38 56075 v1 Distortions to Agricultural Incentives in Sri Lanka Jayatillake Bandara and Sisira Jayasuriya Griffith University, Brisbane J.Bandaralage@griffith.edu.au La Trobe University, Melbourne s.jayasuriya@latrobe.edu.au Agricultural Distortions Working Paper 31, December 2007 This is a product of a research project on Distortions to Agricultural Incentives, under the leadership of Kym Anderson of the World Banks Development Research Group. The authors are grateful for helpful comments from workshop participants and for funding from World Bank Trust Funds provided by the governments of Ireland, Japan, the Netherlands (BNPP) and the United Kingdom (DfID). This Working Paper series is designed to promptly disseminate the findings of work in progress for comment before they are finalized. The views expressed are the authors alone and not necessarily those of the World Bank and its Executive Directors, nor the countries they represent, nor of the institutions providing funds for this research project. 1 Distortions to Agricultural Incentives in Sri Lanka Jayatillake Bandara and Sisira Jayasuriya Sri Lankas agricultural policies since independence in 1948 have reflected changes on overall development strategy, as well as the interplay of conflicting domestic political considerations, including the need to balance producer and consumer interests, government revenue needs, and ethnic and regional factors. Sri Lanka was a relatively affluent open agricultural economy at the time of its independence from colonial rule in 1948, with one of the highest levels of per capita income in Asia, a vibrant democracy, and levels of health, education and other human development that rivalled even those of many developed industrial economies. It was an oasis of peace, stability and order in a turbulent region (de Silva 1981). Prospects for rapid economic development appeared rosy. To many observers it seemed the country with the best prospects for development in Asia (Athukorala and Jayasuriya 1994). However, five decades later it is clear that it has failed to live up to its early promise. It remains a low-income economy, having slipped well below the high-performing East Asian economies in per capita income terms, and mired in seemingly intractable violent ethnic and social conflicts. Since independence Sri Lanka has experimented with a wide variety of policy regimes, switching from open ,,non-interventionist free-market policies (up to 1959) to dirigiste import-substituting industrialization (ISI) (1960 to 1977) and then to export- oriented liberalization (after 1977). Although its economic growth performance from the late 1970s has been relatively satisfactory in comparison to many similar developing countries, its agricultural sector performance ­ with nearly stagnant per capita agricultural output ­ has been disappointing and has hampered poverty alleviation. Its agricultural policies through this period have reflected the changes in the thrust and 2 direction of its broader development strategy. However, there are also elements of enduring continuity in agricultural policy throughout this period.1 Sri Lankas agricultural sector has two sub-sectors: an import-competing food crop sector dominated by rice but also including a range of ,,subsidiary food crops, and an export crop sector dominated by tea (Sri Lanka is the worlds largest tea exporter) but also including rubber, coconut and several minor export crops (cinnamon, spices, etc.).2 Fruits and vegetables, livestock and diary are mostly import-competing (and generally protected) although small quantities of specific commodities are exported. Agricultural policy until well into the 1980s taxed export agriculture while providing various forms of input subsidies (irrigation, fertilizer, R&D and extension) to protected import-competing food agriculture, particularly rice. The shift to inward- oriented development strategies from the late 1950s, with protection for import- substituting industries and accompanying real exchange overvaluation (leading to exchange controls and a discriminatory dual exchange rate regime) exacerbated the burden on export crops. Despite assistance and protection for import-competing food agriculture, the policy regime during this period probably had an overall anti-agricultural bias. The process of policy liberalization from the late 1970s eroded the former overall anti-agricultural bias. Direct taxation of export crops was sharply reduced in the 1980s and largely eliminated by the early 1990s, and manufacturing trade liberalization drastically lowered the indirect burden on agriculture flowing from industrial protectionism. But policy reforms have been both limited and selective in the import- competing parts of the agricultural sector. Many import-competing agricultural industries, including not only rice but also several others such as sugar, potatoes and dairy, have continued to enjoy both direct input subsidies and, to varying degrees, protection through the trade regime. Thus the overall policy regime ­ in general far more liberal than at any 1 For reviews of agricultural policy in Sri Lanka, see Thorbecke and Svejnar (1987), Bhalla (1991), World Bank (1995), Athukorala and Kelegama (1998), Anderson (2002) and Sanderatne (2004). 2 These ,,minor export crops are often referred to as ,,non-traditional agricultural exports, distinguishing them from the ,,traditional exports of tea, rubber and coconut, even though many of these crops have been exported for thousands of years while tea and rubber were introduced to the country only in the late 19 th century. 3 time since the late 1950s ­ now tends to have a pro-agricutural bias because of the reduction in taxes on agricultural exports. Protection from import competitions and direct assistance to food agriculture have failed to achieve their stated goals of adequately encouraging production of staples and reducing the growing gap between rural and urban household incomes. True, domestic rice production increased substantially, but overall food production has almost stagnated (growing less than half as fast as in other developing countries over the 1990s) such that food self-sufficiency has kept declining and rural poverty alleviation has been slow. It is clear that a comprehensive re-assessment of agricultural policies should be on the policy agenda as a priority issue. Economic growth and structural changes Since independence in 1948, Sri Lanka has had regular changes in government with distinctly different economic policy orientations. As indicated earlier, it has experimented with a wide variety of economic policy regimes under different governments. However, since 1977 the basic direction of policy has not changed despite several changes of government, although there have been differences in the pace and scope of liberalization measures. Figure 1 shows real GDP growth rates and the changing policy regimes under different governments.3 Relatively slow per capita growth during the 1950s and 1960s was followed by particularly traumatic experiences during 1970-77, when policy responses to the impact of the first oil shock of 1973 and to a youth rebellion resulted in severe import compression and shortages of essential goods. This generated widespread popular discontent against the ISI policies and extensive government intervention in the economy. As a result, there was massive popular support for a shift in policy that brought into power a new United National Party (UNP) government in 1977 that pledged to adopt 3 For more details of the policy regimes and growth experience, see Athukorala and Rajapatirana (2000), Athukorala and Jayasuriya (1994), World Bank (2004), Weerakoon (2004) and Kelegama (2004). 4 ,,open economy policies. A major liberalization effort was launched in 1977, marking a decisive break with the previous policy regime. As a result of progressive pro-market reforms from 1977, Sri Lanka became and still is the most open economy in South Asia. Since 1977 it has averaged a reasonably healthy real average annual GDP growth rate of 4.75 percent (3.5 percent per capita), attaining a per capita income of US$1200 by 2006 at the official exchange rate (and over $4,000 in PPP dollars). Despite ongoing ethnic and social conflicts that have plagued the country for nearly two decades, this is a higher growth rate than achieved by most countries that had similar per capita incomes in the mid-1970s, with only Botswana recording faster growth. Starting with above average human development indicators across a wide range of indicators from its early years of independence (achieved through investments in health and education financed by taxes on plantation crop exports), Sri Lanka has managed to maintain its position: life expectancy at birth for males and females, for example, averages 72 and 76 years ­ higher than the middle-income country average ­ and its literacy rate is over 90 percent. Nevertheless, Sri Lankas overall developmental performance is disappointing when compared with the high-performing East Asian economies, given that Sri Lanka had comparable or higher real incomes and human capital endowments in the 1950s and early 1960s. This poor performance is at least in part due to the ongoing civil war and ethnic conflict that has caused enormous damage to the economy and to the wider socio-political environment.4 The overall position of agriculture within the national economy is shown in Table 1. The shares of the agricultural sector in GDP, employment and exports has declined progressively from the 1950s, although the sector remains a major source of income and employment of a large proportion of the population, and a significant source of national export earnings.5 In contrast to the gradual shrinking of agricultures share of GDP, the fall in its share in exports in the 1970s and 1980s has been sharp (Figure 2). Until the late 4 The cost of the war between 1983 and 2000 was conservatively estimated at twice the value of Sri Lankas 1996 GDP (Arunatilaka, Jayasuriya and Kelegama 2001). 5 Note that processing of tea, rubber and coconut products ­ usually included under manufacturing ­ is included in this table under Agriculture, Forestry and Fisheries to indicate the overall contribution of the sector to the economy. 5 1960s agricultural commodities accounted for more than 90 percent of Sri Lankas exports but since the early 1990s its share has been below 20 percent, with manufacturing ­ particularly garments ­ emerging as the major export category.6 The shares of different agricultural products in the value of agricultural production and in household consumption expenditure are shown in Figure 3. Paddy accounted for around 25 percent of agricultural and fisheries output in the early 1980s, but its share is now barely half that. Likewise, the share of rice in household spending has halved over that period and currently is around one-tenth. The agricultural sector is widely considered to have contributed to the poor performance of the wider economy: "In terms of sectoral contributions to growth, agriculture has been a continual drag....", and "the long-term average growth rate in agriculture has barely exceeded the rate of population growth, which has contributed to the persistence of poverty (the headcount ratio stood at 23 percent in 2002, which is relatively high for Sri Lankas per capita income)" (IMF 2005, p. 5). Within agriculture, the output of most crops has either stagnated or declined since the 1980s (Appendix Table 1). Tea is the sole important exception. Tea output expanded from around 200 million kgs per year in the early 1980s to around 300 million kgs at the turn of the century. A combination of area expansion (primarily from an expansion of smallholder cultivation in the low-altitude regions) and higher-yielding new cultivars was stimulated by higher prices for the ,,stronger teas produced in lower altitudes. Production of rice and coconuts ­ the two crops that dominate smallholder agriculture in Sri Lanka ­ has stagnated, while outputs of rubber, minor crops and subsidiary food crops, including income-elastic horticultural crops, have fallen quite significantly. This has meant that non-tea agricultural exports have fallen while food imports have grown. Overall, Sri Lankas per capita food production has fallen by over 12 percent since 1980, compared with a rise of 48 percent in other developing countries on average (World Bank 2001). This raises the question: did policy distortions contribute to this relatively poor performance? 