Sri Lanka has rich experience in pursuing a variety of strategies to achieve food security. They range from the projectionist policies implemented mainly during 1948 to 1977, to open economic policies implemented since 1977. For the paddy/rice sector, the government continues to restrict rice imports due to fears that rice trade liberalization would have adverse impacts on poverty. On the domestic front, however, paddy procurement, milling and distribution are mainly done by the private sector. The production of paddy is mainly carried out by small farmers and, on an average, resource-use is found to be inefficient. While some researchers believe that the paddy-rice marketing system is competitive, many consider that it is characterized by a group of oligopolistic buyers. Finally, on the consumption side, the demand for rice, the staple food, is found to be price and income inelastic.

The overall objective of this research project is to assess the impacts of different types of government policies (pertaining to the domestic and external sectors) on the status of food security in the country, with particular emphasis on the paddy/rice sector. Specifically, the goals of this study are:

1. To describe the present status of paddy production, procurement and distribution system in Sri Lanka, paying special attention to the involvement of government agencies and private sector.
2. To document the evolution of domestic and trade policies affecting the above system showing the extent of liberalization over time.
3. To examine Sri Lanka's position on the Agreement on Agriculture in the WTO and its likely impact on paddy/rice sector.
4. To investigate the impact of rice trade liberalization and privatization of paddy procurement system on prices, supply of paddy, demand for rice, imports of rice and calorie intake at the national level.
5. To investigate the impact of rice trade liberalization at the household level, and for various groups in the population, with a view to understand the implications for poverty.
6. To investigate the likely impacts of elimination of oligopsony power of the paddy collectors on the well being of paddy farmers.

Following the introduction, the second chapter of the study provides a detailed background about paddy/rice in Sri Lanka from the point of view of production, storage, distribution channels and international trade. Some of the important information in the case of paddy and rice has been presented below:

1. Paddy occupied the greatest land under agriculture of all the crops in 2001, contributing 15 percent to the agriculture sector GDP. Seventy percent of paddy farmers are smallholders (with a land area of less than one hectare) and the crop is heavily dependent on rainfall.
2. After milling paddy is known as rice, which is the staple food in Sri Lanka, and the main source of calories in the people's diet.
3. Due to the significance of the paddy sector in the Sri Lankan economy, governments have placed a great emphasis on increasing paddy production in order to achieve self-sufficiency. Large-scale irrigation projects and land development and settlement schemes were undertaken and the government also provided free irrigation water, fertilizer subsidies and guaranteed prices to farmers.
4. Accordingly, productivity more than tripled from 1129 kg/hectare in 1970 to 3954 kg/hectare in 2001. At present, the country is 90% self-sufficient in rice, and imports have dropped to less than 10% of the level in 1970 although it has fluctuated over the years.
5. The government was heavily involved in the marketing of paddy and rice up to 1977. In 1977, with the implementation of economic liberalization policies, private sector participation was allowed and as a result of competition there was a dramatic reduction in the government's share in marketing paddy to just 10% in 1995.
6. With regard to storage, paddy is stored and the carry-over stocks are maintained for rice. Farmers sell about half the produce immediately to meet cash requirements and the balance is stored using a traditional system known as "Bissa", which tends to result in a high rate of post-harvest losses. Paddy is also stored by the collectors and millers.
7. Storage facilities for rice belonging to the government are very poor at present. Due to the low keeping time of rice, storage by private wholesalers also tends to be only for short periods of about a month.

