Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
P.S. Tyler
Tropical Stored Products Centre, Tropical Products Institute, London Road, Slough, Berkshire, UK
Food losses after harvest have been the subject of considerable debate and speculation in recent years. Much of the debate concerns the real magnitude of such losses, while the speculation relates largely to the potential for reducing losses. This paper discusses factors influencing the interpretation of post-harvest losses, with particular reference to cereal grains.
Following the call by the special session of the UN General Assembly in 1975 (1) for a 50 per cent reduction in postharvest losses, an interdepartmental sub-committee reviewed past and current activity and concluded that there was no agreed methodology for post-harvest loss assessment. Moreover, it accepted that there could be no agreed upon single figure for the percentage of post-harvest losses on a global scale or even on a national basis. Recent reviews by the Tropical Products Institute 12) and the Food and Agriculture Organization 13) to determine the extent of available information on post-harvest losses in cereals have demonstrated that there are many examples of extremely high estimates of unsubstantiated losses and that confusion has often arisen because the terminology used by authors to describe losses is sometimes ambiguous. There was a clear need to improve and standardize the methodology for loss assessment, in order to justify more firmly the development and introduction of measures designed to reduce losses in an appropriate and economical way.
Considerable progress towards both establishing a methodology and creating further awareness of the need for reliable data has been achieved:
However, statements referring to enormous losses that are quite unqualified still persist. Aggregated data reflecting losses on a world-wide basis, such as the much-quoted global figure of 10 per cent for post-harvest losses of cereals to insects, are of little value. Similarly, figures such as 35 per cent for grain losses in India and 46 per cent for sorghum losses in Nigeria (7), if applied to the whole of those countries, would be grossly misleading and give a greatly distorted view of the real situation. There is always a temptation to cite extreme figures for loss to dramatize the problem. This can easily occur when a single instance of high loss is mistakenly extrapolated to apply to a whole country, for all crops and seasons. Even when data have been produced as the result of careful measurement, they are vulnerable to mischievous manipulation. For example, high figures for loss may be employed to attract assistance from a donor agency. The true loss may be genuinely unknown, but where this is the case, surely it must be openly admitted. Some embarrassment arises when, on completion of a loss-reduction project, no justification for the effort can be demonstrated.
Misinterpretation of high figures may also be the result of arithmetical errors arising from a misunderstanding of the correct basis for their calculation. In quoting figures at different stages of the post-harvest system, the loss at each stage has sometimes been totalled, leading to an overestimate of loss. This is because it was assumed that each loss figure is a percentage of the original weight of material, whereas in fact each figure for loss is a percentage of the amount remaining in the previous stage of the system. Thus, if 10 per cent of the potential crop is lost during harvesting and a further 10 per cent occurs during threshing, the cumulative loss over both stages totals 19 per cent.
Farm-storage losses are particularly susceptible to exaggeration if the pattern of grain withdrawn from the store is overlooked. If grain remains untouched throughout the storage period, and at the time of removal the estimated loss caused by insects is 10 per cent, then this figure indeed represents the loss over the storage period. However, in most cases grain is removed at intervals for consumption. Consequently, each quantity removed will have suffered a different degree of loss, since it will have been exposed to deterioration for a different length of time, and allowance has to be made for this when determining the total estimate of loss (8).
It is not disputed that losses of food after harvest can often be serious. The delegates at the TPI Seminar on Post Harvest Grain Losses in London in 1978 (9) concluded, among other things, that situations arise in most developing countries where post-harvest losses in cereals and grain legumes are likely to be unacceptably high and that a worthwhile reduction of these would be possible and should be attempted.
The seminar conclusions clearly indicated that the primary requirement for a loss-reduction programme is that there should be a national commitment, first to determine where the major losses occur, what causes them, and what their magnitude is, second, to reduce losses, and third, to develop a co-ordinated national plan of action supported with the necessary resources for its effective implementation. However, the changes needed to reduce losses will not be widely adopted until they are acceptable to, and clearly benefit, those who are suffering from the losses. Reliable data on the locations, causes, and consequences of losses are needed by those concerned with the implementation of post-harvest programmes in order to determine priorities for loss reduction in a positive way.