6 Agricultures contribution to net exports is greater than implied by these gross export data because exports of garments that dominate manufactured exports have a large import content (Athukorala and Bandara 1989). 6 Policy evolution At the time of independence, Sri Lankas three plantation crops ­ tea, rubber and coconut ­ dominated Sri Lankas exports, while significant quantities of rice, wheat and other food products were imported. This basic distinction between exportable and import- competing agricultural products is critical to understanding Sri Lankan agricultural policy. Export dependence varies greatly, however. For example, more than 80 percent of the coconut crop is domestically consumed, with both nuts and oil being essential parts of the Sri Lankan diet. The share of domestic consumption is quite low in the case of tea (less than 15 percent) while for rubber it is around 35 percent thanks to an expansion of rubber-based manufacturing industries. The tea and rubber sectors had a pronounced dualistic structure. Foreign-owned large plantations contributed a large share of output at the start of the post-independence era, but they were taken over by the state in 1974. Then from the early 1990s they were progressively handed back to the private sector, including foreign companies, but the share of large plantations has been in gradual decline. By contrast, the import-competing food crops as well as the coconuts have always been dominated by smallholders, although a substantial proportion of the coconut industry was held in the form of large plantations until land reforms in the early 1970s. Export taxes on agricultural crops (tea, rubber and coconut products) initially were a major source of government revenue, accounting for around 30 percent of all government tax revenues during the 1950s. They helped finance expenditures on public education and health, and on food subsidies. Levies on exports also financed agricultural research, extension and replanting programs for plantation crops. The structure of export taxes not only reduced average producer revenues but also had the effect of greatly lowering gains from any price increases. When domestic prices rose, whether due to world price increases or because of currency devaluations, government tax revenues siphoned off the bulk of the price increase. This meant that while producers had to bear the cost increases linked to domestic inflation, they were largely deprived of the benefit 7 of any exchange rate adjustments that were made ­ as happened on several occasions from the late 1960s onwards ­ to restore international competitiveness of tradeables sectors. In common with many other developing countries, Sri Lanka followed a food self sufficiency policy, narrowly interpreted as one of encouraging ,,rice self sufficiency. On the production side this involved major investments in irrigation (irrigation water being supplied at no cost to farmers), fertilizer subsidies, and the provision of rice and other agricultural research and extension services. There was also a public distribution system for procurement and marketing of paddy and other commodities, aimed at making rice more affordable for consumers. Staple food prices were heavily regulated until the 1977 liberalization, and many of them have remained subject to strong government interventions ­ including regulation of import volumes ­ aimed at maintaining price stability. From the late 1950s, the structure of incentives was further biased against export crops with the adoption of import and exchange controls in response to growing balance of payments problems caused partly by a secular downward movement of Sri Lankas international terms of trade. The adoption of these measures, rather than a currency devaluation, led to real exchange rate overvaluation. The import restrictions (tariffs and, increasingly, non-tariff measures) and exchange controls were further strengthened subsequently as a result of the ideological shift to an ISI development strategy. The result was high manufacturing protection, severe import compression, pervasive state controls in trade, marketing and distribution, persistent exchange rate overvaluation, exchange controls and a formal dual exchange rate system between 1968 and 1977.7 Under the dual exchange rate system, there was a basic rate and a so-called Foreign Exchange Entitlement Certificate Scheme (FEECS) which was initially (in 1968) set at 44 percent higher (more depreciated) rate, but adjusted to 55 percent in 1969 and then to 65 percent in 1972. The main plantation crops (tea, rubber and coconut) ­ the ,,traditional exports ­ had to convert export earnings at the less favorable official exchange rate while non- 7 See Athukorala and Jayasuriya (1994) for a description of these developments. Athukorala and Rajapatirana (2000) provide an analysis of the manufacturing sector developments. 8 traditional exports were eligible for the FEECS rate.8 The highly overvalued official exchange rate, rather than the somewhat more realistic FEECS rate, was also applied to some imports of the major agricultural products such as rice, wheat and sugar. However, such imports were heavily regulated and under direct government control. Hence rice producers were still shielded from import competition. And there is other evidence (for example, the high premium in black market rates of exchange) that there was substantial real exchange overvaluation which discriminated against exportable industries and favored import-competing ones, both in agriculture and elsewhere. In short, despite the fact that the rice sector was granted special incentives and other import-competing agricultural products also gained significant protection from the import substitution strategy, studies such as that of Bhalla (1991) conclude that there was an overall bias against agriculture because of the severity of the anti-export impact of the overvalued currency, the large weight of exports in farm output, and the high protection granted to manufacturing. This started to change only from 1977, with the policy shift away from the ISI strategy and its replacement with more liberal pro-market policies. After the 1977 policy liberalization food subsidies to consumers were sharply reduced,9 and the emphasis on food self-sufficiency was enhanced. Public investments in major irrigation systems were expanded, for example. The government implemented a huge irrigation-cum-hydropower scheme (the Accelerated Mahaweli Development Project) with substantial foreign assistance, and around one-third of all government capital expenditure was devoted to this single project for several years from 1979. The project was explicitly rationalised as a major step towards achieving rice self sufficiency ­ the coveted national goal ­ and thereby it appealed to the popular imagination which partially blunted the political impact of cuts in food subsidies.10 8 The large plantation crop sector also suffered much investor uncertainty from the mid-1950s, including facing the threat of nationalization before they were finally taken over by the state in the mid-1970s. 9 In the past the government encouraged substitution in consumption of wheat for rice, to reduce the fiscal burden of the consumer rice subsidy. 10 However, the project was resented by sections of the minority Tamil community: not only was an appeal for extending irrigation to the existing Tamil farming areas rejected, but the newly irrigated lands were settled largely with members of the majority Sinhalese community. This has been highlighted in numerous studies. For example, a study by the OECD Development Assistance Committee pointed out how the project exacerbated ethnic tensions: "The conspicuous absence of consideration of the projects possible negative impact on simmering tensions is striking ­ considering that it had glaring ethno-political implications: (1) there was an ethnic overlay to the geographical areas which would benefit (or not) from 9 The 1977 reforms also reduced manufacturing sector trade protection, and explicit export taxes on plantation crops were largely eliminated in the 1980s. There were both fiscal pressures and political imperatives driving the reduction of export taxes on plantation crops: nationalisation of large foreign-owned plantations had transferred them to state ownership from mid-1970s, and the land reforms of the early 1970s had broadened their ownership base, largely among Sinhalese smallholders (Moore 1985). Trade liberalization was not uniform though: it extended to some import- competing agriculture, but excluded others. Crops that were widely cultivated in the North (e.g. red onion, chillies, grapes) were subject to liberalization while protection for other crops (such as potato) were maintained and even increased. After the initial economic growth stimulus following the liberalization of 1977 and the huge public sector investment boom (assisted by a massive flow of foreign aid), growth started to slacken even though the economy was cushioned for a while by a tea price boom in the early 1980s. Simmering ethnic tensions erupted into a secessionist war following anti-Tamil riots in 1983, and social tensions in the south led to a highly disruptive rural youth rebellion in the late 1980s. Economic and political conditions worsened, and the economy lost steam. In 1990/91, responding an emerging balance of the project; and (2) the government decision to resettle displaced Sinhalese villagers in traditionally Tamil regions. The decision by the Jayawardene government to compress and accelerate the 30-year program into six years further exacerbated ethnic tensions. The original version of the programme had included irrigation projects in the Tamil-majority Northern Province; but this was removed from the accelerated programme with the argument that it would be too expensive and problematic technically" (Bush 1999). These facts are well known, extensively documented and not contested by any serious scholar of Sri Lanka. "Since the 1930s and especially the 1940s resettlement projects have been implemented in Sri Lanka to alleviate the growing shortage of land in the south-west, where the population is very largely Sinhalese. Sri Lanka's Tamils have opposed these projects because they threaten to change the ethnic majority in the provinces concerned to the disadvantage of the Tamils and Muslims. With the Mahaweli project, which has been planned since the 1960s and consists of a large number of subsidiary energy generation projects, the country's largest scheme was launched, the aim being to use at least 74 percent of the settled area ­ where Tamils previously formed the majority of the population ­ for Sinhalese. The Sinhalese settlement projects became one of the decisive motivating factors in the Tamils' resistance. This is not least evident from the many attacks on colonies of new Sinhalese settlers during the civil war." (Klingebiel, 2001, p. 10). Peebles (1990) points out how the scheme was re-designed in a way that excluded the largely Tamil populated Northern Province: "The choice of projects to be developed also reflects the focus on Sinhalese settlement. Under the Water Resources Development Plan systems J, K and L and part of system I fell within the Northern Province and were to irrigate 232,000 acres by a Northern Central Province canal. None of these systems were included in the Accelerated program..." (p.43). See also Moore (1985), Manogaran (1987). The World Bank, a major donor, subsequently acknowledged the problems related to this projects perceived ethnic bias in a World Bank study: ".....donors may have missed a significant opportunity to promote equitable participation through the huge Mahaweli power, irrigation and resettlement scheme" (Kreimer at al. 1998, p. 22). 