56073 v1 Distortions to Agricultural Incentives in Indonesia George Fane and Peter Warr Australian National University George.Fane@anu.edu.au Peter.Warr@anu.edu.au Agricultural Distortions Working Paper 24, December 2007 This is a product of a research project on Distortions to Agricultural Incentives, under the leadership of Kym Anderson of the World Bank's Development Research Group (www.worldbank.org/agdistortions). The authors gratefully acknowledge the excellent research assistance of Arief Ramayandi and the helpful comments and assistance with data of Rina Oktaviani, Neil McCulloch, Peter Rosner and Peter Timmer, helpful comments from workshop participants, intensive spreadsheet assistance from Marianne Kurzweil and Ernesto Valenzuela, and 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. Distortions to Agricultural Incentives in Indonesia George Fane and Peter Warr In Indonesia, agricultural trade policy is a politically charged subject. The staple food, rice, is a net import and this one commodity has been a central focus of Indonesian food policy throughout the post-Independence period. Self-sufficiency in rice, meaning the elimination of rice imports, has been a cherished goal of agricultural policy for all of this time. It is an emotive subject, closely linked in the public imagination to Indonesian nationalism. When asked his proudest single achievement, Soeharto, Indonesia's president for the 32 years from 1966 to 1998, cited the (temporary) achievement of self-sufficiency in rice.1 This dominance of rice will become evident in this study, which documents the changing structure of agricultural protection in Indonesia and attempts to explain the forces that have driven it. Our four central themes can be summarized as follows. First, variations in distortion by sector have been driven by the government's wishes (a) to be self-sufficient in food, (b) to stabilize food prices at acceptable levels, and (c) to promote manufacturing. Food processing has been an important component of manufacturing and was even more important in the 1970s and 1980s than it is now. These aims led to taxes on unprocessed exports and subsidies to processing. Two import-competing industries have been significantly protected ­ sugar and rice ­ and their rates of protection have increased in recent years. The rate of protection of the sugar industry is particularly high. Growers of rice have also received protection, but until about the year 2000 this occurred mainly through input subsidies rather than through the product price. This can be explained in terms of our hypotheses both as the main element of food self-sufficiency and because rice prices affect manufacturing wages through the cost of living. Since 2000 the rice industry has become more highly protected, with imports banned. The second theme is that during the long period of the Soeharto government, in good fiscal times aims (a) to (c) above could be afforded, but in bad times they could not. So good 1 Soeharto's New Order regime began in March 1966, even though Soekarno nominally remained president for a further twelve months. The reported statement was made during Soeharto's visit to the Food and Agriculture Organization headquarters in Rome in 1985, at which time rice imports were temporarily zero. 2 fiscal times meant `bad' trade policies, meaning more protection, and bad fiscal times meant `good' policies, meaning less protection. Third, accumulating evidence from around the world convinced policy makers in Indonesia and most other East Asian countries to rely more on markets and less on government intervention. This evidence was based on theoretical arguments, statistical studies, and simple two-country comparisons such as Thailand/Burma, West Germany/East Germany, Austria/Hungary, and South Korea/North Korea. The policy shift was also influenced by the collapse of the communist system and the victory of capitalism in the Cold War. This third factor may have contributed to the long-term shift towards lower manufacturing protection and agricultural export taxation, world-wide, from the mid-1980s. Fourth, following the democratic reforms that occurred in the wake of the Asian financial crisis of the late 1990s, agricultural protectionism has increased somewhat in Indonesia. Aggregate measures of protection indicate that these changes, along with reduced protection of manufacturing, caused the agricultural sector as a whole to switch from being a net taxed to a (slightly) net subsidized sector, relative to manufacturing. But these aggregate measures mask the fact that the agricultural protection is concentrated in just two crucial industries ­ sugar and rice. The following section describes the changing structure of the Indonesian economy, with emphasis on the agricultural sector. The next two sections provide overviews, for the period since independence, of government economic policy in general and then of government policies towards agriculture in particular. An attempt is then made to provide a political econometric explanation for the changing structure and pattern of agricultural distortions over the last three decades, estimates of which are provided for individual farm industries and for agriculture as a whole. The final section concludes. Economic growth and structural change From 1968 to 2005, Indonesia's gross domestic product (GDP) grew in real terms at an average annual rate of 6.3 percent. The broad characteristics of this growth are summarized in Table 1. 3 For ease of comparison with other Asian economies, the table distinguishes between the `pre- boom' period prior to 1987 and the following `boom' decade, which preceded the Asian crisis of 1997-99. For many Asian countries, the pre-crisis decade of 1987 to 1996 was one of extraordinary growth, far more rapid than in preceding decades. Indonesia also grew rapidly during this decade but, as the table shows, only marginally faster than during the two decades (`pre-boom') before it. Indonesia's economic growth had been sustained over several decades. Output contracted during the `crisis' years of 1997 to 1999, and during the subsequent `recovery' period growth it has averaged a moderate 4.6 percent (see Appendix Figure A1). As is typical of rapidly growing economies, agricultural output grew more slowly than GDP, implying a declining share of agriculture in aggregate output (Appendix Figure A2). The agricultural sector accounted for 56 percent of GDP in 1965. By 2004 this share had declined to 15 percent. Over the same period the GDP share of manufacturing and other industry rose from 13 to 44 percent and the share of services rose from 31 to 41 percent. For more detailed study of the changing composition of the agricultural sector it is convenient to use the input-output tables, which are available for the years 1971, 1980, 1990 and 2000. As incomes rise, the share of spending on starchy staples typically falls, while the share of spending on meat, fruit and vegetables typically rises. Indonesian experience fits this common pattern. It must be recognized that output growth within agriculture was achieved with rapidly diminishing shares of the national supplies of labor and capital. Furthermore, while agriculture grew more slowly than other sectors during boom periods, during the crisis its growth rate declined less than other sectors. Indonesia's agricultural sector has played a `shock absorber' role, and this was particularly important during the crisis years, when agricultural employment absorbed large numbers of people laid off from the urban centers. Although GDP grew much more slowly during the `recovery' period from 2000 to 2005 than during the `boom' decade, agricultural growth was undiminished. Table 2 summarizes the changing composition of value added in agriculture since 1971, using data from Indonesia's input-output tables. Paddy production (unmilled rice as produced at the farm level) contracted from 46 to 31 percent of agricultural value added, while the share of vegetables and fruit increased from 14 to 22 percent and the share of livestock rose from 0.6 percent to 5 percent. 4 It is somewhat surprising that the shares of intermediate inputs used in agriculture actually contracted (Appendix Table A1). The reason is apparently that fertilizer and pesticide usage was subsidized from the late 1960s until the late 1980s under a program called Bimas, discussed below. When the subsidies were phased out, fertilizer and pesticide use contracted markedly, especially in rice production. Most intermediate goods used in Indonesian agriculture are domestically produced (Appendix Table A2). Between 1980 and 2000 the share of imported intermediate goods in total intermediate good use increased from only 3.8 to 10.2 percent. In 1971, sales of paddy from farmers to intermediate users (rice millers) accounted for 56 percent of the total value of paddy output, implying that almost half of paddy output was milled by households themselves. By 2000 sales to intermediate users accounted for 98 percent of the total value of paddy output. Similar trends occurred in maize, rubber, sugarcane, palm oil, coffee and tea (Appendix Table A3). The international trading position of the major agricultural commodities can be seen in Table 3. The most important import-competing agricultural products are rice, sugar, maize and soybean. Major export-competing products include coffee, rubber, tobacco, tea, oil palm, copra, shrimp and spices. Paddy is neither exported nor imported, but milled rice has historically been an important import item for Indonesia. Since 2002 imports have been officially banned, but some imports have still occurred. Cassava is mainly non-traded, although there are exports of its derivatives, manioc and tapioca. Much of the livestock sector is also non-traded, although chickens are exported, while beef and dairy products compete with imports. Some sectors have moved over time from one trade category to the other. The most important example is sugar, which was the most important export during the colonial era, but has become one of the most highly protected import-competing products in the post-independence period. Another example is maize, which switched from net export to net import status during the 1990s. Vegetables and fruits have become important net imports, as have soybeans (included in `other food crops' in these tables). Policy evolution 5 Indonesia obtained its independence from the Netherlands in 1949. The next two decades were chaotic. The post-independence government of President Soekarno pursued a nationalistic, quasi- socialist economic policy that produced hyperinflation and economic stagnation. In 1966 Soekarno was displaced amid economic chaos by one of his generals, Soeharto, whose regime, called the `New Order', lasted until the macroeconomic crisis of 1998. Soeharto pursued more market-oriented economic policies than his predecessor. Upon assuming power, Soeharto speedily introduced a macroeconomic stabilization program and then began liberalizing Indonesia's trade and investment policies. In 1967 foreign investors were guaranteed the right to repatriate both capital and profit and from 1970 onwards the capital account was almost completely open. As we shall see below, trade policy under Soeharto's government was much less open. It was characterized by taxation of exports, especially non-food agricultural exports, and protection of imports, including some food imports. In the wake of the commodity boom of 1972-73 and the oil price shocks of 1973-74 and 1979-80, trade policy became increasingly inward-looking. These external events tripled the ratio of Indonesia's export prices to its import prices (Appendix Table