An important objective of agricultural policy in less-developed countries across the world is to promote food security for the national populations. Widespread hunger, undernourishment, and malnutrition in less-developed countries have meant that providing access to food continues to be a key policy objective for these countries. According to the Food and Agriculture Organisation (FAO), of the 800 million people in the world who were undernourished in 2012–14, 791 million lived in less-developed countries. A number of factors have aggravated food shortages and access to food in these countries – stagnation in farm production, low farm productivity, low levels of food stocks, and volatility in international commodity markets, as well as currency fluctuations, weather-induced fluctuations in output, and low levels of incomes. It is against this backdrop that the book under review examines a key additional – and imminent – threat, the threat that international trade rules established under the auspices of the World Trade Organisation (WTO) pose to food security in less-developed countries.

One of the most important components of the WTO framework of international trading rules has been the Agreement on Agriculture (AoA). This treaty, negotiated during the Uruguay Round of negotiations of the General Agreement on Tariffs and Trade (GATT), was, in 1995, among the first treaties to come into force within the WTO regime. From the perspective of less-developed countries, the rules framed under the AoA have been the most contentious aspect of the WTO regime, not only because of their implications for the ability of countries to ensure food security, but also for their asymmetric design, which has different impacts on less-developed and developed countries.

The structure of the AoA rests on the premise that the international trading regime for “a fair and equitable market oriented agriculture trading system” should favour certain kinds of state support to producers and consumers. Consistent with this approach, the AoA mandates that certain types of support qualify to be in the Green Box, i.e., forms of support that countries can provide without any limits. This type of support was justified on the grounds that it did not directly distort trade and production. For example, state support for general services in agriculture, research and development, environmental protection, and food aid qualify for the Green Box. Decoupled income support, or state support that is distanced from immediate production and trade, qualify for this box as well. On the other hand, state support for programmes that aim to limit production are placed in the Blue Box.

But the most contentious aspect of the AoA pertains to the Amber Box, which sets limits on state intervention and support that is aimed at ensuring stable and remunerative prices for producers, while also enabling the state to supply food at fair prices to consumers. This is typically achieved by setting a floor price, state intervention in procurement, and organising the distribution of foodgrains to consumers. However, the binding commitments that countries have entered into under the provisions of the AoA threaten such modes of state intervention. Less-developed countries have long demanded a review of these rules, especially those that set limits on state expenditure on procurement, stockholding, and distribution, all of which fall in the Amber Box.

The AoA sets limits for forms of support in the Amber Box. These are set out either in terms of a de minimis level that is prescribed in terms of the value of national agricultural output, or a negotiated Aggregate Measurement of Support (AMS), which was set in value terms for each country during the Uruguay Round negotiations. Most less-developed countries, however, do not have the flexibility that is characteristic of the AMS mechanism, and are therefore required to contain support levels below de minimis levels. Limits to state support are crop-specific and for overall support to agriculture; specifically, less-developed countries are allowed to contain both product-specific and non-product-specific support levels within 10 per cent of the value of national output. Crucially, and somewhat curiously, the de minimis limits in the Amber Box are measured using a fixed External Reference Price (ERP) that is based on 1986–88 prices.

The book under review is significant for the manner in which it quantifies the extent to which a group of 12 diverse countries – five Asian countries (China, India, Pakistan, Indonesia, and Jordan), six African countries (Egypt, Morocco, Tunisia, Kenya, Zambia, and Zimbabwe), and one European country, Turkey – have breached, or are in danger of breaching, WTO-mandated limits on state support for specific food crops (mainly wheat and rice). Over half the undernourished population in the world lives in these countries, making Sharma’s selection immediately relevant in terms of vulnerability to shocks caused by the withdrawal of state support in order to comply with WTO rules under the AoA. In five countries – India, Indonesia, Pakistan, Zimbabwe, and Kenya – hunger levels were categorised as “serious,” according to the Global Hunger Index for 2014, and in Zambia the level was “alarming.” The hunger level in China was categorised as “moderate,” and “low” in the others countries selected by Sharma. Significantly, all 12 countries have some state-run programme that ensures a minimum floor price, and most of these countries also have a distribution mechanism that delivers foodgrains to consumers at controlled prices. This is important because Sharma demonstrates that these programmes may be in jeopardy as many of these countries may have already breached – or are on the verge of breaching – the limits set by WTO rules.

The first three chapters of the book explore the background to the AoA and food security in the countries under discussion. Chapter 1 provides an introduction to the problem, focusing on how the Doha Round of negotiations at the WTO have addressed food security concerns of less-developed countries and how these concerns in less-developed countries have been addressed by providing price support to farmers. Chapter 2 is devoted to the AoA of the WTO and its provisions, which set limits on the extent of state support to agriculture. Chapter 3 examines the extent of food security in select countries, while also studying the modes of state intervention in agriculture in order to promote food security. Chapters 4–15 are each devoted to a country that is examined in detail. Each chapter estimates the flexibility a country has under the limits prescribed by the WTO. Sharma considers these constraints in the context of the extent of undernourishment in the country under discussion. He provides a brief summary of the historical evolution of state support policy as well as of the support apparatus in each country, thus giving an idea of what is in danger of being dismantled as a result of non-compliance with WTO rules. The hallmark of the volume is that it provides a measure of the policy space each country has before it violates WTO rules. However, the lack of an extensive bibliography is a limitation of the book, although there is a brief list of references at the end of each chapter. This limitation is serious especially because there is an extensive literature that is available on the broader issues pertaining to international trade in agriculture and to the positions adopted by countries in negotiations at the WTO.