10 payments crisis, the government devalued the currency and initiated a second wave of liberalization (Dunham and Kelegama 1997). Liberalization proceeded in an uneven way following the initial measures. Export duties on plantation crops, already reduced substantially from the mid-1980s, were eliminated in 1992, while high rates of nominal (and effective) protection for import- competing agriculture continued (Edwards 1993, World Bank 1995). The government frequently used regulatory controls on imports, manipulating licensing and variable tariffs, to achieve not only protection but also price stability. The latter was particularly important in the case of the highly politically sensitive commodities such as rice and, at times, also coconut. Tariffs on agricultural imports were gradually reduced through to 1993 and, as part of the GATTs Uruguay Round Agreement on Agriculture, Sri Lanka bound all its tariffs on agricultural goods at a uniform rate of 50 percent from January 1995 ­ although this was well above applied tariffs. Subsequently it removed quantitative restrictions on all agricultural imports except wheat and wheat flour. In fact some of the most important changes since the early 1990s have been not so much in the level of restrictions as in the move away from various forms of nontariff barriers including regulatory interventions to more transparent tariff-based import restrictions. The general picture, which is confirmed by the various indicators of assistance provided below, is that the policy regime in recent years has provided assistance to import-competing agriculture and at rates similar to the levels of assistance to manufacturing industry.11 The general trend in overall policy ­ despite phases of slow progress and occasional backsliding ­ has been towards progressive liberalization, and Sri Lanka now has perhaps the most open trade regime in South Asia. The evolution of the overall structure of incentives can be seen by looking at the changes for the manufacturing sector and the two agricultural sub-sectors. At the time of the 1977 reforms, the implicit rates of protection for manufacturing as well as many import-competing agricultural commodities 11 Athukorala and Kelegama (1998) suggest that the actual level of protection for manufacturing tended to be overestimated by the use of gazetted tariff rates which were often higher than actual tariffs because of various tariff loopholes and exemptions. But the same may be true of earlier estimates of effective assistance to agriculture. For example, the World Bank (1995) produced estimates that attempted to take into account the provision of free or subsidized inputs such as irrigation water used by many import- competing crops grown under irrigated conditions, including rice. We believe the approach used to derive that implicit subsidy associated with free irrigation water tended to overestimate it. 11 were extremely high. Even after major reforms, effective protection of the manufacturing sector in 1979 has been estimated to be 137 percent (Cuthbertson and Athukorala 1991). This had fallen to 90 percent in 1981, and was estimated to have come down to 77 percent in 1991 and 43 percent in 1994 (World Bank 2004). By 2005, liberalization in the manufacturing sector had proceeded further. In addition to reforms in manufacturing, there have been important moves towards liberalizing the food import-competing sector. Wheat imports were liberalized by ending a government-granted long term monopoly on flour milling to a Singaporean based company (PRIMA) that also gave them a virtual monopoly on animal feed supply. Government trading enterprises play a role in both domestic distribution and international trade, but it is minor. Trade policy continues to protect rice and several other import-competing food crops (such as potato) with the use of seasonally varying tariffs and specific duties. Those interventions respond to domestic price and supply conditions. Further, despite the shrinking share of agriculture in both GDP and employment, agricultural producer subsidies remain important as governments have responded to intense pressure to maintain and even expand them. For example, subsidies such as for fertilizer (targeted largely though not solely towards import-competing crops, particularly those cultivated by smallholders) continue, in a very volatile political environment. Periodically the fertilizer subsidy has been a component of government assistance to agriculture, and populist governments have used it to appeal to the politically important farming community. Until recently its aggregate assistance effect would have not been large, but the fertilizer subsidy doubled in 2006 compared with 2005, and its cost may have risen by a further one-third to reach Rs 11 billion ($1200 million) or more in 2007 (Figure 4). There are also frequent ad hoc changes to import policies particularly in the case of ,,subsidiary food crops such as potato and onion, where small but politically powerful farm groups exercise much political clout. Thus the broad contours of present agricultural policies in Sri Lanka appear to resemble some aspects of the early agricultural policy evolution in more-developed East Asian economies: overall a relatively liberal trade regime, but granting significant protection for particular import-competing agricultural industries. In the next section we 12 present the estimates of a number of indicators of incentives for major agricultural products and different commodity categories (such as exportable and import-competing products). We then attempt to explain the changing pattern of agricultural taxation/assistance since independence that has led to the current outcome. Direct and indirect distortions to incentives The main focus of the present studys methodology for estimatingthe extent of distortions to agricultural incentives (Anderson et al. 2008) is on government-imposed measures that create a gap between actual domestic prices and what they would be under free markets. Since it is not possible to understand the characteristics of agricultural development with a sectoral view alone, the projects methodology estimates the effects not only of direct agricultural policy measures (including distortions in the foreign exchange market), but also of distortions in non-agricultural sectors that compete with farmers for mobile resources such as labor and capital. More specifically, this study computes Nominal Rates of Assistance (NRAs) for farmers and also generates an NRA for nonagricultural tradables, for comparison with that for agricultural tradables via the calculation of a Relative Rate of Assistance (RRA ­ see Anderson et al. 2008). We present in Table 2 our estimates of temporal patterns of distortions to agricultural incentives from 1955 to 2004 for seven major commodities: three exportables (rubber, coconut and tea) and four import-competing products (rice, potato, onion and chillies).12 In Table 3 we also include a guesstimate of the NRA for non-covered farm products, which account for around one-third of the overall value of agricultural production. That guesstimate assumes non-covered products are equally divided between exportables, nontradables and import-competing products (one-third each of the residual 12 Some commodities, such as livestock including dairying, are not included in this exercise because of severe data limitations. For the livestock sector, for example, consistent time series data on domestic prices of meat products are not available, and international prices for comparable quality meat products are also not available. The quality differences between domestic and world market products, except perhaps in the case of chicken, are so large as to make them almost completely different products. 13 value of farm production), and that for nontradables the NRA is zero while for exportables and importables they are equal to the estimated NRAs for the two sub-sets of covered products (all of which are tradable). The farm input assistance was minor before 2005 and so no estimate is included for such non-product-specific assistance. The NRA for non-agricultural tradables is based on the import duty collection rate for import- competing manufacturing and an assumption that direct assistance to non-agricultural exportables is zero.13 This can substantially underestimate the actual rate of manufacturing protection, because high tariffs (or binding non-tariff barriers) may result in low rates of import duty collections because they lead to lower import volumes. We believe the non-agricultural NRA before liberalization in 1977-78 would be significantly higher than indicated by these data, and that the RRAs may therefore correspondingly underestimate the policy bias against agriculture. (Annual time-series NRAs and value of production shares of different commodities are tabulated in the Appendix.) Export crops We begin by focusing on export crops before turning to import-competing ones, and in doing so refer to the NRA five-year average estimates in Table 2 and the annual estimates depicted in Figure 5. Tea The annual NRA estimate for tea, which is based on the ratio of the domestic Colombo Auction price and the average fob export price of bulk leaf tea, suggests that this industry has been taxed by more than 30 percent up to the mid-1980s. Since then the estimated rate of taxation has declined sharply as export taxes were progressively lowered and largely abolished by the early 1990s. The NRA has fluctuated around that trend (see Figure 5), primarily because the export tax rate was on a sliding scale such that taxation was higher during periods of high international prices. There was a steep increase in taxes 13 These NRAs for tradables include an estimate of the trade tax effect of the overvalued exchange rate. As outlined in the methodology, that estimate uses the black market exchange rate premium (see Easterly 2006) and assumes that only half of exporters foreign exchange rate earnings are sold to the government at the official rate. See Anderson et al. (2008) for details of this methodology. 