Colombia - Commercial policy survey

If- - Report No. 7310.CO Colombia Commercial Policy Survey Decembwer 15, 1989 Trade, Finance and Industry Operations Country Department IlIl Latin America and The Caribbean Region FOR OFFICIAL USE ONLY Document of the World Bank This docr ment has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Fiscal Year January 1 to December 31 Currency Zoulvalent Currency Unit: Peso (Col$) Exchange Rate Effective March, 1988 US$1.00 - Col$290 Col$l.O0 - US$.003448 Abbreviations AEB (NPV) Anti-Export Bias (Ignoring Plan Vallejo Incentives) AEB (PV) Anti-Export Bias (Considering Plan Vallejo Incentives) BR Banco de la RepOblica CARBOCOL Carbones de Colombia S.A. CAT Certificado de Abono Tributario CERT Certificado de Reembolso Tributario CKD Completely Knocked-Down (Parts for Assembly Industries) CPI Consumer Price Index DANE Departamento Administrativo Nacional de Estadistica DTF Fixed-term Deposit Rate ECOPETROL Compaftia Colombiana de Petroleos GATT General Agreement of Tariffs and Trade GDP Gross Domestic Product GNFS Goods and Non-factor Services IDEMA Instituto de Mercadeo Agropecuario IMF International Monetary Fund INCOMEX Instituto de Comercio Exterior NABANDINA Nomenclatura Arancelaria de los Paises Miembros del Acuerdo de Cartagena PLAN VALLEJO Duty Drawback Program for Exporters PROEXPO Fondo de Promocion de Exportaciones QR Quantitative Restriction SIEX Sistemas Especiales de Importaci6n - Exportacion  FOR OMFCIL USE ONLY COLOMBIA: COMMERCIAL POLICY SURVEY Table of Contents Pate No. EXECUTIVE SUMMARY ............................................... i I. BACKGROUND ............................................... 1 ... Economic Overview. 1 B. Evolution of Trade Policy ............................ 4 C. Present Adjustment Program ........................... 8 First Phase ...................................... 8 Export Incentives .............................. 9 Import Restrictions ............................ 9 Second Phase ..................................... 10 II. THE EXCHANGE RATE AND MARKET INCENTIVES .................. 11 A. Assessment ........................................... 13 III. IMPORT RESTRICTIONS ...................................... 16 A. Import Licenses ...................................... 16 Effects of Licensing Changes ..................... 22 B. Tariffs .............................................. 24 Surcharges and Exemptions ........................ 29 C. Effective Protection ................................. 31 Continuing Tariff Reform ......................... 33 D. Assessment ........................................... 34 IV. EXPORT PROMOTION ......................................... 36 A. Plan Vallejo ......................................... 36 B. Export Tax Rebates (CERT) ............................ 38 C. PROEXPO Credit ....................................... 41 D. Anti-Export Bias ..................................... 42 E. Assessment ........................................... 45 V. CONCLUSIONS AND RECOMMENDATIONS .......................... 47 A. Exchange Rate Management ............................. 50 B. Import Controls ...................................... 51 C. Export Promotion ..................................... 52 This report is based on a mission to BogotA by Bruce Fitzgerald (Mission Leader) and the findings of Kate Whitmore, April 18-May 5, 1988. Farida Khan provided research assistance. This document has a restrictea distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed wi-hout World Bank authorization.  APPENDICES .................................................. 54 BIBLIOGRAPHY .................................................. 59 MAP IBRD 18370R  COLOMBIA: COMMERCIAL POLICY SURVEY List of Tables Page No. TABLE 1.1: Colombian Output, Absorpt:on and Income, 1980-87 .......................................... 2 TABLE 1.2: Exports of Goods and Services, 1981-87 ............. 3 TABLE 1.3: Principal Agricultural and Livestock Exports, 1980-85 .......................................... 4 TABLE 1.4: Colombian Industrial Exports, 1980-86 .............. 5 TABLE 1.5: Percentage of Industrial Output Exported, 1975-86 .......................................... 6 TABLE 1.6: Export Growth Rates, 1965-85 ....................... 7 TABLE 2.1: Relative Price Indices, 1975-87 .................... 12 TABLE 3.1: Import Licensing Requirements, 1980-88 ............. 