A3). Between the early 1970s and the mid-1980s, the government taxed or banned some traditional exports, pursued self-sufficiency in rice, and used part of the burgeoning oil revenues to establish import- substituting manufacturing industries, which it then protected. In the early to mid-1980s several traditional export industries were subjected to quantitative trade restrictions. These included a ban on log exports, conferring very high rates of effective protection on the plywood manufacturing industry, for which raw timber is the principal input. Licensing systems were introduced for exports of vegetable oils, several spices, coffee and some grades of rubber. In the case of palm oil, domestic refiners were protected by a tax on exports and a requirement that growers supply these refiners with part of their output at low, controlled prices. From 1982 onwards, the price of petroluem began to decline. By the mid-1980s it had fallen from US$28 to $10 per barrel. Many oil-exporting countries, including Nigeria and Venezuela, were unable to adjust to these external changes without devastating domestic consequences, but Indonesia responded quickly by cutting public spending and devaluing its currency, partly to promote non-oil exports. In addition, a value added tax (VAT) was introduced between 1983 and 1986. At first, trade policy became increasingly oriented towards import substitution, and the system of import licensing was extended. But after this initial protectionist 6 response to lower petroleum export revenues, trade policy was significantly liberalized from 1985 onwards. With the stated goal of promoting non-oil exports, the government introduced a series of reforms which reduced tariffs and non-tariff barriers (NTBs). Following tariff cuts in 1985 the government transferred most customs functions from the Indonesian Customs Service to an international inspection company, SGS of Switzerland. The role of SGS was phased out by 1995. NTBs on imports were progressively relaxed from 1986 onwards, and the system of providing exporters with duty-free inputs was extended. According to the estimates in Table 4, the effective rate of protection in agriculture declined from 24 percent in 1987 to 14 percent in 1994, and that in manufacturing declined much further, from 86 percent to 29 percent over the same period. Since there was probably more `water' in the tariffs in 1987 than in 1995, the true reductions in protection were probably somewhat smaller than these numbers indicate, but the decline was still substantial. In addition to the lowering of tariff rates, many NTBs were replaced by tariffs. The coverage of `restrictive' NTBs declined from 44 percent of total value added in all traded industries in 1986 to 23 percent in 1995. This switch from NTBs to tariffs was somewhat more extensive in manufacturing than in agriculture, where the coverage of NTBs declined from 67 to 48 percent. In the wake of the financial crisis of 1997­98, the government was obliged to allow free imports of both rice and sugar as a condition for borrowing from the IMF. However, with the ending of the IMF program, imports of rice and sugar have again been restricted by both tariffs and NTBs. Agricultural protection by sector The distinction between import-competing products and export products is not always entirely clear cut, but is nevertheless crucial to any discussion of Indonesian agricultural policies. Whereas import-competing production has generally been protected by government policies, export-competing production has generally been taxed. For centuries, Indonesia's major staple food crops have been net imports. Both the Dutch colonial government and the government of Indonesia in the post-independence period generally tried to control the price of rice and other 7 important food crops to balance the competing interests of domestic producers and consumers. Except when world prices of food crops have been unusually high, imports have been directly restricted, or subject to tariffs, or both. On the rare occasions when world prices have been so high that growers would have had an incentive to export food crops such as rice, they have usually been prevented from doing so by non-tax measures, including outright export bans. In contrast, export crops have been seen by successive governments over the last two centuries as a useful source of tax revenue. Under the Cultivation System introduced by the Dutch in 1830, production of cash crops for export was stimulated by imposing taxes on villagers that could be most easily paid in kind by handing over crops that the Dutch East India Company then processed and exported. By far the most important of these exports was sugar; other important exports in the nineteenth century were coffee, tea, indigo and cinnamon. Booth (1988 p. 202) reports that in the late 1830s, 40 percent of the total income of the Dutch government was derived from the Cultivation System in its Indonesian colonies. During the post-independence period, the Indonesian government's revenue from export crops has been obtained by export taxes that have tended to depress both domestic production and exports of the relevant crops. The main reason for raising export revenue in this quite different way has presumably been the government's desire, in the period since independence, to promote the development of the manufacturing sector, of which food processing is an important part. Rice, sugar and soybean have been protected from import competition by non-tariff barriers. These and maize are discussed in the next four subsections. The remainder of import- competing agriculture has been protected by tariffs and tariff surcharges. It and export agriculture are discussed in the final sub-sections. Rice The most important and most enduring non-tariff barriers have been those on rice and sugar. Figure 1(a) shows estimates of domestic wholesale prices and border prices for rice.2 All the price series in Figure 1 are in rupiah per kilogram, divided by the GDP deflator, indexed at 2005 2 The border price of rice in Figure 1(a) has been converted to make it as nearly comparable as possible to the wholesale price. The fob price was adjusted to the cif level by adding freight and insurance costs; the resulting cif price was then adjusted to the wholesale level by adding margins to allow for the estimated handling, warehousing and interest costs. 8 = 1. While there have been enormous nominal increases in rice and sugar prices since the early 1970s, the charts show that any trends in the real prices of these products have been relatively small. It is clear from Figure 1(a) that the domestic wholesale price of rice has fluctuated much less than the border price and that domestic prices have not differed greatly, on average, from the trend level of border prices. Price stabilization was achieved by giving the state logistics agency, Bulog (Badan Urusan Logistik), a monopoly over international trade in rice and directing it to build up buffer stocks to smooth out fluctuations in domestic supply. It is significant that this stabilization of domestic prices was achieved while keeping the trend value of domestic prices roughly in line with the trend of world prices. Average rates of protection in the output market were very low, but input markets were another matter. From the 1970s onwards, the Soeharto government used part of its new oil wealth to promote self-sufficiency in rice, by subsidising the adoption of high yielding varieties that had been made available by the `Green Revolution'. These new varieties required greatly increased use of irrigation, fertilizers and pesticides, which the government helped to finance. An important motivation for this policy was fear of a repetition of the riots precipitated by high food prices in 1965. Under the Bimas program, introduced with the explicit goal of rice self-sufficiency, farmers received agricultural extension services and subsidized credit, seeds, fertilisers and pesticides. The government also paid for increasing and upgrading irrigation facilities. The resulting increase in the profitability of rice growing, together with some coercion of those farmers who were reluctant to extend the area of rice cultivated, led to a 17 percent increase in gross3 harvested area in the decade to 1985. This increase in area, together with a 50 percent increase in average yields in the same period, allowed Bulog to reduce domestic rice prices relative to the CPI between the late 1970s and 1985 while gradually phasing out imports and then halting imports altogether in 1985. Lower world oil prices and advice from the World Bank contributed to the reduction in agricultural input subsidies in the late 1980s and early 1990s. Figure 1(b) shows a fall in the real price of urea from the late 1970s to the early 1980s and a subsequent rise in the domestic wholesale price of urea relative both to the CPI and to the border price in the late 1980s and early 4 "Gross" indicates that a hectare which is harvested twice in a year is counted as 2 hectares. 9 1990s. Exports of urea require special approval from the Ministry of Trade, under an export licencing scheme. The year-to-year determination of the magnitude of these licences is non- transparent, but the Ministry tends to place priority on ensuring that domestic supplies are stable at a price lower than world market prices and this results in the negative nominal rates of protection shown in Figure 1(b). In the late 1980s, the strict policy of zero imports of rice was replaced by a policy of `borrowing' rice from Vietnam in times of shortage and repaying the rice loans in times of surplus. These `loans' were conducted in bilateral government-to-government deals in which Bulog acted for the Indonesian government. In the early 1990s, it gradually became apparent that Indonesia was unable to maintain rice self-sufficiency, even on average and over a period of years. To satisfy domestic demand at `acceptable' prices, Bulog was forced to undertake substantial net imports. When the Asian crisis forced Indonesia to borrow from the IMF in 1997, one of the loan conditions to which the government agreed was the removal of Bulog's monopoly on rice imports. Until 1999, there was also no import duty on rice but the IMF's aim of free trade in rice proved illusory because the financial crisis briefly converted rice into a potential export and the government banned exports to reduce pressure on domestic prices. Figure 1(a) shows that in 1998 border prices, converted to rupiah at the devalued exchange rate, were far above domestic prices. The reason for this was that the massive depreciation of the exchange rate between mid- 1997 and mid-1998 initially outweighed the much more gradual rise in domestic prices. This episode clearly demonstrated that the government's policy has always been to stabilize food prices at `acceptable' levels, rather than simply to protect growers. The general increase in domestic prices in 1998-99 and the stabilization of the exchange rate after mid-1998 removed the incentive to export rice. Bulog's monopoly on imports was not immediately re-imposed, but a 20 percent tariff on rice imports was introduced in 1999. Problems with under-invoicing by importers resulted in this tariff being converted to a specific tariff at Rp 430/kg. In 2002, Bulog's monopoly over imports was restored and since 2004 imports have officially been banned, although Bulog has occasionally been issued with special import permits. Sugar 10 The Indonesian sugar industry is dominated by the state-owned mills, mainly on Java, that were acquired by the nationalization of the formerly Dutch-owned sugar estates in 1957. Investment and technical progress