FAO GENEVA ROUND TABLE ON SELECTED AGRICULTURAL TRADE POLICY ISSUES 21 March 2001 Room XII Palais des Nations Geneva, Switzerland Towards Improving the Operational Effectiveness of the Marrakesh Decision on the Possible Negative Effects of the Reform Programme on Least-Developed and Net-Food Importing Developing Countries Discussion paper no. 2 Food and Agriculture Organization of the United Nations Rome, 2001

Towards Improving the Operational Effectiveness of the Marrakesh Decision on the Possible Negative Effects of the Reform Programme on Least-Developed and Net-Food Importing Developing Countries

I. Background

One of the subjects extensively debated during the Uruguay Round (UR) negotiations was the possible negative impact of agricultural trade liberalization on least developed and net food-importing developing countries. This reflected a concern that the impact of the liberalization could be felt relatively strongly on world market prices of food commodities, which are imported in large and growing amounts by these countries, than on tropical agricultural products, which are the main agricultural exports of these countries. Other sources of possible negative impact on the terms-of-trade were erosion of trade preferences as tariffs are reduced and doubts about the ability of these countries to raise their agricultural export volumes due to supply-side constraints. Several model-based studies had come to the conclusion that this group of countries could experience negative effects on account of these factors, at least during the reform process. Many of them are among the poorest and most food insecure in the world.

These concerns were recognized at the political level. Included within the package of the UR Agreements was a specific Ministerial Decision called the Decision on Measures Concerning the Negative Effect of the Reform Programme on Least-Developed and Net Food-Importing Developing Countries (Decision henceforth), which articulated the concern as follows:

“Ministers recognize that during the reform programme leading to greater liberalization of trade in agriculture least-developed countries and the net-food importing developing countries may experience negative effects in terms of availability of adequate supplies of basic foodstuffs from external sources on reasonable terms and conditions, including short-term difficulties in financing normal levels of commercial imports of basic foodstuffs”.

To deal with the negative effects, the Decision provided for four response mechanisms:

♦ Food aid

♦ Short-term financing of normal levels of commercial imports

♦ Favourable terms on agricultural export credits

♦ Technical and financial assistance to improve agricultural productivity

The two groups of countries eligible for assistance under the Decision are the least-developed countries (LDCs, currently 48) and the net food-importing developing countries (NFIDCs, currently 19). All relevant statistics clearly differentiate these two groups from the rest of the developing countries with regard to food security situation and the capacity to import food.¹

The implementation of the Decision has remained unsatisfactory and a matter of concern for all, both donors and beneficiaries, despite the fact that the political commitment to its implementation has been stressed from time to time at major Conferences, such as the World Food Summit, UNCTAD and WTO. Many analysts feel that its non-implementation, even in a period such as 1995-96 when food prices in world markets rose sharply and food import bills surged (see Box 1), could indicate some inherent, serious problem with the design of the Decision.

More importantly now, in the context of the new agricultural trade negotiations, it is essential that a climate of confidence and trust be developed for continuing the reform process. The fulfillment of all commitments, the Decision included, is one way forward. For all food-insecure developing countries, food security remains a top priority. Having a contingent mechanism that can be trusted during difficult times would both create a conducive environment for trade negotiations as well as encourage countries to liberalize further their food markets.

The objective of this Note is to discuss some ideas to facilitate the implementation of the Decision. With this background, the next section defines the nature of the “negative effects” alluded to in the Decision and provides a rough order of magnitude of the “excess” food import bills. Section III provides information on the three instruments envisaged in the Decision for assisting the target countries facing short-term food import-related difficulties. Section IV discusses two proposals for implementing the Decision.

