This internal working paper is prepared for STAFF USE ONLY. The views expressed are not necessarily those of the World Bank. AGREP Division Working Paper No. 116 THE PROCESS OF FORMULATION AND IMPLEMENTATION OF ECONOMIC AND AGRICULTURAL POLICY IN THE DOMINICAN REPUBLIC G. O'Mara Economics and Policy Division Agriculture and Rural Development Department June 1986 CURRENCY EQUIVALENTS Currency Unit: Dominican Peso (DR$) US$1;00 - DR$ 3.00 (estimated annual average of the floating exchange rate for 1985) WEIGHTS AND MEASURES Metric System 1 pound (lb) = 0.45 kilograms (kg) 1 hundredweight (cwt) = 100 pounds (lbs) = 45.36 kilograms (kg) 1 tarea (ta) = 0.065 hectares (ha) 1 hectare (ha) = 15.9 tareas (ta) a ABBREVIATIONS ADACA - Aseguradora Dominicana Agropecuaria (Dominican Agricultural Insurance Company) BA - Banco Agricola (Agricultural Bank) CAPA - Comite de Analisis de Politica Agropecuaria - (Committee for Analysis of Agricultural Policies) CCC - Commodity Credit Corporation CDE - Corporacion Dominicana de Electricidad (Dominican Electricity Corp) CEA - Consejo Estatal de Azucar (State Sugar Council) CEDOPEX - Centro Dominicano de Promocion de Exportaciones (Dominican Export Promotion Center) CNA - Consejo Nacional de Agricultura (National Council for Agriculture) CVMA - Centro de Venta de Materiales Agropecuarios (Farm Input Supply Center) DGCP - Direccion General de Control de Precios (Directorate of Price Control) FIDE - Fondo de Inversiones para el Desarrollo Economico (Investment Fund for Economic Development) IAD - Instituto Agrario Dominicano (Dominican Agrarian Reform Institute) IDB - Inter-American Development Bank IMF - International Monetary Fund INAZUCAR - Instituto Nacional del Azucar (National Sugar Institute) INDRHI - Instituto Nacional de Recursos Hidraulicos (National Water Resources Institute) INESPRE - Instituto Nacional de Estabilizacion de Precios (National Price Stabilization Institute) MD - Molinos Dominicanos (Dominican Wheat Mill) ONAPLAN - Oficina Nacional de Planificacion (National Planning Office) ONAPRE$ - Oficina Nacional de Presupuesto (National Budget Office) ONE - Oficina Nacional de Estadistica (National Statistical Office) PNA - Programa Nacional de Afiliados (Affiliated Stores Program) PROSEMA - Programa de Servicio de Mecanizacion Agricola (Agricultural Mechanization Service) PVA - Programa de Ventas Populares (Popular Sales Program) SEA - Secretaria de Estado de Agricultura (Ministry of Agriculture) STP - Secretariado Tecnico de la Presidencia (Technical Secretariat to the Presidency) USAID - Unites States Agency for International Development FINANCIAL YEAR January 1 - December 31 CEA: October 1 - September 30 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Volume II - Table of Contents The Process of Formulation and Implementation of Economic and Agricultural Policy in the Dominican Republic Page No. 1. Introductionri..................... ........... ........... 1 II. The Importance of Macroeconomic Policy.................... 1 III. Interaction of Macro and Sector Policy................... 6 A. Inconsistency between Macro and Sector Policy......... 6 B. Effects of Policy Inconsistency....................... 7 C. Secondary Role of the Ministry of Agriculture......... 7 IV. The Costly Intersection of Policies in Rice Production.... 8 A. The Involvement of the Dominican Agrarian Reform Institute (IAD) .....,,,. ............................. 8 B. The Role of the Price Stabilization Institute (INESPRE) in Rite Price Suppression.................. 9 C. .The Countervailing Roles of the Agricultural Bank and the National Water Resources Institute...... 9 V. Efforts at Diversification Away from Suga................ 10 A. The Evolving Crisis in Sugar........................ 10 B. The Diversification Program of the State Sugar Cou cil ........................... ... . ............ 11 VI. An Efficient Role for the tinistry of Agriculture......... 12 A. The Present Situatio...............2................. 12 B. How to Achieve More Efficiency in Public Agencies Serving Agriculture........................ 14 VII. Resource Mobilization In and Out of Agriculture........... 14 A. The Still Distorted Present Situation................. 14 B. Toward a More Neuzral and Stable Tax Revenue System... 16 VIII. Conclusions and Recommendations................... 17 - ii - Annex I (coat'd) Tables Page No. 1.1 Rates in Dual Foreign Exchange Market, 1970-84........... 20 1.2 Financing of Current Account Balance, 1970-84............ 21 1.3 Consolidated Accounts of Public Sector................... 22 1.4 Structure of Taxes, Central Government, 1970-84.......... 23 1.5 Structure of Taxes, Central Government, 1970-84.......... 24 1.6 Linear Regression Tests of Fiscal Response to Balance of Payments Shocks, 1970-84.......................... 25 1.7 Structure of the Dominican Sugar Industry................ 26 1.8 Cane and Sugar Production and Export Data, 1970-84....... 27 1.9 Sugar Export Prices, 1970-97-28 ........ .............. 28 1.10 CEA Unit Costs and Profits per Export Sale from Outgrower Sugarcane................................... 29 1.11 Dominican Republic: Expenditures by Ministry of Agriculture, 1978-84.............................. 30 1.12 Public Enterprises Related to Agriculture, Current and Capital Expenditures, 1979-84...............9.8.. 31 1.13 Dominican Republic: GDP from Primary Production, 1970-84..........................4............ 32 1.14 Calculation of Implicit Taxes on Traditional Agricultural Exports ...............**************# **... 33 1.15 Export and Import Taxes on Traditional Agricultural Exports, 1970-85....................................... 34 1.16 Calculation of Lower Bounds for Recent Agricultural Tax Shares.....................*******.*******.***.**** 5 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY The Process of Formulation and Implementation of Economic and Agricultural Policy in the Dominican Republic I. Introduction 1. This paper is concerned with 4J4LJnaiaton and iMR_etation as a decision-making process. It recognizes that this process is a dynamic and adaptive response to unforeseen change that endeavors to balance off com- peting claims on resource in a way that achieves the highest feasible level of social welfare, according to some politically determined scheme of weights for the claims of various groups. However, the analysis will be presented from the viewpoint of economic efficienc in order to proviie a sharper ana- lytic focus. Since economic eff4iency is inevitably a concern of government in a world of scarce resources,a knowledge of the opportunity costs of al- ternative policies is always an important part of policy analysis. 2. Although this study is primarily concerned with the agricultural sector, it will empha§ize the overwhelming importance of macroeconomic policy in determining the impact of the more microeconomically oriented sector level policies. It follows that there is little point in discussing sector-speci- fic policies in isolation from the conditioning impact of macro policy. From this viewpoint, consistency between macro policy and sector policy is essen- tial to implementing an effective sector policy. 3. L n accord with this insight, the paper will discuss macro policy in the Dominican Republic prior to a general analysis of the reasons for the ineffectiveness of agricultural policy in that country. In-succeeding sections, case studies of policies toward rice Production and diversification away from sugar will be examined in order to consider in detail the inter- action of macro and sector policy. This discussion will also provide and opportunity to look at some of the key2b i s,(netitutions involved in imple- mentation of agricultural policy. The following sections will then present a discussion of efficiency in public sector institutions serving agriculture and a view of resourc i llQn in and out of agriculture. Finally, con- clusions and recommendations will be presented. II. The Importance of Macroeconomic Policy 4, The Dominican Republic is a small, relatively open economy for which export earnings are heavily dependent (i.e., 88%) on natural resources, with traditional agricultural exports, e.g., sugar, coffee, cocoa and tobacco, providing over 60% of total export earnings. However, the country 2- has developed sufficiently that the population and labor force are predom- inantly urban. This transformation has occurred under a'consistent policy of import substitution. Inevitably, sustained recourse to such a policy impli- citly taxes the export sectors through the effects of import suppression on the foreign exchange rate. This is true even where the exchange rate is freely floating or is adjusted frequently to new equilibrium levels. It applies a fortiori when foreign exchange controls overvalue the domestic cur- rency. 5. Such was the case with the Dominican Republic from the early 1960s through 1984. The official exchange rate was fixed at one Dominican peso per U.S. dollar throughout this period. In the late 1960s a dual foreign ex- change market emerged through official tolerance of a parallel market which absorbed excess demand for dollars at a floating rate. This parallel rate provides some measure, however imperfect, of the overvaluation of the peso. Table 1.1 presents time series estimates of the average parallel rate for 1970-84. These data indicate (i) increasing overvaluation of the peso; and (ii) an increasing proportion of trade financed through the parallel market over the period covered. Beginning in 1979, the government gave exporters of specified "non-traditional" goods partial or complete exemption from a requirement to surrender foreign exchange earnings at the official rate. By 1984, the government discontinued sale of foreign exchange at the official rate for all imports other than petroleum. In that same year, traditional exporters received a rate of 1.48 pesos per U.S. dollar of export value. In January 1985, this arrangement was superceded by an ostensible unification in which the parallel market rate is determined as before by the supply and demand for dollars in that market, mainly derived from the private sector, and an official "market" at the Central Bank in which the parallel market rate is used for all transactions. Thus, only one foreign exchange rate exists at any one time, but the rate is set in a market where only part of the foreign exchange transactions occur. Of course, if all agents have rational expectations and access to information on the central bank market, then profit maximizing behavior by traders should result in price determina- tion in the parallel market which reflects the overall supply of and demand for dollars. In the absence of these conditions, it is not clear that the parallel market rate is an equilibrium rate. 6. The data of Table 1.1 suggest that overvaluation of the peso impli- citly taxed traditional exports at a rate that varied from 12 to over 80% during the 1970-84 period. However, these estimates are clearly only lower bounds since exporters would have been implicitly taxed even if the currency was not overvalued, given the existence of a structure of protective tariffs, quotas, and prohibitions on imports. To estimate the actual level of implicit taxation would require calculating the equilibrium exchange rates in the absence of the structure of protection, and the implied level of taxation could be much larger than the implicit taxation due to foreign exchange overvaluation via exchange controls. 