14 in the immediate aftermath of the 1977 reforms, for example, associated with the exchange rate depreciation that accompanied the liberalization (which increased the domestic currency price on which the tax was based) and the international tea price increase of the early 1980s. There is no evidence of significant market imperfections within the domestic market for tea production, processing and wholesale marketing through the auction. The Colombo Tea Auction is considered to be quite competitive though there have been some criticisms of the system in the past. Hence the rate of export taxes and cesses provides a reasonable estimate of NRAs given the fact that it is not possible to use domestic and border price data to directly compute NRAs because of the growth of importance of ,,value added teas (tea bags etc) for which a reliable border price is not available. What is not reflected in these figures is that tea imports into Sri Lanka have been effectively subject to a near total ban until recently, ostensibly to ensure that cheap foreign teas are not re-exported as quality ,,Ceylon Tea. Though there has been some relaxation of this recently, a prohibitive tariff has effectively ensured that tea imports are negligible. The outcome has been that Sri Lanka lost the chance to develop a tea blending industry, resulting in Dubai emerging as the centre of the lucrative tea blending centre using significant quantities of both Sri Lankan tea and imported Sri Lankan labor (Ganewatta 2002). Rubber The annual NRA for sheet rubber, which is based on the percentage gap between the average domestic Colombo Auction price for sheet rubber and the average fob export price of rubber, suggests that this industry has been taxed by even more than the tea industry, by more than 50 percent up to the early 1980s. Since then the rate of taxation has declined fairly rapidly, and reflecting the virtual elimination of export taxes, the NRA estimate has averaged clse to zero in recent years. The NRA has fluctuated around this trend (see Figure 5), again primarily because the export tax rate was on a sliding scale such that taxation was higher during periods of high international prices. As with tea, the rate of export taxes and cesses provide a reasonable estimate of NRAs in recent years 15 given that it is difficult to compute reliable NRAs from available price data because of the changes in the composition of exported rubbers. Coconut Coconut products (copra, oil, fibre and coir products, etc.) were a major export product category at the time of independence, but export volumes have since fallen. Because of its importance as a food crop (in the form of nuts) and for coconut oil for household consumption (it is the most widely used cooking oil), it attracted significant direct price and non-price interventions by the government with the aim of stabilising domestic prices while maintaining ,,reasonable producer prices (de Silva 1979). However, over time imported palm oil has emerged as a viable substitute for coconut oil in many domestic uses. Thus in more recent years coconut oil may be considered an import-competing product. Up to the early 1980s, the NRA estimates suggest the coconut industry was taxed to a similar degree to tea (hence less than rubber), but since then its taxation has averaged closer to zero and in the late 1990s/early 2000s the industry enjoyed a positive NRA. This reflects the fact that imported palm oil has become a viable substitute for coconut oil in many uses, and protection from that competition raised the domestic price of coconuts above what it would have been without palm oil import restrictions. In the case of coconuts, the reductions in export taxes did not produce corresponding reductions in NRAs because of trade restrictions on coconuts and coconut products (such as export bans) imposed to maintain low consumer prices (and ,,reasonable producer prices). Import-competing farm products Rice The pattern of NRAs for rice reflects the fluctuating impacts of direct government interventions in response to changes in domestic supplies and international prices. Apart from the 1980s, the NRA average each decade has been positive (Table 2), which is 16 consistent with the fact that both average consumer and producer domestic rice prices have been usually above the average cif import price.14 When domestic prices fall, whether due to a bumper harvest at home or due to a fall in the price of imports, pressures for assistance emerge. Rather than increase domestic prices by direct purchases, the response to such political pressures from the rice farming lobby usually has been to raise import barriers and expand input subsidies for fertilizer, R&D, etc. Those ad-hoc policy changes in rice import tariffs since 1995 are shown in Appendix Table A3, while the Figure 4 shows the dramatic increase in the fertiliser subsidy in recent years. A fuller picture of rice policy requires that it be placed in the context of the wider cereal staples policies, particularly for wheat which is the closest substitute for rice in consumption even though it is not produced domestically. Rice consumption in Sri Lanka averages around 100 kg per capit per year, but wheat consumption has increased with urbanization from less than 25 kg during the 1950s to nearly 50 kilogram per capita at present. The consumer tax equivalent on wheat flour from import measures was low and sometimes even negative until the early 1980s, but then it increased significantly and has since fluctuated widely around an average value of more than 50 percent. Chillies, onions and potatoes High rates of import protection were enjoyed by producers of chillies and onions in the pre-liberalization protectionist period, but this protection came down after the liberalization in 1977/78. In contrast, protection for potatoes increased sharply in the post-liberalization period, particularly in the 1990s. Although imports were briefly liberalised around 1996, potatoes ­ grown by only a few thousand farmers ­ have continued to enjoy a pre-eminent position among protected crops.15 14 For those familiar with Sri Lankas history of government subsidized rice provision for consumers, this may seem rather puzzling. But those subsidies via ration shops apply only to a small fraction of national consumption, such that the weighted average of the (very low) price of government-provided rice rations and the open market price consumer price is close to the latter. 15 Tobacco is another crop that enjoys a high level of protection. According to the WTO Trade Policy Review 2004, the average tariff for the tobacco sub-sector was 149 percent in 1998 and 153 percent in 2003 ­ well above the bound rate of 50 percent applying to most agricultural products. 17 Sugar NRAs are not provided for sugar because it is predominantly imported and domestic production is quite small. This is despite the fact that sugar has been heavily protected. The level of protection has come down since the 1980s but remains quite high. During the period up to the late 1970s when consumer food subsidies were not uncommon, sugar was an exception: its high tariff brought in much needed government revenues that helped maintain other food subsidy expenditures. Agricultural versus nonagricultural assistance The products covered in Table 2 account for about two-thirds of the overall value of agricultural production. Their weighted average NRA was around -30 percent in the 1960s and 1970s, but following the reform of the late 1970s it rose to around -15 percent in the 1980s, to -2 percent in the first half of the 1990s, and to an average of around 10 percent since then. The picture does not change much when our assumptions about the NRA for non-covered products are included (top of Table 3), allowing us to conclude that the direct taxing of Sri Lankas agricultural sector has been gradually phased out over the past three decades. This is not to say the sector is without price distortions, as there is still some dispersion in the product NRAs and, in particular, the NRA for exportables remains below that for importables. However, both of those indicators of dispersion are now well below what they were in earlier decades. This is depicted in Figure 6, and is captured also in the trade bias index shown in the middle of Table 3. Also important for intersectoral resource allocation is the extent to which non- agricultural tradables have been assisted by the government. Prior to the 1977 reforms, protection for import-competing manufacturing was extremely high. Even taking into account the lower assistance to producers of other tradables, the nominal rates of assistance to all non-agricultural tradable sectors was well over 100 pecent prior to the 1970s. By the late 1970s it had fallen to below 60 percent, and since the early 1990s it has fallen further and is now les than 25 percent. When these estimates are combined 18 with those for agriculture to generate the relative rate of assistance (RRA), the full extent of the discrimination against farmers becomes evident. As shown in the middle rows of Table 3, those RRA values suggest that during the 1960s, Sri Lankan farmers received only one-third of what they would have received had markets for both farm and non-farm goods been free (RRA average of -67 percent). In the 1970s the extent of discrimination was not much less (RRA average of -53 percent), but then the average RRA continued to fall to around -45 percent in the 1980s, to -25 percent in the 1990s, and to just under -10 percent on average in recent years. As is clear from Figure 7, the decline in protection to manufacturing did more to reduce distortions harming farmers then did the changes in direct agtricultural policies. And the exchange rate distortions were not a major contributor to this trend in average NRA and RRA values, but they did affect substantially the anti-trade bias of past policies (see bottom rows of Table 3) The political economy of agricultural policies What were the political forces behind the governments agricultural policy choices? We consider first the export subsector, and then the import-competing subsector. Export crops At the time of independence relatively large scale, foreign owned companies dominated tea and, to a lesser extent, rubber sectors. Even the coconut sector, though largely smallholder based, still had a significant number of large ,,estates. There was little political sympathy for the plantation sector from the left of centre coalitions that ruled the country from 1956 to 1965 (with a brief interruption in 1960) and then again from 1970- 77. The foreign plantations were under threat of nationalization from 1956 onwards, and finally they were nationalized in the mid-1970s when all large holdings were subjected to a land reform. The majority of the workers in the large plantations, who had migrated from South India from the mid-19th century, had been politically marginalized by being 19 disenfranchised by the first post-independence government. And, most importantly perhaps, government revenues depended heavily on foreign trade taxes, with export duties alone contributing nearly one-third of all government revenues in the first decade after independence. With slow economic growth, governments were continually under fiscal pressure and had little scope, even if they desired, to reduce the export taxes on plantation crops. The combination of high taxes, secular falls in real world prices, and the threat of nationalization led to a slow but steady decline of the plantation sector, particularly the large plantations. The sector became more and more smallholder-based, a process that accelerated after the land reforms. The large foreign-owned firms passed into state ownership. The 1977 liberalization did not have an immediate positive impact on these industries, however. The exchange rate reforms that involved a significant nominal devaluation failed to have much of a positive impact on the sector because the structure of progressive export taxes ­ involving a sliding scale of taxes ­ effectively taxed away most of the gains from the devaluation (and periodic relative price improvements). The plantation crop sector, and its role within the economy, had changed in quite fundamental ways by the mid-1980s. Its share of the national economy had shrunk, and it became quite clear that new investment in replantings and factory modernization were essential if the decline of the sector was to be arrested. The political tensions that emerged in the aftermath of the 1977 liberalization made the government more sensitive to the political importance of plantation crop cultivators. The plantation crops were cultivated almost entirely in the wetter central, western and southern regions of the country, and the cultivators were predominantly from the majority Sinhalese community, whose support was critical if the governing political parties were to stay in power. Also, export taxes on plantation crops directly reduced revenues of state-owned large plantations. These factors combined to erode the economic and political incentives for maintaining the traditional high export tax regime. Import-competing crops 20 We have seen that government policies towards import-competing crops have not been uniform. Further, in the case of key staple food crops, most importantly in the case of rice, there has been