in this sector has been extremely sluggish and the industry has languished behind protective barriers. The finished product of these antiquated factories, known as `plantation white sugar', is not exactly comparable to either the refined or the raw sugar traded on the world market. Plantation white contains more impurities--mainly molasses--than internationally traded raw sugar, but has already undergone some of the bleaching processes that separate refined from raw sugar in more technologically advanced sugar industries. Most firms in the food and beverage sectors cannot use plantation white sugar because of its relatively high level of impurities; their needs are mainly met by imports of raw sugar, although there is a small amount of raw sugar produced domestically. As in the case of rice, the main motive behind government policy for sugar appears to be the desire to stabilize the domestic price at an `acceptable' level. In addition, in the case of sugar, the government has tried to protect the sugar factories that it owns. This may explain why, at least since 1957, the sugar industry has been more tightly regulated than any other agricultural sector, with the government monopolising not only imports but also domestic marketing. Government ownership also helps to explain newspaper reports that, in the 1970s, farmers in traditional sugar growing areas were regularly forced to grow sugar to supply local factories, subject to threats that other crops would be burnt. Figure 1(c) compares the border price of raw sugar (after allowing for margins between the fob price and domestic wholesale prices) with the domestic wholesale price of plantation white sugar. The chart shows that for much of the period since 1970, domestic prices were about twice the border price, implying a nominal rate of protection (NRP) of about 100 percent. However, in 2006 this gap has been have been greatly narrowed by the abrupt rise of world prices. Our estimates of the NRP for sugar ignore two factors, the first of which makes our estimates tend to understate the true NRP, while the second goes in the opposite direction. The first factor is that our estimates make no adjustment for the relatively low polarity (high level of impurities) of plantation white sugar. The offsetting factor is the neglect of the cost of bleaching to obtain plantation white sugar. Experts on the sugar industry have suggested that the low 11 polarity effect is probably more important than the effects of bleaching, but that the difference is small. Soybeans Until 1996, the government protected soybean growers by giving Bulog a monopoly on imports. Since 1996, soybean imports have been unrestricted and the tariff is currently zero. Figure 1(d) shows the domestic and border prices in rupiah/kilogram, deflated by the GDP deflator indexed at 2005 = 1.00. The excess of domestic price of soybeans over the border price was reduced in 1988, when a local soybean crushing plant, operated by PT Sarpindo Industri, began to operate. However, Bulog prevented domestic prices of beans from falling as rapidly as world prices in the period 1988-94 and Sarpindo was protected by a local content scheme that required the domestic feedmills to source at least 20 percent of their total usage of soybean meal from local supplies-- which meant Sarpindo, since it was the only local supplier of soybean meal. The high cost of feed inhibited the growth of the increasingly powerful poultry industry and in 1996 the local content scheme was abandoned and Sarpindo was allowed to go out of business. Maize From being a substantial net export industry in the 1970s, maize subsequently became a net import item. The transition coincided with a movement from negative protection during the export phase to a small amount of positive protection since the early 1980s (Figure 1(e)). The tariff on imports of processed maize in the form of pellets and flour is currently only 5 percent, but during the presidency of Megawati Sukarnoputri (2001 to 2004) the then Minister of Trade (Rini Suwandi), supported by BULOG, created import licences which restricted imports, raising average nominal rates of protection well above 5 percent. The rents created by this measure accrue to members of the maize importers' association. Other import-competing agriculture: tariffs and tariff surcharges 12 The growth of protection during the 1970s and early 1980s, followed by the reduction in protection in the late 1980s and 1990s, were mainly achieved by changing the rates of import tariff surcharges (bea masuk terbahan), rather than the rates of the import tariff (bea masuk).4 In terms of their economic effects, the surcharges were equivalent to tariffs, but, unlike tariffs, the rates of the surcharges could be changed by administrative decree, without the need to amend the law. The rates of import duty shown in Appendix Table A4 are the combined rates of tariffs plus tariff surcharges. A comparison of Apendix Table A5 (tariff changes) with Appendix Table A6 (changes in tariff surcharges) shows that much of the growth of protection between 1974 and 1979, and almost all the much larger increases in the period 1979­85, was achieved by raising tariff surcharges rather than tariffs. When protection was reduced between 1985 and 1989, about half of the reduction was achieved by largely eliminating tariff surcharges, which were negligible by 1989, but had been a very important part of total import duties in 1985. By 1994, tariff surcharges had been totally abolished and from 1994 onwards, there is no need to distinguish between tariffs and total import duties. The import duties on food processing have always been higher than those on agriculture: in every year, the average rate of import tax on food processing alone (HS chapters 15­24) is higher the corresponding average rate on the entire agriculture and food processing products category (chapters 1­24). Within the sectors defined here as making up agriculture (HS chapters 1­14), flowers, particularly orchids, and vegetables--have always been the most highly protected. Among the more traditionally agricultural sectors, livestock has always received relatively high protection from imports, and so have estate crops. However, whereas livestock is mainly import-competing (but also in part non-traded), many estate crops are mainly export- competing. In these cases, of which coffee, tea and spices are important examples, there is a great deal of `water' in the tariff--that is, the tariff overstates the extent to which the overall system of protection raises their domestic price. In the 1970s, the total rates of import tax on tea, coffee, vanilla, cinnamon, nutmeg and ginger were 70 percent and by 1985 this rate had 4 The term `tariff surcharge' is a misnomer in the sense that the base to which the rates of the tariff surcharge applied was the tariff, but the border value (cif) of the imports subject to the tariff surcharge. For example, in the case of live animals (other than pure bred) in 1985, the tariff was 30 percent and the tariff surcharge was 15 percent, giving a total rate of import duty of 45 percent of the border value (cif). 13 increased to 100 percent. However, by 1989, the total import duty rates for all these products had been reduced to 30 percent and by the mid-1990s they had fallen to 5 percent. The rates of import duty on food crops have generally been relatively low, at least in the period preceding 2000. However, these rates understate the extent of protection of food crops for two reasons. Producers of food crops have received input subsidies, and food crops have also been protected by non-tariff barriers, hence their separate treatment above. Estate crops: rubber, copra, coffee and tea Rubber, copra, coffee and tea are all produced by perennial plants and tend to be produced on large estates in Indonesia, except that copra is also produced by smallholders. All have been export crops and all have been taxed, but at varying rates. Export volumes of all these commodities have declined since the 1980s. High rates of export taxation are a significant part of the explanation. Figure 2(a) to 2 (d) show the calculations of the nominal rates of protection for each of these four commodities, respectively. For rubber, the rate of export tax has been low, but the data show high rates of export taxation for copra and tea. For coffee the rate of export taxation has declined from very high rates prior to the 1990s. The political economy of protection: Do good times produce bad policies and vice versa? Some key characteristics of Indonesia's political circumstances provide background to attempts to explain the changes in trade policy summarized above. First, among the Indonesian elite, confidence was instilled by the economic successes achieved elsewhere in East and Southeast Asia from the 1970s onwards. There seemed no fundamental reason why Indonesia should not succeed also and this confidence meant that bold strategies could be contemplated. This confidence can be contrasted with the timidity and lack of confidence in the international trading system that characterized most of South Asia at the same time. Second, Soeharto's political authority within Indonesia was unchallenged until the very end of his regime. Even policies that were unpopular, at least initially, could be contrmplated if Soeharto considered them necessary. 14 Observers of economic reform in Indonesia have coined the phrase `Good times produce bad policy, bad times produce good policy', where `good times' means favorable external conditions and `good policies' means deregulation in general, and lower barriers to international trade and investment in particular. This summary does indeed describe much of Indonesia's history of economic reform. The oil price booms of the 1970s were followed by a series of trade restricting import-substitution policies aimed at protecting some at least of the traded goods industries that were potentially harmed by the `Dutch disease' effects of the petroleum booms (that is, the decline in the domestic competitiveness of traded goods industries due to a rise in the prices of non-traded goods and services relative to traded goods ­ see Corden (1984) and Warr (1986)). Trade liberalization followed the adverse terms of trade effect of the decline in petroleum prices from the early 1980s onwards. But while the `Good times - bad policy, bad times - good policy' summary describes the Indonesian experience, it does not provide an explanation for it. Why do good times produce bad policy, and why the converse? Observers of policy formation under Soeharto reported the contest for Soeharto's attention between the technocrats on the one hand and the nationalists on the other (Hill 2000). At different times, either of these groups might have ascendancy, which meant that Soeharto was heeding their messages. The technocrats, many of whom were professional economists trained in the United States, favored a market-oriented economy, a strong emphasis on macroeconomic stability, and a relatively open trade policy. This group dominated the Ministry of Finance and the National Planning Agency (Bappenas) and had considerable influence on the Bank of Indonesia. The World Bank used its influence directly in support of the technocrats and the Bank's resources and technical expertise also assisted the technocrats to make their case in a convincing way. The economic nationalists were more diverse. They included in particular the `engineers', led by the Minister for Research and Technology, Dr B.J. Habibie, a German-trained engineer with a strong preference for `crash-through' economic programs based on advanced technology. This group promoted large-scale, capital