UNCTAD paper reveals distortions of Green Box subsidies (Article 1 of 3 articles)

8 February, 2007

The current efforts towards a breakthrough in the stalled Doha negotiations at the World Trade Organisation have focused firstly on getting the United States to make a new offer to put a maximum limit on its total "trade-distorting" domestic support (TDS) in agriculture that is more acceptable than its previously announced offer.

According to various media reports, the US has indicated its willingness to have a cap on overall TDS at about $17 billion. This is an improvement from its previous offer of $23 billion, but hardly significant when compared to the actual spending (i. e. the applied level) of $19-20 billion in its overall TDS in 2005.

If this offer is finally put forward, it means the US is pledging to bring down its actual spending on TDS by some $2-3 billion. That is indeed a small amount. It is actually surprising that such a small offer had not been made much earlier; the talks might then not have gone into "suspension" mode.

In exchange for this meagre "concession", the US is demanding that its corporations obtain big benefits through opening the markets of others. The developing countries are asked to pay a very heavy price by cutting their agricultural tariffs by much more than they did at the Uruguay Round, and by slashing their industrial tariffs as well as opening up key service sectors to foreign firms.

This already obviously imbalanced proposed package is actually even more unequal. For the US (as well as the EU, Canada, Japan and other developed countries) will be able to escape from really cutting their agricultural domestic support by the mechanism of shifting more of their subsidies to the "Green Box".

The Green Box (GB) subsidies are supposed to be non trade-distorting, because they are said to have no or minimal effect on the market or on world agricultural trade. This is unlike the "trade-distorting" subsidies in the Amber Box, the Blue Box and the de minimis category, which are linked to production and to market factors such as prices.

Present WTO rules discipline the trade-distorting subsidies, but there are no disciplines to cap or reduce the GB subsidies. In the current Doha talks, the developed countries have successfully insisted that there be no new rules that would place a cap on the GB subsidies nor to have disciplines (as there are for the other subsidies) to reduce their levels.

However, developing countries in the WTO have long been suspicious of the claim made by the developed countries that the GB is "non trade distorting". They have expressed concern that there will merely be "box-shifting" by the major subsidising countries, i. e. reducing subsidies that are "trade distorting" (and thus subject to caps and reduction) while increasing their subsidies in the Green Box.

Indeed, the US has already shifted the bulk of their domestic support to the Green Box, and the EU has planned to do the same in the next few years under their Common Agricultural Policy reform.

Having lost the opportunity to have caps or reductions to the GB, the developing countries have put on the table some proposals to tighten the criteria for the GB, so as to limit the type and number of subsidies that can be placed in this category. But it remains to be seen whether these proposals - mild as they are - will be acceptable to the developed countries.

While the suspicion is strong among developing country diplomats that the developed members of the WTO are largely engaging in a magician's trick of box-shifting without in fact reducing their subsidies and their effects, there has up to now not been a comprehensive detailing of how the Green Box distorts trade, nor empirical evidence of the extent of distortion.

Recently, the UNCTAD team in India issued a paper describing the ways in which GB subsidies distort the agricultural trade and production system. It also provides estimates showing the large extent of this distortion.

The figures from this study are so significant that they should lead to a basic re-thinking of the nature of the Green Box and the claim that its subsidies are non trade-distorting. Moreover, in the light of this study, the kind of offers being made by the US and EU to cut the "bound levels" of their overall "trade distorting" domestic support appears to be merely symbolic or deviously deceptive - pulling wool over the eyes of the world public which are told that the major developed countries are making big sacrifices, while serving to pressurise developing countries to make very serious concessions in exchange for illusory gains.

Even if the US were to meet the demands of the Group of 20 (that the US cap be at $12 billion), it would not be a significant offer as the US would only have to cut its actual "trade distorting" support by a few billion dollars while it would still have recourse to increase its Green Box subsidies without limit.

If the US and the EU were to maintain and indeed expand their Green Box subsidies, this would have far reaching effects on world agricultural trade and on the agricultural production and trade of developing countries.

The paper, Green Box Subsidies: A Theoretical and Empirical Assessment" was prepared by the UNCTAD India Team under an on-going trade research project jointly hosted by the Indian Commerce Ministry, UNCTAD and the United Kingdom's international aid department (DFID). The paper has also undergone a "critical appraisal" by several outside economists, including the Nobel Laureate and former World Bank chief economist Joseph Stiglitz.