7. The essence of import substituting industrialization is that it applies strong disincentives to production for export and strong incentives to production for domestic consumption of selected goods marked for import substitution. If the import substitution goods are not wisely selected, or -3- if the protection is excessive, several decades of such a policy can result in a high cost, capital intetsive industrial structure that cannot compete internationally and yields very low social returns to domestic investment. At the same time, several decades of disincentives to export will have retarded development of either the traditional export sectors or new ones and provoked increasingly severe balance of payments crises. From this perspective, successful use of an inward oriented policy lies in recognizing when it has started to become counter productive and a shift to an outward oriented policy is necessary to sustain development. 8. The critical magnitude in the analysis of the balance of payments over a sequence of years is the current account balance since this number measures the real resource transfer to or from the country. The current account balance has been negative for the Dominican Republic in every year since 1961, implying a real resource transfer to the country for each of those years. Now there is nothing wrong or even exceptional in a negative current account balance that persists for decades. It may simply mean that attractive investment prospects have induced a sustained inflow of external capital. However, unless the capital inflows are pure gifts, there will eventually have to be a period of sustained capital outflows as the returns from previous investments pay-divilehds or retire debt. The overriding con- cern is that capital inflows are used in productive investments. Clearly external borrowing to finance current consumption simply transfers the burden of external debt service to another generation without providing an increase to the flow of income that exceeds the burden of the debt. 9. The sources of funds utilized to finance current account deficits can provide some insight into the uses of external capital. Table 1.2 pre- sents such an accounting of sources of financing for 1970-84. Note that over the years 1970-77, private capital inflows always exceed official capital inflows (i.e., new external debt by Dominican government), usually by a sub- stantial margin, and reserve adjustments are relatively small. Whereas over the years 1978-82, official capital inflows always exceed private capital in- flows by a substantial margin and reserve adjustments were much larger rela- tive to the other financing sources, For the years 1983-84, official capital inflows fell dramatically and these decreases were not offset by private capital inflows. Instead, a large reserve adjustment was required in 1983; and in 1984, external credit collapsed and extraordinary measures were required. This was followed in 1985 by strong adjustment measures -- e.g., foreign exchange market unification, a large increase in revenues from new taxes and fiscal stringency -- and access to an IMF standby credit. Thus, the period 1970-84 breaks down into a sub-period with dominant private capital inflows, years when investment opportunities stemming from the import substitution policy were financially attractive, followed by a subperiod when government external borrowing dominated, which culminated in years of balance of payments crises. 10. The question that naturally arises from the review of balance of payments financing over the past 15 years is the disposition of the external funds borrowed by government. This question is answered by reference to the consolidated accounts of the public sector, which are presented (as percent of GDP) in Table 1.3. Please refer to the two lines at the bottom of the -4 table, which present averages for the subperiods, 1970-77 and 1978-84. Turning first to current revenue, which declined form 28% of GDP in the fitst subperird to 23.2% in the second subperiod. Rost of this decrease occurred- in tax revenues, which dropped from 15.4% to 11.4% of GDP in the second subperiod. In contrast, current expenditures increased from 20.7% of GDP in the first subperiod to 23.2% in the second. On the other hand, capital expenditures decreased from 7.0% in the first subperiod to 4.2% in the second. In simple consequence, the budget went from an overall surplus of 0.3% in the first subperiod to a deficit of 4.1% in the second. Thus, the second subperiod saw significant erosion of the tax revenue base, due in some measure to liberal tax holidays and exemptions granted compared to the previous subperiod, a significant expansion of government consumption and a marked decrease in government investment. In other words, the increase in official external debt in the second subperiod 'was not matched by an increase in government capital formation. Rather, the debt covered a shortfall in domestic resource mobilization. Put another way, external borrowing permitted an increase in government consumption without the need to impose new domestic taxes. 11. The issue of exactly how and why the tax revenue base of the Government of the Dominican Republic (GODR) was eroded, and why nothing effective was done to prevent it, remains to be explored. In order to assess the nature of the erosion, a look at the structure of taxes is necessary. Tables 1.4 and 1.5 present time series data on the tax structure in terms of current pesos and percent of GDP, respectively. With regard to the catego- ries of taxes shown, sales taxes include a value-added tax, enacted in 1983, which yielded DR$ 88 million or 0.9% of GDP in 1984, and specific excises on cigarettes, alcoholic beverages, petroleum products (since 1973), sugar (for 1974-80) and other goods and services. Although initially small, the tax on petroleum products contributed DR$ 173 million or 1.8% of GDP in 1984. Taxes on property consist of taxes on transfers, motor vehicle fees and minimal gift and inheritance taxes. There are no taxes on property values. Income and profit taxes broke down in 1984 into 1.8% of GDP from corporate profits and 0.8% from individual income taxes. 12. Thus, the taxes, which are used by most governments as principal sources of revenue, are either at very low levels or non-existent in the Dominican Republic -- e.g., value-added, individual income, corporate profits and property taxes. Moreover, when the revenues for each type of tax (as a percent of GDP) are compared, all show a decrease in 1984 as compared with 1970, except for sales taxes. However, if petroleum products sales taxes, which were assessed as import tariffs through 1972, and reclassified as domestic sales taxes after the opening of a domestic refinery in 1973, are removed from the sales tax category and added to the import taxes category for comparability, then all tax categories show decreases (as percent of GDP) over 1970-84, The greatest decline in relative revenue generation is for import taxes, and this is due in good measure to the exemptions and exonera- tions granted to import substituting firms in the late 60s and 70s. On the other hand, the general erosion of revenue generation by all types of taxes strongly suggests a political process that is unable to enact taxes to pay for the costs of government. After all, the taxes that most governments principally use to finance expenditures are used only marginally in the Dominican Republic. -5- 13. This conclusion is re-enforced by the response of the GODR to the IMF's demand that substantial new revenues be generaged as a condition for a large credit in 1985. The Government proposed, and the IMF agreed to a 36% tax on "traditional" exports, i.e., sugar, coffee, cocoa, tobacco. This tax, which yielded DR 550 million, or 25% of estimated tax revenues in 1985, has powerful disincentive effects on exports and represents a continuation of the anti-export bias of preceding policies. It has to be viewed as a strictly temporary measure for a country with a significant external debt, and in fact it was reduced to 18% at the start of 1986. Thus, the absence of a political process that can articulate to potential taxpayers the need to finance government on a sustaining basis.remains a serious problem. Perhaps the liberal use of exonerations has created a climate-of opinion which holds that only fools pay taxes. 