The earlier often unsubstantiated figures for post-harvest losses have perhaps served some useful purpose. Without them there might not be the recognition that post-harvest losses are deserving of attention and that the prevention of potential losses will contribute positively to the alleviation of the world food shortage. It may well be that some of the exaggerated loss figures result from the fact that postharvest losses are extremely variable. They may be due to a variety of factors, the importance of which varies from commodity to commodity, from season to season, and from country to country, and to the enormous variety of circumstances under which commodities are grown, harvested, stored, processed, and marketed. The methodology to measure post-harvest losses is of necessity complex because consideration has to be given to the assessment of the various component causes of loss.
Durable foods such as cereals, grain legumes, and most oilseeds may suffer physical losses during threshing and handling after harvest; there may be consumption by pests such as insects and rodents during storage, spillage during handling, or losses due to inefficiencies in processing; or the food material may become totally inedible through microbial spoilage or contamination. There are, however, other forms of loss such as a reduction in quality or nutritive value or, in the case of seeds, viability. All of these may be important in social or economic terms.
Loss in quality is of particular importance in relation to perishable commodities such as fruits, vegetables, root crops, meat, and fish. Perishables are more susceptible to injury than durables because of their constituents and structure, and reduction in quality through mechanical injury in the case of fruits and vegetables or rapid microbial deterioration may lead to total rejection by the consumer. It is not the purpose of this paper to discuss in detail the various levels of losses found, but attention must be drawn to the care that is required in assessing and interpreting the losses measured.
Taking the commercial section of grain storage, where large volumes of grain are held for long periods, the potential for loss is extremely high and is directly summable in financial terms. Instances of high loss do arise, but when taken in the context of overall operation and profitability, and provided certain basic steps are taken to safeguard grain condition, the losses are often held at an acceptable level.
It must be emphasized, however, that loss assessments in this sector require a high level of skill and precision in order to be meaningful. That so few examples can be quoted is evidence of our general inability to monitor loss effectively at the present time.
At the farm level good progress has been made. Long-term studies in Zambia (10) and India (11) using reliable methodology have clearly shown that, in surveyed areas, storage and related post-harvest losses are much lower than previously reported, and evidence from current studies elsewhere (table 1) supports this view. It should, however, perhaps be noted that even in 1964 the high estimates for farm-level storage losses were being questioned (13), but the suggestion that losses were lower was generally disregarded. Sadly, even now, those responsible for postharvest loss-reduction programmes sometimes choose to ignore the evidence and continue to quote high figures for the loss, perhaps in the belief that lower figures will not attract the resources necessary to mount a loss-reduction programme. It is unfortunate if this is the case. Low figures for loss may confirm the excellence of the system, and although loss-reduction as such may not be directly justifiable, the case for direct support through a farmer extension service is sound. By this means encouragement will be given for the continuation of use of proven and/or traditional methods that are recognized to have merit.
TABLE 1. Some Reliable Estimates of Post-harvest Grain Losses in Farm-Level Storage
| Estimated weight loss (%) and standard error | Storage
period (months) |
Cause of loss | Grain | Country | Reference |
| 1.7 ± 5.6 | 7 | insects | maize | Zambia | ref. 9 |
| 4.26 ± 1.33 | 7 | insects, rodents, and moulds | paddy | India (Andhra Pradesh) | ref. 10 |
| 3.53 ± 0.25 | up to 9 | insects end rodents | maize | Kenya | ref. 12 |
| 3.2 ± 3.4 | up to 9 | insects | maize | Malawi | ref. 14 |
| 1.8 ± 3.5 | up to 9 | ||||
| 1.7 ± 0.5 | up to 9 | insects | sorghum | Malawi | ref. 14 |
| 5.5 | 6 | insects, rodents, and moulds | maize | Nepal | Boxall* |
* Private communication.