According to Dr Veena Jha, coordinator of the UNCTAD India programme, the study was undertaken at the request of the Indian Commerce Department. "Using different econometric approaches, the paper shows that the Green Box subsidies are production distorting and trade distorting," she said.

Dr Jha added that reducing the Green Box subsidies would have positive effects on developing countries' agricultural sector. In addition, the reduction would lead to poverty alleviation as it would have positive effects on the most vulnerable sections of society. "The paper recommends that the current negotiations address the issue of eligibility criteria of Green Box subsidies in developed counties on an urgent basis with a view to restricting them," she concluded.

The study reiterates that the WTO's Agreement on Agriculture was aimed at globally reducing trade-distorting subsidies, but measures identified as having no or minimal trade-distorting effects were to be categorized as Green Box measures, and these were exempt from reduction commitments and these subsidies could even be increased without any limitations under the WTO.

"However, recent research shows that current Green Box subsidies do not meet the criterion of 'no or at most minimal' trade-distorting effects and that the so-called 'decoupled' programmes under Green Box do in fact distort trade," says the report.

While there is sufficient theoretical justification for the trade-distorting effects of Green Box, little empirical work has been undertaken to substantiate this claim. Using computable general equilibrium modelling, that is, the Global Trade Analysis Project (GTAP), and Data Envelopment Analysis (DEA), the paper shows empirically that Green Box subsidies do have significant distortive effects on trade and production.

The paper's most striking conclusion is that if the Green Box (GB) subsidies were reduced or removed, the developed countries such as the US, EU, Canada, Switzerland and Japan would suffer major declines in their agricultural exports. Their production would also be reduced.

The study points out that Annex 2 of the Agreement on Agriculture sets out a number of general and measure-specific domestic support criteria which, when met, allow measures to be placed in the GB. They must be provided through a publicly-funded government programme (including government revenue foregone) not involving transfers from consumers and must not have the effect of providing price support to producers.

However, says the report, recent research shows that in some instances current GB subsidies do not meet the criteria of 'no or minimal trade distortion'.

"Studies have shown that the so-called 'decoupled' programmes under GB could distort trade. A large amount of money paid to farmers, decoupled with current production, nevertheless is likely to distort trade and production because of the wealth and risk effects associated with it.

"This problem is further compounded as these payments are not transitory measures, and are therefore permanently incorporated into the cash flows of farmers, thereby increasing their creditworthiness and serving as an instrument for hedging against risk. In addition, the practice of updating base acres, number of heads and payment yields, as well as changing eligible crops under Farm Assistance Programmes, tend to raise expectations of future assistance thereby influencing their future production decisions.

"Other important channels through which direct payments and insurance programmes can affect output are through their effects on capital and labour markets. Programmes that reduce income variability can increase farm investment by lowering the risk of loan default, thereby increasing rural credit availability. Investment aid takes different forms in different countries. In France and Germany it is used to a large extent to subsidize interest rates for farmers. French farmers, for instance, paid 274 million euro less in interest in 2003 than they would have done without the assistance."

The paper examines the theoretical debates around GB subsidies, particularly the various ways in which such subsidies might have a distorting effect on trade and production.

The two broad categories of GB subsidies are Government service programmes and Direct Payments to producers.

The Government service programmes include general services provided by governments, public stockholding programmes for food security purposes and domestic food aid. The GB thus provides for the continuation (and enhancement) of programmes such as: research, pest and disease control, agricultural training and extension, inspection services, infrastructural services, holding of public food stocks for food security purposes, and domestic food aid.

The GB provides direct payments to producers which are not linked to production decisions on the type or volume of agricultural production (this is termed as 'decoupling'). This precludes any linkage between the amount of such payment, on the one hand, and production, prices or factors of production in any year, after a fixed base period. In addition, no production is required in order to receive such payments.

Measures may include decoupled income support measures; income insurance and safety- net programmes; natural disaster relief; structural adjustment assistance programmes; and payments under environmental programmes and regional assistance programmes.

The paper stresses that recent research has identified various ways through which GB subsidies can affect agricultural trade and production through wealth and risk effects, cost effects, insurance effects, expectation effects and productivity increases.