II. The Nature and Magnitude of the Food Problem Addressed by the Decision

Defining “negative effects”

The Decision defines the nature of the food problem in a general manner, stating that during the reform programme, the LDCs and NFIDCs “may experience negative effects in terms of the availability of adequate supplies of basic foodstuffs from external sources on reasonable terms and conditions”. Since inadequacy of supplies of basic foodstuffs in world markets, in the physical sense of the term, is hardly an issue in these times, the reference to the problem must be on economic access, or to accessing food in world markets “on reasonable terms and conditions”, which is indeed a widely recognized problem for many net food-importing developing countries. That this is the correct interpretation of the “negative effects” is also supported by the fact that the Decision goes on to prescribe three² response mechanisms all of which could contribute to improving the economic access to food.

What could be the “reasonable terms and conditions” alluded to in the Decision? How does one identify “unreasonable” access terms? There could be many different ways of defining this problem, but a simple interpretation of the “unreasonable” access term for net food-importing developing countries would be food import bills over and above normal levels.

The trigger to activate the response for assistance

Triggers are essential in order to decide when and how to respond. The text of the Decision provides some guidance on this. Specifically, the Decision does not rest on a narrow interpretation of the negative effects based upon individual sources of the problem, e.g. sharp rises in world market prices, sharp declines in food aid shipments, and so on. Rather, it refers to the broader effects – i.e. the “unreasonable” or “unexpected” outcomes, as noted above.

These considerations suggest that the triggers should not be based on the individual factors that lead to the problem (e.g. higher world prices) but rather on the totality of the outcome itself, i.e. the unexpected level of food import bills. This has the further implication that the triggers refer to country-specific outcomes such as exceptional increases in a particular country’s import bill rather than global triggers such as abnormal increases in world prices. Thus, what seems to have been envisaged is the need for assistance for some countries, albeit usually different ones, every year, rather than all countries in one year like 1995/96 when world cereal prices spiked.

Quantifying unreasonable or excess food import bills

Although cereals represent a high proportion of total food imports for most net food-importing developing countries, non-cereal foods particularly vegetable oils, meat and dairy products, and sugar, also contribute a significant proportion to the diet of poor people and to the total food import bill. For this analysis, in order to demonstrate the order of magnitude involved, excess food import bills were computed separately for cereals only (Food I) and for a broader definition of food (Food II), which also most basic food items, namely vegetable oils, dairy and meat products and sugar.

Two alternative thresholds were defined to measure the excess import bill: i) import bills more than 5% above the trend level; and ii) more than 10% above the trend level.³ The difference between actual import bills and these threshold levels are considered as “excessive” or “unreasonable” import bills, to use the term of the Decision.⁴

In the calculations, the absolute values of the excess food import bills that met the above criteria are added across countries (separately for 46 LDCs⁵, 19 NFIDCs and all 65 countries). The mean values shown in Table 1 are the averages of these totals for the 10 years covered (1989-98). The maximum value shown in the table is the largest annual total. For example for the 10% rule, the mean excess import bill of the LDCs for the 1989-98 period was US$179 million for Food I and US$224 million for Food II. The corresponding maxima were US$421 and US$441 million, which occurred in 1992 and 1996 respectively. Under the 5% rule, the means and maxima were higher as these include excess bills above 5% also. For the entire sample of 65 countries, under the 10% rule, the mean import bill was US$401 million for Food I and US$432 million for Food II. The maximum level reached was just over one billion dollars. In only three out of the 10 years did excess import bill exceed US$500 million – in 1995 (US$984), 1996 (US$1037) and 1998 (US$502). At the other extreme, excess food import bill for all 65 countries taken together was lowest in 1991 (US$84 million).

Table 1: Food import bills in excess of 5 and 10 percent above trend import levels (in million US$)

	In excess of 5% above trend		In excess of 10% above trend
	Mean	Maximum (year)	Mean	Maximum (year)
Food I (cereals only) LDCs (46) NFIDCs (19) All 65 Food II (cereals and others) LDCs (46) NFIDCs (19) All 65	216 300 515 290 336 627	466 (1992) 914 (1996) 1 252(1996) 549 (1996) 978 (1996) 1 527(1996)	179 222 401 224 208 432	421 (1992) 714 (1996) 998(1996) 441 (1996) 647 (1995) 1 037(1996)

Notes: Mean is the average for 1989-98 of the total excess import bills (summed for all those countries with food import bills more than 5% or 10% over trend). Maximum is the largest value for any year during 1989-98. Food I includes cereals only. Food II includes cereals, dairy and meat products, vegetable oils and sugar.