14,. The existing structure of taxes has one other serious defect. It is excessively dependent on trade related taxes. Even before the export surcharge on traditional exports was enacted, about 45% of tax revenue was from duties on traded goods, to say nothing of profits and sales taxes whose revenue yield varies with activity in traded goods. This structure makes government revenues extremely vulnerable to balance of payments shocks. In practice when confronted with revenue shortfalls from such shocks, GODR has attempted to sustain public expendi¾ures by recourse td external borrowing and domestic credit creation (i.e the inflation tax),. Thus, government deficits have become linked to current account deficits. In order to assess the strength and the magnitude of this linkage, several regression tests were performed; and the results are presented in Table 1.6. These results can be summarized as follows: (i) Variation in the current account balance explains 55% of variation in tax revenues, and the regression coefficient indicates that on average every peso of current account deficit reduces tax revenues by 0.97 peso. (ii) Variation in the current account balance explains 71% of overall government surplus or deficit (OGD) and 76% of the government's total fiscal surplus or deficit (TED), and the regression coefficients indicate that on average every peso of current account deficit increases 0GD by 0.72 pesos and TFD by 0.93 pears. (iii) Variation in current account balance explains 55% of variation in foreign deficit financing (FDF) by the government external debt by 0.31 peso. 15. While the goodness of fit statistics (i.e., F and t statistics) are excellent, the regression results assume (a) continuation of the government policies in effect during the base period; and (b) any other influences on revenues, deficits and foreign financing are purely random. To the extent that systematic factors affecting these variables have been omitted, the results are biased. In spite of these qualifications, the strength of the past correlation between current account deficit and the fiscal variables is striking; and these results constitute a strong argument for diversification of government revenues away from heavily trade dependent sources. 16. To recapitulate the discussion of this section, over the past two decades an import substituting economic policy has been in place in the Dominican Republic with the typical anti-export bias emerging through an overvalued currency and the imposition of.a heavy protective fence of import tariffs, quotas and prohibitions. In more recent years, there has been a -6- policy shift towird government consumption and away from government investment. Equally importantly, the tax revenue base has been seriously eroded over the past 15 years, due in part to liberal exemptions and exoq,erations granted to imoort substituting firms, and government has responded by accumulating external debt and creating money. This mix of policies suffers from a lack of coherence that has been exacerbated by the effects of external shocks on the Dominican economy. While the lack of coherence certainly has political roots, some of its manifestations are clearly the consequences of weak overall economic management. III. Interaction of Macro and Sector Policy A. Inconsistency between Macro and Sector Policy 17. As the discussion in section Ii indicated, for several decades the Dominican Republic has featured a macroeconomic policy that stresses import substituting induotrialization. The primary instruments for implementing this policy have been trade policy, e.g., import tariffs, quotas and prohibi- tions, and exchange rate policy. Both of these instruments have operated to implicitly tax agriculture exports by means of their effect on the rate of which traditional agricultural exporters surrender foreign exchange earn- ings. For the past ten years, this implies lower bound estimates of implicit tax rates of traditional agricultural exports that range from 20 to 80%. Ef- fective implicit taxation rates were perhaps twice the lower bound esti- mates. Thus, the basic thrust of macro policy has been anti-export. 18. Consistent with this basic policy, the Institute for Price Stabilization (INESPRE) has exnrcised broad control over prices, imports and marketing of basic foods, e.g., rice, sugar, milk, edible oils. In theory, these powers were to,have been used to buffer domestic prices for important foods from international price variation. In practice, they have been used to suppress domestic prices for rice, milk, and sugar below world levels. Thus, another set of disincentives for agricultural production have been added to the basic anti-export bias of macro policy. 19. In contrast to prevailing macro policy, which has acted to suppress the profitability of agriculture and therefore the levels of resource use and output of the sector, prevailing sector policy has aimed at stimulating pro- duction, at least for selected basic foods and non-traditional agricultural exports. Thus, a basic inconsistency exists between macro and sector level policy. 20. The extent to which sector level policy is directed toward stimula- ting output is not immediately obvious without a review of the actions of the public enterprises serving the sector. Thus, the Agricultural Bank makes subsidized credit available selectively to farmers growing cartain crops -- e.g., rice; the Water Resources Institute (INDRHI) delivers subsidized water to farmers with irrigated lands; and the Agrarian Reform Institute (IAD) pro- vides free land and subsidizeu services -- e.g., inputs, machinery services -- to farmers working agrarian reform lands. The Ministry of Agriculture -7- (SEA) provides research and extension services, subsidized machinery services (PROSEMA) and subsidized agrochemicals (CVMA). However, the SEA units providing these services are underfunded and the levels of subsidized services have diminished under fiscal pressure. B. Effects of Policy Inconsistency 21. The interaction of inconsistent macro- and sector-level economic policies has been costly for the Dominican Republic. As already noted, macro policy has implicitly taxed traditional agricultural exports at rates that were estimated to be at least 20 to 80% and were probably substantially higher. This was a means of transferring an economic surplus from producers of traditional exports to favored economic sectors by means of a shift of the internal terms of trade. However, by lowering returns to traditional crops, macro policy discouraged production of these commodities for export. Valued at social opportunity costs, the effect of the policy was to shift resources at the margin from high to lower productivity uses. Similarly, suppression of the prices of basic foods below world levels is a disincentive to produ- cers that tends to shift resources at the margin from high to lower producti- vity and consumption uses. 22. In contrast, the thrust of sector level policy has been to enhance incentives for transferring resources to the production of agricultural 'com- modities that have been penalized by macro policy. The instruments for this policy have been cost reducing subsidies of one kind or another. Thus, macro and sector level policies have essentially worked at cross purposes. Depend- ing on which policy has the strongest effect on the profitability of a given commodity, its production may be stimulated or depressed. Since all of the policy instruments mentioned have administrative and delivery costs, the net effect has been a very inefficient and often ineffective set of policies. Moreover, the inconsistency of policies has sent mixed signals to producers that have tended to discourage long-term investments in agriculture. C. Secondary Role of the Ministry of Agriculture 23. The institution most likely to play a lead role in formulating and impiementing agricultural policy, the Ministry of Agriculture (SEA), has been unable to do so. The public enterprises which implement policies which impact on agriculture, e.g., INESPRE, CEA, IAD, Agricultural Bank, INDRHI, operate more or less autonomously, with informal inter-agency cooperation, and coordination of budgeting and planning at the level of the Secretariat of the Presidency. While there is a cross membership among the boards of execu- tive directors for the several public enterprises, SEA and the National Agricultural Council (CNA), these bodies have no executive power and little influence. 24. The prevailing climate of opinion on economic policy treats macro and sectoral policy as separate and distinct functions, and it is this atti- tude that has permitted the emergence of policies at different levels that work at cross purposes. Thus, it is not surprising that the Ministry of Agriculture is not represented on the chief macro policy group, the Monetary Council. Clearly, if sector level policies are to become efficient and -8- effective, high level policy makers must come to recognize that inconsistency between macro and sector policy produces inefficiency and tends to raise false expectations4 It is somewhat like pressing both the accelerator and brake pedals of an automobile -- nothing much happens but the stress on the system can be costly. IV. The Costly Intersection of Policies in Rice Production A. The Involvement of the Dominican Agrarian Reform Institute (IAD) 25. The Institute is an autonomous agency that was established in 1962 with an endowment of 192,000 ha -xpropriated by the state from the holdings of the former dictator, Trujillo. Later, other lands were expropriated and entrusted to IAD to manage -- e.g., in 1972, by law, all rice lands over 500 tareas (31.44 ha) in one holding were expropriated; later, all holdings over 100 tareas, which benefited from government financed irrigation systems, were required by law to turn over to governmdnt a variable fraction of the land, 50% if the land was cultivated prior to irrigation and 80% if was not. The latter law has been resisted by legal manuevers, stalling, refusal to nego- tiate, and political pressure. In the meantime, the status of the lands in question is uncertain, inhibiting investments on these properties. 