The third chapter traces the evolution of domestic and trade policies affecting the food procurement, storage and distribution system in Sri Lanka over time. They are chronologically enumerated below:

1. In the 1930s and 1940s policies were directed at developing the dry zone for the cultivation of rice. However, the main problem associated with paddy was low yield due to the large number of smallholdings. Therefore, successive governments focused efforts on promoting rice production in order to achieve food self-sufficiency.
2. The universal rice-rationing scheme (RSS) was initiated in 1942 to ensure equitable distribution of available food resources. The distribution system comprised consumer co-operative societies with an island-wide network of retail shops. There were two main sources of food to meet the RSS requirements—domestic procurement and imports. The government encouraged domestic food production and strengthened the food procurement system in order to try to reduce the dependence on the international market and develop domestic agriculture.
3. Internal Purchase System (IPS) — The government introduced this in 1942 to meet the demand for rice by the RSS.
4. IPS was replaced by a scheme called -Marketing of Home Grown Foods" in 1948, in order to improve the functioning of domestic marketing channels.
5. In 1961 the government introduced the Domestic Produce Purchasing and Storage Act of 1961, in response to the greater marketable surplus available with farmers. The purpose was to strengthen the role of co-operative societies in marketing.
6. The 1960s also witnessed the setting up of several state enterprises to promote non-paddy agricultural activities. Examples are the National Milk Board, Ceylon Fisheries Corporation and Fisheries Harbor Corporation.
7. Trend in international rice prices have tended to affect international procurements. During the three decades 1950-70 the government entered into bilateral trade agreements with China and Burma to secure rice in reliable quantities.
8. From the 1970s onward, the government was confronted with high international rice prices and insufficient foreign exchange, and therefore had to procure more rice domestically.
9. The Paddy Marketing Board (PMB) was established in 1971 in order to purchase more paddy through domestic channels. Disappointingly, despite the monopoly power vested with PMB, paddy procurement fell under the Guaranteed Price Scheme to 27% of total production in 1974 from 48% in 1971. The situation did not improve even though the government instituted measures such as the prohibition of storage of greater than a certain minimum amount of paddy by farmers, and also increased procurement price. This was on account of high prices paid by the private sector for rice. The government also imposed rules whereby rice meals were to be substituted by non-rice meals in schools and public institutions.
10. The Food Stamp Scheme was introduced in 1978, which aimed at targeting the poorest households in the country. Government intervention in promoting production and marketing of food grains dropped markedly after this scheme was implemented.
11. The RSS was discontinued in 1979.
12. After 1977 there was a marked change in policies toward greater economic liberalization and private sector participation.
13. With respect to rice the changes involved allowing private sector companies to import rice from 1988 onward under a strict quota system.

In 1995 rice imports were liberalized and the licensing system was abolished and replaced with a tariff rate of 35%. This rate was reduced to 20% later that year. The tariff rate remained unchanged but waivers were granted at different rates in specified periods to curtail increasing prices in the domestic market due to production shortages in the market. In the area of procurement, PMB's procurement fell as low as 1.3% of total production for the period 1993-1995.

In the fourth chapter, the author examines Sri Lanka's position under the Agreement on Agriculture (AoA) and analyzes the likely implications. There are three pillars in the Agreement on Agriculture—market access, domestic support and export subsidies.

1. Regarding market access provisions, Sri Lanka has bound nearly 99 percent of its tariffs on agricultural products including rice at 50% under the AoA, and it has 10 years from 1995 onward to bring down these tariffs. AoA requires member countries to use import tariffs to restrict imports rather than quantitative restrictions.
2. Sri Lanka does not have commitments with regard to domestic support and export subsidies, as these are based on the base levels of such support during 1986-88 and Sri Lanka did not have any trade distorting programs during this time. Due to commitments with the AoA Sri Lanka will not be in a position to implement any new subsidies.
3. As a result, impacts of AoA on the paddy/rice sector will be negligible. If Sri Lanka wishes to implement new subsidy programs on paddy, they may be requested under the special and differential treatment.