Source: Computed from FAOSTAT data.

For such a large sample, excess import bills would obviously vary widely from country to country, and from year to year for the same country. But note that in a particular year only a sub-sample of countries face excess food import bills, and this is what matters for a compensation scheme.⁶

III. The Three Response Mechanisms in the Context of the Decision

In order to evaluate the alternative proposals discussed in Section IV, it is important to understand how the three response mechanisms operate currently. This section provides that information, particularly on how they relate to the Decision.

Food aid

The Decision calls for action at two levels. First, the Committee on Food Aid under the Food Aid Convention (FAC) is called upon to establish a level of food aid commitment sufficient to meet the legitimate needs of developing countries during the reform programme. Second, it calls for “adopting guidelines to ensure that an increasing proportion of basic foodstuffs is provided to LDCs and NFIDCs in fully grant form and/or on appropriate concessional terms in line with Article IV of the Food Aid Convention”.

On the first point, the FAC 1999 determined 4 895 000 tonnes in wheat equivalent (plus US$130 million) as the minimum FAC level. Actual volumes of food aid granted have mostly exceeded these minimum levels.

On the second point, and which is important in the present context, Annex B of the 1999 FAC lists eligible recipients (countries and territories) under four categories (Box 2) as follows: (a) all LDCs; (b) 24 low-income countries (6 are NFIDCs); (c) 52 Lower Middle-Income Countries (9 are NFIDCs); and (d) 4 NFIDCs not included above.

Article VII of the Convention addresses prioritization and criteria for food aid allocations. Its Article VII (c) calls for giving priority to categories (a) and (b), i.e. LDCs and low-income countries (which include 6 NFIDCs). Article VII a(iii) sets some criteria for other two groups which includes the remaining 13 NFIDCs (see Box 2).

In summary, there are some considerations to note from the standpoint of the Decision. One is that the FAC 1999 did not treat all LDCs and NFIDCs equally in terms of priority and criteria for food aid allocation. As a result of the priority assigned to some groups, it could lead to a situation where the needs of the remaining 13 NFIDCs are not fully taken into account. Two, and related to the above, the food aid allocation criteria suggested for groups (c) and (d) do not fully conform with those of the Decision, notably that the principle of compensation implicit in the Decision is related strictly to the reform process and not to considerations such as international financial crises and feeding vulnerable groups. Therefore, it is doubtful whether the FAC fully reflects the concerns of the Decision.

Compensatory financing facility

In paragraph 5 of the Decision, Ministers recognized that “as a result of the Uruguay Round certain developing countries may experience short-term difficulties in financing normal levels of commercial imports and that these countries may be eligible to draw on the resources of international financial institutions under existing facilities, or such facilities as may be established, in the context of the adjustment programmes, in order to address such financing difficulties”. The subsequent sentence refers to consultations by GATT Director-General with the Managing Director of the IMF and the President of the World Bank. In view of this reference, most attention has gone to existing facilities of the IMF and the World Bank, in particular the former which already had a cereal financing facility. Very little debate has taken place so far on the other option stated above, i.e. “or such facilities as may be established”. To make progress in this area, therefore, it is first necessary to understand the operation of the IMF cereal facility. The main features of the facility are summarized in Box 3.

The reasons why the CFF has been little used have a bearing on the implementation of the Decision. One key factor appears to be the clause “in the context of the adjustment programmes” (see the quote above), i.e. the conditionality attached with the access to this facility. That the eligibility criteria are very difficult to meet by most developing countries is clear from the following quotes, from recent reviews of the facility by the IMF staff and Directors.