26. By the end of 1984, IAD possessed 392,000 ha, of which 142,000 were lands distributed to agrarian reform settlers prior to 1960. The remaining 250,000 ha were distributed in the bpast 25 years to 517 settlements, of which about 20% are farmed collectively, and the remainder by farmers working indi- vidual plots. The Organization of American States estimated in 1984 that 150 settlements with a total of 50,000 ha were practically abandoned, including many in irrigated areas. It is estimated by a Dominican consultant, AGROKORTI, that only 104,000 of the 250,000 ha are consistently cultivated, of which 65% are on irrigated lands. Crops grown on the consistently cultivated lands are rice, 40%; beans, 10%; maize, 8%; minor fruits, 20%; and miscellaneous crops, 22%. Since rice is produced on only 120,000 ha and cul- tivated overwhelmingly by small farmers, it is clear that the 40,000 ha of rice cultivated by land reform farmers is a disproportionate representation of this group among rice farmers. Moreover, since rice lands are overwhel- mingly irrigated, 97%, of which half is from state-owned canals, it is also clear that consistently cultivated agrarian reform lands are overrepresented in the approximately 223,000 ha of presently irrigated land in the Dominican Republic. In fact, knowledgeable Dominicans argue that the state deliberate- ly sited irrigation projects to service areas where IAD possessed significant holdings and the agro-climatic conditions were suitable for rice production, a labor intensive, smallholder crop. In this way, a faltering land reform program could be propped up and production of an import substituting crop could be stimulated. The irrigation investments and introduction of high yielding varieties induced rice production to increase at an annual average rate of 6.5% between 1973 and 1979. At present, the Dominican Republic is self-sufficient in rice in some years, but not consistently so. 9A- 27. Certainly, it is generally conceded that the results of the land reform program have been disappointing. While the reasons for this situation are no doubt multtple, one factor mentioned by many observers is the insist- ence on cooperative organization and the fact that the state (i.e., IAD) has kept title to all lands it distributes to land reform beneficiaries. This policy, it is believed, weakens the farmer's incentive to make long-term investments and diminishes access to commerctal (non-subsidized) credit. In an effort to stimulate better cultivation of the land that it has distribu- ted, IAD offers a variety of services to land reform farmers -- land clear- ing, construction of feeder roads, building small canals, subsidized credit and extension. However, the Government of the Dominican Republic (GODR) has been unwilling to fund these activities at a level that would make them available to all or even most IAD farmers. B. The Role of the Price Stabilization Institute (INESPRE) in Rice Price Suppression 28. INESPRE has an effective monopoly over domestic marketing of rice, sugar and wheat bran, and with the Directorate of Price Control (DGCP) has been able to set ceiling prices for many staple goods at various stages of distribution. It also operates retail outlets, including some designed to sell subsidized basic consumer goods to the poor. INESPRE is the sole legal importer of foodstuffs, which together with ceiling price control has been sufficient to suppress domestic prices for basic foods below border prices. 29. From 1973 to 1983, INESPRE's price for rice averaged 16% below the comparable border price (valued at the parallel market exchange rate). It should be noted that this subsidized price was available to all consumers, and in absolute terms it benefitted the rich more than the poor. (Cf. An- nex 3). The disincentive effect of the depressed rice prices on farmers was quite negative, and in turn this disincentive required a piling on of subsidies in order for the Government to achieve its objective of self-sufficiency. C. The Countervailing Roles of the Agricult-ral Bank and the National Water Resources Institute 30. The Agricultural Bank (BA) and the National Water Resources Institute (INDRHI) have contributed significantly to the piling on of subsi- dies needed to offset the disincentive of supressed rice prices and achieve self-sufficiency in rice. BA supplies subsidized credit to farmers, and loan policy has directed a large and growing proportion of lending to rice. As fiscal pressure has squeezed the funds supplied to BA, the proportion of loans going to finance rice production has increased to the point that half of all BA lending goes for rice produccion. Credit for IAD's land reform farmers is channeled through the cooperatives. 31. The source of of subsidy from INDRHI is classically simple -- heavily subsidized water charges. Currently, INDRHI recovers less than 10% of operating and maintenance costs through water charges. Since a large pro- portion of irrigated lands are in, rice production, much of the water charge subsidy goes directly to stimulate rice production. Thus, both BA and INDRHI have become deliver mechanisms for subsidizing rice production. --d0 - 32. To recapitulate, GODR has offset the price disincentive on rice production caused by INESPRE's consumer consumption subsidy by piling on subsidies delivered via IAD, BA and INDRHI. A more efficient and just policy would be to eliminate the rice price suppression and to target explicit rice consumption subsidies to the very poor, perhaps through INESPRE's retail outlets for the poor. V. Efforts at Diversification Away from Sugar A. The Evolving Crisis in Sugar 33. Sugar has been produced and exported from the Dominican Republic since the 16th century. The land currently in sugar accounts for nearly one third of total crop land, and sugar exports accounted for 37% in value terms of merchandise exports for 1978-84. The sugar subsector has only three producers, who collectively control estate lands of 170,000 ha and purchase sugar cane from outgrowers on another 105,000 ha (Table 1.7). The three organizations comprise the State Sugar Council (CEA) with 12 sugar mills, and two private companies, one with three mills and the other one with one mill. These 16 mills have a daily processing capacity of 67,000 tons. This translates into an annual capacity of 10 million tons of cane, convertible into 1.1 mill-ion tons of raw sugar, of which an average of over 950,000 tons was exported during 1970-82. Most of the sugar exports went to the United States. Tables 1.7 and 1.8 present basic information on the structure of the sugar industry, including production and export time s6ries data. 34. The large volume of sugar exports to the U.S. have also enjoyed preferential access to that market via quotas and prorations that have resul- ted in sugar imports to the U.S. receiving an average price that exceeded the world price by some 70% over 1970-84. Table 1.9 presents calculations which indicate that the average Dominican raw sugar export price (in DRS) when con- verted to U.S. dollars at the official and parallel exchange rates exceeded the world price by 20 and 70% respectively on average over 1970-84. While valuation at the parallel rate may somewhat overstate the social value of the foreign exchange earnings from sugar, this estimate is surely much closer to the actual social value than the overvalued local currency. Note also that the difference implies implict taxation of sugar exports from the overvalued currency of 50%. 35. Several developments in the international sugar market in the 1980s have diminished the prospects for Dominican export earnings from sugar. The first is the development of sugar substitutes, particularly fructose produced from maize. This development reduced the U.S. demand for sugar by 2 million tons in 1982-83. The reduction in U.S. sugar demand has been taken up by a reduction in import quotas; and, in consequence, the total amount, quota and proration, that the Dominican Republic received dropped from a range of 600-900 thousand tons (over 1970-81) to an average of around 400,000 tons in 1982-84. The quota for 1986 was further reduced to 290,000 tans. While fructose derived from maize may be less attractive economically at the very low relative world prices for sugar of the past several years, its cost tends - 11 - to provide a backstop (ceiling) price Eor sugar since virtually unlimited amounts of maize could be diverted to fructose production worldwide. This means that international sugar prices will probably never again see the peaks in real prices that have occurred in the past. A second unfavorable long-run trend is a shift in preferences in developed countries away from sugar consumption as health and weight conscious consumers switch to products using low calorie sugar substitutes. A third such trend is the recent decline in the relative price of petroleum, which has diminished the economic appeal of alcohol derived from sugar cane as a substitute fuel. In consequence, as time passes, much of the sugar cane output used for alcohol production will be transferred to sugar production. A final unfavorable development has been the decision by the U.S. Congress to eliminate the sugar import quotas, and the associated high domestic price for sugar, by 1991. All of these develop- ments have made necessary a reduction in the scale of the Dominican sugar industry and a diversion of resources toward other agricultural commodities. 36. The need to diversify into other crops and retrench in sugar raises the important question of which of the three producers and which sugar-produ- cing areas should cut back. As Table 1.7 indicates, a public enterprise, the State Sugar Council (CEA), has over 60% of processing capacity and processes cane from over 60% of the sugar acreage. For this reason, market processes cannot be counted on to eliminate the high cost capacity. Informed Dominican opinion argues that the most inefficient mills are owned by CEA and that CEA sugar lands are better situated (in agroclimatic terms) for diversion to other crops. However, CEA has access to subsidies through the central government, and the pressure for adjustment has been greater on the two private.sugar producers. B. The Diversification Program of the State Sugar Council (CEA) 37. Nonetheless, bowing to the inevitable, CEA has prepared guidelines for a diversification program. These feature: (i) closing six of the older, smaller sugar mills; (ii) diversifying the lands supplying the six mills to production of pineapples, citrus fruits, African oil palm, vegetables, avocados, mangos, cotton, and feedlot beef, mostly for export. Other alter- natives proposed include (iii) promotion of tourism (in scenic coastal areas with good beaches); and (iv) export processing zones (for export of value added from domestic labor inputs). For the six remaining mills, the plan envisions: (v) production of by-product electric power through thermally efficient combustion of bagasse; (vi) greater production of by-products from sugar cane such as paper (from bagasse), furfural and citric acid; and (vii) using cane to produce ethyl alcohol as a fuel for domestic and export markets. Thus, the diversification alternatives being considered are numerous; and among them are certainly some commercially viable options. 