tension between the twin objectives of producer support and maintenance of low consumer prices. As mentioned earlier, the enduring theme that runs through Sri Lankan import-competing agricultural policy in the post-independence period is this difficult balancing act between producers and consumers in staple food crops. This is best illustrated when we examine policies towards the rice sector. Rice policy Rice is the most sensitive political commodity in the country. The immediate post- independence governments ­ aided by the proceeds from high commodity prices during the Korean War ­ were able to maintain and even expand the policy of providing cheap subsidized rice to consumers through a universal rice ration, which they had inherited from the British colonial rulers. It formed a central component of a wider political strategy that aimed to undercut the potential threat from a Marxist left with strong roots in trade unions. An attempt to reduce the rice subsidy in 1953 brought the country to the verge of revolution. Subsequent attempts in this direction almost invariably led to the downfall or political humiliation of the government. Subsidized rice distributed through the ration system became a symbol of the political power of the powerful left-wing parties in Sri Lanka. Rice has enormous symbolism and emotive power in Sri Lanka. Rice self sufficiency has been a slogan that appealed to deep seated nationalist aspirations, particularly among the Sinhalese, whose ancient civilization was based on irrigated rice cultivation and who are stirred by the vision ­ irrespective of its historical veracity ­ of a Golden Age when Sri Lanka was supposedly the granary of the east and exported rice to other parts of Asia. At a more mundane level, the landed rice cultivators and politically powerful rice millers exert much clout in terms of garnering electoral support for particular political parties. In Sri Lanka, the drive to attain rice self sufficiency was not simply one of achieving a Green Revolution on existing rice lands. From the 1930s, promotion of rice cultivation through land settlement schemes ("Colonisation Schemes") were linked to the restoration of the ancient glory of the Sinhalese in the sparsely 21 populated so-called dry zone of the country which had been the cradle of the ancient hydraulic civilization. Restoration of old irrigation systems ­ already initiated during colonial times by the British ­ and the building of new and grandiose irrigation systems were seen as a vehicle for new Sinhalese farmer settlements in those parts of the country that had been largely abandoned by previous generations.16 As Brunton (1992, p. 82) pointed out, "the paddy society was almost entirely Sinhalese". Thus the rice self sufficiency drive became associated with the rising Sinhala nationalism of the post- independence era. In addition to free or subsidized provision of irrigation, fertilizer, seeds, research and extension, etc. to assist producers, the government also purchased paddy from farmers at a guaranteed price through a state trading entity. It distributed varying quantities of rice rations to consumers at subsidized prices until the late 1970s, including a period (1966-78) when some rationed rice was distributed free. In addition farmers sold rice to private buyers who either milled it themselves or sold it to millers who then sold it in the ,,open market. In 1972, facing rising international prices and hoping to increase domestic supplies at lower prices, the government granted monopoly procurement rights to the state-owned Paddy Marketing Board ­ but the monopoly could not be enforced, and was abolished in 1975.17 The government has regulated international trade and 16 The President and other government leaders were quite explict about the goal of land settlement. For example, the Minister responsible for the Mahaweli project, Gamini Dissanayake, wrote that it represented "a return of the people to the ancient homeland ...in the Rajarata" (Dissanayake 1983, p. 6, cited in Tennekoon 1988). Rajarata is the popular name for the area, largely in the North Central province, that was the heartland of the ancient Sinhalese kingdoms. Nimal Sanderatne, former Director of Economic Research at the Central Bank of Sri Lanka, writes: "Colonization had another political significance in pluralist Sri Lanka. It gave the majoity ethnic community the opportunity to resettle Sinhalese in the ancient historical capitals and ancient kingdoms and thereby confirm the area as a Sinhalese rather than a Tamil region. The land settlement issue has been a most controversial issue and was an underlying cause for the ethnic conflict" (Sanderatne 2004, p. 211). Moore (1985, p. 96). adds: " ....Dry Zone development has been explicitly viewed as a means of increasing the Sinhalese population in the historic heartland of Sinhalese civilization....Between 1946 and 1971 the Sinhalese proportion of the population of the five ,,frontier districts ­ Amparai, Batticaloa, Polonnaruwa, Trincomalee and Anuradhapura ­ increased from 33 to 51 percent. The main cause was the migration of Sinhalese settlers to new irrigation schemes." 17 These measures included prohibiting of storage and transportation of paddy in bulk, and procurement price increases. But with open market prices high and private traders active in procuring rice despite restrictions, the state agency was unable to procure expected quantities. 22 supplies to maintain stable consumer prices, and that remains a central goal of food policy.18 The non-uniform pattern of trade liberalization from 1977 onwards can only be understood by recognizing the central political role of rice, both in consumption and in production. The reform government of 1977 was led by Mr J.R Jayawardena, a lifelong opponent of the left, strongly pro-western in foreign policy, and a man who had tried and failed once to dismantle the rice subsidy scheme. This was where the Accelerated Mahaweli Development Project played a critical political role. It provided crucial popular political support that enabled him to dismantle the rice subsidy scheme and crush the traditional political left, while implementing major liberalization measures. But the end of the rice subsidy scheme did not mean that domestic rice prices were ,,left to the market, whereby rice producers would be exposed to the pressure of cheap imports. So post-1977 government policies continued to aim at consumer price stabilization around a ,,reasonable price, while assisting producers through input subsidies and import protection. The NRAs in rice also reflects the impact of international price and weather-related domestic supply changes. Because non-tariff barriers insulate domestic markets from international markets, changes in domestic supply change domestic prices and, even in the absence of any change in the policy regime, change estimated NRAs. Hence good harvests depress domestic prices and generate pressures for assistance and increased protection. Sri Lankas signing of the Uruguay Round Agreement on Agriculture in 1994 did not constrain the government from exercising a high degree of discretion in changing tariffs, because applied tariffs were significantly lower than bound tariffs. The government has frequently chosen to change import tariffs in response to domestic pressures. For example, from 1995 rice imports were subject to a tariff of 35 percent, but an import licensing requirement was imposed in July 2000. In July 2001 when the domestic rice price increased because of a production shortfall, the government allowed the state agency (the Cooperative Wholesale Establishment, CWE) as well as private traders to import rice duty free, waiving the 35 percent tariff. Then in January 2002 a 18 For a comprehensive recent discussion of the paddy-rice sector, see Weerahewa (2004). Note that a variety of both consumer and producer rice prices have existed, and the wide variety of different rice varieties compound the problem of estimating average prices. 23 specific duty of Rs 7 per kg replaced the tariff ­ but the CWE was allowed to import rice at a specific duty of Rs 4 per kg. In March 2002 the licensing requirement was removed and the specific duty was raised to Rs. 5 per kg.19 It was raised further to Rs. 7 in March 2003, to Rs. 9 in August 2003, and to Rs. 20 in January 2006 (Appendix Table A3). These policy gyrations reflect the political power of the large rice producers and processors, but also the sensitivity of governments to consumer opposition to increases in the price of the main staple food. The policy liberalization in 1977 clearly reduced protection from the previous high levels, but they have not trended down since then. Sri Lanka is considered a high-cost rice producer at the margin, so the sector would come under considerable pressure, particularly in marginal areas, if trade were to be fully liberalized.20 It would be a mistake to assume that the negative rice NRAs observed in 2003 and 2004, for example, represent a permanent policy shift. The difficult balancing act between serving producer and consumer interests will continue. The policy responses will be most sensitive to domestic price movements: high domestic prices will tend to induce import liberalization, while low domestic prices will tend to induce import restrictions and/or input subsidies (constrained of course by fiscal deficit considerations). Wheat and sugar import policies The changes over time in government policies towards wheat have been closely linked to rice policies. During the 1950s and 1960s, when the government was providing a subsidized rice ration to consumers, a significant proportion of the ,,ration rice had to be imported as domestic procurement was inadequate. The government had an incentive to encourage substitution of wheat for rice in consumption to reduce the fiscal burden of the rice subsidy, because wheat was relatively cheaper in international markets than rice (and some came through US government aid). With the phase-out of the subsidized rice ration 19 This discussion is based on the WTO Trade Policy Review (2004) and IPS (2003). 20 The extent to which Sri Lanka has any comparative advantage in rice production has been the subject of several studies and much debate. See Abeyratna et al. (1990), Shilipi (1995), Rafeek and Samarathunage (2000), Kikuchi et al. (2000 and 2001), Weerahewa, Gunatilake and Pitigala (2003), and Thibbotuwawa (2004). 24 scheme starting in 1979, incentives changed. Thus from the early 1980s rice was implictly protected via tougher restrictions on wheat imports.21 Sugar imports were very heavily taxed in the pre-liberalization period. Tax levels have come down since 1977, but are still high. The past heavy taxation has a simple explanation: almost all sugar was imported as there was little domestic production, and import taxes contributed to reducing the fiscal burden imposed by other food subsidies. The prevailing domestic price of sugar was accepted by consumers though much higher than the international price, and so long as domestic prices were not increased too much, there was little political opposition to the maintenance of this implicit protection. The continuing high protection of the sugar industry in recent years is not so much for government revenue reasons as to honor a government-foreign private enterprise agreement signed in the early 1980s.22 Chillies and onions As noted previously, the liberalization of 1977 involved import liberalization that was selective and discriminatory. The agricultural sub-sectors chosen for liberalization (and continuing protection) had a clear regional dimension, which also overlapped with ethnicity. The previous protectionist policies benefited so-called subsidiary food crops (chillies, red onions) as well as crops like grapes. This encouraged expansion of these 21 The state monopoly on wheat imports and distribution that prevailed until 2001 (CWE) enabled the government to influence domestic grain prices. PRIMA, a Singapore-based company had a monopoly on wheat milling from 1980 under a 20-year agreement signed by the government (and extended for a further 5 years in 2000). CWE supplied wheat to this monopoly miller until 2001 when PRIMA was granted right to import wheat. PRIMA also gained a virtual monopoly in the animal feed sector because it could retain wheat bran and other milling by-products and had bought out the only viable competitor (a state agency that was privatised). During the contract period, PRIMA could import wheat grain and mill-related equipment duty-free, and also enjoyed income-tax exemptions. PRIMAs monopoly in wheat imports was ended in 2006 but is being challenged in courts. The official justification given for the agreement with PRIMA is that it helped to attract a wheat miller to Sri Lanka. This has certainly come at a large cost to the country. Clearly these special privileges given to a foreign owned entity impose a burden on consumers while not offering any benefits either to producers (no wheat is produced in the country) or in the way of government revenues. We refrain from speculating on why the government provided these special privileges for a large foreign firm with no obvious compensating benefits to the country. For a fuller discussion of this, see Athukorala and Kelegama (1998). 