Concepts of food grain stock policy

Contents - Previous

An Indonesian case study

DJB Calverley, ODNRI

At the beginning of this decade Indonesia made a major policy decision to become self-sufficient in rice and stop impoting rice. Government involvement in rice marketing and price control was carried out by BULOG, a parastatal organisation with considerable autonomy. As a result of price incentives, rice production increased sharply. BULOG was called on to intervene to support producer prices but it had little opportunity to dispose of surplus stocks through price stabilisation operations at the upper level. It did have the role of supplying about 1,5 million tonnes annually to, so called, budget groups; hospitals, schools, civil servants, etc., but this was insufficient to dispose of its annual procurement. Consequently as production increased so did its stocks, at times to over 4 million tonnes. ODNRI was asked to advice on the minimum stock levels needed for BULOG to meet its commitments, and what were the costs incurred of producing and holding these stocks.

The problem, basically, was one of achieving a high degree of food security at minimum cost. Food Security was defined as:

Having food (rice) available to meet the demands of those who need it, anywhere, at any time and at resonable cost.

The definition implies a sense of dynamism, of having continually to address the problem of food supplies. A failure of food security was considered to occur if stocks became exhausted anywhere in Indonesia at any time. This was a very limiting consideration of failure, since in most instances short term exhaustion could be anticipated and prevented by emergency action.

The study considered the national aggregated stock requirements, and those in each of Indonesia's 27 provinces. Some provinces were identified as being regularly in rice surplus, others in rice deficit. A simulation model was developed which hat 5 major components.

1. A set of economic models of BULOG's marketing activities of procurement and distribution of paddy and rice in each of the provinces.
2. A model of the provincial stock levels and changes based on a set of re-order levels for deficit provinces from those in surplus.
3. A model for calculating the probability of failure of BULOG's operations in provinces which are normally in deficit.
4. A model for calculating the likelihood of BULOG purchasing adequate aggregate national stocks from provinces normally in rice surplus (or from exports).
5. A model of aggregate BULOG storage costs.

It was assumed that BULOG's stock levels follow a conventional pattern; stock levels rising due to procurement immediately after harvest, and later falling to meet disbursement to budget groups and for price support operations. The net change of stocks on a monthly basis being the difference between outgoing and procurement. When the stocks in deficit provinces reached stated minimum levels, re-order rules were established for ordering more rice from provinces in surplus. The period of minimum stocks occurs immediately before procurement begins. Adequate provincial and national stocks should be carried over to prevent exhaustion of stocks at this critical period, and this level of stocks is regarded as the minimum needed by BULOG.

An analysis of the provincial marketing operations, on the basis of the standard deviations of rice production about the trend production, made it possible to estimate the probability of cumulative stock changes and whether these would be outside the limits of acceptability. Since it was a political decision that there should be no imports of rice, stock changes were required, as far as possible, never to result in an aggregate national deficit. From these probabilities could be calculated the degree of confidence that system failure will be prevented, i.e. the degree of confidence that food security can be sustained.

Clearly the nearer domestic rice production approaches the national demand, the smaller will be the carry over stock levels needed for any required level of food security. It was found that a 2 million tonne carry over in 1983 would give 69% confidence that the system would not fail; with the higher trend production in 1985/86, 1,5 million tonnes of carry over, with the increased BULOG procurement gave 97% confidence.

Two further important points are illustrated in Figure 1. First at low levels of confidence, small increases in stock levels have significant effects on improving food security. Beyond about the 90% confidence limits, very substantial increases in stocks increase food security by very small margins. In 1985/86 increasing the stock level from 1.5 million tonnes to 5 million tonnes increased confidence levels from about 95 to 98%. Second is the high annual cost of procuring and holding stocks. Market operations in supporting specified producer and consumer prices and in carrying over 1,5 million tonnes in 1985/86 would cost about US $180 million. This is in contrast to the Indonesian Government's rule of thumb guideline of 5 million tonnes which would cost some US $600 million.

The model is a simulation of BULOG's operations. It was a first attempt to quantify a dynamic concept of a food stocks policy and as such it is very specific to Indonesia Considerable work is still needed to refine and simplify the model, make it more user-friendly and adapt it for a wider application. This is presently on-go ing.