38. However, implementation of this program is proceeding slowly. Notwithstanding CEA's intentions, the Government (CEA's shareholder) has been reluctant, for political reasons, to announce publicly any plans for closing the six antiquated factories. Much of the proposed diversification activi- ties are planned for areas of the affected sugar estates that have not been planted to sugarcane. The conditions for diversification laid down by CEA are joint ventures in which the other partner provides the capital, technolo- gy and entrepreneurial risk-taking and CEA provides the land and improvements - 12 - and has guaranteed minimum profits. Title to the land would remain with CEA, but it would be assigned to thh joint venture by means of long term lease. These terms have not proved very attractive to potential partnets, and CEA's negotiations with them have proven long and tedious. 39. At first glance, it seems strange that the GODR would be content to entrust the important structural adjustment task of agricultural diversifica- tion to an organization with little experience or special skills for the job. After all, the future export earnings capacity of the Dominican economy depends significantly on the success of this adjustment. However, closer examination reveals the existence of several significant groups who have benefitted from the status quo and are concerned that the adjustment not damage their prospects, or at least does so with minimal damage. For exam- ple, the outgrowers selling sugar cane to the sugar producers are organized politically and in effect engage in bilateral negotiations over cane prices that have allowed them to capture some of the rents from sugar production. As Table 1.10 indicates, CEA had operating losses in processing of outgrower cane in five of the last seven years. In part, these losses are due also to the legal formula for paying outgrowers for their delivered cane, the formula is based on an assumption that outgrowers' cane has a 13% sucrose content, whereas the actual figure Ls probably closer to 8%. Similarly, livestock producers were able to pressure government into setting prices for molasses 'sold for cattle feed well below border prices (i.e., 11% of the world price in mid 1985). Then, there are the 100,000 employees of CEA itself, of which some 87,000 are employed in the sugar mills and administration. The manual labor of cane cutting is largely done by migratory laborers from Haiti. The number of sugar mill workers significantly exceeds commercial norms for CEA's level of capacity, and this suggests that one management objective has been to maximize employment, perhaps as a social welfare measure. 40. Thus the existence of organized groups that have captured some of the rents from sugar production has impeded effective political action on diversification. However, the rents are disappearing; CEA has accumulated large debts to its outgrowers; and Austained delays in initiating necessary structural adjustments will only make it more painful when external events force some kind of action. VI. An Efficient Role for the Ministry of Agriculture A. The Present Situation 41. The Ministry of Agriculture (SEA) is nominally responsible for implementing and coordinating the agricultural policies set out by the Presidency and the National Agricultural Council. Yet its administrative control over the decentralized agencies it is supposed to review and whose funding is formally through SEA, is no more than theoretical. Important public enterprises such as the State Sugar Council fCEA) and the Price Stabilization Institute (INESPRE), both of which have a significant impact on agriculture are outside even its nominal review. The activities that SEA directly controls are agricultural research, extension, training, planning, data collection and analysis. It should be noted that these are primarily technical, skill-intensive activities. - 13 - 42, To understand the role of SEA better, it is instructive to examine its pattern of expenditures, which is given in Table 1.11 for 1978-84. Note first the relatively small part of the total budget assigned for operations. The largest expenditures are for current financial and capital support to other agencies, which is overwhelmingly for public sector agencies serving agriculture, e.g., INORHI, IAD. In a real sense, SEA is simply the conduit through which such funds pass to other agencies. To put the matter in pers- pective, it is useful to compare SEA's expenditures with those for semi- autonomous public enterprises with functions related to agriculture. Expend- itures data for these enterprises are presented in Table 1.12. When current and capital expenditures by these enterprises are compared with current oper- ational and capital expenditures for machinery and construction of SEA, it is clear that these agencies individually in general have greater capital expen- ditures and almost as large (1AD, INDRRI, BA) or larger (INESPRE, CEA) cur- rent expenditures. Collectively, they control the disposition of a much greater amount of resources than SEA. Clearly, SEA cannot be held account- able for the state of Dominican agriculture if most of the public sector resource expenditures related to agriculture are controlled by others. This situation implies that clear accountability for agriculture must effectively rest only at the level of the Presidency. It is desirable that clear accoun- tability should exist at a lower level since the Presidency must cope with integrating the many important affairs of state and should not be required to delve into the intricacies of the agricultural sector. 43. There is evidence that the present institutional arrangement for decision-making with respect to agriculture is not serving the Dominican Republic well. Consider the time series data (at constant prices) for the contribution to GDP from primary production, 1970-84, given in Table 1.13. Taking the contribution from crops, which provide 70% of export earnings and most of domestic food supply, it is clear that value added in crop production actually declined over the 1974-84 period from DR$ 888 to DR$ 849 million, or 4.4% in real terms. Thus, over 10 years and after the expenditure of quite large sums by government to stimulate agricultural production, the crop pro- duction value actually declined. In spite of a massive expansion of SEA: employment from 7,600 to 17,000 people over 1979-84, there is no evidence of a tangible impact on agricultural production. However, this judgment is per- haps too harsh. As we have seen, macroeconomic policy over this period was providing powerful disincentives for agricultural production. In a real sense, SEA and the public enterprises serving agriculture (and not disserving as in the case of INESPRE) were only trying to offset the very negative effects of macroeconomic policy. In the absence of such efforts, it can be argued that the situation would have been much worse. 44. Moreover, as Table 1.11 indicates, operational expenditures (which include the compensation of employees) only increased nominally from 32 to 51 million pesos, or 60% over 1979-84. Since the Consumer Price Index increased by 80% over the same period, real operational expenditures actually decreased while employment was increasing by 120%. Very clearly, the real wages of the employees of SEA fell dramatically on average over that period. This pattern of sharp decreases in real wages was quite general in the Dominican govern- ment as the recent World Bank Economic Memorandum for the Dominican Republic has documented, as employment policies promoted an explosive growth in public - 14 - sector employment. The resultant demoralization of poorly paid public employees is not conducive to efficient, responsible government. In particu- lar, it works iagainst effectiveness in skill-intensive public services such as SEA is mandated to provide to agriculture. B. How to Achieve More Efficiency in Public Agencies Serving Agriculture? 45. Quite clearly, the major issue at hand is posed by the following question: Does it make sense on the one hand to expend large amounts at the sectoral level to stimulate agricultural development if on the other hand macro policy is repressing the sector? Or to express the matter metaphori- cally, is it rational to press the accelerator and brake pedals simultaneous- ly? The answer is obviously no to both questions. Macro and sector policy should be consistent if effective and efficient policies are desired. The build-up of public services to agriculture over the past decade has been ineffectual in actually increasing production because it has run counter to a strongly repressive macro policy. Now that macro policy is less repressive, stimulative sector-level policies should be more effective. 46. Even if policy consistency is obtained, the.rapid growth of SEA employment combined with rapidly falling real wages would not have been conducive to effective service to agriculture. The existence of low real salaries and a consequently demoralized staff tends to drive away the most productive people. While the size of the staff may make it difficult to raise wages for all, short of a massive reduction in the.scale of the organization, a policy of rewarding outstanding performance and punishing non-performance would gradually raise the average productivity of staff. 