22 That agreement has some quite striking similarities with the government agreement with the PRIMA wheat agreement. This agreement has imposed a huge cost on domestic consumers while delivering extremely high profits to the subsidiary of a giant US-based sugar producer. As in the case of PRIMA, this too was ,,justified by the government on the grounds that it attracted a foreign investor. Again, there are other explanations possible as to why the government entered into this agreement, raising issues of good governance, and complicity in rent extraction and rent sharing. 25 crops in particular regions, such as the Jaffna penisula. Jaffna district produced two-thirds of the countrys entire production of red onions. The dependency of farmers in these areas on these high-value crops was such that by 1977/78 red onions and chillies accounted for around 75 percent of the total area devoted to non-rice (minor food crop) cultivation in the Jaffna district.23 It was also the only dry zone district that was a rice- deficit region. The 1977 liberalization produced a sharp fall in protection. Despite the negative income effects on producers, their opposition to the reform was ineffective. The ability of Jaffna farmers to obtain any response to their demands was constrained by the "marginality of the Sri Lankan Tamils to the electoral system as a whole" (Moore 1985, p. 109). Even so, this issue played an important role in Jaffna in the 1982 presidential elections. The escalation of the ethnic tensions into a full-scale secessionist war from 1983 onwards has had devastating effects on the agricultural economy in the northern and eastern provinces. During the past two decades, commercial ties with the rest of the country have been massively disrupted. The upshot of these developments has been that farmers in the southern regions expanded their production of minor food crops, and this has changed the political dynamic in relation to minor crop policies. This is seen most clearly in the case of policy towards crops such as potatoes, where a small group of producers have been able to exercise much political clout, and consumer resistence to high prices could be largely ignored. As seen in Table 2, despite occasional declines, protection for potatoes has tended to increase over time (with increases also ­ although to a lesser degree ­for crops such as red onions and chillies). From many viewpoints, a strong case can be made for removing assistance to potato growing, given its destructive environmental impact on the fragile ecosystems of the countrys most important river catchments and repositories of sources of bio-diversity. As Bruton (1992, p. 170) points out, "the strong commitment to the Sinalese culture made it dificult to design an economic policy that was equally appropriate for both major ethnic groups"; and there can be little doubt that the pattern of liberalization in the 1977/78 period is suggestive of a strong ethnic bias. However, the near-complete marginalization of Tamil farmers in the North and East from mainstream political life, and the collapse of the agricultural 23 Based on data from Department of Census and Statistics, Statistical Abstract ­ 1979. 26 economy of these regions from the mid-1980s due to the secesionist war, has meant that the role of ethnicity has diminished in agricultural policy formulation. Tensions between producer and consumer interests have thereby become the dominant political economy issue. Future prospects Our discussion of the political economy of agricultural policies has pointed out a range of factors that have influenced policy formation in agricultural crops. The broad sweep of trade liberalization since the late 1970s has not by-passed the agricultural sector. Further progress along these lines is likely, as many of the smaller import-competing sectors may find it difficult to withstand the pressures for reform and liberalization. But liberalization will be difficult in the case of rice, the most politically sensitive industry. The history of agricultural policy in Sri Lanka and the experience of other countries in Asia suggest that this will remain the strongest bastion of protectionist pressures in the country for some time to come. Even in recent times, whenever the rice industry has been stressed by falling prices, the government has come under pressure to placate the industry. Almost invariably the government has done so, maintaining the historical pattern of special treatment for this sector. Arguably the course of future agricultural trade liberalization will depend on the extent to which rice producers in Sri Lanka are able to maintain their political clout. But as long as rice (and wheat) remain major items in consumption baskets, the level of direct price protection will also be constrained by the need to keep an eye on consumer interests, particularly when international prices spike upwards as in 2008. Fiscal pressures will also place a limit to the extent to which governments can provide assistance in the form of input subsidies. It will also be interesting to see if governments will be able to maintain current levels and forms of protection for livestock, chicken and diary industries (past data for which were not adequate for including in this study). 27 References Abeyratne, F., E. Neville, W.G. Somaratne, and P. Wickramaarachchi (1990), "Efficiency of Rice Production and Issues Relating to Protection", Sri Lanka Journal of Agricultural Economics 1(1): 16-25. Anderson, K. (2002), "Agricultural Development and Trade Reform: Their Contributions to Economic Growth, Poverty Alleviation and Food Security in Sri Lanka", Report following a World Bank Mission to Sri Lanka, Washington DC, April. Anderson, K., M. Kurzweil, W. Martin, D. Sandri and E. Valenzuela (2008), "Methodology for Measuring Distortions to Agricultural Incentives", Agricultural Distortions Working Paper 02, World Bank, Washington DC, revised January. Posted at wwwr.worldbank.org/agdistortions Arunatilake, N., S. Jayasuriya and S. Kelegama (2001), "The Economic Cost of the War in Sri Lanka", World Development 29(9): 1483-1500. Athukorala, P.-C. and J.S. Bandara (1989), "Growth of Manufactured Exports, Primary Commodity Dependence, and Net Export Earnings: Sri Lanka", World Development 17(8): 897-903. Athukorala, P-C. and S. Jayasuriya (1994), Macroeconomic Policies, Crises and Growth in Sri Lanka, 1969-90, Washington DC: World Bank. Athukorala, P.-C. and S. Kelegama (1998), "The Political Economy of Agricultural Trade Policy: Sri Lanka in the Uruguay Round", Contemporary South Asia 7(1): 7-36. Athukorala, P.-C. and S. Rajapatirana (2000), Liberalisation and Industrial Transformation: Sri Lanka in International Perspective, Oxford and Delhi: Oxford University Press. Bhalla, S. (1991), "Sri Lanka", pp. 195-235 in A.O. Krueger, M. Schiff and A. Valdes (eds.), The Political Economy of Agricultural Pricing Policy, Vol 2, Baltimore: John Hopkins University Press. Brunton, H.J. (1992), Sri Lanka and Malaysia: The Political Economy of Poverty, Equity, and Growth, New York: Oxford University Press for the World Bank. 28 Bush, K. (1999), "The Limits and Scope for the use of Development Assistance Incentives and Disincentives for Influencing Conflict Situations: Case Study for Sri Lanka", Development Assistance Committee, Paris: OECD. Central Bank of Sri Lanka (various issues), Annual Report, Columbo. Central Bank of Sri Lanka (various issues), Economics and Social Statistics of Sri Lanka, Columbo. Cuthbertson, A.G. and P. Athukorala (1990), "Sri Lanka", Part 3 in D. Papageorgiou, M. Michaely and A.M. Choksi (eds.), Liberalizing Foreign Trade: Indonesia, Pakistan and Sri Lanka, pp. 287-414, Oxford: Basil Blackwell. de Silva, K.M. (1981), A History of Sri Lanka, Delhi: Oxford University Press. de Silva, H. and W. Sumith (1979), " The Coconut Industry in Sri Lanka: An Analysis of Government Intervention Measures", unpublished Masters Thesis, Australian National University, Canberra. Department of Census and Statistics (various issues), National Accounts of Sri Lanka, Columbo. Department of Census and Statistics (1979), Statistical Abstract of the Democratic Socialsit Republic of Sri Lanka -1979, Colombo. Dissanayake, G. (1983), "Foreword", in Mahaweli Projects and Programme 1984, Ministry of Lands and Land Development and Mahaweli Development, Colombo. Dunham, D. and S. Kelegama (1997), "Does Leadership Matter in the Economic Reform Process? Liberalization and Governance in Sri Lanka, 1989-93", World Development 25: 179-190. Easterly, W. (2006), Global Development Network Growth Database, accessible at http://www.nyu.edu/fas/institute/dri/global%20development%20network%20gro wth%20database.htm Edwards, C. (1993), A Report on Protectionism and Trade Policy in Manufacturing and Agriculture, Sri Lanka, Colombo: Institute of Policy Studies. Ganewatta, G. (2002), The Market for Value Added Tea Products of Sri Lanka: an Economic Analysis, unpublished PhD Thesis, LaTrobe University, Melbourne. 29 Gunasinghe, N. (1984), "The Open Economy and its Impact on Ethnic Relations in Sri Lanka", in Committee for Rational Development in Sri Lanka (CRD), Sri Lanka: the Ethnic Conflict, Navarang, New Delhi. International Monetary Fund (2005), Sri Lanka: Selected Issues and Statistical Appendix, IMF Country Report No. 05/337, Washington DC: IMF. Institute of Policy Studies (2003), Sri Lanka: State of the Economy 2002, Colombo: Institute of Policy Studies. Jayanetti, S. and G. Tilakaratna (2005), "Impact of Trade Liberalization on Poverty and Houehold Welfare in Sri Lanka", Research Studies: Poverty and Social Welfare Series No. 6, Colombo: Institute of Policy Studies. Kelegama, S. (2004) (ed.) Economic Policy in Sri Lanka: Issues and Debates, New Delhi: Sage Publications and London: Thousand Oaks. Kikuchi, M., R. Barker, M. Samad, and P. Weligamage (2000), "Comparative Advantage of Rice Production in an Ex-rice Importing Country: The Case of Sri Lanka", Paper presented at the III Conference of Asian Society of Agricultural Economists at Jaipur, 18-20 October. Kikuchi, M., R. Barker, M. Samad, and P. Weligamage (2001), "Comparative Advantage of Rice Production in Sri Lanka with Special Reference to Irrigation Costs", Paper presented at the Workshop on Medium- and Long-Term Prospects of Rice Supply and Demand in the 21st Century 3-5 December. Klingebiel, S. (2001), "The OECD, World Bank and International Monetary Fund: Development Activities in the Crisis Prevention and Conflict Management Sphere", German Development Institute (GDI), Bonn. Kreimer, A., J. Eriksson, R. Muscat, M. Arnold and C. Scott (1998), Experience with Post-Conflict Reconstruction, A World Bank Operations Evaluation Study, Washington DC: World Bank. Ministry of Plantation Industries (various issues), Plantation Sector Pocket Book, Columbo. Moore M. (1985), The State and Peasant Politics in Sri Lanka, Cambridge, Cambridge University Press. 