“Indeed, recognition that it was rare in practice for a country to have an exogenous and temporary export shortfall without also having a more persistent BoP problem was the main reason why CFF conditionality was tightened in the 1980s and why the complex system of access limits under the facility evolved. It is notable in this respect that no country has been judged under the CFF to have a BoP problem limited exclusively to the effect of a temporary export or cereal import shock since the early 1980s, and that the great majority of purchases under the CFF have taken place in conjunction with arrangements”. These considerations would suggest that “stand-alone” CFFs would be suitable only for circumstances – where there is no need for adjustment – which are, in practical terms, unlikely to hold” (paragraph 15 of the review ⁷).

There are also some other problems with using the CFF for purposes of the Decision. One is that the coverage of the food in the CFF is limited to cereals only, whereas the Decision refers to basic foods. Also, the CFF explicitly links excess cereal imports to export earnings – an excess in cereal import cost may be compensated only to the extent that it is not offset by an excess of export earnings. There are also problems of interpretation of whether shocks are exogenous and short-lived events (requiring no policy adjustments) or otherwise, requiring an adjustment programme.⁸

The status of this facility within the IMF’s various facilities was also never comfortable, as its origin and continuity were linked to humanitarian and food insecurity considerations. To quote the above IMF staff review again:

“The cereal import element was added in 1981 (to the original CFF), following the increased volatility of food prices in the 1970s, initially for a fixed term of four years. The Board had rejected this idea in 1978, out of concern that it was inappropriate for the Fund to single out food imports as a BoP problem, but reconsidered upon receipt of requests from the World Food Council and the FAO, and giving weight to the “human considerations associated with this issue”. The cereal element has since been routinely extended”.⁹

Given these difficulties, the future of the CFF also seems to remain uncertain. A recent (early 2000) review of the facility by the IMF Executive Board summarized the situation as follows: “No Director has argued for retention of the CFF as it stands now, and the debate has focussed on the two main options discussed in the staff paper: i) elimination of the CFF; or ii) substantial amendments to the facility, limiting it to cases where an arrangement is in place or no other BoP problem is present”. The conclusion reached was that “we leave the CFF for now, pending the broader review (of all facilities), on the understanding that if it is decided to retain the CFF in the context of that review, there is strong sentiment for modifying and streamlining it along the lines the staff has proposed.”¹⁰

In summary, the CFF may require considerable adjustments if the facility were to be used as intended by the Decision. At the very least, these include: i) extending the coverage from cereal import costs to all basic foods; ii) relaxation of the BoP and other adjustment conditionality, in recognition that shocks are exogenous and temporary; iii) uniform eligibility criterion for all LDCs and NFIDCs, in line with the Decision; and iv) raising the rate of concessionary financing for making the facility more attractive, e.g. closer to the IMF Poverty Reduction and Growth Facility (PRGF).

Agricultural export credits

The Decision envisaged to “ensure that any agreement relating to agricultural export credits takes appropriate provision for differential treatment in favour of LDCs and NFIDCs”. Negotiations on export credit Understanding covering agricultural products have been taking place at the OECD for some years. No agreement has been reached as yet, and therefore it is not known how the “differential treatment” called upon by the Decision would be provided for in the eventual agreement.

A recent study by the OECD Secretariat provides some information on agricultural export credits.¹¹ It showed that agricultural export credits provided by OECD countries increased markedly in recent years, from US$5.5 billions in 1995 to US$7.9 billions in 1998. The subsidy element in these credits was generally small, about US$300 million in 1998. Second, it was found that most export credits were used for basic foods, bulk cereals being the main followed by vegetable products (including oilseeds and wheat flour) and livestock products. Third, as regards recipients, both LDCs and NFIDCs received very small shares of agricultural export credits (0.2% and 9% of the total, respectively). One of the implications of this finding - also stressed by the OECD study – was that current agricultural export credit programmes do not necessarily target countries facing liquidity constraints to purchase food in world markets, as is often claimed for their justification.