47. There is much evidence from many countries that the private sector is more efficient at providing rental services of machinery and at distribu- ting seeds and agrochemicals. There is no real evidence that the Dominican Republic is an exception to this rule, and dropping such services from SEA's tasks should improve efficiency without harming agriculture at all. 48. With respect to the public enterprises related to agriculture, the activities and organizations for INDRHI and BA are analyzed in Annexes 6 and 4 respectively, while IAD and CEA have already been discussed in this paper. However, the general principle stated above remains valid for these enterprises as well: If the policies they are implementing run counter to prevailing macro policy, they are unlikely to be effective. VII. Resource Mobilization In and Out of Agriculture A. The Still Distorted Present Situation 49. In Section II of this paper, the state of public resource mobiliza- tion was reviewed and two general observations emerged: (i) a consistent ero- sion over time of the tax revenue base; and (ii) an unhealthy dependence on trade-related taxes. The trend toward erosion of the tax base has induced successive governments reluctant to curtil expenditures to accumulate - 15 - external debt and expand domestic credit (i.e., print money) to finance domestic consumption. These fiscal responses in conjunction with serious balance of payments shocks led to the creditworthiness crisis of the past two years. The dependence on trade-related taxes assured that balance of payments shocks were instantly transmitted to the government's budget, with the unfortunate consequences already noted. Quite clearly, a tax base that relied more heavily on taxation of nontradeables would have helped buffer government finances from balanr of payments crises. As a small country whose export earnings are overwhelmingly dependent on exports of primary commodities that have historically shown considerable price fluctuation, the Dominican Republic can expect to encounter balance of payments shocks with some regularity. For this reason, it makes sense to construct a tax base that is not strongly coupled to tradeables. 50. . In Section II, it was also noted that existing taxes in the Dominican Republic include levies on personal incomes, corporate profits and value added, but that these taxes are levied at very low rates on restricted revenue bases. There are no taxes on real property values. All of the just mentioned tax sources are significant for more developed and neidly industria- lized countries. Moreover, except for real property taxes, the infrastruc- ture is already in place to expand the relative contribution of these broadly based taxes in the Dominican Republic. 51. Let us now consider the share that agriculture has contributed to tax revenues and whether the level or the types of taxes are appropriate. This is a matter requiring some economic understanding; because, as was noted in Section II, much of the taxation on agriculture in the past has consisted of implicit taxation via a structure of tariffs, quotas and prohibitions on imports as well.as an overvalued currency. As a first approximation, we calculate an estimate of the revenue due to the overvalued currency rate from traditional agricultural exports. To do this, the dollar volumes of these exports are valued in Dominican pesos using (1) the official rate; and (2) the parallel market rate. The difference between these valuations provides an estimate of the implicit tax revenue. Details of the calculation for the years 1970-84 are given in Table 1.14. 52. While it kan be objected that the parallel rate is not necessarily the equilibrium foreign exchange rate that would have prevailed in the absence of the dual market with an overvalued official rate, the movement of the freely floating rate since unification of the foreign exchange market in January 1985 lends support to the judgment that the parallel rate is a reaso- nable estimator, if perhaps somewhat biased on the upward side. Moreover, the true equilibrium rate for an implicit tax incidence calculation would be one prevailing in the absence of both the overvalued official rate and the protective 'fence of tariffs, quotas, etc. around industrial imports. It is quite likely that the parallel market rate is an underestimate of this conceptually correct equilibrium rate. The issue cannot be resolved without doing an extensive analytical exercise in which equilibrium rates are calcu- lated for the conceptually correct situation. The position taken here is that the parallel rate is with high probability an underestimate of the correct equilibrium rate, and thus our calculations provide lower bound esti- mates for implicit taxation of traditional agricultural exports. - 16 - 53. To derive estimates of total taxation on traditional agricultural exports, it is necessary to compute the level of explicit taxes and add this total to the total for implicit taxes. This calculation is done in Table 1.15. Note that in the five years before unification, the proportion of implicit taxes ranged from 69 to 98% of the total taxation. 54. If the reader is wondering how the government collected the revenue from the implicit taxation, he need only recall that foreign exchange earnings from traditional agricultural exports had to be surrendered at the overvalued currency. Government then sold or transferred these foreign exchange earnings at the nominal official rate to preferred importers such as INESPRE. Such transactions subsidized preferred importers by the difference between the official rate and the equilibrium rate (hereia estimated by the parallel rate). 55. To complete the estimation of the share of total tax revenues con- tributed by traditional agricultural exports, total explicit tax revenues must be augmented by implicit taxes on traditional agricultural exports; and total taxes on these exports computed as a percent of the augmented total tax revenues. This calculation is presented for the 1980-84 period in Table 1.6, The calculated contribution of traditional agricultural exports ranges from 21 to 38%, averaging 28%. 56. The implicit taxation of traditional agricultural exports by an overvalued foreign exchange rate ended in January 1985. However, this form of taxation was replaced by an explicit export tax of 361, which generated estimated revenues of 548 million pesos, or 25.9% of estimated total tax revenues of 2,117 million pesos in 1985. Thus, only the form of the tax and not its relative burden was changed. B. Toward a More Neutral -nd Stable Tax Revenue System 57. A prescription for improved resource mobilization in the Dominican Republic should incorporate a change to a tax code which eschews the exemp- tions and exonerations that have eroded the revenue generating capacity of the old tax rate structure; and it should embody a shift away from taxing tradeables toward taxing nontradeables in order to buffer government revenues from balance of payments shocks. It should also move toward horizontal equity in the sense that no sector pays a disproportionate share of the tax burden and should embody fairness in the sense that the poor have a lighter burden that the rich. These prescriptions clearly indicate a shift toward greater reliance on real property, personal incomes, corporate profits and value added as the bases for the improved tax structure. 58. Turning to a consideration of the proper share for agriculture in such an improved tax code, note first that the horizontal equity and fairness principles imply that agriculture should contribute not more than its share in GDP, which is itself a permanent income proxy for share of national wealth. Since agriculture now contributes about 17% of GDP, this should be its share of the revenue base. Given that traditional agricultural exports account for about one half of agriculture's share of GDP, the share of these exports in the revenue base ought to be about 8.5%, which is significantly less than the 28% that has'been paid on average in recent years. - 17 - 59. However, the existing forms of taxation of agriculture are ineffi- cient since both export taxes and suppressed domestic prices are strong dis- incentives to produce. If instead of these taxes a tax on the value of agri- cultural land were levied, the distortion of resource allocation would be removed since a tax on a factor in fixed supply is neutral in its effect on resource allocation. In addition, such a tax would provide an incentive to use agricultural land that is lying idle. Since idle land has a social cost in the form of lost output, a land tax would impose a private cost on idle land and eliminate a perverse private incentive. Such a tax would in effect be a form of agricultural income tax since land values are determined by the economic rents they can earn, and these in turn are related to agricultural production and incomes. Indeed, if land values are monitored and assessed values are kept up to date, then a land tax can serve as vehicle for cost recovery of irrigation investments since the increased rents from improved water supply due to irrigation would be capitalized into land values. Of course, such a method of cost recovery does not provide an incentive for efficient incremental use of water, but then a system of water charges does not do this either unless it is based on volumetric use. At present, meter- ing of water use is not practical in the Dominican Republic. VIII. Conclusions and Recommendations 60. A recurring theme -throughout this paper has been the importance of consistency and coherence between macro and sector policy. By and large, agricultural sector policy has achieved very little in the past because the conditioning macro policy has been repressive, rather than stimulative as agricultural policy has attempted to be. There is little chance that agri- cultural sector policy will be effective if it is not consistent with the basic thrust of macro policy. By the same token, a macro policy that fails to assess realistically the country's comparative advantage will not serve well the interests of the people. Therefore, the following recommendations are humbly offered as suggestions with respect to future policies at the macro and sector levels: 1. To assure policy consistency and avoid placing all of the burden on the Presidency, a minister directly responsible for integrating agricultural policy with macro policy and for supervising its implementation should be designated. This official might logically be the Minister of Agriculture, but conceivably could be in some other position in government. He would review the activities of semi-autonomous public enterprises affecting agriculture, and he would have a veto over actions by these enterprises affecting agriculture in ways that he judged to violate basic agricultural policy guidelines. In order to monitor and ensure consistency, he should be a member of the Monetary Council and any other group formulating macro policy. 