(1) Wealth and risk effects: Direct payments, used for decoupled income support, are exempt from reduction commitments, since they are supposed to be lump sum transfers with no effect on production decisions. Further, since direct payments are based on a past, fixed base period, farmers cannot affect payment size through current behaviour. Therefore. their current production decisions could only be based on market considerations. This is the rationale behind compensation programmes such as the 'production flexibility contract (PFC) payments' in the US and Canada's WGTPP (Western Gains Transition Payment Programme).

However, says the paper, the mere fact that a decoupled support is not linked to the current volume of production however does not make it non-trade distortive. Income support may influence production decisions in a manner that could be inconsistent with the principles of free market and fair competition. Recent estimates (by Bouet et al 2003) suggest that 'decoupled' payments do have a positive impact on output, given their role on reducing risk.

The effect of a fixed direct payment on production is strongly determined by the risk behaviour of the producer. For a risk-averse producer, the direct payment can produce a wealth effect. Studies show how the increase in wealth created by a direct payment might increase the farmer's capacity to take risk or expand production by planting crops that would otherwise be viewed to be too risky. Policies that affect initial wealth, impact profits both directly and indirectly - the problem is accentuated when decoupled payments are provided in conjunction with Amber Box or Blue Box subsidies.

A study by Chavas and Holt (1990) found that both wealth and risk perceptions were important determinants of acreage allocation decisions for corn and soybeans. Under some conditions, the wealth effect was even more important than the direct price effect on acreage planted. In another study, Roe, Somwaru, and Diao (2004) found that PFC payments increased land values which in turn could lead to increased production due to increased access to credit. This finding that direct payments might lead to increased production is supported by a number of other studies on PFC payments.

(2) Insurance Effects: GB subsidies such as government-provided safety nets can act as insurance and affect production. A safety net stabilizes variables such as farm income by assuring a minimum income level which acts as a floor.

When safety nets exist, expected net income might create an incentive to increase production. Farmers who are protected by insurance from downward income fluctuations have more risk taking capacity in agricultural production related decisions. Further, the level of the floor also affects the levels of production.

Crop insurance also may encourage producers to move risky marginal land into production and the crop mix may be biased towards production of more risky crops. Agricultural incomes being subject to instabilities and fluctuations, risk-averse farmers may benefit considerably from income stabilization measures such as counter-cyclical subsidies. These income-stabilization programmes and fixed payments may have a risk- reducing effect.

(3) Cost, price and expectation effects: There are situations where the direct payment can indirectly affect the decisions of risk neutral producers as well. Such payments help producers to overcome their credit constraints.

There are other potential effects on production such as increases in production executed in anticipation that the higher level of production now will form the new base for a new programme (expectation effect). For instance, the 2002 Farm Security and Rural Investment Act in the US allowed farmers to update their base acres from the 1981-85 planting history to a more recent (1998-2001) production level. The subsidy can, among others, affect the decisions to exit or enter.

A study by Rude (1998) shows how production might be affected by GB payments in three ways. The first is placed in an increasing returns to scale scenario. A fixed direct payment would increase average revenue lowering average cost and increasing level of production.

The second is in the context of behavioural theories of the firm. Direct payments loosen constraints and encourage continuation of the status quo, possibly increasing production. The third is in situations where the producer faces debt constraints. The payment leads to increased future production in the debt constrained case.

(4) Productivity effects: GB measures could also have a distorting effect on production through increasing productivity. For instance, measures such as research and extension services, conservation, environmental and natural resource programmes, structural programmes to adjust farm size and numbers, infrastructure under GB may have production effects on long-term agricultural production. These programmes cause productivity improvements. If the GB provider is not a small country then, world prices of agriculture products can decline with possible augmentation of global production.

The UNCTAD paper points out that the criteria specified in Annex 2 of the Agreement on Agriculture on classifying GB subsidies try to ensure they are decoupled from production and hence have no or minimal trade and production distorting effects.

"But a review of the literature shows that this may not necessarily be correct as they may influence the risk behaviour of the producer, have an impact on productivity increases or reduce the cost of production."

These issues are analysed empirically in the rest of the paper.

(This is the first of a series of articles in the SUNS on the Green Box. A second article will provide data from the UNCTAD India paper showing the extent to which the Green Box subsidies distort global output and trade.)