However, it would be wrong to claim that all farm-level post-harvest losses are low; they are not. There are examples of situations encountered, even within some of the studies quoted above, where losses were significantly higher. These tend to arise with the introduction of new, improved varieties or multiple cropping. Such innovations disturb the traditional capability to conserve grain, and the farmer has to face new and intractable problems with an increased quantity of produce. The traditional handling, drying, and storage systems are found to be inadequate; the characteristics of the crop are changed (often to a softer grain that is more susceptible to mould and insect damage); and the timing of harvest and subsequent operations may be switched from a favourable dry to an unfavourable wet season. Unless the grain can be marketed rapidly the inevitable result is higher losses for the farmer, and unfortunately these are a direct consequence of inappropriate or inadequately planned "improvements."
Thus, the plea has to be made for a careful consideration of the suitability of new crop varieties before they are forced on the farmer. With increasing pressure on farmers to adopt new varieties, their problems are likely to increase rather than decrease. Therefore, great care must be taken to ensure that the farmer is not put at a disadvantage by the "improvement," and the low baseline of losses encountered in the unimproved traditional situation should be recognized as an acceptable level of loss that should be attainable under all circumstances. Losses need to be related to the situation and location. It should be recognized that the magnitude of the loss is important but not all-important. In order to decide whether or not to undertake a loss-reduction programme, how to implement it, and where to concentrate effort, a number of other factors must be considered. These will include the value of the grain in economic terms, the social changes that may result from loss-reduction activities, and the overall effects of a programme on producers, storers, traders, and consumers.
The National Academy of Sciences (5) concludes its excellent report by recommending more systematic approaches to loss estimation in developing countries by:
It is to be hoped that the United Nations University will serve to stimulate and catalyse this work on food losses.
References
1. UN General Assembly, Resolution of the Seventh Special Session, Sept. 1975.
2. J.M. Adams, "A Review of the Literature concerning Losses in Stored Cereals and Pulses Published since 1964," Trop. Sci., 19:1 (1977).
3. Analysis of an FAO Survey of Post-harvest Crop Losses in Developing Countries, AGPP: Misc/27 (Food and Agriculture Organization, Rome, 1977).
4. K.L. Harris and C.J. Lindblad,Post-harvest Grain Loss Assessment Methods (Association of Cereal Chemists, St. Paul, Minn., USA, 1978).
5. Post-harvest Food Losses in Developing Countries (National Academy of Sciences, Washington, D.C., USA, 1978).
6. Group for Assistance on Systems Relating to Grain After Harvest (GASGA), Priorities for Action in Grain Post-harvest Loss Reduction (1978).
7. N.S. Scrimshaw, "Global Use of the Instruments of Scholarship for the Conquest of Hunger: The World Hunger Programme of the United Nations University," Food and Nutrition Bulletin, 1, no.1: 6-14 (1978).
8. J.M. Adams, "A Guide to the Objective and Reliable Estimation of Food Losses in Small Scale Farmer Storage," Trop. Stored Prod. Inf., No. 32, pp 5-12 (Tropical Stored Products Centre, Tropical Products Institute, Slough, England, 1978).
9. "Tropical Products Institute Seminar on Post-harvest Grain Loses, Held in London, 13-17 March 1978," Trop. Stored Prod. Inf., no. 36, special issue (1978).
10. J.M. Adams and G.W. Harman, The Evaluation of Losses in Maize Stored on a Selection of Small Farms in Zambia with Particular Reference to the Development of Methodology, Report of the Tropical Products Institute, G109 (1977).
11. R.A. Boxall, M. Greeley, and D.S. Tyagi, with M. Lipton and J. Neelankanta, The Prevention of Farm Level Food Grain Storage Lasses in India: A Social Cost Benefit Analysis, I DS research report (Institute of Development Studies, University of Sussex, Brighton, England, 1978).
12. C.P.F. De Lima, "The Assessment of Losses due to Insects and Rodents in Maize Stored for Subsistence in Kenya," Trop. Stored Prod. Inf., no. 38, pp. 21-26 (1979).
13. M. P. Miracle, Maize in Tropical Africa (Univ. of Wisconsin Press, Madison, Wisc., USA, 1966).
14. P. Golob, "A Practical Appraisal of On-Farm Storage Losses and Loss Assessment Methods in Malawi. 1: The Shire Valley Agricultural Development Area," Trop. Stored Prod. Inf., no.40,pp.5-13(1981).