2. To remove a recurrent source of macroeconomic instability as well as eliminate sources of disincentives to agricultural production, the structure of the tax code should be modified so as to: - 18 - (i) significantly decrease the dependence of public revenues on taxation of tradeable goods; (ii) shift the revenue base toward taxation of noatradeable goods and services; and (iii) reverse the alarming deterioration of the tax revenue base. Several broadly based taxes, widely used as significant revenue sources in many other countries, are available to accomplish these objectives: personal income, corporate profits, value added and real property taxes. 3. With respect to the contribution of agriculture to public resource mobilization, the sector should not exceed its share in GDP, cur- rently about 17%. In particular, traditional agricultural exports should not be made to bear a greater burden than their share in GDP, currently about 8.5%. 4.- The form of the taxation of agriculture should be shifted to taxes that do not provide disincentives to agricultural production or otherwise distort the sectoral resource allocation. Accordingly, it is recommended that a tax on the value of agricultural land should become the basic method of revenue generation from agricul- ture. A tax on a factor like land that is fixed in supply does not distort the resource allocation, and in addition shifts the revenue source to a nontradeable. 5. The effort to diversify away from dependence*on sugar exports should be a top national priority. Since the basic concern is as much with agro-processing as with agricultural production, the sub- ject is intersectoral and should be addressed at the highest le- vel. Letting CEA have responsibility indicates a lack of the pol- itical will to tackle the issues, which extend beyond CEA's reach, and is simply a form of inaction. Therefore, we recommend that a Presidential Commission on Economic Diversification be formed to lay out a program of action and subsequently to supervise the implementation of the program by an independent agency set up for that purpose. The minister responsible for integrating macro and agricultural policy (see Recommendation 1) should be a member of the Commission, as should the Minister of Commerce and Industry. 6. The failure of the IAD to achieve a land distribution to small far- mers on terms that permit most of them to become self-sufficient and productive is appallingly clear. The cost has been excessive and the end is nowhere in sight. Perhaps the chief flaw of IAD is that it has been unable or unwilling to cut the umbilical cord and let the beneficiaries manage on their own. Accordingly, it is recommended that legislation permitting a new approach be enacted. In particular, it should made legally possible that beneficiaries be given title individually to the land in exchange for signing a note which commits them to amortize a fair purchase price over a number of years in the form of annual payments. If a farmer misses - 19 - a payment, he is in default and loses title to the land. All of the uncommitted IAD lands should be distributed in this way as rapidly as possible. 7. Recommendations with respect to BA and tNDRHI, which have also been heavily involved in the costly promotion of rice production, are . presented in Annexes 4 and 6 respectively. Recommendations for the remaining public enterprise involved in that situation, INESPRE, are presented in Annex 3. 8. Consistent with the recommendations for the public enterprises con- cerned with agriculture, the role for SEA needs to be redefined. If as seems likely, the minister with overall responsibility for agriculture is also the head of SEA, the planning, coordinating and integrating functions of SEA will need to be strengthened by hiring of highly qualified professionals. In addition, the "classical" functions of agricltural research, extension and data collection will need to be re-invigorated by a reform that clearly assigns tasks and responsibilities, and is accompanied by a system of incentives (i.e., rewards and penalities) set up to enforce the assignment of tasks and job responsibilities. Many of the clas- sical functions can and should be decentralized away from Santo Domingo. Hiscellaneous functions, such as providing machinery services, distributing seeds and agrochemicals, which are more efficiently performed by the private sector, should be discontinued oy SEA. The problem of excessive employment and low wages for SEA employees needs to be addressed on a long term basts. Strict con- trols on further hiring should be enfor4ed, and staff size allowed to decrease by attrition. Redundant people should be encouraged to seek better prospects elsewhere. - 20 - Table 1.1 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Rates in Dual Foreign Exchange Market, 1970-84 Parallel Estimated Official Market Proportion of Trade Weighted Exchange Exchange Trade Through Average Rate Rate Parallel Exchange Year (DRS/US$) (DR$/US$) Market Rate (DR$/US$) 1970 1.00 1.15 .08 1.012 1971 1.00 1.14 .08 1.012 1972 1.00 1.12 .07 - 1.008 1973 '1.00 1.13 .09 1.012 1974 1.00 1.14 .07 1.010 1975 1.00 1.18 .08 1.014 1976 1.100 1.20 .10 1.d19 1977 1.00 -1.22 .13 1.028 1978 1.00 1.25 .15 . 1.039 1979 1.00 1.23 .15 1.034 1980 1.00 1.26 .16 1.043 1981 1.00 1.28 .26 1,073 1982 1.00 1.46 .42 1.191 1983 1.00 1.62 .52 1.322 1984 1.00 2.79 .70 2.253 Source: Central Bank, Monthly Bulletin, January 1985. - 21 - Table 1.2 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Financing of Current Account Balance, 1970-84 (in millions of US$) Curreat Errors and Account Private Official Reserve Unidentified Year Balance Capital Capital Adjustment Financing 1970 -126.0 109.0 29.5 -6.0 -6.5 1971 -124.1 106.7 25.1 -2.3 -5.4 1972 -59.7 72.7 45.7 -14.2 -44.5 1973. -80.4 53.3 25.2 -8.5 10.4 1974 -331.8 134.0 77.2 37.5 83.1 1975 -82.5 78.6 45.7 -45.9 4.1 1976 -293.6 132.5 116.4 0 44.7 1977 -264.5 202.2 133.9- -71.6 -- 1978 -311.9 43.6 157.4 110.9 -- 1979 -331.3 13.4 195.4 122.5 -- 1980 -669.8 155.5 365.5 148.8 -- 1981 -405.9 88.6 174.1 143.2 -- 1982 -441.9 -65.7 212.1 295.6 -- 1983 -421.1 79.7 -15.9 357.3 -- 1984* -218.3 57.6 30.4 -130.3 260.6 * = Preliminary. Source: World Bank Report No. 5965-DO, `Dominican Republic - An Agenda for Reform," February 26, 1986 (green cover), pages 195-7. - 22 - 'able 1.3 DOKINICAN REPUBLIC 'AGRICULTURAL SECTOR STUDY Consolidated Accounts of Public Sector (percent of GDP) Overall Current Tax Current Capital Surplus or Year Revew- Revenue Expenditure Expenditure Deficit 1970 24.9 15.8 19.0 6.1 -0.2 1971 28.2 16.2 20.6 6.8 0.8 1972 26.9 15.4 18.8 6.7 1.4 1973 26.7 14.9 8.1 -0.7 1974 31.6 15.6 Zo.U 7.5 -2.0 1975 32.7 17.1 21.2 8.1 3.2 1976 28.0 14.3 21.4 6.3 0.3 1977 25.2 13.5 19.1 6.6 -0.6 1978 23.2 12.3 20.6 5.1 -2.4 1979 22,6 11.8 22.2 6.0 -5.6 1980 24.8 12.2 25.4 4.6 -5.2 1981 24.7 11.7 26.0 3.2 -4.4 1982 21.1 9.7 23.2 3.2 -5.3 1983 22.4 10.6 23.5 3.3 -4.4 1984* 23.8 11.6 24.6 3.8 -1.6 Average 1970-77 28.0 15.4 20.7 7.0 0.3 1978-84 23.2 11.4 23.2 4.2 -4.1 * Preliminary. Source: World Bank Report No. 5965-DO, "Dominican Republic - An Agenda for Reform," February 26, 1986 (green cover), pg. 238. - 23 - Table 1.4 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Structure of Taxes, Central Government, 1970-84 (DRS million) Sales Tax on Goods Total, Income Property and Import Export Other Tax Year Taxes Taxes Services Tariffs Taxes Taxes Revenue 1970 45.6 8.6 48.2 97.9 9.0 5.0 214.3 1971 53.1 8.4 52.7 111.3 12.7 5.7 243.9 1972 62.4 9.0 57.5 118.6 19.7 7.0 274.2 1973 72.9 10.4 58.6 33.7 30.4 9.4 315.4 1974 99.6 12.2 70.2 165.3 64.8 9.9 422.0 1975 126.9 14.7 94.9 178.9 153.5 10.3 579.2 1976 123.9 16.3 118.5 186.8 67.9 10.6 524.0 1977 108.3 17.6 140.5 205.3 90.3 11.9 577.7 1978 111.0 18.0 144.1 211.1 39.3 12.4 537.0 1979 131.5 18.0 164.5 220.8 54.2 17.5 606.5 1980 182.1 20.5 187.2 226.7 59.2 20.7 696.4 1981 186.2 21.4 234.9 182.8 88.0 21.1 734.4 1982 181.4 24.0 253.3 174.9 10.3 17.4 661.3 1983 199.6 24.6 299.6 233-2- 6.4 19.0 782.4 1984* 262.0 31.6 475.3 276.5 19.4 25.0 1089.8 * = Preliminary. Source: World Bank Report No. 5965-DO, "Dominican Republic - An Agenda for Reform," February 26, 1986 (green cover), pg. 221. h, - 24 - Table 1.5 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Structure of Taxes, Central Government 1970-84 (Percent of GDP) Income Sales Tax and on Goods Total Profit Property and Import Export Other Tax Year Taxes Taxes Services Taxes Taxes Taxes Revenue 1970 3.1 0.6 3.2 6.6 0.6 0.3 14.4 1971 3.2 0.5 3.2 6.7 0.8 0.3 14.6 1972 3.1 0.5 2.9 6.0 1.0 0.4 13.8 1973 3.1 0.4 2.5 5.7 '1.3 0.4 13.5 1974 3.4 0.4 2.4 5.6 2.2 0.3 14.4 1975 3.5 0.4 2.6 5.0 4.3 0.3 16.1 1976 3.1 0.4 31.0 4.7 1.7 0.3 13.3 1977 2.4 0.4 3.1 4.5 2.0 0.3 12.5 1978 2.3 0.4 3.0 4.5 0.8 0.3 11.4 1979 2.4 0.3 3.0 4.0 1.0 0.3 11.0 1980 2.9 0.3 3.0 3.7 1.0 0.3 11.3 1981 2.7 0.3 3.5 2.7 1.3 0.3 10.8 1982 2.4 0.3 3.4 2.3 0.1 0.2 8.9 1983 2.5 0.3 3.7 2.9 0.1 0.2 9.7 1984* 2.6 0.3 4.7 2.7 0.2 0.2 10.8 Difference: 1970-84 -0.5 -0.3 1.5 -3.9 -0.4 -0.1 -3.6 * Preliminary, Source: World Bank Report No. 5965-DO, "Dominican Republic - An Agenda for Reform", February 26, 1986 (green cover). - 25 - Table 1.6 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Linear Regression Tests of Fiscal Response to Balance of Payments Shocks, 1970-84 CAB Dependent Inter- (Independent Variable cept Variable) R2 F Prob. F OGD 103.62 0.722 .710 35.20 .0001 (2.30) (5.93) TFD L10.60 0.935 .760 45.22 .0001 (2.15) (6.72) FDF 16.593 0.310 .553 18.35 .0009 (0.62) (4.28) TXR 287.02 -0.973 ..555 18.49 .0009 (3.43) (4.30) Ke: OGD - Overal Government Surplus or Deficit (-) TVD - Total Fiscal Surplus or Deficit (-) FDF - Foreign Finance of Government Deficit TXR - Tax Revenues, Public Sector CAB - Current Account Balance (in RD$) R2 - Adjusted R-Squared Statistic F - F Statistic t - Statistics are in parentheses below coefficient estimates Data Source: World Bank Report No. 5965-DO, "Dominican Republic - An Agenda for Reform," February 26, 1986 (green cover). - 26 - Table 1.7 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Structure of the Dominican Sugar Industry Average Average Daily Cane Number Cane 1/ Raw Sugar Processing of Sugar Production Input Prod'n Capacity Organi- Out- Lands (K. Ha) 4/ 1979-80 1979-80 (shor) zation growers Estate Outgrowers (K.Mt) (K. Mt.) tons) / State 4,360 101.1 77.8 6,798 727 3 46,500 Sugar Council Central 1,540 43.9 27.4 15,000 Romand 3,130 335 3/ Casa 0 24.7 0 5;850 Vicini Total 5,900 169.7 105.2 9,928 / 1,062 / 67,350 1/ Source: CEA Estado Financieros, and Informe de Cultivos Reports, various years. 