Reverting back to the Decision, which calls for “differential treatment” for LDCs and NFIDCs in the eventual agreement, two points may be noted in particular. First, special attention may be required on monitoring the share of the total agriculture export credits flowing in favour of LDCs and NFIDCs – current shares are very small. Second, since it is the subsidy element in credits, and not necessarily the volume of credit itself, that represents “assistance” in the true sense of the term, some attention is required to ensure that the rate of the subsidy remains most favourable for LDCs and NFIDCs, e.g. a form of most-favoured nation treatment (e.g. best credit term given to any country by a donor).

IV. Implementing the Decision: Two Alternatives

In what follows, two alternatives are discussed, one under which the Decision is implemented by donors/agencies in a decentralized fashion as currently but with some strengthening. The second is to create a special Fund for the purposes of the Decision.

Alternative 1: Strengthening the Current Mechanisms

Each of the three short-term response mechanisms represents transfers from the donor to the recipient country. Some of them function multilaterally, e.g. the IMF’s CFF and World Food Programme’s food aid, while others work bilaterally, i.e. export credits and food aid under the FAC. Each has a different time frame for disbursement and eventual repayments, if required, and different accounting procedures.

In order to relate the contributions under the various programmes and the needs of recipients under the Decision, some centralized accounting and reporting would be required. After all, the three sources of assistance all have their own objectives to meet in addition to those stemming from the Decision. The IMF facilities are for IMF members facing various types of BoP difficulties. Food aid is given for a variety of other reasons not just exceptional import bills (e.g. nutrition programmes, emergency assistance). Export credits are given to different commodities on different terms for different reasons. To keep track of all these different forms of assistance and how they relate to the Decision requires some sort of mechanism to inform the relevant WTO Committee of actions taken in response to the Decision.

At the very least, all donors would have to report their assistance on meeting exceptional food import bills to some central secretariat whose job it would be to standardize the data on a common reporting basis, expressed in comparable units of value (e.g. the grant element) compiled by donor and recipient and assessed against the estimated excess food import bill. This secretariat would report regularly to the WTO Committee on Agriculture. Box 4 illustrates a format for reporting this information. Using this approach, it is to be hoped that the shortfalls in assistance for exceptional food import requirements would be identified early enough in the year for donors to adjust their levels of assistance as appropriate. This format is also useful to identify areas where improvements may be required, in the sense of fully matching assistance to the needs.¹²

A strong reporting mechanism that provides information on assistance from the three sources against the needs, as in Box 4, should gradually lead to improvements as the gaps in the assistance are identified and implementation difficulties noted. Yet, there is no guarantee about the predictability of resources, i.e. that in their totality the resources made available through them would be adequate and timely to meet the needs arising under the Decision. Moreover, this alternative is unlikely to lead to binding commitments at the WTO but only amount to "best endeavour " commitment, as it is now.

It is for these weaknesses in the decentralized approach that has led several countries to prefer a separate food security facility with commitments bound in the WTO and a disbursement system that results in an equitable share of the exceptional food import bills being met. This is discussed next.

Alternative 2: A Special WTO Food Security Facility

This alternative is to create a separate facility, say a WTO Food Security Facility (FSF), solely for the purpose of implementing the Decision. Although details would have to be worked out, the FSF could operate in general as follows.

The FSF to be a centralized, implementation machinery

In contrast to the current, decentralized approach, the FSF would be a centralized machinery specially created for the purpose of the Decision. It would have its own resources to draw upon as necessary and an administrative machinery for implementation. In addition, unlike at present, assistance commitments would be bound in the WTO.

Financial requirements of the FSF

The quantification of excess food import bills in Section II provides a basis for calculating the rough order of magnitude of the resources required for the Facility. The mean annual level of compensation was estimated to be roughly US$600 million. Assuming that the reform process lasts for six (ten) years, the total requirement would be US$3.6 billion (6 billion). Donor commitments can be both in cash and kind, e.g. food aid, as assistance is expressed in grant equivalent basis.