intensive projects in industries such as aeronautics, shipbuilding, steel, fertilizers and petrochemicals. To ensure the profitability of these projects, high rates of protection were advocated on `infant industry' grounds. In addition to Dr Habibie's own department, this group was influential within the state-owned petroleum company, Pertamina. A second group of nationalists were the advocates of self-sufficiency in 15 food in general and rice in particular. This group dominated the Ministry of Agriculture and the food logistics agency, Bulog.5 More general support for import-substitution based policies was concentrated in the Ministry of Industry. During bad economic times, the technocrats tended to gain Soeharto's attention. During good times, he listened to the nationalists. The central dynamic derived from the role of external shocks to the Indonesian economy, operating through petroleum prices. During the Soeharto period, petroleum was both a principal source of foreign exchange, through direct oil exports, and a major source of government revenue, through the royalties received by the government on those exports. Reduced oil prices implied both balance of payments and budgetary stresses. In addition, the majority of Indonesia's foreign debt was public debt. When the price of oil fell, the fiscal burden of debt servicing became more painful. This increased the influence of the World Bank, whose willingness to extend concessional loans to Indonesia was important directly and also as a signal to other potential foreign lenders. At such times the government needed these loans to `balance' its budget. The only alternative was inflationary financing, the consequences of which had been experienced under Soeharto. Increased influence for the World Bank meant increased influence for the technocrats and the policies they advocated. In addition, reduced oil prices meant reduced influence for Pertamina by lowering its contribution to government revenue. It also meant increased influence for the Ministry of Finance, whose tax reforms, designed by technocrats and like-minded foreign advisors, helped make up for lost oil revenues. In other countries, a deterioration in the terms of trade might be met by exchange controls, import licensing and other import-substitution policies. In Indonesia, a protectionist response also occurred briefly in response to the oil price declines of the early 1980s. But it did not last long because it did not address the simultaneous fiscal problem. An example of the tax reforms which emerged from this dynamic was the introduction of a value-added tax (VAT) in 1986 and the simultaneous reduction in import duties. An import duty (tariff) is equivalent to both a tax on consumption and a subsidy on production, set at the same ad valorem rate. The tariff raises positive net revenue because, for an import commodity, the volume of consumption exceeds that of production. A value-added tax is a tax on consumption alone and it can raise the same amount of revenue as a tariff, but at a lower rate of tax, because it does not expend revenue on subsidizing production. Similarly, the switch from non-tariff barriers (NTBs) to tariffs 5 See, for example, Warr (1992) and Timmer (1996). 16 generates revenue. NTBs can be thought of as privately levied tariffs, making no contribution to government revenue. A final example was the phasing out of the Bimas scheme, designed to help rice growers achieve self-sufficiency. The budgetary cost of the fertilizer and pesticide subsidies and subsidized lending of Bimas became serious with the fiscal deterioration of the 1980s. During times of reduced petroleum prices, such as the early to mid-1980s, illiberal trade policies were unaffordable in fiscal terms and this reinforced the argument that trade liberalization would promote improved foreign exchange earnings from non-oil exports. The technocrats then held sway. In contrast, during the euphoria of the 1970s, induced by high petroleum revenues, the import-substitution schemes advocated by the nationalists seemed affordable and were politically attractive. At such times, the nationalists captured Soeharto's attention. The Asian financial crisis of 1997­98 was the worst of times and produced the best of economic policies, given that `best' is being used here to mean more closely in conformity with the laissez-faire advice of neo-classical economists. The Asian crisis also provides the clearest illustration of the causal link between bad times and laissez-faire policies: the reforms that the government introduced in the wake of the crisis were explicitly adopted as conditions for borrowing from the IMF, when all other sources of external lending had dried up. Following Soeharto's political demise in 1998 and the subsequent move to a much more democratic form of government, the president no longer holds absolute authority and policy determination is therefore no longer simply a contest between the technocrats and nationalists to influence the president. The parliament, a token institution under Soeharto, now has teeth and the president cannot ignore its will. Populist economic nationalism has tended to dominate the parliament and this has reduced, but not eliminated, the influence of the technocrats. In addition, the conspicuous reluctance of the major industrialized countries to reduce protection for their own agricultural sectors has weakened the influence of those technocrats who argue against Indonesian restrictions on trade. The increased protection of the rice and sugar sectors that followed the ending of the IMF program was a direct consequence of these political changes. Movement towards rice self-sufficiency and protection of farmers are both politically attractive in Indonesia and, in the public imagination, both are strongly associated with the national interest. Protection for the rice industry is supported by all major political parties. With 17 democracy, both rice and sugar farmers therefore receive more protection from imports than they did under Soeharto. Imputed protection at the farm level The above discussion of protection rates has focused to the effects that policy interventions have at the wholesale market level. In this section, we extend the analysis to consider the way protection (or its opposite) at the wholesale level produces price effects at the farm level. Theory One of the intentions of agricultural protection policy is to influence prices at the farm level. But the goods produced directly by farmers seldom enter international trade themselves. The raw commodities produced by farmers are generally non-traded, whereas the commodities which enter international trade are the processed or partially processed versions of these raw products. Between the non-traded raw product produced by the farmer and the traded processed commodity which enters international trade, there may be several steps of transport, storage, milling, processing and re-packaging. The significance of this point is that border protection policy operates directly on the goods which actually enter international trade, either exported or imported, not the raw commodities produced by farmers. Protection at the farm level is therefore a derived effect. It depends on the extent to which policies applied to trade in processed agricultural goods induce changes in their prices which are then transmitted to the prices actually faced by farmers. The question thus arises as to what extent price changes at the wholesale level, induced by protection policy, affect the prices actually received by farmers for the raw products they sell. We construct a simple econometric model to investigate this issue. We use the notational convention that upper case Roman letters (like X ) will denote the values of variables in their levels and lower case Roman letters (like x ) will denote their natural logarithms. Thus x = ln X . Protection at the wholesale level is defined as 18 PitW = Pit* (1 + TitW ) , (1) where PitW denotes the level of the wholesale price of commodity i at time t, Pit* is the corresponding border price, expressed in the domestic currency and adjusted for handling costs in getting the commodity from the cif level to the domestic wholesale level, in the case of an import, and for the cost of getting it from the wholesale level to the fob level in the case of an export. The nominal rate of protection at the wholesale level is given by TitW . In this discussion, both the border price and the nominal rate of protection are treated as exogenous variables. The border price is determined by world markets and the country concerned is presumed to be a price taker. The nominal rate of protection is determined by the government's protection policy. F The farm gate price of the raw material is denoted by PitF and its logarithm, pit , is related to the logarithm of the wholesale price by pit = ai + bi pit + u it , F W (2) where ai and bi are coefficients and uit is a random error term. The coefficient bi is the `pass- through' or `transmission' elasticity. The estimated values of the coefficients ai and bi are ^ ^ denoted ai and bi , respectively. The econometric estimation of these parameters is discussed below. The estimated coefficients are used as follows. We estimate the logarithm of the farm price that would obtain in the absence of any protection as pit * = ai + bi pit * , ^F ^ ^ W (3) W where pit * is the estimated value of the wholesale price that would obtain in the absence of protection, pit * = ln PitW * . This is then compared with the estimated value of the wholesale price W in the presence of protection 19 pit = ai + bi pit . ^F ^ ^ W (4) ^F ^F ^ ^ Denoting the anti-logs of pit and pit * by PitF and PitF * , respectively. The nominal rate of protection at the farm level is then estimated as TitF = ( PitF - PitF * ) / PitF . ^ ^ ^ ^ (5) It is important to observe that the value of the protection-inclusive farm level price used in these calculations is the level estimated from the econometric model (equation 4) rather than the actual price given by the raw data. The reason is that our intention is to use the model to estimate the change in the farm gate price caused by protection at the wholesale level. Thus both the protection-inclusive and the protection-exclusive prices used in (5) are their predicted values, obtained from the model. The implied nominal rate of protection at the farm level can be related to the nominal rate ^ of protection at the wholesale level, as follows. Substituting PitF = Ai ( PitW ) bi and ^ ^ ^ PitF * = Ai ( PitW * ) bi into equation (5), where Ai is the anti-log of ai , rearranging, and using ^ ^ ^ ^ equation (1), we obtain the simple expression ^ TitF = (1 + TitW ) bi - 1 . ^ (6) Obviously, if TitW = 0 , then TitF = 0 , regardless of the value of bi . Similarly, if bi = 0 , then ^ ^ ^ TitF = 0 , regardless of the value of TitW . Also, if bi = 1 , then TitF = TitW . It can readily be seen that ^ ^ ^ when TitW > 0 , T F TitW as bi 1 and T F TitW as bi 1 . When TitW < 0 , T F TitW as bi 1 ^ ^ ^ ^ ^ ^ and T F TitW as bi 1 . ^ ^ Econometric application 20 ^ The purpose of the econometric analysis is to estimate the parameter bi for each commodity. Here the results are summarized, but details of the econometric analysis are available upon request. For each commodity we conduct the analysis using time series price data with each variable expressed in logarithms and each deflated by the GDP deflator for Indonesia: the farm gate price (LFP), the wholesale price (LWP), and the log of the international price, adjusted by the nominal exchange rate and transport and handling costs (LIP). The data extended from 1976 to 2001. The seven commodities for which