In the fifth chapter the author uses a partial equilibrium model to assess the national level impacts of rice trade liberalization, treating Sri Lanka as a small net importer of rice. The details of the model are:

1. The consumer side is represented by a demand system for cereals (of the AIDS functional form) and the producer side by a paddy supply function. The marketing functions are captured by an equation to link paddy prices and rice prices. Government procurement is considered as an exogenous variable affecting marketing functions. A tariff barrier and other border charges link the retail price of rice with the world market price of rice.
2. The identities used in the model are as follows: Calorie intake from rice, wheat and millet were used to obtain total calorie intake. Rice production has been considered a constant proportion of paddy production, and rice imports have been considered as the difference between domestic production and consumption.
3. Changes in trade policy are simulated by changing tariff levels and by eliminating other border charges that affect the retail price of rice, and changes in paddy procurement policy are simulated by changing the proportion of paddy procured by government institutes.
4. The data used was obtained from food balance sheets and the Central Bank of Sri Lanka.
5. The findings of the simulation exercises are as follows:

With trade liberalization, calorie intake increases, suggesting greater food security. Moreover, producer prices would fall, which would lower incomes of paddy farmers and reduce their capacity to purchase food. Since most of these farmers are small holders who are below the poverty line, the repercussions would be worse. The central question is whether the gains to consumers will exceed losses to producers and the underlying assumptions of the model suggest that they will. However, this result must be viewed with the caveat that the model does not study the how the impact of lower wages of the small holders would affect their consumption.

In the absence of government procurement, paddy prices will be further depressed, as private purchasers of paddy would exercise oligopsony power on paddy sellers/producers. This is because paddy sellers need to sell their produce immediately in order to obtain cash to meet repayment obligations and because they lack storage facilities. Therefore they are in a weak bargaining position. Moreover the market structure in the case of paddy is such that there are fewer millers of paddy than suppliers, which allows the millers to exert oligopsony power.

The policy recommendations that arise from this part of the study are that rice trade liberalization should be encouraged as a strategy to ensure food security, since it lowers prices and increases calorie consumption, both of which would be directly beneficial to consumers. The government should devise some method to monitor the oligospsony power that the millers are able to exert over the farmers.

In the sixth chapter, household level impacts of trade liberalization are assessed on various population groups. The analysis is based on the method used by McCulloch (2000), and the relative importance of rice and paddy in the total expenditure and income respectively of the particular population group is estimated in order to indicate the extent to which they will be affected by rice trade liberalization. The population groups are specified on the base of expenditure class, province, sector (rural, urban, estate), and by size of land holding. The policy shock introduced in this exercise is a 25% drop in rice and paddy prices. The findings are that there is a net gain to all the income groups, provinces and sectors. The biggest gainers are the lowest expenditure, those residing in the Northwestern province and those who are part of the estate sector, where the incidence of poverty is found to be the highest. Paddy farmers with relatively bigger holdings, on the other hand, are found to be negatively affected. Therefore, overall it appears from the results that rice trade liberalization is a pro-poor policy.

Finally, this study examines the impacts of trade liberalization in imperfectly competitive markets. The objective is to assess the impacts of rice trade liberalization in the presence and absence of market power. For this purpose, a partial equilibrium model is developed for the paddy market in Sri Lanka, under oligopsony, since the bargaining power of paddy suppliers is smaller than that of paddy buyers. Using this model, impacts of trade liberalization were simulated. Results reveal that losses to paddy producers due to trade liberalization can be minimized if oligoposony power can be eliminated simultaneously. Therefore, strategies to eliminate oligopsony power along with trade liberalization are recommended to achieve efficiency gains.

The study concludes by drawing policy implications, based on the analysis and findings of the study. Some of the important ones are that rice trade should be liberalized, as it leads to an increase in efficiency and net gains at the national level. (As mentioned earlier, this result should be interpreted keeping in mind the caveat of the model, which is that the impact on the consumption of the smallholders whose wages are adversely affected by trade liberalization has not been considered). A second policy implication is that government procurement should continue but private sector participation in imports should be allowed. Finally, the findings of this study indicate that methods to help farmers combat the oligopsony power of millers should be provided. These may include provision of storage and credit facilities and strengthening of farmer organizations.

To order a discussion paper, please fill out an online order form, email IFPRI-MTI@cgiar.org or send requests to Markets, Trade, and Institutions Division, IFPRI, 2033 K Street, N.W., Washington, D.C. 20006, U.S.A.