Meanwhile it remains an important tool in its own right. With it, stock level requirements can be calculated regularly as circumstances change, policy decisions can be more critically analysed, and the consequences of policy changes more sharply focussed. When, therefore, ODNRI was asked to predict optimal levels of rice production to minimise BULOG's accumulation of stocks, this was done by adapting the simulation model to examine the impact of policy variables (paddy prices, input prices) which previously were taken as fixed, and the opportunities offered by relaxing other constraints, particularly the possibility for linking food security strategies to agricultural food production policies where imports and exports can be considered as complementary to stock holding operations.

There can be no question of directly transferring the existing model to, say, an African country but the deterioration in food security presently evident in many African countries, demands that concepts and techniques of managing staple food stocks be applied to a much needed policy development in matters of food security. The institutional context and the database are much weaker in Africa than in Indonesia. Nevertheless, there is scope in African countries for the application of a model similar to that developed in Indonesia. New and recent initiatives towards regional cooperation in food security need support and offer scope for an application of a model on a regional basis.

Food security and the cost of holding stocks

Utilization of small grains in the SADCC-region - Report on a workshop

G.K. Tuengerthal, B. Bohnstedt

1. Foreword

At the 19th annual meeting of GASGA, held in Germanyin 1987, several papers were presented which highlighted the importance of small grains (millet, sorghum) in East Africa and outlined a number of problems, especially in marketing, which hinder the promotion and increased consumption of these crops.

It was the common feeling that the problems and their interrelation have not been fully understood so far. It was therefore decided to organise a regional workshop in order to look into the various problems of small grains in the SADCC region (Southern African Development and Coordination Council) and possibly to issue recommendations on how to overcome existing obstacles.

The workshop was organised by members of GASGA and was held from 17th to 21st October 1987 at the Matopos Research Station of ICRISAT, Matopos, Zimbabwe. The participants of the workshop would like to thank ICRISAT for their organisational support and hospitality.

2. Method applied

The workshop was conducted with the aid of the ojectives-oriented project planning method (ZOPP). This method is based on the logical-framework approach, which is widely used in the development community.

ZOPP is a systematic approach to planning projects and their objectives. Starting with the identification of the core problem (which the project is intended to help solve) and the specification of other factors causally linked with this in the project environment, objectives are identified which will have to be achieved in order to solve the specified problems.

These objectives are used to establish alternative solutions, which are investigated in feasibility studies. The project and its objectives are then developed for the most suitable approach. A clear, logical and unambiguous statement is prepared showing why the project should be implemented, what other factors-may endanger the success of the project, what state implemeltation of the project should bring about and how data should be collected in order to identify achievement of objectives. The planning for the project and objectives resulting from this process may be presented in a project planning matrix.

The method consists of three inter-supportive components:

• the procedure, which is the guideline for work in the planing group;
• the team approach as the framework for studying multisectoral problems;
• the visualisation technique, which is used to document the contributions by the individual participants and the results of discussions.

A three-day-workshop does not provide sufficient time to deliberate extensively on each of the steps, especially in cases where participants have no experience with the method. Nevertheless all workshops members agreed that using the method had made it much easier to structure a goal-oriented discussion, reach a group consensus and agree on final recommendations.

3. Participation Analysis

The first stage in the ZOPP planning procedure aimed at clarifying the interests and relationships underlying the problems. The terms in which a problem is perceived or stated depend on the interests involved or the point of view. In many situations the participation analysis is a provisional substitute for direct participation of partners or target groups in the planning process. It is an attempt to identify more closely with the situation of the participants, in order to acquire a different perspective and develop a better understanding of the problem.

From the documentation (cf. Annex 1) it is apparent that the production and consumption of small grains affects an extraordinarily wide range of groups and organisations in one way or another. In addition to alarge part of the public sector, especially the grain marketing boards and the Ministers of Agriculture, the private sector (production, processing and marketing) is heavily influenced by any change in the consumption of small grains. The workshop members realised that some of the groups (middleman traders, branches of processing industry, policy makers) may be apposed to projects boosting small grauns production and consumption.