these data were available were: rice, maize, soybeans, sugar, rubber, coffee and tea. We first test each of the series (each deflated by the GDP deflator) for the existence of a unit root. For rice, the null hypothesis of a unit root was rejected for all three price series (recalling that they are real, not nominal, price series, using the GDP deflator) at the 10 percent level of significance. The price series are thus considered stationary. For other commodities the results are more mixed. For maize, the null hypothesis of a unit root could not be rejected for farm level prices (LFP), but was strongly rejected for the other two price series. For soybeans, the null hypothesis of a unit root could not be rejected for the wholesale price series (LWP) but was rejected at the 10 percent level for the other two series. For sugar, the null hypothesis of a unit root could not be rejected for any of the three series, especially the farm level price series (LFP). For rubber, coffee and tea the results were similar. The null hypothesis of a unit root marginally failed to be rejected for the farm level price series (LFP), but was rejected for the other two series. Ordinary least squares (OLS) estimates of equation (2) were first produced. In most cases, autocorrelation was a problem and an AR(1) correction term was included to eliminate it, which it did effectively. The OLS estimates assume that LFP is endogenous and LWP is exogenous. These assumptions were tested using Hausman's endogeneity test, although it is recognized that the test has low power when the number of data points is small, as in this case. In the case of each commodity, the null hypothesis that LWP was (weakly) exogenous to LFP failed to be rejected, confirming the validity of the OLS estimates. Reverse Hausman's tests were also conducted and the null hypothesis that LFP was exogenous to LWP was rejected in the cases of maize, sugar, rubber, coffee and tea. It marginally failed to be rejected for rice and soybeans. These results roughly support the validity of using the OLS framework to estimate the transmission elasticity from LWP to LFP, treating LWP as exogenous. 21 Usable estimates were produced for five commodities: rice, soybeans, sugar, rubber and tea. The estimated elasticity had the expected positive sign and was significantly different from zero, with the estimated equation performing well. Table 5 summarizes the estimates. For maize and coffee, the estimated elasticity was not significantly different from zero and the estimated equation performed poorly. It is often asserted that middlemen prevent commodity price changes at the wholesale level, resulting from protection or from international price movements, from being transmitted to farmers. This hypothesis is rejected by the Indonesian data, at least for the five commodities mentioned above. The transmission elasticities are not zero. Economists often assume that the transmission elasticities are unity. But the estimated values are generally less than unity, lying between 0.2 and 0.8. The lower values are obtained in the case of perennial crops rubber and tea, which have high processing costs. The other values all exceed 0.5. It is likely that the true transmission elasticities change over time, but the limited data available for this exercise made it necessary to assume that the true values are constant. Estimation of protection at farm level Given the estimated value of the transmission elasticity, equation (6) is used together with the estimated nominal rates of protection at the wholesale level, discussed above, to produce estimates of imputed NRPs at the farm level. These are shown in Figures 3 and 4. Because usable estimated of the transmission elasticity could not be obtained for three commodities ­ maize, coffee and copra ­ the estimated values for rice, tea and rubber, respectively, were used instead, as proxies for the true elasticities for these commodities. Because the transmission elasticities lie between zero and unity, the imputed nominal rates of protection at the farm level are somewhat lower in absolute value than the nominal rates at the wholesale level, but (because of the assumption of constant transmission elasticities) they track the pattern of the wholesale level results closely. Aggregate measures of assistance 22 In this section we calculate aggregate measures of rates of protection using the information assembled from the preceding analysis and following, as much as possible, the methodology outlined in Anderson et al. (2008). The annual calculations reported fluctuate somewhat from year to year, because international and domestic price changes alter the protective effects of all instruments of protection except ad valorem tariffs. In addition, the time taken for domestic prices to adjust to international price changes means that annual data on price differences indicate some variation from one year to the next. Calculations of nominal rates of protection at the wholesale and farm levels described in the previous section are used to calculate nominal rates of assistance (NRAs) at the farm level, taking account of assistance to fertilizer inputs in addition to output price distortions. That is, the NRA for a particular commodity is calculated as its NRA on output plus the product of the cost share of fertilizer in production of the commodity concerned and the nominal rate of subsidy to fertilizer use (which was as high as two-thirds the fertilizer price in the early 1970s but has declined over time and in recent years has been only one-fifth). The aggregate nominal rate of assistance therefore exceeds the nominal rates on output for each commodity that uses fertilizer as an input. The calculations of nominal rates of assistance confirm that during the period 2000 to 2004, import-competing commodities were significantly protected, notably rice and especially sugar (Table 6). The rates of protection for these two commodities increased significantly compared with the 1990s, whereas the assistance to maize and soybean producers has fallen. Tea is still moderately taxed, but export commodities such as rubber, copra and coffee are either only lightly taxed or slightly assisted today, having been taxed ­ sometime significantly ­ prior to the late 1990s. The average NRA for import-competing farm products is always above that for export products, and the extent of that anti-trade bias within the agricultural sector has not diminshed over time (Figure 3, and row 6 of Table 7). Nor has the dispersion of NRAs of covered products, as measured by the annual standard deviation around the weighted mean, declined (bottom of Table 6), which means the inefficiency of resource use across industries within the sector remains non-trivial. Finally, the relative rate of assistance (RRA) to agriculture is a function of the difference between the nominal rate of assistance to tradable agriculture and the nominal rate of assistance to non-agricultural tradables such as manufactures but also non-farm primary products from 23 fishing, forestry and mining.6 Our RRA estimates show that agriculture has moved from being effectively taxed by between one-quarter and one-third to being a slightly net subsidized sector, with that transition occurring shortly after the Asian crisis of 1997-99 (Table 7 and Figure 4). Since we have erred on the side of understating rates of manufacturing protection prior to 1987, better estimates of manufacturing protection during this period would show larger negative RRA numbers prior to the 1990s, and would thereby reinforce rather than undermine the above finding. Conclusions and prospects for future reform Having taxed agriculture relative to the non-farm sectors throughout the post-independence period, since around 2000 Indonesia's trade policies are no longer taxing agriculture, on average. This change occurred following the Asian crisis of the late 1990s. The switch took the form of: (a) increases in protection of the import-competing commodities sugar and rice; (b) declines in taxation of agricultural exports, especially rubber and copra; and (c) declines in manufacturing protection. The movement to a more democratic form of government has weakened the influence of Indonesia's `technocrats', who have generally favored liberalized trade policies. Greater protection of some key agricultural commodities has been a consequence. Protection of agriculture primarily takes the form of protection for the import-competing sugar and rice sectors. Other output sectors receive virtually no direct protection. Subsidies to fertilizer and other inputs have been an indirect source of protection to agriculture in the past, but these rates of subsidy have declined. 6 The NRA for non-agricultural tradables is estimated using predominantly the effective rates of protection for manufacturing estimated by Fane and Condon (1996), who estimate the effective rate of protection for manufacturing for 1987 and 1995 at 48 and 20 percent, respectively. They also project the corresponding effective rate for 2003, at 13 percent, based on the May 1995 tariff reduction package, which was to be implemented by 2003, and which was in fact largely implemented. For all years before 1987 we have used the 1987 values, even though some tariff reduction had occurred during the few years before 1987. For the years between 1987 and 1995 and between 1995 and 2003, we have interpolated linearly. For 2004 we have used the 2003 value. As noted above, the focus of this discussion is to identify broad long-run trends over time in the structure of protection more than annual fluctuations. 24 The political explanations for protection of the sugar and rice industries are quite different. Protection of the sugar industry is a consequence of the political power of the highly concentrated sugar refining industry, including the state-owned component of this industry, closely linked with large-scale sugar plantations. By contrast, Indonesia's farm-level production of rice (paddy) is dominated by small scale farm-level producers. The rice milling sector is much more concentrated and better organized, however, and this is relevant because imports compete with milled rice rather than the raw, unmilled product (paddy) produced by the farmer. The political power of rice millers has been an important source of support for protection of the rice industry. The enhanced political power of the Indonesian parliament since the upheavals induced by the Asian crisis, together with the economic nationalism that dominates the membership of the parliament, have strengthened the support for protection of the rice industry. Since 2004 imports of rice have officially been banned. In part, this policy has reflected the mistaken claim, advanced by supporters of rice industry protection, that restricting rice imports reduces poverty. A general equilibrium analysis presented in Warr (2005) shows that the policy is more likely to have increased poverty, within both rural and urban areas, because the poverty-increasing effects of increasing the consumer price of rice far exceed the poverty-reducing effects of increasing the producer price. Given the politics at work, trade liberalization in Indonesia's sugar and rice industries seems unlikely in the foreseeable future. On the contrary, increasing protectionism seems a more likely outcome, and this could conceivably extend to industries other than sugar and rice. References Anderson, K., M. Kurzweil, W. Martin, D. Sandri and E. Valenzuela (2008), "Methodology for Measuring Distortions to Agricultural Incentives", Agricultural Distortions Working 25 Paper 02, World Bank, Washington DC, revised January. Posted at