after the 1979 drought finding that while 1.7% of households in a given region owned tractors, 11% of households had ploughed their land by this means (Jones, 1980). Migrants' earnings were especially important in financing the hire of tractors by poor households.
3.30 Change in crop composition. Different crops vary in their tillage requirements. For example, in the ease of Ethiopia, teff needs a finely-worked seedbed; in Mali, much of the millet crop may be sown on unploughed land whereas groundnuts require a prepared seedbed. Farmers can moderate the effects of work oxen loss by switching to less tillage-intensive crops, the possibilities for doing this depending on their access to seed, their family's consumption requirements and their marketing possibilities.
3.31 Fertiliser purchases. 'Farmers can reduce their shortfall in output arising from a decline in area cultivated by higher levels of fertiliser use leading to higher yields. The effectiveness of this option in maintaining total crop production depends on crop response to fertiliser use and the relative costs of purchase, transport and application of fertiliser. The adoption of a short term policy of heavy fertiliser use in the years immediately following drought may be a reasonable policy for governments to promote, as will be investigated later. Lack of cash available to the farmer in the post-drought period is the main constraint on this option being widely pursued in the absence of extensive government subsidies on the purchase and distribution of this input.
3.32 Supplement remaining animals. Where some livestock are still left to a farmer, their work capacity can be increased by giving them supplementary fodder. This fodder could come from crop by-products or natural pasture and browse - both likely to be in short supply following drought - or from purchases of locally available agro-industrial by-products, such as cotton-seed, molasses and bran. However, the latter are also likely to be heavily in demand in the last few months of the dry season, when oxen most need supplementation, as many livestock owners will be trying to carry their stock through to the next rainy season. Exports of these commodities will further raise their prices and restrict supplies. Work is currently in progress in Ethiopia and in several Sahelian states to develop supplementation programmes based on blocks made from a mixture of molasses and urea (FAO 1985 a,b,c). While urea must usually be imported, molasses are in abundant supply in many countries as a by-product of the sugar processing industry. These molasses blocks have the advantage of being relatively easy to transport and store, in contrast to molasses in their liquid form. However, the molasses-urea mix is not meant to be a substitute for other sources of fodder but rather to stimulate consumption of low quality roughages like natural hays and crop straw. Preliminary estimates within the Sahelian context put production and sales costs at 35-40 FCFA per kg, equivalent to US$0.07-0.08 per kg (FAO 1985 a,b,c).
3.33 Earn income elsewhere to buy replacement oxen. Farm households may be able to finance the purchase of new stock by income earned from other sources. For example, migration earnings are a major source of income to many farming areas of southern Africa and the Sahel. Migration may either be seasonal or take the form of a male household member being away for a number of years, during which time cash remittances are sent back to the agricultural sector to purchase both food and farm inputs. The ease with which these earnings car be used to finance the purchase of new work animals depends on the relative value of remittances, the price of work oxen and the urgency of other calls upon cash income. In times of drought, urban labour markets experience a large number of job-seekers and real wage levels tend to be lower than at normal times For this reason, the availability of migration earnings will be lower in the post-drought period and possibilities for acquiring the funds for oxen purchase more limited than in normal times. It also will be harder for farmers to reconstitute their holdings where both the arable and the livestock sectors of the economy have been simultaneously hit by drought losses. In this case, there will be heavy demand from both the farming community and the meat market for a limited supply of young male animals and prices will rise accordingly. Delgado (1980) noted in the case of Mali that farmers provide strong competition for young oxen with those involved in meat-fattening schemes, substantially constraining the availability of animals and the profits to be made from meat enterprises.
3.34 Wait for herd to regrow: For those households with their own breeding herd, the offspring of these animals can provide future working animals. In this case, the speed of recovery in work oxen numbers depends on the number of oxen required for ploughing, the size of the breeding herd and its rate of increase. Usually one would expect both work oxen and breeding herds to have been adversely affected by drought, although there may be circumstances in which households prefer to sell their oxen rather than lose their breeding herd. In such an event, the household must wait for a sufficient number of male calves to be born and brought to an age when they are ready to work, although in the intervening years some arrangement must be made to either plough by other means or to purchase food from elsewhere. Where some cattle must be sold to buy food on a regular basis, the household may never be able to reestablish itself as a viable farming unit and it will be faced by the prospect of increasing impoverishment (Vierich and Sheppard, 1980).