4. Problem Analysis

This involves the separation and linking of cause and effect. Taking the core problem as a starting point, the attempt is made to investigate and present the causality underlying the problem complex in the environment. It is not enough to stay at the level of symptoms or superfi cial phenomena the effort must be made to pursue the problems to their roots.

The workshop identified "insufficient production of food and cash crops in harsh environments" as the core problem (cf. Annex 2). Below the core problem the causes which lead to this problem are listed.

Above the core problems its effects are described. If this "problem tree" is elaborated properly, the causes and effects of one level should be necessary and sufficient to explain those of the next level. In some cases a cause may also be an effect. This may be valid e.g. for "Too many cheap grain imports from overseas". The most important and possibly central consequence of the core problem was considered to be "sustained survival in dry areas endangered".

5. Objectives Analysis

This comprises the analysis of relationships between inputs and objectives on the basis of reformulation of problems as objectives. Questions to be considered here are:

• whether the objectives as formulated are morally or politically acceptable and
• whether the intensity and scale of the objectives as formulated and the relationships between inputs and objectives appear plausible.

Defined potential objectives (target states expressed as completed actions) and their breakdown and interrelationships are presented in terms of means and ends in a causal model. The "objectives tree" permits the presentation of not only monocausal structures but also more complex relationships through multiple entries for objectives, arrows showing relationships or numerical input/output points.

Proceeding from the core problem the workshop agreed on the objective "sustained production of food with marketable surplus in harsh areas" and converted the problem tree to a hierarchy of objectives (cf. Annex 3). The result showed that two different project types could be considered in order to reach the objective, a "crop and product development approach" and a "production system development approach". The participants were perfectly aware that general policy decisions such as e.g. protection of small grains against cheap grin imports will have a substantial influence on all possible progress in this field. It was accepted that marketing is clearly one of the most relevant contributors to a wider use of small grains and that every effort must be made to widen marketing possibilities.

In the course of the objectives analysis it became apparent that a pure R & D effort will not be sufficient to overcome existing obstacles. It was pointed out by some members that worldwide R & D has already provided answers to many if not most production, processing and marketing issues, but that either the lack of international R & D cooperation or adverse national agricultural policy decisions delay the utilization of such already established knowledge.

In this context the workshop returned to the discussion of small grains use in the SADCC region, outlining relations between producers and consumers (cf. Annex 4). This led to the conclusion that the market for small grains should be split into two areas, one covering the food sector and the other industrial utilization. This in turn may be divided into different sub-markets, since the desired properties of sorghum for malting differ from required for starch production, to mention only two examples.

6. Project Planning Matrix

The project planning matrix presents in matrix form the main project elements and objectives (i.e. project purpose and overall goal) emerging from the ZOPP process. The matrix has four levels, "overall goal", "project purpose", "results" and "activities", and four columns, "summary of objectives and activities", "objectively verifiable indicators", "means of "verification" and "important assumptions".

Due to lack of time the workshop concentrated on the results and activities and on important assumptions. On the basis of the extensive problems and objectives analysis the workshop decided to choose "sustained production of food with marketable surplus in harsh areas ensured" as the overall goal and "expanded utilization of small grains achieved" as project purpose (cf. Annex 5).

The workshop took great care to scrutinise important assumptions. The list of assumptions indicates the concern about the necessity of integrating any small grains project in an overall agricultural policy, e.g. "good policy allows economic production", or "prices encourage farmers to plant new varieties" etc.

7. Final recommendations

After establishing the project planning matrix the workshop members split up into sub-groups to discuss priority issues-in the following areas: institution building, marketing and R & D (c f. Annex 6). Due to the R & D background of most workshop members, researchoriented priorities were prominent in the ensuing list. After a renewed and sometimes most controversial discussion, a list of final recommendations (cf. Annex 7) was agreed upon. These recommendations cover the areas of policy, R & D and marketing.

Participation analysis - Annex 1

Problem analysis - Annex 2

Objectives analyses - Annex 3

Diagram of small grains uses - Annex 4

Project Planning Matrix - Annex 5

Annex 6

Final recommendations - Annex 7

Contents - Previous