www.worldbank.org/agdistortions. Booth, A. (1988), Agricultural Development in Indonesia, Sydney: Allen and Unwin. Central Bureau of Statistics (1971, 1980, 1990 and 2000), Input-Output Tables of Indonesia, Jakarta: Central Bureau of Statistics. Corden, W.M. (1984), "Booming Sector and Dutch Disease Economics: A Survey", Oxford Economic Papers 36: 359-80. Fane, G. and T. Condon (1996), "Trade Reform in Indonesia, 1987-1995", Bulletin of Indonesian Economic Studies 32: 33-54. Hill, H. (2000), The Indonesian Economy, (2nd ed.), Cambridge: Cambridge University Press. Nicita, A. and M. Olarreaga (2006), "Trade, Production and Protection, 1976-2004", mimeo, World Bank, Washington DC. Timmer. C.P. (1996), "Does BULOG Stabilize Rice Prices in Indonesia? Should it Try?", Bulletin of Indonesian Economic Studies 32: 45-74. Warr, P. (1986), "Indonesia's Other Dutch Disease: Economic Effects of the Petroleum Boom", pp. 288-320 in J.P. Neary and S. van Wijnbergen (eds.), Natural Resources and the Macroeconomy, Oxford: Basil Blackwell. Warr, P. (1992), "Comparative Advantage and Protection in Indonesia", Bulletin of Indonesian Economic Studies 28: 41-70. Warr, P. (2005), "Food Policy and Poverty in Indonesia: A General Equilibrium Analysis", Australian Journal of Agricultural and Resource Economics 49: 429-451. World Bank (various issues), World Development Indicators, Washington DC: World Bank. 26 Figure 1: Border and domestic prices of import-competing farm products, relative to the GDP deflator, Indonesia, 1975 to 2005 (Rp/kg, 2005=1 and percent) (a) Rice Real price NRP 5000 40.0 4500 30.0 4000 20.0 3500 10.0 3000 0.0 2500 -10.0 2000 1500 -20.0 1000 -30.0 500 -40.0 0 -50.0 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 2005 Left axis: Border price Domestic price Right axis: NRP (%) 27 Figure 1 (continued): Border and domestic prices of import-competing farm products, relative to the GDP deflator, Indonesia, 1975 to 2005 (Rp/kg, 2005=1 and percent) (b) urea fertilizer Real price NRP 40.0 5000 20.0 4000 0.0 3000 -20.0 2000 -40.0 1000 -60.0 0 -80.0 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 Left axis: Border price Domestic price Right axis: NRP (%) smoothed NRP 28 Figure 1 (continued): Border and domestic prices of import-competing farm products, relative to the GDP deflator, Indonesia, 1975 to 2005 (Rp/kg, 2005=1 and percent) (c) Sugar Real price NRP 10000 390 9000 340 8000 290 7000 240 6000 190 5000 140 4000 3000 90 2000 40 1000 -10 0 -60 1971 1974 1977 1980 1983 1986 1989 1992 1995 1998 2001 2004 Left axis: Border price Domestic price Right axis: NRP (%) 29 Figure 1 (continued): Border and domestic prices of import-competing farm products, relative to the GDP deflator, Indonesia, 1975 to 2005 (Rp/kg, 2005=1 and percent) (d) Soybean Real price NRP 6000 50 40 5000 30 20 4000 10 3000 0 -10 2000 -20 -30 1000 -40 0 -50 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 Left axis: Border price 2003 Domestic price Right axis: NRP (%) 30 Figure 1 (continued): Border and domestic prices of import-competing farm products, relative to the GDP deflator, Indonesia, 1975 to 2005 (Rp/kg, 2005=1 and percent) (e) Maize Real price NRP 2300 40.0 30.0 1800 20.0 10.0 1300 0.0 -10.0 800 -20.0 -30.0 300 -40.0 -50.0 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999 2002 -200 -60.0 Left axis: Border price Domestic price Right axis: NRP (%) Source: Authors' calculations. 31 Figure 2: Border and domestic prices of export crops relative to the GDP deflator, Indonesia, 1967 to 2005 (Rp/kg, 2005=1 and percent) (a) rubber Real price NRP 12000 30 10000 20 10 8000 0 6000 -10 4000 -20 2000 -30 0 -40 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 Left axis: Border price Domestic price Right axis: NRP (%) 32 Figure 2 (continued): Border and domestic prices of export crops relative to the GDP deflator, Indonesia, 1967 to 2005 (Rp/kg, 2005=1 and percent) (b) copra Real price NRP 8000 40 7000 20 6000 0 5000 -20 4000 -40 3000 -60 2000 1000 -80 0 -100 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 Left axis: Border price Domestic price Right axis: NRP (%) 33 Figure 2 (continued): Border and domestic prices of export crops relative to the GDP deflator, Indonesia, 1967 to 2005 (Rp/kg, 2005=1 and percent) (c) coffee Real price NRP 40,000 10 35,000 0 30,000 -10 25,000 -20 20,000 -30 15,000 -40 10,000 -50 5,000 -60 0 -70 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 Left axis: Border price Domestic price Right axis: NRP (%) 34 Figure 2 (continued): Border and domestic prices of export crops relative to the GDP deflator, Indonesia, 1967 to 2005 (Rp/kg, 2005=1 and percent) (d) tea Real price NRP 30000 100.0 80.0 25000 60.0 20000 40.0 20.0 15000 0.0 10000 -20.0 -40.0 5000 -60.0 0 -80.0 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 2003 Left axis: Border price Domestic price Right axis: NRP (%) Source: Authors' calculations. 35 Figure 3: Nominal rates of assistance to exportable, import-competing and all covered agricultural products, Indonesia, 1970 to 2004 (percent) 30 Import-competing Exportables 20 Total 10 0 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 -10 -20 -30 Source: Authors' spreadsheet 36 Figure 4: Nominal rates of assistance to all non-agricultural tradables, all agricultural tradable industries, and relative rates of assistancea, Indonesia, 1970 to 2004 (percent) 40 NRA ag tradables NRA non-ag tradables 30 RRA 20 10 0 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 -10 -20 -30 a The RRA is defined as 100*[(100+NRAagt)/(100+NRAnonagt)-1], where NRAagt and NRAnonagt are the percentage NRAs for the tradable parts of the agricultural and non- agricultural sectors, respectively. Source: Authors' spreadsheet 37 Table 1: Real growth of GDP and its sectoral components, Indonesia, 1968 to 2005 (percent per annum) Whole Pre-boom Boom Crisis Recovery period 1968-1986 1987-1996 1997-1999 2000-2005 1968-2005 Total GDP 7.4 7.7 -2.5 4.6 6.3 Agriculture 4.4 3.4 0.6 3.5 3.7 Industry, including mining 10.6 9.8 -2.3 4.2 8.5 Services 7.8 7.9 -4.0 5.7 6.6 Source: Author's calculations using data from World Bank (various issues). 38 Table 2: Industry shares of agricultural value added, Indonesia, 1972 to 2000 (percent) Sector 1971 1980 1990 2000 Paddy 46.1 38.0 37.5 30.8 Maize 3.1 3.7 4.1 5.9 Root crops 7.2 6.8 7.6 8.9 Vegetables and fruits 14.1 14.5 21.7 21.8 Other food crops 3.3 4.4 6.4 3.9 Rubber 5.5 5.2 2.0 5.5 Sugarcane 2.2 2.4 2.1 2.5 Coconut 5.2 4.3 3.3 3.7 Palm oil 2.9 2.1 2.4 2.3 Tobacco 2.5 1.7 0.7 0.3 Coffee 2.6 4.3 1.5 0.9 Tea 1.4 1.9 0.5 0.3 Cloves 1.4 3.0 1.6 0.9 Other agriculture 1.8 1.7 3.5 7.3 Livestock 0.6 6.0 5.0 4.9 Total 100.0 100.0 100.0 100.0 Source: Central Bureau of Statistics (1971, 1980, 1990 and 2000) 39 Table 3: Sales to export users as a share of total sales and imports as a share of total usage, Indonesia, 1971 to 2000 (percent) Export shares Import shares Sector 1971 1980 1990 2000 1971 1980 1990 2000 Paddya 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Maize 11.4 0.3 1.8 0.3 0.0 1.1 0.2 11.3 Root crops 4.7 3.7 0.7 0.3 0.0 0.1 0.0 0.1 Vegetables and fruits 0.1 0.2 0.2 0.3 0.6 1.1 0.9 4.9 Other food crops 4.3 0.3 1.9 0.4 0.0 22.7 18.1 47.7 Rubber 57.5 58.4 6.5 0.6 0.1 0.0 0.1 0.7 Sugarcane 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Coconut 6.6 1.2 0.3 1.5 0.0 0.0 0.0 0.1 Palm oil 2.3 33.5 29.5 0.2 0.0 0.0 0.1 0.2 Tobacco 7.4 12.3 7.6 0.0 1.6 7.3 15.6 0.0 Coffee 21.5 58.7 13.7 0.0 0.0 0.0 0.0 0.0 Tea 24.5 18.5 32.5 0.0 0.0 0.0 3.9 0.0 Cloves 0.0 0.0 0.1 3.0 47.4 12.7 0.0 24.5 Other agriculture 40.5 15.9 26.7 21.2 24.1 43.8 1.4 30.3 Livestock 1.2 0.0 1.8 2.1 0.0 0.5 0.9 8.4 Total Agriculture 7.5 9.3 2.8 2.6 1.3 2.8 2.0 8.7 Milled ricea 0.0 0.8 0.02 0.01 13.5 12.5 0.2 3.9 Fertilizer and pesticides 0.0 4.4 14.6 22.9 66.0 18.1 9.6 23.3 a The input-output tables classify unmilled rice (paddy) as an output of the agricultural sector and milled rice as an output of the manufacturing sector. Source: Central Bureau of Statistics (1971, 1980, 1990 and 2000) 40 Table 4: Estimates of effective rates of protection, Indonesia, 1987, 1995 and 2003a (percent) Sectors 1987 1995 2003 Agriculture 24 12 9 Manufacturing (excluding oil & gas) 86 24 16 Manufacturing (including oil & gas) 48 20 13 All tradable sectors 18 8 4 Anti-trade biasb 52 28 20 a Estimates for 1987 and 1995 are from Fane and Condon (1995). They measure ERPs just before the reform package of December 1987 and just after the reform package of May 1995. The estimates for 2003 apply the same methodology, using the tariff cuts announced in the May 1995 package to be implemented by 2003. With relatively minor exceptions, the plans announced in 1995 were in fact implemented. The projections of ERP for agriculture in 2003 shown in this table make no allowance for the increased protection of rice and sugar which occurred from 2000 onwards. b The anti-trade bias (ATB) is defined by: 1 + ATB = (1 + gm)/(1 + ge), where gm and ge denote, respectively, the average ERPs in all import-competing and all export-competing sectors. Source: Fane and Condon (1996), updated by Fane to 2003. 41 Table 5: Estimates of transmission elasticities from wholesale to farm prices, Indonesia Commodity Estimated elasticity (t-statistica) Rice 0.7345 (5.24) Soybeans 0.5294 (3.17) Sugar 0.6128 (2.29) Rubber 0.4365 (2.60) Tea 0.2607 (2.65) a t-statistics are shown in parthentheses. Source: Author's calculations, using data and methodology discussed in the text. Estimates shown relate to the parameter bi in equation (2). 42 Table 6: Nominal rates of assistance to covered agricultural products including subsidy to fertilizer use, Indonesia, 1970 to 2004 (percent) 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 Exportables a -3.3 -0.3 -7.0 -16.5 -24.6 -17.2 -3.0 Coffee -7.1 -3.7 -8.6 -2.2 -0.5 2.3 3.0 Tea -6.3 -1.9 1.8 -2.3 -2.5 -13.9 -15.5 Coconut -5.9 2.2 -6.1 -22.0 -45.6 -29.4 -8.1 Rubber 15.2 -3.4 -16.2 -20.5 -31.9 37.0 16.7 Palmoil -14.5 -9.2 22.2 -1.1 11.9 -18.3 -3.8 Import-competing products a -2.3 16.5 19.5 5.1 -0.7 -5.8 24.7 Rice 21.9 13.9 7.5 -0.9 -8.7 -13.0 18.7 Maize -15.4 10.2 18.6 21.9 22.5 24.6 10.8 Soybeans -5.9 31.8 49.0 17.0 17.7 17.5 1.2 Sugar 2.1 23.5 53.8 8.5 3.9 11.3 49.4 Poultryd 72.8 144.3 147.5 86.8 94.9 87.9 99.8 Total of covered products a -3.1 11.1 12.2 -0.3 -5.5 -9.1 16.4 Dispersion of covered products b 28.7 49.4 53.6 35.0 40.5 49.0 36.3 % coverage (at undistorted prices) 65 68 64 61 64 63 59 a Weighted averages, with weights based on the unassisted value of production. b Dispersion is a simple 5-year average of the annual standard deviation around the weighted mean of NRAs of covered products. c The output-subsidy equivalent of this fertilizer input subsidy is incorporated in the NRA shown for each crop, estimated by multiplying it by the share of fertilizer in the cost of production of each product and adding it to the output NRA. d First and last periods refer to 1971-74 and 2000-03. Source: Authors' spreadsheet 43 Table 7: Nominal rates of assistance to agricultural relative to non-agricultural industries, Indonesia, 1970 to 2004 (percent) 1970-74 1975-79 1980-84 1985-89 1990-94 1995-99 2000-04 a Covered products -3.1 11.1 12.2 -0.3 -5.5 -9.1 16.4 Non-covered products -1.9 5.4 4.2 -3.8 -8.4 -7.7 7.1 All agricultural products a -2.6 9.3 9.2 -1.7 -6.6 -8.6 12.0 Non-product specific (NPS) assistance 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Total agricultural NRA (incl. NPS) b -2.6 9.3 9.2 -1.7 -6.6 -8.6 12.0 Trade bias index c 0.00 -0.14 -0.21 -0.20 -0.24 -0.12 -0.22 Assistance to just tradables: All agricultural tradables -3.0 10.4 10.5 -1.9 -7.5 -9.7 13.9 All non-agricultural tradables 27.7 27.7 27.7 26.5 17.6 10.6 8.1 Relative rate of assistance, RRA d -24.1 -13.6 -13.5 -22.5 -21.3 -18.3 5.4 a NRAs including product-specific input subsidies. b NRAs including product-specific input subsidies. Total of assistance to primary factors and intermediate inputs divided to total value of primary agriculture production at undistorted prices (%). 