3.35 Table 3.4 gives rough estimates of the likely length of time taken for the farming sector to recover from a range of drought-induced losses in the absence of outside intervention. The spread in the number of years given for each case results from the dependence of reconstitution speeds on a variety of factors discussed earlier. Reconstitution will be more rapid in the following circumstances:
- where sharing of animals provides a temporary means by which those without draft power can continue to cultivate their land;- where the agricultural sector is sufficiently productive for farmers to have access to a regular surplus for investment and where the relative price of crops to oxen is in favour of the former, so that a good harvest may enable the farmer to replace his lost animals in a single year;
- where there are external sources of income available to members of the household which can be used to buy new animals and equipment and which can provide for the household's food needs in the intervening
Table 3.4 Estimated Time Taken for Recovery of Farming Sector after Drought in the Absence of Outside Intervention.
|
Work oxen losses |
Number of years |
|
< 20% |
0 - 2 |
|
20 - 40% |
2 - 5 |
|
> 50% |
5 -10 |
3.36 Conversely, rehabilitation speeds will be longer, the heavier are oxen losses, the greater the area affected by crop losses and livestock deaths and sales' and the higher the price of oxen relative to crop output and migration earnings.
3.37 Post-drought surveys provide some data on the evolution of harvests over subsequent years and on the time taken for crop production to get back to normal levels However, there are often a number of factors affecting total harvest size in the post-drought period which makes it hard to assess the relative importance of work oxen losses as compared with other variables, such as lack of seed, rainfall levels, social and political dislocation, etc.
3.38 In the case of Ethiopia, data on crop output and areas cultivated is scarce for both the pre- and the post-1973 period. One-source estimates a fall of 60% in arable land planted in the worst year of the drought in the regions of Wollo and Tigre (IBRD, 1979) due to shortages of work oxen and seed. Wolde Mariam (1984) cites a report showing that previous losses of work oxen accounted for between 44% and 87% of the cases in which land was not cultivated in 1973, seed shortages being mentioned as being of only secondary importance. Crop production figures suggest that 1974 saw a return to more normal levels of output at the national levels with the total crop harvest up by 14% over the previous year (IBRD, 1980). However, these figures must be treated with care because they are based on a slender sampling frame and apply to national production rather than being limited to the most drought-affected provinces. In subsequent years' additional factors such as land reform and political instability have themselves contributed substantially to the difficulties faced in assessing changes in farm output in the drought-prone areas of Wollo and Tigre.
3.39 For Botswana, harvests in the post-drought years of the late 1960s show very high variability, attributable to continued fluctuations in rainfall. While figures on total areas cultivated are not available, those on total output suggest that the farm sector was not operating at normal levels of output until 1971 (Jones, 1979). The drought year of 1978/79 in Botswana saw an estimated 60% fall in area cultivated in one region surveyed, caused by the poor condition of work oxen and the poor timing of rainfall at the start of the farming season. However, the following farming season appears to have seen an increase in area cultivated above normal levels, farmers having an incentive to increase their field size, given shortages in food supplies (Jones, 1980). This increase in area was made possible by the relatively low losses (10-15%) suffered among draft animals and the extensive systems for loaning draft power between households.
3.40 In the ease of the Sahel, no material is available on the speed with which areas cultivated returned to normal after the drought of 1968-73. However, grain production figures would suggest that harvests were back to pre-drought levels in 1974 which would imply no significant adverse impact on farm productive capacity as a result of the drought. This may be explained by the relatively low losses of harvests and livestock in most -farming areas, so that area farmed and output could return rapidly to normal once rainfall conditions improved. It may also be the case that producers were able to minimise the consequences of drought-induced crop losses with the help of migration earnings and systems for help and redistribtution within peasant society thus avoiding sales of work oxen. By contrast, recent material from north west Mali would suggest a fall in area cultivated of between 30 and 50% in 1984, due to heavy losses among draught animals the previous year (FAO, 1984c).