30 Peebles, P. (1990), "Colonization and Ethnic Conflict in the Dry Zone of Sri Lanka", The Journal of Asian Studies 49(1): 30-55. Rafeek, M.I.M. and P.A. Samarathunga (2000), " Trade Liberalization and its Impact on the Rice Sector of Sri Lanka", Sri Lankan Journal of Agricultural Economics 3: 143-154. Sanderatne, N. (2004), "Agricultural Development: Controversial Issues", pp. 195-212 in S. Kelegama (ed.), Economic Policy in Sri Lanka: Issues and Debates, New Delhi: Sage Publications and London: Thousand Oaks. Shilpi, F. (1995), "Policy Incentive, Diversification and Comparative Advantage of Non- Plantation Crops in Sri Lanka", Working Paper 2, World Bank, Washington DC. Tennekoon, S.N. (1988), "Rituals of Development: The Accelerated Mahawali Development Program of Sri Lanka", American Ethnologist 15(2): 294-310. Thibbotuwawa, M. (2004), "Analysis of the Comparative Advantage and Income from Paddy Farming in Anuradhapura, Kurunagala and Gampaha Districts with Special Emphasis on Scale of Cultivation", unpublished Project Report, Department of Agricultural Economics and Business Management, University of Peradeniya. Thorbecke, E. and J. Svejnar (1987), Economic Policies & Agricultural Performance in Sri Lanka: 1960-1984, Development Centre Studies, Paris: OECD. Weerahewa, J. (2004), "Impacts of Trade Liberalization and Market Reforms on the Paddy/Rice Sector in Sri Lanka", MITD Discussion Paper No. 70, IFPRI, Washington DC. Weerahewa, J., H.M. Gunatilake and H. Pitigala (2003), "Future of Paddy Farming in Sri Lanka: Scale, Comparative Advantage and Rural Poverty", Sri Lanka Economic Journal 3(2): 104-144. Weerakoon, Dushni (2004), "The Influence of Development Ideology in Macroeconomic Policy Reform Process, pp. 54-70 in S. Kelegama (ed.), Economic Policy in Sri Lanka: Issues and Debates, New Delhi: Sage Publications and London: Thousand Oaks. World Bank (1995), Sri Lanka: Non-Plantation Crop Sector Policy Alternatives, Report No: 14564-CE, Washington DC: World Bank. 31 World Bank (2003), Sri Lanka: Promoting Agricultural and Rural Non-farm Sector Growth, Report No. 25387-CE, Washington DC: World Bank. World Bank (2004), Trade Policies in South Asia: An Overview, Report No. 29949, Washington DC: World Bank. World Trade Organization (2004), Trade Policy Review: Sri Lanka, Geneva: World Trade Organization. 32 Figure 1: Real GDP growth rate and political episodes and policy regimes,a Sri Lanka, 1951 to 2005 a The United National Party (UNP) has been a right of centre party; the Sri Lanka Freedom Party (SLFP) has been a left of centre party. Both parties have frequently entered into coalition governments with other parties. The Peoples Alliance (PA) is a SLFP- dominated coalition. Source: Authors compilation Figure 2: Share of agriculture in GDP and exports, Sri Lanka, 1950 to 2005 (percent) 100 90 80 70 Percentage 60 50 40 30 20 10 0 1950 1953 1956 1959 1962 1965 1968 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 Years Ag Share of GDP Ag Share of Exports Source: Central Bank of Sri Lanka, Annual Report, Various Issues 34 Figure 3: Shares of covered crops in value of agricultural production at undistored prices, Sri Lanka, 1966 to 2004 (percent) 100 90 Residual 80 Onion 70 Chillie 60 Potatoe 50 Rubber 40 30 Coconut 20 Tea 10 Rice 0 66 70 74 78 82 86 90 94 98 02 19 19 19 19 19 19 19 19 19 20 Source: Derived from FAO data and authors spreadsheet 35 Figure 4: Fertilizer subsidy, Sri Lanka, 1962 to 2007 $US Millions 120 100 80 $US Millions 60 40 20 0 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 Years Source: Central Bank of Sri Lanka (various issues). -80 -70 -60 -50 -40 -30 -20 -10 0 -60 -50 -40 -30 -20 -10 0 1955 1955 1957 1957 1959 1959 1961 1961 1963 1963 1965 1965 1967 1967 1969 1969 1971 1971 1973 1973 1975 1975 1977 1977 36 Tea 1979 1979 Rubber (percent) 1981 1981 1983 1983 1985 1985 1987 1987 1989 1989 1991 1991 1993 1993 1995 1995 1997 1997 1999 1999 2001 2001 2003 2003 Figure 5: Nominal rates of assistance to tea, rubber, coconut and rice, Sri Lanka, 1955 to 2004 -60 -40 -20 0 20 40 60 -40 -20 0 20 40 60 80 100 1955 1955 1957 1957 1955 to 2004 1959 1959 1961 1961 1963 1963 1965 1965 Source: Authors spreadsheet 1967 1967 1969 1969 1971 1971 1973 1973 1975 1975 1977 1977 37 1979 Rice 1979 (percent) Coconut 1981 1981 1983 1983 1985 1985 1987 1987 1989 1989 1991 1991 1993 1993 1995 1995 1997 1997 1999 1999 2001 2001 Figure 5 (continued): Nominal rates of assistance to tea, rubber, coconut and rice, Sri Lanka, 2003 2003 38 Figure 6: Nominal rates of assistance to exportable, importable and all covered agricultural products, Sri Lanka, 1955 to 2004 (percent) 100 Import-competing Total 80 Exportables 60 40 20 0 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 -20 -40 -60 Source: Authors' spreadsheet 39 Figure 7: Nominal rates of assistance to all agricultural tradable industries, all nonagricultural tradables, and relative rates of assistance, Sri Lanka, 1955 to 2004 (percent) NRA ag tradables 160 NRA non-ag tradables RRA 120 80 40 0 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 -40 -80 t t t a The RRA is defined as 100*[(100+NRAag )/(100+NRAnonag )-1], where NRAag and t NRAnonag are the percentage NRAs for the tradable parts of the agricultural and nonagricultural sectors, respectively. Source: Authors spreadsheet 40 Table 1: Agricultures share of GDP, employment and exports, Sri Lanka, 1950 to 2005 (percent) 1950-51 1960-61 1970-71 1980-81 1990-91 2000-02 2005 Sectoral shares of GDP Agriculture 44.5 34.6 35.1 33.7 26.3 21.9 18.9 Plantation Agriculture 26.3 17.8 15.8 13.9 8.1 5.2 4.4 Tea 7.7 6.5 3.2 2.6 2.4 1.4 1.2 Rubber 5.5 2.3 1.8 1.4 0.6 0.4 0.4 Coconut 7.1 4.8 4.0 3.8 2.4 1.4 1.1 Processing of tea, 6.5 4.2 6.8 6.1 2.7 2.0 1.7 rubber, coconuta Non-plantation Agriculture 14.6 14.5 16.7 15.0 14.8 12.1 11.5 Paddy 1.5 5.7 7.2 6.1 4.9 3.1 3.0 Other 13.1 8.8 9.4 8.9 9.9 9.0 8.5 Forestry 1.9 1.4 1.4 2.0 1.6 1.9 1.7 Fishing 1.2 0.9 1.2 2.7 1.8 2.7 1.3 Industries 8 9 14 18 23 26 27 Services 53 55 51 48 51 52 55 Sectoral shares of exports Agricultural Exports >90 >90 94.6 61.6 36.3 19.3 18.2 Tea >50 >50 55.5 35.1 24.9 14.3 12.8 Rubber >20 >20 22.0 14.7 3.9 0.5 0.7 Coconut >15 >15 14.5 7.0 3.5 1.7 1.8 Other 3.1 4.8 4.0 2.8 2.9 Manufactures <2 <2 1.7 31.2 56.6 77.1 80.3 Unclassified 3.9 2.7 7.1 1.0 1.5 Sectoral shares of employment 1953 1963 1971 1981 1991 2001 2005 Agriculture 53 53 50 45 42 33 31 Industry 12 12 12 15 20 22 24 Services 35 35 38 40 38 45 45 a Processing of tea, rubber and coconut products is usually included under manufacturing. Here it is included under Agriculture, Forestry and Fisheries in GDP and export shares, but not in employment shares. Source: Central Bank of Sri Lanka (various issues) Table 2: NRAs of covered agricultural products, Sri Lanka, 1955 to 2004 (percent, five-year averages) 1955-59 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 a Exportables -22.8 -40.0 -38.6 -41.1 -45.2 -31.1 -21.4 -24.2 -2.0 5.9 Rubber -15.8 -51.5 -48.9 -56.2 -59.6 -51.4 -37.7 -21.8 -4.4 -0.2 Coconut -28.6 -29.6 -24.9 -32.8 -36.5 -19.6 -5.1 -34.6 -1.3 16.7 Tea -22.3 -39.4 -39.1 -36.8 -37.4 -30.4 -25.6 -12.6 -1.5 -1.2 Importablesa 62.5 11.9 -5.9 9.0 -3.7 -0.6 -2.1 22.4 31.8 12.8 Rice 62.5 11.9 -5.9 9.0 -7.7 -5.8 0.1 8.6 19.2 3.7 Potato - - - - 77.6 43.3 32.6 157.7 124.8 205.8 Onion - - - - -11.6 28.7 -12.6 43.7 79.3 53.4 Chillies - - - - 52.6 33.4 6.9 62.1 76.9 67.2 All covered productsa -10.3 -29.9 -30.0 -20.3 -31.9 -19.2 -12.6 -1.7 11.5 8.6 Dispersion of covered 44.2 28.7 20.9 31.9 26.2 22.8 22.4 25.0 20.8 12.6 productsb % coverage (at 66 66 67 65 64 63 67 65 62 64 undistorted prices) a Weighted averages, with weights based on value of production at undistorted prices. b Dispersion is a simple 5-year average of the annual standard deviation around the weighted mean of NRAs of covered products. Source: Authors spreadsheet Table 3: Nominal rates of assistance to agricultural and non-agricultural industries and relative rate of assistance, Sri Lanka, 1955 to 2004 (percent) 1955-59 1960-64 1965-69 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 a Covered agric. products -10.3 -29.9 -30.0 -20.3 -31.9 -19.2 -12.6 -1.7 11.5 8.6 Non-covered agric. products 13.2 -9.3 -14.8 -10.7 -16.3 -10.6 -7.8 -0.6 9.9 6.2 All agricultural productsa -2.4 -23.0 -24.9 -16.9 -26.4 -16.0 -11.1 -1.4 10.9 7.7 Non-product-specific input assistance (NPS) 0.0 0.2 0.4 0.6 0.8 2.4 1.1 0.1 1.3 1.7 Total agriculture (incl. NPS)b -2.4 -22.8 -24.5 -16.3 -25.5 -13.5 -9.9 -1.2 12.2 9.5 Agricultural trade bias indexc -0.52 -0.45 -0.35 -0.45 -0.43 -0.31 -0.18 -0.38 -0.25 -0.05 Tradables NRA, All Agriculture -2.7 -25.7 -27.6 -18.5 -29.0 -15.4 -11.2 -1.3 14.0 10.8 NRA, All Non-Agriculture 104.9 124.6 138.4 70.7 52.9 57.1 59.0 47.1 36.4 22.9 Relative rate of assistance (RRA)d -52.5 -66.6 -68.0 -51.6 -53.5 -46.2 -44.3 -32.9 -16.3 -9.8 MEMO, ignoring exchange rate distortions:d NRA, all agric. products -2.4 -12.9 -15.2 -11.0 -19.7 -11.0 -8.0 -0.9 11.7 9.5 Agricultural trade bias index -0.52 -0.21 -0.06 -0.23 -0.23 -0.22 -0.09 -0.34 -0.24 -0.05 RRA -52.5 -61.7 -63.9 -48.0 -49.2 -44.3 -42.9 -32.6 -16.8 -9.8 a NRA including product-specific input subsidies. b NRAs including product-specific input subsidies and non-product-specific (NPS) assistance. Total of assistance to primary factors and intermediate inputs divided by total value of primary agriculture production at undistorted price, expressed as a percentage. c Trade bias index is TBI = (1+NRAagx/100)/(1+NRAagm/100) ­ 1, where NRAagm and NRAagx are the average percentage NRAs for the import-competing and exportable parts of the agricultural sector. d The RRA is defined as 100*[(100+NRAagt )/(100+NRAnonagt)-1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and nonagricultural sectors, respectively. Source: Authors spreadsheet Appendix: Key quantity and price data, assumptions and sources for Sri Lanka Volume of production, exports and imports data for agricultural products From various publications of Central Bank of Sri Lankas Annual Report (various issues) and Economics and Social Statistics of Sri Lanka (various issues) and FAO. Farm-gate product prices data Data for paddy/rice are from Bhalla (1987) for the period of 1955 to 1986 and data are from the data series of producer price of paddy of Central Bank of Sri Lanka Annual Report (various issues) and adjusted following the method of Bhalla (1987). Data for other products are from Central Bank of Sri Lankas Annual Report (various issues) and Economics and Social Statistics of Sri Lanka (various issues) and FAO. Wholesale and retail product prices From Central Bank of Sri Lankas Annual Report (various issues) and Economics and Social Statistics of Sri Lanka (various issues) Border prices Fob and cif prices are directly from Central Bank of Sri Lankas Annual Report (various issues) and Economics and Social Statistics of Sri Lanka (various issues) and FAO. Exchange rates Official exchange rates are from Central Bank of Sri Lankas Annual Report (various issues) and Economics and Social Statistics of Sri Lanka (various issues) and, for black market premia, from Easterly (2006). Production, consumption, input and trade taxes and subsidies These data are from Central Bank of Sri Lanka's Annual Report (various issues) and Economics and Social Statistics of Sri Lanka (various issues) and the Ministry of Plantation Industries (various issues). 