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 non-agricultural sectors, respectively. Source: Authors' spreadsheet 44 Appendix Figure A1: Annual growth rate of real GDP, Indonesia, 1965 to 2005 (percent per annum) 15 10 5 0 1965 1970 1975 1980 1985 1990 1995 2000 -5 -10 -15 Annual GDP growth: Indonesia (%) Source: World Bank (various issues). 45 Appendix Figure A2: Share of agriculture, industry and services in GDP, Indonesia, 1965 to 2005 (percent) 60 50 40 30 20 10 0 1965 1970 1975 1980 1985 1990 1995 2000 Agriculture share of GDP Industry share of GDP Services share of GDP Source: World Bank (various issues). 46 Appendix Figure A3: Terms of tradea, Indonesia, 1965 to 2004 (2000 = 100) 140 120 100 80 60 40 20 0 1965 1970 1975 1980 1985 1990 1995 2000 Terms of trade a The terms of trade are here calculated as the ratio of average unit value of exports (value relative to volume) to the average unit value of imports. Source: World Bank (various issues). 47 Appendix Figure A4: Estimation of imputed NRA on output of import-competing crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (a) rice Real price NRP 2500 30 20 2000 10 1500 0 -10 1000 -20 -30 500 -40 0 -50 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Left axis: Actual farm level price Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (%) Source: Authors' calculations based on data used in Figure 1 (a) and Table 5. 48 Appendix Figure A4 (continued): Estimation of imputed NRA on output of import-competing crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (b) sugar Real price NRP 3500 180 160 3000 140 2500 120 100 2000 80 1500 60 40 1000 20 0 500 -20 0 -40 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Left axis: Actual farm level price Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (%) Source: Authors' calculations based on data used in Figure 1 (b) and Table 5. 49 Appendix Figure A4 (continued): Estimation of imputed NRA on output of import-competing crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (c) soybean Real price NRP 6000 80 5000 60 4000 40 3000 20 2000 0 1000 -20 0 -40 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 Left axis: Actual farm level price Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (%) Source: Authors' calculations based on data used in Figure 1 (c) and Table 5. 50 Appendix Figure A4 (continued): Estimation of imputed NRA on output of import-competing crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (d) maize Real price NRP 250000 50 40 200000 30 20 150000 10 0 100000 -10 50000 -20 -30 0 -40 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 Left axis: Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (NRP) Source: Authors' calculations based on data used in Figure 1 (d) and Table 5. 51 Appendix Figure A5: Estimation of imputed NRA on output of export crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (a) rubber Real price NRP 8000 140 7000 120 100 6000 80 5000 60 4000 40 3000 20 0 2000 -20 1000 -40 0 -60 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 Left axis: Actual farm level price Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (%) Source: Authors' calculations based on data used in Figure 2 (a) and Table 5. 52 Appendix Figure A5 (continued): Estimation of imputed NRA on output of export crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (b) copra Real price NRP 250000 20 10 200000 0 150000 -10 -20 100000 -30 -40 50000 -50 0 -60 1970 1973 1976 1979 1982 1985 1988 1991 1994 1997 2000 Left axis: Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (%) Source: Authors' calculations based on data used in Figure 2 (b) and Table 5. 53 Appendix Figure A5 (continued): Estimation of imputed NRA on output of export crops at the farm level, Indonesia, 1970 to 2001 (Rp/kg, 2005=1 and percent) (c) tea Real price NRP 18000 20 16000 15 10 14000 5 12000 0 10000 -5 8000 -10 6000 -15 -20 4000 -25 2000 -30 0 -35 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 Left axis: Actual farm level price Predicted price - with protection Predicted price - without protection Right axis: Imputed nominal rate of protection (%) Source: Authors' calculations based on data used in Figure 2 (c) and Table 5. 54 Appendix Figure A6: Agricultural production value shares by farm product, Indonesia, 1966 to 2003 (percent at distorted prices) 100 90 Non-covered Tea 80 Soybeans 70 Coffee 60 Rubber 50 Poultry 40 Palmoil Sugar 30 Maize 20 Coconut 10 Rice 0 19 6 19 7 19 8 19 9 70 19 1 19 2 19 3 74 19 5 19 6 19 7 78 19 9 19 0 81 19 2 19 3 19 4 19 5 19 6 19 7 19 8 19 1 19 9 19 0 19 2 19 3 94 19 5 19 6 19 7 98 20 9 20 0 20 1 02 03 6 6 6 6 7 7 7 7 7 7 7 8 8 8 8 8 8 8 8 9 8 9 9 9 9 9 9 9 0 0 19 19 19 19 19 19 19 20 Source: Authors' calculations using FAO data. 55 Appendix Table A1: Intermediate inputs as a share of output of agricultural industries, Indonesia, 1971 to 2000 (percent) Sector 1971 1980 1990 2000 Paddy 39.8 28.6 11.8 16.4 Maize 7.4 11.8 14.3 14.8 Root crops 26.9 14.0 5.1 6.0 Vegetables and fruits 5.7 5.9 7.3 8.4 Other food crops 12.9 11.3 14.4 16.0 Rubber 57.9 50.4 28.1 28.9 Sugarcane 37.7 35.6 25.3 24.5 Coconut 4.4 5.2 9.0 17.7 Palm oil 54.0 43.0 26.7 32.9 Tobacco 57.2 35.4 42.4 46.7 Coffee 35.7 26.2 33.0 32.4 Tea 47.3 11.3 11.5 14.1 Cloves 4.9 7.0 12.1 14.9 Other agriculture 8.6 7.2 10.3 23.1 Livestock 8.3 9.7 25.7 29.5 Total Agriculture 34.2 23.3 13.3 17.0 Source: Central Bureau of Statistics (1971, 1980, 1990 and 2000). 56 Appendix Table A2: Imported intermediate input as a share of total intermediate input use in agriculturea, Indonesia, 1980 to 2000 (percent) Sector 1980 1990 2000 Paddy 2.4 3.3 10.1 Maize 9.6 3.4 8.9 Root crops 2.1 4.0 5.1 Vegetables and fruits 8.4 8.2 6.4 Other food crops 3.9 4.4 17.2 Rubber 3.7 33.5 14.6 Sugarcane 42.1 5.2 9.3 Coconut 3.8 11.0 12.6 Palm oil 5.8 1.2 12.1 Tobacco 10.6 0.5 23.2 Coffee 1.5 0.7 11.4 Tea 8.3 0.8 17.7 Cloves 8.1 9.1 22.8 Other agriculture 8.8 0.9 8.3 Livestock 7.6 5.7 5.7 Total Agriculture 3.8 5.7 10.2 a Data for 1971 are not available because the input-output tables did not distinguish between imported and domestically sourced intermediate inputs until 1980. Source: Central Bureau of Statistics (1971, 1980, 1990 and 2000) 57 Appendix Table A3: Sales to intermediate users as a share of total sales, Indonesia, 1971 to 2000 (percent) Sector 1971 1980 1990 2000 Paddy a 56.3 70.9 99.9 97.5 Maize 12.2 14.1 35.1 52.7 Root crops 24.8 17.2 11.1 20.3 Vegetables and fruits 10.1 6.9 7.8 11.3 Other food crops 56.6 58.0 43.6 67.8 Rubber 42.5 38.2 92.5 99.3 Sugarcane 75.5 74.5 98.7 98.1 Coconut 68.1 52.2 35.5 55.6 Palm oil 46.6 26.1 70.3 99.7 Tobacco 53.1 78.6 87.9 87.5 Coffee 46.3 19.7 54.1 93.4 Tea 33.3 22.5 58.7 86.6 Cloves 99.8 95.6 96.7 93.5 Other agriculture 50.9 66.2 31.9 73.0 Livestock 79.2 88.9 90.2 82.0 Total Agricultureb 49.0 52.1 61.0 66.1 Milled ricec 3.6 6.8 7.8 11.4 a The input-output tables classify unmilled rice (paddy) as an output of the agricultural sector and milled rice as an output of the manufacturing sector. b Milled rice excluded. c Data for 1980 refer to milled cereal. Source: Central Bureau of Statistics (1971, 1980, 1990 and 2000) 58 Appendix Table A4: Total taxes on imports, Indonesia, 1974 to 2006 (percent) (a) 1974 to 1994 1974 1979 1985 1989 1994 01 Live animals 40 30 45 15 15 02 Meat and poultry 40 70 70 30 30 03 Fish 40 40 70 30 30 04 Milk and eggs 32 49 73 28 28 05 Offal 20 20 30 10 10 06 Flowers 30 60 60 30 30 07 Vegetables 60 60 86 28 24 08 Nuts and fruits 60 60 90 30 24 9 Coffee, tea, spices 70 73 94 24 24 10 Milled rice, other un-milled cereals 5 6 6 4 4 11 Other products of milling 0 30 45 13 11 12 Oil seeds and misc. grains 10 25 33 18 10 13 Gums, saps etc. 0 30 40 10 10 14 Wood 0 20 30 10 10 15 Edible fats and oils 2 39 59 21 16 16 Processed meat 72 130 130 60 33 17 Sugar 54 69 75 14 14 18 Cocoa 60 75 70 50 18 19 Prepared cereals 44 60 70 36 23 20 Prepared vegetables, fruit 50 60 80 33 28 21 Miscellaneous edible products 40 55 66 26 29 22 Beverages 73 77 81 50 37 23 Residues of food processing 30 30 35 11 9 24 Tobacco 45 50 83 38 23 Average, Chapters 1­24 37 51 66 26 20 59 Appendix Table A4 (cont.): Total taxes on imports, Indonesia, 1974 to 2006 (percent) (b) 1995 to 2000 1995 1996 1997 1998 1999 2000 01 Live animals 6 6 4 3 3 3 02 Meat and poultry 23 19 19 5 5 5 03 Fish 20 17 14 5 5 5 04 Milk and eggs 21 17 17 5 5 5 05 Offal 5 5 5 4 4 4 06 Flowers 16 15 12 11 11 11 07 Vegetables 18 18 15 5 5 5 08 Nuts and fruits 16 16 15 5 5 5 09 Coffee, tea, spices 21 21 17 5 5 5 10 Milled rice, other un-milled cereals 2 2 2 2 6 7 11 Other products of milling 5 5 5 4 5 5 12 Oil seeds and misc. grains 4 5 4 4 4 4 13 Gums, saps etc. 5 5 5 5 5 5 14 Wood 5 2 2 2 2 2 15 Edible fats and oils 10 9 7 5 5 5 16 Processed meat 25 20 20 5 5 5 17 Sugar 11 10 8 4 9 9 18 Cocoa 13 13 9 5 5 5 19 Prepared cereals 23 19 19 5 5 5 20 Prepared vegetables, fruit 28 22 22 5 5 5 21 Miscellaneous edible products 20 18 16 7 7 7 22 Beverages 35 35 35 33 33 33 23 Residues of food processing 4 4 4 4 1 1 24 Tobacco 18 16 13 10 10 10 Average, HS Chapters 1­24 16 14 13 6 6 6 60 Appendix Table A4 (cont.): Total taxes on imports, Indonesia, 1974 to 2006 (percent) (c) 2001 to 2006 2001 2002 2003 2004 2005 2006 01 Live animals 3 3 3 2 2 2 02 Meat and poultry 5 5 5 5 5 5 03 Fish 5 5 5 5 5 5 04 Milk and eggs 5 5 5 5 5 5 05 Offal 4 4 4 4 4 4 06 Flowers 11 11 11 11 13 12 07 Vegetables 5 5 5 5 5 5 08 Nuts and fruits 5 5 5 5 6 6 9 Coffee, tea, spices 5 5 5 5 5 5 10 Milled rice, other un-milled cereals 7 7 7 12 10 11 11 Other products of milling 5 5 5 5 5 5 12 Oil seeds and misc. grains 4 4 4 4 4 4 13 Gums, saps etc. 5 5 5 5 5 5 14 Wood 2 2 2 2 2 2 15 Edible fats and oils 5 5 5 5 5 5 16 Processed meat 5 5 5 5 5 6 17 Sugar 9 12 12 12 10 8 18 Cocoa 5 5 5 5 5 12 19 Prepared cereals 5 5 5 5 5 6 20 Prepared vegetables, fruit 5 5 5 5 5 6 21 Miscellaneous edible products 7 7 7 10 10 10 22 Beverages 33 33 33 33 33 34 23 Residues of food processing 1 1 1 1 1 2 24 Tobacco 10 10 10 11 11 11 Average, HS Chapters 1­24 6 6 6 7 7 7 Source: Directorate General of Customs and Excise, Import Duties, Indonesia, Jakarta (various years). 61 Appendix Table A5: Increases in tariffs (`bea masuk'), Indonesia, 1974 to 1994 (percentage points) 1974­79 1979­85 1985­89 1989­94 01 Live animals -10 0 -15 0 02 Meat and poultry 0 0 -10 0 03 Fish 0 0 -10 0 04 Milk and eggs 14 0 -18 0 05 Offal 0 0 -10 0 06 Flowers 0 0 0 0 07 Vegetables 0 0 -34 -3 08 Nuts and fruits 0 0 -30 -6 9 Coffee, tea, spices 0 0 -46 0 10 Milled rice, other un-milled cereals 2 -1 -1 0 11 Other products of milling 30 0 -18 -2 12 Oil seeds and misc. grains 15 0 -14 -1 13 Gums, saps etc. 30 0 -20 0 14 Wood 20 0 -10 0 15 Edible fats and oils 37 0 -24 0 16 Processed meat -2 0 -15 -22 17 Sugar 6 0 -46 0 18 Cocoa 0 0 -10 -32 19 Prepared cereals 6 -6 -13 -9 20 Prepared vegetables, fruit 0 0 -20 -2 21 Miscellaneous edible products 0 0 -14 3 22 Beverages -3 0 -14 -7 23 Residues of food processing 0 0 -21 0 24 Tobacco 0 0 -23 0 Average, Chapters 1­24 9 0 ­21 ­4 Source: Directorate General of Customs and Excise, Import Duties, Indonesia, Jakarta (various years). 