2/ Estimated by assuming each outgrower has as much land on average as the outgrowers of CEA. 3/ Estimated by assuming that each organization inputs came with the same sucrose content and achieves equal efficiency in conversion to raw sugar as the national average. 4/ Source: INAZUCAR (980). This source underestimates data from the CRIES Land Cover/Use Inventory (1980) which repoited 420,500 ha planted in sugarcane. Quoted from AGROHORTI S.A., pers. comm. 1986. See also Annex 5, Table 5.4. - 27 - Table 1.8 DORINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Cane and Sugar Prod-action and Export Data, 1970-84 (1) (2) (3) (4) (5) (6) Raw Sugar Apparent Raw Sugarcane Raw Sugar 4/ Purchases 5/ Change in Sugar Production Sugar / Exports INESPR5 Stocks Exports (000's Production (000's (000's (000's, MT) to U.S. / Year MT)1/ (000's MT) MT)T) ) ((2)-(3)-(4)] (000's, MT) 1970 8,654.0 1,014 792.6 1971 8,652.0 1,098 .1,010.9 656.0 1972 9,987.4 1,139 1,141.3 737.0 1973 10,089.0 1,143 1,069.5 741.0 1974 10,128.0 1,194 1,054.9 876.0 1975 9,334.2 1,136 975.3 160.4 0.3 700.1 1976 11,277.0 1,2;t0 998.4 164.2 107.4 900.4 1977 11,073.3 1,222 1,116.6 169.0 -63.6 921.2 1978 11,093.2 1,164 936.7 177.1 50.2 601.6 1979 11,093.2 1,164 992.4 183.6 -12.0 317.7 1980 8,763.6 961 802.0 205.6 -46.6 525,1 1981 9,321.5 1,005 847.5 195.0 -37.5 711.2 1982 10,191.5 1,154 833.3 205.5 115.2 329,0 1983 -- -- 984.9 207.9 415.0 1984 -- -- -- 247.9 Average: 1970-82 9,973.7 1,128 967.0 1/ Source: INAZUCAR, 1970-82. 2/ Source: INAZUCAR, 1970-82. 4/ Source: National Statistics Office, External Trade, Various Years. 5/ Source: INESPRE Sugar Accounts, Various Years, 6/ Sources: U.S. Department of Agriculture, Agricultural Statistics, various years. International Sugar Journal, various years. DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Sugar Export Prices, 1970-84 (1) (2) (3) (4) (5) (6) (7) Ratio DR Ratio, US Ratio DR Raw Sugar U.S. Import World Export Quota Price DR Sugar Export Price Export Quota Market Price to to World Export 3/ Price to Price Price Price World Price _/ Price Price World Price Year (DR$JMT) 1/ (US$/MT) 2 US$/T) 2/ ((1)/(3)) [(2)/(3)] (US$/MT) ((6)/(3)) 1970 135.5 177.9 82.7 1.64 2.15 155,8 1.88 1971 132.8 187.8 99.6 1.33 1.89 151.4 1.52 1972 145.4 200.4 163.8 0.89 1.22 162.4 0.99 1973 180.4 226.9 211.9 0.85 1.07 203.9 0.96 1974 319.1 650.4 661.2 0.48 0.98 363.8 0.55 1975 520.2 495.4 451.7 1.31 1.10 696,4 1.54 1976 261.8 293.4 255.3 1.03 1.15 314.2 1.23 1977 199.0 242.3 181.0 1.10 1.34 242.8 1.34 00 1978 190.2 307.1 172.2 1.10 1.78 237.8 1.38 1979 192.4 343.0 213.0 0.90 1.61 236.7 1.11 1980 316.8 663.8 639.8 0.50 . 1.04 399.2 0.62 1981 605.6 435.0 373.2 1.63 1.17 775.2 2.08 1982 318.6 439.2 185.6 1.72 2.37 465.2 2.51 1983 287.2 485.9 187.2 1.53 2.60 465.3 2.49 1984 329.4 479.3 114.2 1.95 4/ 4.20 621,0 4/ 5.44 Average: 1970-84 1.197 1.711 1.709 1/ Monthly Bulletin, Central Bank, Various Years. 2/ USDA/ERS, Sugar and Sweetener Situation Reports, Various Years. 3/ Computed using parallel market exchange rates from Table 1.1. 4/ Adjusted to reflect 1.48 DR$/US$ rate applied to traditional agricultural exports In 1984. 5/ Export price converted to dollars at official rate. - 29 - Table 1.10 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY CEA Unit Costs and Profits Per Export Sale from Outgrower Sugarcane (U.S. cents/1b) (1) (2) Export Unit Avg. Operating Avg. Oper. Profit Cost, Using Raw Sugar or Loss Year Outgrower Cane Export Price [(2)-(1)] 1978 10.8 8.6 -2.2 1979 10.8 8.7 -2.1 1980 15.1 16.4 1.3 1981 20.1 27.5 7.4 1982 16.9 14.5 -2.4 L983 16.7 12.8 -3.9 1984 19.4 14.9 -4.5 Source: CEA and World Bank (1985), p. 18. - 30 - Table 1.11 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Dominican Republic: E%,penditures by Ministry of Agriculture, 1978-84 (RD$ millions) Item 1978 1979 1980 1981 1982 1983 1984 Current Expenditures: 19.42 57.01 66.26 60.41 57.29 68.03 87.07 Operations 8.28 31.91 36.23 32.41 29.46 35.29 50.91 Current Financial Support Other Agencies 9.84 24.97 30.01 28.27 27.82 32.74 36.17 Miscellaneous 1.29 0.11 0 0 0 0 0 Capital Expenditures 2.61 39.84 87.78 112.86 64.40 58.64 97.52 Machinery and Equipment 0.79 4.62 4.09 1.31 0.78 1.48 3,,54 Construction 0.23 5.02 1.69 0.01 0.15 0.08 Q.87 Agricultural Development 0 0 15.41 11.55 7.34 10.00 14.85 Capital Support to Other Agencies 0 26.61 66.51 99.98 56.13 44.23 73.76 Miscellaneous 2.31 3M58 0.07 0.01 0 2.84 4.50 Total 22.03 96.85 154.04 173.27 121.69 126.68 184.59 Source: National Budget Office (ONAPRES): Budget Execution, 1980-84. - 31 - Table 1.12 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Public Enterprises Related to Agriculture, Current and Capital Expenditures, 1979-84 (In Millions of Current DR$) Year BAGRICOLA I.A.D. INDRHI INESPRE C.E.A. Current 1979 15.9 16.1 8.5 243.2 167.9 1980 21.0 15.6 12.1 310.3 242.9 1981 20.1 20.8 12.2 366.7 307.3 1982 20.1 17.2 9.6 309.7 262.2 1983 20.7 18.8 10.4 341.4 266.2 1984 23.3 24.9 14.8 478.3 299.1 Capital 1979 153.2 3.4 12.6 119.9 79.6 1980 167.8 6.8 14.1 63.3 180.2 1981 147.1 14.1 34.9 57.6 372.5 1982 136.7 6.0 20.6 120.6 125.5 1983 167.3 6.4 22.7 94.6 91.4 1984 155.0 10.9 30.5 30.6 80.3 Total 1979 169.1 19.5 21.1 363.1 247.5 1980 188.8 22.4 26.2 373.6 423.1 1981 167.2 34.9 47.1 424.3 679.8 1982 156.8 23.2 30.2 430.3 387.7 1983 188.0 25.2 33.1 436.0 357.6 1984 178.3 35.8 45.3 508.9 379.4 Source: National Budget Office (ONAPRES): Budget Execution, 1k98O-84. - 32 - Table 1.13 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Dominican Republic: GDP from Primary Production, 1970-84 (DR$ Millions at Constant 1980 Prices) Forestry and Year Crops Livestock Fishing Mining Primary 1970 740.9 212.7 21.1 64.1 1,038.7 1971 788.0 221.6 20.8 66.3 1,096.7 1972 807.9 233.4 22.6 178.9 1,242.9 1973 886.6 242.6 28.3 283.0 1,440.5 1974 887.7 244.4 27.4 310.1 1,469.7 1975 822.8 251.7 26.5 343.5 1,444.4 1976 913.1 269.4 26.7 414.1 1,623.3 1977 911.5 288.8 23.8 403.7 1,627.8 1978 935,4 313.1 25.4 323.1 1,597.0 1979 916.6 322.6 39.4 413.5 1,692.2 1980 771.1 282.4 34.2 351.7 1,439.4 1981 808.7 300.8 36.2 376.9 1,522.5 1982 839.2 319.1 37.8 240.0 1,436.2 1983 856.4 334,4 38.1 351.8 1,580.6 1984 848.8 336.0 40.3 381.4 1,606.6 Source: World Bank Report No. 5965-DO, "Dominican Republic - An Agenda for Reform," February 26, 1986 (green cover), pg. 189. DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Calculation of Implicit Taxes on Traditional Agricultural Exports Traditional Exports Valued in DR$ million Implicit Official Parallel Market Tax Traditional Exports (US$ Million) Exchange Exchange Revenue Year Sugar Coffee Cocoa Tobacco Total Rate (A) Rate (B) (B-A) 1970 110.8 28.9 19.6 14.0 173.3 173.3 199.3 26.0 L971 139.1 23.3 13.5 20.0 195.9 195.9 223.3 27.4 1972 167.9 29.8 18.3 28.6 244.6 244.6 274.0 29.4 1973 197.9 46.5 24.2 29.8 298.4 298.4 337.2 38.8 1974 340.2 45.7 48.0 39.0 472.9 472.9 539.1 66.2 1975 577.2 43.2 29.0 34.5 683.9 683.9 807.0 123.1 1976 269.3 100.8 44.9 39.3 454.3 454.3 545.2 90.9 1977 231.6 184.7 93.1 29.0 538.4 538.4 656.8 118.4 1978 211.1 96.9 87.7 46.4 442.1 442.1 552.6 110.5 1979 233.5 157.7 78.4. 55.5 525.1 525.1 645.9 120.8 1980 330.7 76.8 55.8 35.6 498.9 498.9 628.6 129.7 1981 560.4 75.8 50.1 67.3 753.6 753.6 964.6 211.0 1982 308.6 95.6 59.0 24.0 487.2 487.2 711.3 224.1 1983 299.0 76.4 60.9 24.0 460.3 460.3 745.7 285.4 1984 320.3 94,3 76.2 36.0 526.8 779.7 1/ 1469.8 690.1 1/ Rate for traditional exports was RD$ 1.48/US$ in 1984. Source Traditional Exports: World Bank Report No. 5965-DO, "Dominican Republic - An Agenda for Reform," February 26, 1986 (green cover), Official and Parallel Exchange Rates: Table 1.1 this report. W~ (D DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Export and Import Taxes on Traditional Agricultural Exports, 1970-85 (RD$ Million) Explipit Taxes Cocoa Total Total and Implicit Explicit and Percent Year SIgar Coffee Tobacco Total Taxes Implicit Taxes Implicit 1970 0 0 0.50 0.50 26.0 26.5 98.1 1971 12.89 0 0.50 13.39 27.4 40.8 67.2 1972 19.14 0 0.50 19.64 29.4 49.0 60.0 1973 29.79 0 0.60 30.39 38.8 69.20 56.1 1974 63.78 0 1.00 64.78 66.2 131.0 50.5 1975 151.69 0.75 1.70 154.14 123.1 277.2 44.4 1976 50.17 10.98 6.70 67.85 90.9 158w 57.2 1977 35.70 29.70 1.10 66.50 118.4 184.9 64.0 1978 19.20 27.70 1.10 48.00 110.5 158.5 69.7 1979 12.20 40.80 1.20 54.20 120.8 175.0 69.0 1980 40.50 17.60 1.10 59.20 129,7 . 188.9 68.7 1981 83.00 4.20 0.80 88.00 211.0 299.0 70.6 1982 2.15 7.50 0.70 10.35 224.1 234.4 95.6 1983 0 5.90 0.50 6.40 285.4 291.8 97.8 1984 0 19.00 0.40 19.40 690.1 709.5 97.3 1985* -- -- -- 547.80 0 547.8 0 * = Preliminary. Total is an DMF estimate of the revenue from the 36% tax on traditional agricultural exports. Sourcei Explicit Taxes: State Secretariat for Finance, Statistical Bulletin, 1980-85 Implicit Taxes, Table 1.14, this report. Ln - 35 Table 1.16 DOMINICAN REPUBLIC AGRICULTURAL SECTOR STUDY Calculation of Lower Bounds for Recent Agricultural Tax Shares (DRS Killion) Total Total Total Total Implicit Explicit Taxes ExplUcit & Total Taxes Explicit Taxes on Plus Implicit IkAlicit on Trad. Tax Trad. Agr. Taxes on Trad. Taxes on Trad. Agr. Exp. as Year Revenues Exports Agr. Exp. Agr. Exp. % of Total 1980 755.3 129.7 885.0 188.9 21.3 1981 592.2 211.0 1003.2 299.0 29.8 1982 722.7 224.1 946.8 234.4 24.8 1983 854.7 285.4 1140.1 291.8 25.6 1984 1164.5 690.1 1854.6 709.5 38.3 Average, 1980-84 28.0