44 Appendix Table A1: Production of major agricultural crops, Sri Lanka, 1948 to 2005 Minor Exports Sub.Food Tea Rubber Coconut Paddy Crops Crops Sugar Chillies &Peper Red Onion B'Onion Potatoes Prod Pro(Kg, Year (Mn.Kg) Mn) Prod(Mn.nuts) Prod(MT,000) Prod(MT'000) Prod(MT'000) Prod(MT'000) Prod (MT) Prod (MT) Prod (MT) Prod(MT) 1948 136 96 1765 390.7 na na 1949 135 91 1763 482.7 3 14 1950 139 115 1982 461 4 37 1951 148 107 2238 459.7 4 39 1952 144 98 2455 603.9 na 44 1953 137 100 2288 457.6 na 36 1954 166 95 2203 649.8 na 0 1955 172 95 2420 745 8 73 1956 170 97 2374 561 8 68 1957 180 99 2108 652 8 75 1958 187 102 2109 763 21 55 1959 187 93 2313 760 20 59 1960 197 99 2183 897 20 62 1961 206 98 2601 899 21 76 11,000 36,593 3,928 1962 212 104 2811 1001 17 64 12,000 0 1,148 1963 220 105 2549 1026 8 87 13,000 29,216 7,271 1964 219 112 2991 1054 11 59 16,243 27,202 4,721 1965 228 119 2681 757 23 55 20,231 31,956 2,117 1966 222 131 2468 954 14 99 19,367 31,888 7,361 1967 221 144 2416 1145 21 97 28,916 38,425 11,509 1968 225 149 2601 1346 13 68 24,108 34,922 18,270 1969 220 151 2440 1374 15 91 19,812 34,581 26,927 1970 212 160 2410 1616 26 78 32,198 38,218 31,741 1971 218 142 2610 1396 14 106 30,000 41,613 30,500 1972 213 140 2963 1312 15 133 31,980 43,679 31,279 1973 211 155 1935 1312 49 144 40,037 46,011 26,384 1974 204 132 2031 1602 50 144 44,581 52,798 26,896 45 1975 214 149 2398 1154 46 160 37,607 55,680 21,924 1976 197 152 2330 1252 50 171 43,535 58,407 27,876 1977 209 146 1821 1677 52 181 42,972 61,496 29,102 1978 199 116 2207 1891 60 182 38,586 58,426 29,098 1979 206 153 2393 1917 58 173 46,400 67,900 37,900 1980 191 133 2026 2133 63 225 50,987 66,891 51,121 1981 210 124 2258 2230 54 231 37,500 59,100 66,000 1982 188 125 2521 2156 68 247 36,445 67,543 76,893 1983 179 140 2312 2484 69 313 40,600 95,300 89,800 1984 208 142 1942 2420 42 219 73,600 36,700 3,249 68,300 1985 214 138 2958 2661 41 261 98,700 41,700 3,184 89,000 1986 211 138 3039 2588 25 296 105,784 57,124 3,806 82,482 1987 213 122 2292 2127 26 299 29 73,501 56,267 2,723 81,042 1988 227 122 1937 2477 27 329 53 82,700 59,200 4,222 87,500 1989 207 111 2484 2063 31 299 54 67,900 77,000 5,365 83,500 1990 233 113 2532 2538 33 307 57 100,000 61,000 15,903 87,200 1991 241 104 2184 2389 35 258 66 99,507 41,630 14,046 66,737 1992 179 106 2296 2340 28 254 60 73,919 82,340 27,879 78,562 1993 232 104 2164 2570 37 342 69 94,700 72,860 22,838 78,180 1994 242 105 2622 2684 34 293 72 93,014 82,950 34,726 79,385 1995 246 106 2755 2810 20 232 71 101,485 78,110 29,719 81,657 1996 258 112 2546 2061 20 305 70 73,611 63,305 19,367 100,755 1997 277 106 2631 2239 21 257 63 72,231 73,940 29,138 66,484 1998 280 96 2552 2692 24 207 62 62,470 55,480 17,444 25,900 1999 284 97 2828 2857 31 259 66 60,030 105,380 62,729 27,170 2000 306 88 3096 2860 30 248 64 55,860 79,060 36,560 48,410 2001 295 86 2769 2695 28 226 48 49,040 68,830 31,966 57,680 2002 310 90 2392 2856 33 241 38 46,350 66,890 31,560 88,710 2003 303 92 2562 3071 33 244 61 46,190 67,820 32,301 71,740 2004 308 95 2591 2609 63 40,480 76,970 37,508 81,270 2005 2515 50,000 80,000 55,552 76,300 Source: Central Bank of Sri Lanka (various issues of Annual Report) 46 Appendix Table A2: Composition of imports, Sri Lanka, 1070 to 2004 (percent) 1970 1980 1990 2000 2004 Agricultural Imports 46.4 20.6 14.2 11.3 11.2 Rice 15.1 2.2 1.2 0.1 0.8 Flour 9.7 5.3 1.0 0.0 0.0 Wheat & Meslin 1.0 1.6 2.1 1.7 2.3 Sugar 6.7 5.7 4.9 1.9 1.4 Milk and milk products 2.4 2.5 2.2 1.6 1.5 Other 11.5 3.3 2.8 6.05 5.2 Other Imports 53.6 79.4 85.8 88.7 88.8 Total 100.0 100.0 100.0 100.0 100.0 Source: Central Bank of Sri Lanka (various issues) 47 Appendix Table A3: Rice import tariffs, Sri Lanka, 1995 to 2006 Period Statutory Duty Effective Surcharge TT/GST DL/ Total Tax Duty Waiver Import % VAT % NSL Incidence Duty/tariff % Jan 1, 1995-Feb 7, 1995 35% or 0 55% 0 EX 4.5 65.7% Rs7/Kg Feb 8, 1995- April 1996 35% 0 35% 0 EX 4.5 44.6% Apr 15, 1996-Jan 30, 1997 35% 35% 0 0 EX 4.5 7.6% Jan 31, 1997-Nov 20, 1997 35% 0 35% 0 EX 4.5 44.6% Nov 21, 1997- Jan 31, 1998 35% 35% 0 0 EX 4.5 7.6% Feb 1, 1998- Nov 5, 1998 0 35% 0 EX 4.5 44.6% Nov 6, 1998-Oct 23, 1999 35% 0 35% 0 EX 5.5 46.3% Oct 24, 1999-Dec 31, 1999 35% 25% 10% 0 EX 6.5 20.9% Jan 1, 2000-May 10, 2000 35% 0 35% 0 EX 6.5 48.0% May 11, 2000-July 16, 2000 35% 0 35% 0 EX 6.5 48.0% July 17, 2000-Feb 19, 2001 35%+QR 0 35%+QR 0 EX 6.5 48.0% Feb 20, 2001- Mar 31, 2001 35%+QR 0 35%+QR 40 EX 6.5 60.0% Apr 01, 2001-Sep 11, 2001 35%+QR 0 35%+QR 40 EX 7.5 61.7% Sep 12, 2001 ­ Nov 21, 2001 35%+QR 0 35%+QR 40 EX 6.5 60.0% Nov 22, 2001-Dec 8, 2001 35%+QR 35% 0+QR 40 EX 6.5 8.1% Dec 9, 2001-Dec 31,2001 35%+QR 17.5% 17.5%+QR 40 EX 6.5 34.1% Jan 1, 2002- Jan 20, 2002 35%+QR 0 35%+QR 40 EX 6.5 60.0% Jan 21, 2002- Jul 31, 2002 Rs 7/Kg 0 Rs. 7/Kg 0 EX 6.5 55.6% CWE was allowed to import 30,000 MT August 1, 2002 ­ Nov 5, 2002 Rs 7/Kg 0 Rs. 7/Kg 0 10 36.6% Nov 6, 2002-Mar 4, 2003 Rs 5/Kg 0 Rs. 5/Kg 0 10 27.4% Mar 5, 2003-Aug 18, 2003 Rs 7/Kg 0 Rs. 7/Kg 0 10 36.4% Aug 19, 2003-Jan 31, 2006 Rs 9/Kg 0 Rs. 9/Kg 0 10 Na Jan 31, 2006 todate Rs 20/Kg 0 Rs: 20/Kg 0 10 Na Source: UpadatedTable 4 of Jayanetti and Tilakaratna (2005) using Various Notifications of Sri Lanka Customs. a. Includes the defense levy, stamp duty & other surcharges Figures in brackets are Average Duty Collection Rates. 48 Appendix Table A4: Exchange rates, Sri Lanka, 1960 to 2005 (Rs./US$) Year Official Exchange Rate Secondary Market Rate 1960 4.75 6.8 1961 4.76 8.1 1962 4.76 9.6 1963 4.76 11.3 1964 4.78 12.6 1965 4.78 11.5 1966 4.78 16.2 1967 5.93 14.1 1968 5.93 7.9 1969 5.96 7.7 1970 5.96 9.8 1971 5.96 13.4 1972 6.7 16.6 1973 7.75 11.0 1974 6.6 13.9 1975 7.71 16.9 1976 8.83 24.3 1977 15.56 17.7 1978 15.51 21.2 1979 15.45 25.0 1980 18 22.4 1981 20.55 22.6 1982 21.32 27.6 1983 25 36.4 1984 26.28 36.2 1985 27.41 32.8 1986 28.52 31.5 1987 30.76 33.8 1988 33.03 54.5 1989 40 48.3 1990 40.24 53.1 1991 42.58 53.5 1992 46 54.2 1993 49.56 54.8 1994 49.98 54.1 1995 54.05 56.7 1996 56.71 62.1 1997 61.29 68.8 1998 67.78 72.1 1999 72.12 80.1 2000 80.06 80.1 2001 93.16 93.2 2002 96.73 96.7 2003 96.74 96.7 2004 104.61 104.6 Source: Official from Central Bank of Sri Lanka; black market premia from Easterly (2006) 49 Appendix Table A5: Annual distortion estimates, Sri Lanka, 1955 to 2004 (a) Nominal rates of assistance to covered products (percent) Chillies Coconut Onion Potato Rice Rubber Tea All covered 1955 na -35 na na 42 -18.0 -23 -14 1956 na -24 na na 51 -3.6 -18 -8 1957 na -28 na na 70 -19.0 -23 -10 1958 na -29 na na 95 -24.9 -25 -9 1959 na -28 na na 54 -13.5 -22 -11 1960 na -32 na na 42 -38.5 -32 -21 1961 na -37 na na 32 -45.6 -35 -26 1962 na -23 na na -7 -53 -40 -31 1963 na -26 na na 4 -59 -44 -34 1964 na -30 na na -11 -62 -46 -38 1965 na -31 na na -16 -59 -43 -38 1966 na -18 na na -23 -70 -51 -44 1967 na -19 na na -9 -57 -44 -32 1968 na -34 na na 1 -25 -28 -20 1969 na -23 na na 18 -34 -30 -15 1970 na -22 na na 15 -48 -36 -19 1971 na -48 na na 7 -60 -41 -7 1972 na -33 na na 3 -56 -44 -31 1973 na -26 na na 42 -45 -30 -10 1974 na -34 na na -22 -72 -33 -35 1975 na -30 na na -12 -61 -39 -31 1976 na -4 na 91 -16 -70 -40 -35 1977 na -37 na 23 38 -38 -15 -11 1978 36 -53 -36 150 -18 -61 -45 -38 1979 69 -58 13 46 -29 -67 -48 -45 1980 52 -52 -9 -41 -12 -55 -36 -33 1981 36 -25 97 164 -4 -53 -35 -18 1982 48 -9 16 25 -1 -48 -30 -12 1983 35 1 11 28 -10 -48 -26 -15 1984 -5 -13 28 40 -2 -54 -25 -17 1985 22 -14 51 70 8 -30 -30 -7 1986 0 26 59 80 23 -31 -25 5 1987 -22 25 -62 -39 7 -30 -21 -8 1988 30 -21 -87 14 -29 -54 -33 -33 1989 5 -42 -25 38 -10 -43 -20 -20 1990 41 -45 -47 67 4 -46 -25 -14 1991 47 -33 3 178 -1 -40 -28 -8 1992 32 -34 -42 146 8 -15 -5 -5 1993 73 -26 68 144 18 -4 -2 9 1994 118 -35 236 253 14 -4 -2 10 1995 102 -42 61 177 29 -5 -2 10 1996 59 -30 86 66 15 -7 -2 0 1997 74 10 65 83 22 -7 -1 13 1998 83 35 108 149 6 -3 -1 16 1999 67 20 78 150 23 0 -2 18 2000 71 25 65 160 3 0 -1 11 2001 81 50 78 225 7 0 -2 19 2002 63 19 37 181 23 0 -1 17 2003 54 1 34 257 -5 0 -1 2 2004 na -10 na na -9 -1 -1 -6 50 Appendix Table A5 (continued): Annual distortion estimates, Sri Lanka, 1955 to 2004 (b) Nominal and relative rates of assistance to alla agricultural products, to exportableb and import-competing b agricultural industries, and relativec to non-agricultural industries (percent) Total ag NRA Ag tradables NRA Non- All Non-ag Covered products covered products Export- Import- tradables Inputs Outputs products (incl NPS) ables competing All NRA RRA 1955 0 -14 6 -7 -25 42 -8 105 -55 1956 0 -8 12 -1 -16 51 -1 105 -52 1957 0 -10 16 -1 -23 70 -2 105 -52 1958 0 -9 23 2 -26 95 2 105 -50 1959 0 -11 10 -4 -23 54 -4 104 -53 1960 0 -21 3 -13 -34 42 -14 103 -58 1961 0 -26 -2 -18 -38 32 -20 114 -63 1962 0 -31 -15 -26 -39 -7 -29 130 -69 1963 0 -34 -13 -27 -44 4 -30 135 -70 1964 0 -38 -19 -31 -46 -11 -36 142 -73 1965 0 -38 -20 -32 -43 -16 -36 158 -75 1966 0 -44 -24 -37 -50 -23 -42 172 -79 1967 0 -32 -17 -26 -41 -9 -30 165 -74 1968 0 -20 -10 -17 -30 1 -19 124 -64 1969 0 -15 -3 -11 -29 18 -12 73 -50 1970 0 -19 -6 -15 -35 15 -17 77 -53 1971 0 -7 -14 -9 -48 7 -11 93 -54 1972 0 -31 -14 -24 -44 3 -28 93 -63 1973 0 -10 3 -4 -34 42 -5 52 -38 1974 0 -35 -22 -29 -45 -22 -34 39 -53 1975 0 -31 -18 -25 -43 -12 -30 43 -51 1976 0 -35 -20 -29 -45 -13 -33 65 -60 1977 0 -11 3 -5 -28 37 -7 55 -40 1978 0 -38 -22 -31 -52 -13 -37 51 -58 1979 0 -45 -25 -37 -58 -17 -43 51 -62 1980 0 -33 -19 -26 -48 -8 -32 51 -55 1981 0 -18 -10 -11 -34 5 -17 56 -47 1982 0 -12 -7 -8 -25 4 -12 56 -43 1983 0 -15 -9 -11 -22 -4 -14 56 -45 1984 0 -17 -9 -13 -26 0 -16 68 -50 1985 0 -7 -3 -4 -23 15 -7 68 -44 1986 0 5 4 6 -10 22 5 68 -37 1987 0 -8 -5 -6 -6 -10 -8 53 -40 1988 0 -33 -22 -29 -35 -32 -33 53 -56 1989 0 -20 -13 -17 -33 -5 -20 53 -48 1990 0 -14 -9 -13 -35 7 -14 45 -41 1991 0 -8 -6 -7 -32 15 -8 45 -37 1992 0 -5 -4 -4 -23 12 -5 52 -38 1993 0 9 7 8 -14 34 10 50 -27 1994 0 10 9 10 -18 44 11 43 -23 1995 0 10 9 11 -20 47 11 37 -19 1996 0 0 4 3 -16 29 2 40 -27 1997 0 13 12 14 3 32 14 36 -16 1998 0 16 11 16 13 20 16 35 -14 1999 0 18 14 18 10 32 19 33 -11 2000 0 11 8 11 10 14 11 21 -8 2001 0 19 13 19 17 22 19 23 -3 2002 0 17 13 17 8 32 17 23 -4 2003 0 2 2 4 0 6 3 25 -18 2004 0 -6 -5 -4 -4 -9 -6 22 -23 a. NRAs including assistance to nontradables and non-product specific assistance. b. NRAs including products specific input subsidies. 51 c. The Relative Rate of Assistance (RRA) is defined as 100*[(100+NRAagt)/ (100+NRAnonagt)-1], where NRAagt and NRAnonagt are the percentage NRAs for the tradables parts of the agricultural and non-agricultural sectors, respectively. 52 Appendix Table A5 (continued): Annual distortion estimates, Sri Lanka, 1955 to 2004 (c) Value shares of primary production of covereda and non-covered products, (percent) Chillies Coconut Onion Potato Rice Rubber Tea Non-covered 1955 na 15 na na 11 12 27 34 1956 na 16 na na 9 13 28 34 1957 na 16 na na 9 13 27 34 1958 na 18 na na 9 12 27 34 1959 na 21 na na 10 11 24 34 1960 na 15 na na 11 14 26 34 1961 na 15 na na 11 12 28 34 1962 na 12 na na 15 12 26 34 1963 na 12 na na 14 13 27 34 1964 na 14 na na 15 13 25 34 1965 na 16 na na 12 13 25 34 1966 na 11 na na 14 18 23 34 1967 na 11 na na 18 13 21 36 1968 na 20 na na 21 8 19 32 1969 na 17 na na 20 12 18 32 1970 na 16 na na 21 13 18 31 1971 na 5 na na 48 4 7 36 1972 na 12 na na 17 11 22 38 1973 na 12 na na 19 13 16 39 1974 na 14 na na 31 11 12 33 1975 na 10 na na 23 10 15 43 1976 na 9 na 1 20 17 16 37 1977 na 21 na 1 16 8 21 33 1978 2 18 2 0 21 9 16 32 1979 2 19 1 1 18 14 11 35 1980 2 19 1 1 18 10 11 37 1981 2 16 1 1 23 8 12 38 1982 2 14 1 1 23 7 13 38 1983 2 12 1 1 21 8 16 38 1984 3 16 1 1 16 8 20 36 1985 6 16 1 1 20 7 17 32 1986 8 11 1 2 21 9 15 34 1987 6 12 2 4 18 7 15 35 1988 4 10 7 2 22 9 13 32 1989 6 13 1 2 23 7 15 34 1990 5 11 3 1 24 6 18 31 1991 7 13 1 1 24 5 16 33 1992 6 18 4 1 24 4 9 34 1993 5 15 1 1 23 4 13 37 1994 4 16 1 1 21 6 12 39 1995 4 15 1 1 21 7 11 40 1996 4 18 1 1 15 6 14 41 1997 2 19 1 1 16 5 17 40 1998 3 16 1 0 20 4 19 38 1999 3 20 2 0 22 3 16 34 2000 2 15 1 0 22 3 19 36 2001 2 14 1 1 21 3 21 37 2002 2 17 1 1 20 4 19 36 2003 2 14 1 1 23 5 18 36 2004 na 16 na na 22 6 21 36 a At farmgate undistorted prices Source: Authors spreadsheet