Although procurement data for China are not available after 2008, Sharma projects different scenarios by assuming a range of proportions of national output of wheat procured – from 25 per cent to 40 per cent – through state support. He estimates that if 25 per cent of wheat output in China was being procured, China would have breached its de minimis limit of 8.5 per cent of total value of support for a specific product under the WTO rules in 2016. If the Chinese government had procured higher proportions of national wheat output, the de minimis limits may have been breached as early as 2013 or 2014. Similarly, support for maize in Kenya, and for wheat in Pakistan were both above the de minimis limit; Kenya having breached it in 2000 and Pakistan in 2008–9. Although Jordan has claimed that its inflation-adjusted support levels for barley and wheat are within the bounds set by the AMS in 2009–10, the claim has since been challenged by exporting countries such as Australia and Canada. However, Sharma states that Jordan, which is highly dependent on foodgrain imports, may be unable to provide price support in order to safeguard food security.

In the case of India, product-specific support to wheat, which was 132.37 per cent below the de minimis limit in 1995–96, narrowed to less than one per cent (minus 0.73 per cent) in 2010–11. The situation has also changed, but more dramatically, in the case of rice. For product specific support for rice, the level of support was 55.4 per cent below (that is, minus 55.4 per cent) the de minimis limit in 1995–96; by 2010–11 it was over the limit by 7.21 per cent. Significantly, although the dollar-denominated external reference price for rice remained constant between these two time points – (at USD 262.51/tonne) – the applied value of administered price incurred locally in Indian rupees but denominated in dollars increased from USD 168.92/tonne in 1995–96 to USD 329.11/tonne in 2010–11. The administered price of rice remained below the reference price between 1995–96 and 2006–7, but breached the reference price after 2007–8. Sharma observes, “Given the trend in administered price and the level of procurement, it is feared that India would breach the de minimis limit of rice in the near future (p. 80).”

What this means in the immediate term is that procurement agencies such as the Food Corporation of India (FCI) would find it difficult to procure foodgrains, particularly rice, because the country would be breaching the de minimis limits set at the WTO. This is because the administered price of rice has already breached the ERP. Sharma notes that “this would severely constrain the government in fighting hunger and undernourishment affecting 190 million people in India (p. 80).”

It is evident that in the case of both Egypt and India, setting a dollar-denominated limit has played a major role in these countries breaching the limits set in the AoA. Although this concern is mentioned only in passing in several of the 12 cases that Sharma examines, he does not focus on this particular issue, especially its methodological and conceptual fallacies. This is critical because if the limits are defined in dollar terms, and if less-developed country currencies have depreciated significantly (as they have since 1995), the problem of being in danger of breaching the limits set in the AoA arises from a serious methodological flaw that requires urgent correction. In short, the rigidity of the ERP set in 1995 using a benchmark that was almost 10 years old even at the time is a serious flaw in design. The use of a reference price, without reference to the rupee-dollar exchange rate that has been used to measure the extent of support provided by the Indian government in rupees, is a major contributory factor to countries that are in danger of breaching the limits.

The use of a reference price also poses serious methodological problems because of the conversion of currencies. Wise (2004) has argued that the setting of a reference price is problematic for several reasons. First, as he demonstrates from the case of using milk prices from New Zealand as a reference for global export prices, the choice of the reference itself can be unrepresentative. Secondly, the two-step method of currency conversion – first of the reference price into the domestic currency, and then of the extent of price support in national currency converted back in terms of the reference currency – can significantly distort the actual measure of support provided to producers. These errors are likely to be magnified during a period of great currency volatility, as has been the case with many national currencies since the global financial crisis of 2007–8.

Wise observes:

A measure intended to estimate support to producers should take account of the buying power of that support, since it is a measure of opportunity costs to domestic consumers at domestic prices. Exchange rates tend to undervalue the purchasing power in lower income countries, because many inexpensive services are not internationally traded (Wise 2004, p. 9).

Dhar and Kishore (2014) have pointed out that while the ERP benchmark set in 1986–88 prices may have been relevant in 1995 when the WTO came into being, 30 years later, they have lost significance. They cite the case of rice prices in the international market to make their point:

For instance, as against the external reference price of USD 262 (per tonnes) that India notified for rice, international price of this commodity has been well over USD 500 (per tonne) in most of the recent years and had even breached USD 1000 (per tonne) during 2008 (ibid., p. 9).