62 Appendix Table A6: Increases in tariff surcharges (`bea masuk terbahan'), Indonesia, 1974 to 1994 (percentage points) 1974­79 1979­85 1985­89 1989­94 01 Live animals 0 15 -15 0 02 Meat and poultry 30 0 -30 0 03 Fish 0 30 -30 0 04 Milk and eggs 3 24 -27 0 05 Offal 0 10 -10 0 06 Flowers 30 0 -30 0 07 Vegetables 0 26 -25 -2 08 Nuts and fruits 0 30 -30 0 9 Coffee, tea, spices 3 21 -24 0 10 Milled rice, other un-milled cereals 0 0 -1 -1 11 Other products of milling 0 15 -15 0 12 Oil seeds and misc. grains 0 8 -2 -7 13 Gums, saps etc. 0 10 -10 0 14 Wood 0 10 -10 0 15 Edible fats and oils 0 20 -15 -5 16 Processed meat 60 0 -55 -5 17 Sugar 9 6 -15 0 18 Cocoa 15 -5 -10 0 19 Prepared cereals 10 16 -21 -5 20 Prepared vegetables, fruit 10 20 -27 -3 21 Miscellaneous edible products 15 11 -26 0 22 Beverages 7 4 -16 -6 23 Residues of food processing 0 5 -3 -3 24 Tobacco 5 33 -23 -15 Average, Chapters 1­24 6 15 ­19 ­2 Source: Directorate General of Customs and Excise, Import Duties, Indonesia, Jakarta (various years). 63 Appendix Table A7: Increases in total import taxes, Indonesia, 1994 to 2006 (percentage) 1994­95 1995­97 1997­98 1998­06 1994­06 01 Live animals -9 -3 -1 -1 -13 02 Meat and poultry -7 -4 -14 0 -25 03 Fish -10 -6 -9 0 -25 04 Milk and eggs -7 -4 -12 0 -23 05 Offal -5 -1 0 0 -6 06 Flowers -14 -4 -1 2 -18 07 Vegetables -6 -3 -10 0 -19 08 Nuts and fruits -8 -1 -10 1 -18 9 Coffee, tea, spices -3 -4 -12 0 -19 10 Milled rice, other un-milled cereals -1 0 0 9 8 11 Other products of milling -6 0 0 1 -6 12 Oil seeds and misc. grains -6 0 0 0 -6 13 Gums, saps etc. -5 0 0 0 -5 14 Wood -5 -3 0 0 -8 15 Edible fats and oils -6 -3 -2 0 -11 16 Processed meat -8 -5 -15 1 -27 17 Sugar -3 -3 -4 4 -6 18 Cocoa -5 -4 -4 7 -6 19 Prepared cereals 1 -5 -14 1 -17 20 Prepared vegetables, fruit -1 -6 -17 1 -23 21 Miscellaneous edible products -8 -4 -9 3 -18 22 Beverages -2 -1 -2 1 -3 23 Residues of food processing -5 0 0 -2 -7 24 Tobacco -4 -5 -3 0 -12 Average, Chapters 1­24 -4 -3 -7 1 -13 Source: Directorate General of Customs and Excise, Import Duties, Indonesia, Jakarta (various years). 64 Appendix Table A8: Nominal rate of assistance at the wholesale level, by agricultural commodity, and fertilizer consumer tax equivalent (CTE), Indonesia, 1970 to 2004 (percent) Fertilizer Year Rice Sugar Soybean Maize Rubber Copra Coffee Tea CTE 1970 0.0 0.07 0.3 -41.9 -4.3 -36.1 -54.8 -23.4 -67.9 1971 0.0 106.5 -3.8 -46.7 -9.3 -8.2 -59.7 -17.7 -67.9 1972 0.0 26.9 -8.8 -11.0 -6.1 2.6 -65.8 -21.7 -67.9 1973 0.0 26.0 -45.4 -49.0 -2.2 -6.9 -30.7 -17.0 -67.9 1974 0.0 -56.5 -10.4 -53.0 -10.8 -26.5 -40.9 -32.8 -67.9 1975 -36.7 -22.8 31.9 -8.1 -9.5 -13.7 -39.1 -14.4 -67.9 1976 14.8 49.0 33.1 21.9 -3.6 20.9 -28.4 1.9 -50.5 1977 13.8 127.0 27.5 28.1 -2.3 30.8 -36.7 4.1 -47.0 1978 -14.2 140.4 31.8 23.9 -3.2 19.1 -36.6 84.0 -45.9 1979 -8.7 62.5 21.6 9.3 1.0 -36.9 -43.6 41.7 -48.8 1980 -21.6 -28.9 31.2 8.5 -1.2 -11.8 -45.2 0.2 -48.0 1981 -23.9 91.0 39.9 10.8 -6.0 18.1 -43.0 14.4 -54.6 1982 30.3 243.6 46.0 39.8 3.4 14.8 -41.9 -4.5 -57.5 1983 4.4 150.9 22.2 6.1 -24.3 -40.3 -63.5 -48.9 -60.0 1984 9.5 290.4 20.7 4.9 -17.6 -54.0 -65.7 -59.0 -61.6 1985 17.6 394.8 36.3 21.8 -7.7 -38.3 -57.2 -35.4 -60.3 1986 14.5 192.6 37.9 27.8 -9.9 -5.4 -45.0 -23.0 -53.4 1987 -11.9 122.3 29.1 39.7 3.0 -42.9 -15.2 -47.7 -51.6 1988 -19.4 59.2 11.4 -13.7 3.2 -50.5 -32.2 -39.9 -51.9 1989 -24.6 36.7 17.6 -9.8 2.4 -82.9 -47.0 -50.0 -44.6 1990 -16.0 53.0 31.7 25.1 2.7 -74.8 -36.6 -59.1 -39.4 1991 -21.1 115.6 35.8 31.6 4.8 -77.2 -27.1 -49.1 -40.2 1992 -10.2 121.8 37.9 20.8 4.0 -78.6 -24.3 -52.2 -34.3 1993 -2.4 103.0 32.3 29.7 9.2 -75.6 -16.3 -46.1 -21.7 1994 -7.6 63.3 35.9 33.7 10.0 -80.6 -21.4 -36.9 -19.9 1995 -12.9 55.6 33.4 27.4 12.1 -78.1 -21.6 -40.0 -22.1 1996 -9.8 67.7 21.6 19.0 9.8 -78.7 4.1 -61.1 -22.7 1997 -3.6 41.6 15.1 32.0 4.1 -37.2 -1.1 -66.7 -23.5 1998 -42.9 -4.9 -11.2 -0.9 3.6 -55.1 -29.3 -69.1 -37.0 1999 11.4 58.9 33.6 22.6 24.2 -29.8 -6.7 -54.5 -28.4 2000 14.0 25.7 -1.6 -7.9 6.4 -23.8 7.7 -60.0 -25.6 2001 16.0 39.6 6.6 -19.0 11.5 -23.9 -2.7 -57.8 -22.3 2002 29.1 70.5 6.7 -1.6 -8.2 -25.2 -9.6 -50.3 -19.7 2003 33.4 128.5 5.0 33.9 -14.8 -15.9 -9.5 -41.3 -14.3 2004 8.2 88.4 -4.4 8.8 -22.5 -15.9 -8.4 -49.4 -28.7 Source: Authors' calculations. 65 Appendix Table A9: Nominal rate of assistance at the farm level, by agricultural commodity, excluding fertilizer subsidy, Indonesia, 1970 to 2004 (percent) Year Rice Sugar Soybean Palmoil Maize Rubber Copra Coffee Tea Poultry 1970 10.6 15.7 -1.9 -33.5 -9.0 -4.2 -17.7 -18.7 -4.9 na 1971 10.6 15.7 -1.9 -30.3 -22.7 -4.2 -3.7 -21.1 -4.9 89.4 1972 10.6 -10.3 -7.7 25.6 -24.6 4.7 1.1 -24.4 -22.3 62.2 1973 10.6 -42.9 -23.9 7.2 -7.4 9.8 -3.1 -9.1 -29.5 50.4 1974 10.6 -22.8 -13.0 -41.5 -25.4 -14.2 -12.6 -12.8 -31.8 89.2 1975 10.6 16.2 22.9 -15.3 -26.8 -2.3 -6.2 -12.1 -18.0 106.8 1976 10.6 52.7 24.9 6.6 -5.6 7.7 8.6 -8.3 -15.7 92.3 1977 14.6 28.2 26.4 -2.7 19.8 -11.2 12.4 -11.2 -13.4 187.4 1978 -7.9 -4.2 32.0 -12.8 27.4 -40.9 7.9 -11.2 13.0 219.6 1979 -0.1 -23.2 37.1 -21.7 22.2 -39.7 -18.2 -13.9 -3.9 115.2 1980 -17.3 15.6 29.4 9.8 7.4 -42.3 -5.3 -14.5 -0.4 137.7 1981 -14.2 47.1 45.9 14.3 6.7 -36.3 7.5 -13.6 13.0 183.1 1982 24.6 18.9 73.1 64.9 8.8 -27.9 6.2 -13.2 5.6 185.6 1983 2.9 55.1 37.2 35.7 45.2 -29.3 -20.2 -23.1 -13.7 132.2 1984 1.8 78.3 39.9 -13.6 4.8 -18.5 -28.7 -24.3 -14.0 99.0 1985 8.1 28.3 26.1 3.8 3.8 -21.9 -19.0 -19.9 -0.6 144.1 1986 1.5 10.3 33.0 48.4 19.8 -37.5 -2.4 -14.4 -0.7 107.8 1987 -9.7 -10.5 9.3 -12.8 27.1 -46.4 -21.7 -4.2 -15.1 57.9 1988 -14.4 -17.4 -4.5 -33.4 44.9 -30.2 -26.4 -9.6 -7.9 68.1 1989 -17.4 -15.2 -1.8 -11.6 -9.0 -23.8 -53.8 -15.2 -6.4 55.8 1990 -13.5 7.6 16.6 17.1 -6.6 -30.7 -45.2 -11.2 -11.1 68.9 1991 -19.6 8.2 14.2 0.4 23.6 -36.0 -47.6 -7.9 -4.9 147.3 1992 -10.5 -2.2 11.1 4.3 32.2 -39.4 -49.0 -7.0 -6.1 69.2 1993 -5.6 -10.0 14.1 23.3 18.7 -40.6 -46.0 -4.5 -7.0 86.2 1994 -7.4 -10.0 17.2 14.1 29.6 -42.2 -51.1 -6.1 4.3 102.9 1995 -14.2 -10.0 26.0 -17.6 35.3 -35.3 -48.4 -6.2 4.3 102.1 1996 -17.2 -15.4 16.8 1.6 26.5 -24.9 -49.1 1.1 -15.1 136.2 1997 -9.1 -32.6 13.4 -12.0 16.8 52.1 -18.4 -0.3 -24.9 70.9 1998 -37.9 61.2 -3.4 -57.1 32.8 51.0 -29.5 -8.6 -26.4 -8.9 1999 0.8 32.6 21.7 -6.2 -0.6 117.9 -14.3 -1.8 -18.6 139.2 2000 12.6 56.1 -1.2 -2.8 20.7 51.3 -11.2 2.0 -21.2 107.5 2001 19.1 22.7 -7.3 12.4 -5.4 34.4 -11.2 -0.7 -20.1 70.4 2002 20.7 38.7 3.5 -11.6 -9.0 -3.7 -11.9 -2.6 -16.6 120.2 2003 23.6 65.9 2.6 -13.2 -22.7 -6.8 -7.3 -2.6 -12.9 101.2 2004 6.0 47.4 -2.4 na -24.6 -10.5 -7.3 -2.3 -16.2 na Source: Authors' calculations. 66 Appendix Table A10: Nominal rate of assistance at the farm level, covered agricultural products including fertilizer subsidy,a Indonesia, 1970 to 2004 (percent) Palm Weighted Year Rice Sugar Soybean oil Maize Rubber Copra Coffee Tea Poultry average 1970 22.0 26.1 1.7 -33.5 -7.2 12.0 -16.6 -8.5 9.5 na -5.8 1971 22.0 26.5 1.7 -30.3 -20.5 12.3 -2.4 -10.9 8.5 89.4 5.1 1972 21.9 0.8 -4.0 25.6 -22.2 21.5 2.5 -14.2 -9.9 62.2 1.4 1973 21.9 -31.6 -20.1 7.2 -4.8 27.0 -1.7 1.0 -18.0 50.4 -7.5 1974 21.9 -11.1 -9.1 -41.5 -22.5 3.2 -11.1 -2.7 -21.4 89.2 -8.5 1975 21.8 28.3 26.9 -15.3 -23.6 15.5 -4.7 -2.0 -8.6 106.8 14.7 1976 18.8 61.7 27.8 6.6 -3.1 20.9 9.9 -1.0 -9.6 92.3 20.8 1977 22.1 36.8 29.2 -2.7 22.4 1.3 13.6 -4.4 -8.4 187.4 20.2 1978 -0.6 4.4 34.8 -12.8 30.1 -28.5 9.1 -4.5 17.2 219.6 3.2 1979 7.6 -13.8 40.1 -21.7 25.2 -26.3 -16.8 -6.8 -0.1 115.2 -3.6 1980 -9.7 25.0 32.4 9.8 10.6 -28.9 -3.9 -7.5 2.6 137.7 -3.5 1981 -6.0 57.7 49.5 14.3 10.5 -21.5 9.3 -5.3 16.5 183.1 9.3 1982 32.8 29.9 77.1 64.9 12.9 -12.9 8.2 -3.9 9.3 185.6 27.4 1983 11.0 66.4 41.6 35.7 49.7 -14.1 -17.9 -13.0 -9.7 132.2 16.6 1984 9.6 89.9 44.6 -13.6 9.6 -3.5 -26.2 -13.4 -9.8 99.0 11.0 1985 15.3 39.4 31.0 3.8 8.7 -7.7 -16.3 -8.7 3.6 144.1 14.4 1986 7.4 20.1 37.4 48.4 24.3 -25.4 0.2 -4.2 3.2 107.8 10.8 1987 -4.4 -1.3 13.8 -12.8 31.6 -35.2 -19.1 6.1 -11.3 57.9 -5.2 1988 -9.4 -8.2 0.2 -33.4 49.6 -19.4 -23.6 1.1 -4.0 68.1 -9.2 1989 -13.5 -7.4 2.4 -11.6 -4.8 -14.9 -51.2 -5.6 -3.0 55.8 -12.3 1990 -10.4 14.4 20.5 17.1 -2.8 -23.2 -42.8 -2.4 -8.0 68.9 -4.5 1991 -16.3 15.0 18.1 0.4 27.5 -28.4 -44.9 1.0 -1.7 147.3 -6.8 1992 -7.7 3.5 14.5 4.3 35.5 -33.0 -46.6 0.5 -3.4 69.2 -5.3 1993 -3.8 -6.4 16.3 23.3 20.7 -36.6 -44.4 0.2 -5.2 86.2 -4.7 1994 -5.6 -6.8 19.2 14.1 31.4 -38.5 -49.6 -1.9 5.9 102.9 -6.2 1995 -12.2 -6.5 28.2 -17.6 37.4 -31.3 -46.6 -1.5 6.2 102.1 -10.8 1996 -15.1 -11.8 19.0 1.6 28.6 -20.8 -47.1 5.8 -13.2 136.2 -9.8 1997 -6.9 -29.0 15.7 -12.0 18.9 56.3 -16.2 4.5 -23.0 70.9 -6.8 1998 -34.3 66.8 0.2 -57.1 36.1 57.6 -25.8 -1.2 -23.3 -8.9 -24.2 1999 3.6 36.9 24.4 -6.2 2.0 123.0 -11.4 3.9 -16.1 139.2 6.2 2000 15.2 59.9 1.3 -2.8 23.0 55.7 -8.4 7.0 -19.1 107.5 15.5 2001 21.4 26.0 -5.1 12.4 -3.4 38.3 -8.7 3.6 -18.2 70.4 17.0 2002 22.7 41.5 5.4 -11.6 0.6 -0.3 -9.6 1.2 -14.9 120.2 15.4 2003 25.1 68.0 4.0 -13.2 25.2 -4.3 -5.6 0.1 -11.7 101.2 17.9 2004 9.1 51.4 0.4 na 8.9 -5.7 -3.7 3.1 -13.7 na 12.2 a NRA at the farm level for that industry's output plus the product of the cost share of fertilizer for that industry and the subsidy to fertilizer use (from Appendix Tables A8 and A9); averaged using value of production at undistorted prices as weights (Appendix Table A12). Source: Authors' calculations. 67 Appendix Table A11: Nominal rates of assistance to all,a exportableb and import-competing b agricultural industries, to non-agricultural industries, and relative rate of assistance,c Indonesia, 1970 to 2004 (percent) Total ag NRA Ag Covered products Non- All tradables Non-ag covered products NRA tradables Inputs Outputs products (incl NPS) NRA RRA 1970 4 -10 -4 -5 -6 28 -26 1971 4 1 3 4 5 28 -18 1972 5 -4 1 1 2 28 -20 1973 4 -11 -4 -6 -7 28 -27 1974 4 -13 -6 -7 -8 28 -28 1975 8 7 6 12 13 28 -11 1976 6 15 12 18 20 28 -6 1977 5 15 12 17 20 28 -6 1978 5 -2 2 3 3 28 -19 1979 6 -10 -4 -4 -4 28 -25 1980 6 -10 -4 -4 -4 28 -25 1981 7 2 5 8 9 28 -15 1982 7 20 13 22 25 28 -2 1983 7 10 5 12 14 28 -11 1984 6 5 2 7 9 28 -15 1985 6 8 4 10 12 28 -12 1986 6 5 4 8 9 28 -14 1987 5 -10 -6 -6 -7 28 -27 1988 5 -14 -9 -9 -10 26 -29 1989 4 -16 -12 -12 -14 24 -30 1990 3 -8 -7 -6 -6 22 -23 1991 4 -11 -10 -8 -9 20 -24 1992 3 -8 -9 -7 -7 18 -21 1993 2 -7 -8 -6 -7 16 -19 1994 2 -8 -8 -7 -8 14 -19 1995 2 -13 -11 -11 -12 12 -21 1996 2 -12 -9 -10 -11 11 -20 1997 2 -9 -4 -6 -7 11 -16 1998 4 -28 -17 -22 -25 10 -31 1999 3 3 3 5 6 10 -4 2000 3 13 8 13 14 9 5 2001 2 15 9 14 16 9 7 2002 2 13 7 12 14 8 5 2003 2 16 9 14 16 8 8 2004 3 9 3 8 10 8 2 a. NRAs including assistance to nontradables and non-product specific assistance. b. NRAs including products specific input subsidies. 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. Source: Authors' calculations. 68 Appendix Table A12: Value shares of primary production of covered products at farmgate undistorted prices, Indonesia, 1970 to 2004 (percent) Total- Non- Rice Sugar Soybean Palmoil Maize Rubber Coffee Tea Poultry Coconut covered covered 1970 18 11 1 1 4 5 2 1 n.a. 23 66 34 1971 16 15 1 1 4 4 2 1 1 23 68 32 1972 16 18 1 1 4 3 1 1 1 16 62 38 1973 15 17 2 1 3 4 1 0 1 21 65 35 1974 12 13 1 1 3 4 1 0 1 28 65 35 1975 34 11 1 1 4 2 1 0 1 12 69 31 1976 36 8 1 1 3 3 2 0 1 14 70 30 1977 30 8 1 1 2 3 3 1 0 18 67 33 1978 32 11 1 1 2 4 2 0 0 15 68 32 1979 28 11 1 1 2 5 2 0 1 15 67 33 1980 34 7 1 1 2 6 2 0 1 11 65 35 1981 35 11 1 1 2 3 1 0 1 10 65 35 1982 31 14 0 1 2 3 1 0 1 9 63 37 1983 33 10 1 1 2 3 1 0 1 11 63 37 1984 31 7 1 2 3 3 1 0 1 13 62 38 1985 30 10 1 2 3 3 1 0 1 10 60 40 1986 29 11 1 1 3 3 2 0 1 7 59 41 1987 27 11 1 2 2 5 2 0 1 8 60 40 1988 30 12 1 2 1 5 2 0 1 8 63 37 1989 35 12 1 2 2 4 1 0 1 4 64 36 1990 35 11 1 2 4 4 1 0 2 3 63 37 1991 36 11 1 2 3 4 1 0 1 4 63 37 1992 34 14 2 3 2 4 1 0 2 5 66 34 1993 29 14 1 2 2 4 1 0 2 4 61 39 1994 29 13 1 4 2 6 2 0 2 5 65 35 1995 30 10 1 4 2 6 2 0 1 5 61 39 1996 33 10 1 4 3 5 1 0 1 5 62 38 1997 30 10 1 5 3 2 1 0 2 8 62 38 1998 36 3 1 9 2 2 2 0 1 9 67 33 1999 35 4 1 4 3 1 2 0 1 13 64 36 2000 33 4 1 6 2 2 2 0 2 11 63 37 2001 29 7 1 6 3 2 1 0 2 9 61 39 2002 27 5 0 8 3 3 1 0 2 8 58 42 2003 23 4 0 7 3 4 1 0 2 8 54 46 2004 36 6 1 n.a. 4 5 1 0 n.a. n.a. 54 46 Source: Authors' calculations.