Eligibility and the level of compensation

The first task for implementing the proposal would be for the FSF to verify that the country was eligible to draw on the resources of the Facility up to its ceiling entitlement for that particular year. This requires the computation of excess import bills for each beneficiary country.

In order to avoid a situation where excessive drawings in some year could undermine the viability of the Facility itself, it would probably be necessary to apply a cap on the level of compensation for each country. That ceiling could be determined by, for example, a) the volume of food imports in a reference period and the increase in current world prices over some threshold amount, or b) the value of food imports in the reference period.

Should the FSF fully control the resources?

The full control of the resources by the FSF is the key underlying feature of this alternative. However, as a variation to this alternative, it may not be absolutely essential that the FSF controls the resources physically. Resources may be held elsewhere, as in Alternative 1, but these would have to be committed for the purpose of the Decision in order for the FSF to use them as needed. This arrangement essentially requires that commitments are bound in the WTO.

Reporting

Since the sole objective of the Fund is to implement the Decision, it would report directly to the Committee on Agriculture which has the responsibility of monitoring the implementation of the Decision.

¹ See The Food Situation in the Least-Developed and Net Food-Importing Developing Countries, Commodities and Trade Division, FAO, 1999.

² The Decision also envisaged a fourth response mechanism, the provision of technical and financial assistance to the LDCs and NFIDCs to improve their agricultural productivity and infrastructure. This provision addresses food problems of a longer-term nature, in contrast to the other three mechanisms. By also including this instrument in the list, the Decision rightly draws attention to the importance of addressing agricultural productivity and growth in these countries. But as the focus of this Note is on short-term problems, this instrument is not covered in the analysis.

³ Trend levels were computed by fitting linear trend lines to annual food import data covering 1985-98. An alternative and widely used proxy for trend values is moving averages.

⁴ Thus, essentially, excess import bills up to 5% (and 10% in the second case) are excluded from the calculations. Negative deviations from trends are also ignored as these are not the problems addressed by the Decision.

⁵ Data were not available for two other LDCs.

⁶ Annex Table 1 provides information on food import capacity of these countries, in terms of the ratio of food import bill to exports of goods and services, adjusted (subtracted) for debt repayments.

⁷ Review of the Compensatory and Contingency Financing Facility (CCFF) and Buffer Stock Financing Facility – Preliminary Considerations (Staff Review), 9 December 1999, available in http://www.imf.org/external/np/ccffbsff/review/index.htm.

⁸ The distinction between short-lived and long-lived shocks is important in this context. The scope for commodity stabilization in long-lived shocks is limited. Notwithstanding the empirical findings in several recent studies that shocks to commodity prices are typically long-lasting, the duration of shocks in the world market prices of grains and other basic foods (e.g. sugar) are relatively shorter. Moreover, price spikes in world markets are typically of much shorter duration than troughs.

⁹ Box 2 of IMF (1999), op. cit.

¹⁰ Summing Up by the Acting Chairman – Review of Compensatory and Contingency Financing Facility (CCFF) and Buffer Stock Financing Facility – Preliminary Considerations, Executive Board Meeting 00/5, 14 January 2000, also available at the same IMF web site as above.

¹¹ An Analysis of Officially Supported Export Credits in Agriculture, Document COM/AGR/TD/WP (2000) 91/Final, 2000, OECD, Paris. Available online at http://www.oecd.org/agr.

¹² Currently, the WTO Committee on Agriculture monitors the implementation of the Decision by treating the instruments separately. Each relevant donor and agency provides information on its assistance programmes, including the amount of the assistance given to the LDCs and NFIDCs. This information is then compiled by the WTO Secretariat. But no attempt is made to identify and quantify country needs as stipulated by the Decision nor to verify if the total level of assistance provided was adequate or even exceeds the needs.