Moreover, they argue that the value of food stocks acquired by the government ought to be adjusted for inflation, given the relatively high levels of inflation of the prices of food articles in less-developed countries. They also observe that in the run-up to the AoA, less-developed countries were given to understand that the reference price would be applicable for a certain time period and would be periodically reviewed. They note that “[the] external reference price, which was taken as the de facto internationally competitive price, therefore no longer remains the numeraire.”

While the illogical application of the ERP can have critical consequences for a marginal food importing country such as India, it can have a much more severe impact in a country like Egypt, which is heavily import-dependent, especially for wheat, the staple food crop. Although Egypt’s public procurement system became active relatively late compared to India, its state agency, the General Agency for the Supply of Commodities (GASC), now procures a significant proportion of the national wheat output. While the GASC procured 12.3 per cent of the wheat output in 1995–96 (roughly the time when the AoA came into effect), by 2013–14 it was procuring 46.2 per cent of wheat produced in Egypt. Although import-dependence has increased significantly – from about 38 per cent in 1990–91 to almost 46 per cent in 2009–10 (the figure cited by Sharma) – the procurement of a higher fraction of the national wheat output offers Egypt a measure of flexibility in handling periods of national shortage, high international prices, or volatile currency movements that would hamper the ability to procure adequate quantities from overseas markets. Sharma notes that Egypt breached the de minimis limit as early as 1995–96, that is, when the WTO came into effect.

Turkey, the only member country from the Organisation of Economic Cooperation and Development (OECD) in the list of 12 countries, may at first appear to be an outlier. But it has good reason to figure in the list for two reasons. First, agricultural productivity in Turkey is the lowest among all 35 member countries of the OECD, and coupled with the “predominance of small-scale and semi-subsistence family farms,” indicates why Sharma regards the only European country in his list as important from the perspective of food security. Secondly, although the country is a net exporter of agricultural products, its dependence on cereal imports has increased significantly. The cereal dependency ratio in Turkey increased from about 5 per cent in the early 1990s to more than 13 per cent by the end of the last decade. Despite this growth, the provision of a protective tariff barrier and welfare measures allowed Turkey to prevent this dependence from affecting the standard of living in the country, reflected in the relatively low levels of undernourishment and the depth of the food deficit (defined in terms of calories per capita per day).

The state-owned Turkish Grain Board (TMO) was established in 1938, but plays a wider role than the Food Corporation of India (FCI). It not only procures grains such as wheat, barley, rye, oats, and rice, but also sets procurement prices annually, with a provision that guarantees procurement at those prices to farmers who are not able to sell their grain at a higher price. More critically, Sharma notes, the TMO plays an active role in “limiting competitive imports” of grains such as wheat, rice, and corn. Although product specific support for wheat in Turkey, as a proportion of the value of production, breached the 10 per cent de minimis limit only in two years (2005 and 2009), Sharma warns that the food security apparatus in the country, built over many decades, may be in jeopardy. Turkey may violate the limits imposed by the AoA either because it may have to offer higher prices to farmers or if it procures greater volumes of produce. Sharma notes that in either case this may “adversely affect the ability of the government to address hunger by procuring wheat at administered prices.”

The asymmetry that is built into the structure of the AoA is best illustrated by the manner in which it treats state support levels in developed countries and in less-developed countries. Sharma notes that although the United States provided food aid valued at USD 106.8 billion through the Supplemental Nutrition Assistance Program (SNAP), popularly known as the Food Stamp programme, the programme did not qualify as a trade-distorting means of state support. This meant that the SNAP remained in the Green Box. However, less-developed countries that seek to achieve the same objective via a different route – through state-led price support – bear the brunt of restrictive rules under the AoA.

Dhar and Kishore (2016) point out that the United States moved away from an agriculture policy that focused on price support to one that emphasised income support to farmers. This, they argue, was done with the specific objective of appearing to comply with WTO regulations that deemed price support as being trade- and market-distorting. However, the US Farm Act of 2014 has brought back price support through commodity subsidies that provide “price assurance,” protect farm revenues, and set floor prices for major agricultural commodities. They argue that the SNAP, while offering a measure of food security to the most vulnerable in the United States, has historically been used to perform a market-clearing function in a “surplus-ridden food economy.”

They note that

SNAP is creating the demand for providing an opportunity to the farmers to liquidate their stocks of food, and this should be considered as a commercial operation. However, the US has labelled its domestic food aid as a non-market distorting form of subsidy and has included this component as a Green Box measure (Dhar and Kishore 2016, p. 40).

They also state that “the growing importance of farmers’ markets in the US, which have created a direct link between local agricultural production and government benefits under SNAP,” clearly imply that this kind of support ought not to qualify as support permissible under the Green Box criteria.

Sharma blames the “unfair, outdated, and illogical provisions of the AoA” for the plight of the countries under discussion. He urges countries that are similarly placed to adopt a common negotiating position in order to replace the existing rules with ones that allow the implementation of food security policies that are not constrained by rules set by the WTO.