16 TABLE 3.2: Annual Import Licenses by Foreign Exchange Regime ....... .................. 17 TABLE 3.3: Import Licenses Classified by Use .................. 19 TABLE 3.4: Imports by Use, 1980-87 ............................ 20 TABLE 3.5: Consumer Imptrts by Type, 1985-6 ................... 21 TABLE 3.6: Licensing Requ:rements by Stage of Processing, 1988 ............................................. 23 TABLE 3.7: Licensing Requirements Coverage of Manufacturing Output ........................................... 24 TABLE 3.8: Customs Taxes, Imports, and Government Finances .... 25 TABLE 3.9: Summary Comparison of Tariffs, 1984-88 ............. 26 TABLE 3.10: Number of Tariff Positions, by Tariff Rate .26 TABLE 3.11: Tariffs by Stage of Production ..................... 28 TABLE 3.12. Average Tariffs in 22 Sectors of National Accounts .28 TABLE 3.13: Licensing Requ.rements and Average Tariffs by Stage of Processing, 1988 .29 TABLE 3.14: Tariff Exemptiors, 1984 .30 TABLE 3.15: Sectoral Estimates of Protection .32 TABLE 3.16: Effective Protection in a Proposed Tariff System .33 TABLE 4.1: Plarn Vallejo Export Registrations, 1980-86 ......... 37 TABLE 4.2: Average Indirect Tax Rates and CAT and CERT Rates .... .. ............. 40 TABLE 4.3: CERTs Payments, 1985-87 ............................ 41 TABLE 4.4: Domestic Credit and PROEXPO Credit, 1981-86 ........ 43 TABLE 4.5: PROEXPO Credit Disbursements, 1981-86 .............. 43 TABLE 4.6: Average Anti-Export Bias by Sector, 1986 ........... 44 List of Figures FIGURE 1: Trade Liberalization Index ......................... 9 FIGURE 2: Distribution of Tariff Rates ....................... 27  i COLOMBIA: COMMERCIAL POLICY SURVEY Executive aSumary i. In 1984 the Colombian Government began an adjustment program to facilitate export-led growth and encourage export diversification. The principal elements of the program have been: maintenance of a competitive exchange rate; reduction of import and export restrictions; reduction of tariff rates and dispersion; and changes in export incentives. During the program exports have grown and the foreign exchange budget for imports has been increased. The purposes of this paper are to: (a) review of the effects of the program; (b) summarize the accomplishments of the programs; and (c' consider further trade reforms. The Exchange Rate and Economic Incentives ii. One of the major accompl'.. 'ents of the adjustment program has been the maintenance cf a competitive exchange rate. In 1984 the Government devalued the peso and used a crawling peg to achieve the real exchange rate which prevailed in 1975. The target was reached by the end of 1985, and by 1987 the index was 15Z above the 1975 level.1 iii. The depreciation of the peso has been substantial, but it has not been fully translated into the domestic terms of trade, and prices of traded goods relative to nontraded goods are below those in 1975. Rather than being 152 above their 1975 levels, the prices of imports and noncoffee exports relative to nontraded goods are lower by 15Z. Only in the case of manufactured exports have relative price incentives improved. iv. There are at least two possible explanations why the continuing depreciation of the exchange rate in 1986 and 1987 may not have been fully reflected in these relative prices: (1) the price indices may not be accurate for these purposes, or (2) the underlying structure of the economy may have changed in a way which attenuated the impact of the nominal exchange rate movement on the relative price of traded goods. Because of the need to coordinate exchange rate policy with commercial policy reforms to provide adequate incentives to the traded goods sector, additional research into the discrepancy between the exchange rate and domestic terms of trade indices is desirable. Import Restrictions v. Licenses. In 1984, less than one percent of tariff positions were importable without a license. In 1988, 38Z were unrestricted. This represents significant progress from the trade regime which prevailed in 1984, but as recently as 1980, 69Z of tariff items were unrestricted. vi. The coverage of domestic manufacturing by restrictions fell from 1002 in 1984 to 82Z today. The largest percentages of freely importable items are in intermediate inputs, raw materials, and capital equipment. On 1/ The exchange rate is defined as number of Colombian pesos per unit of foreign currency.  - ii - the other hand, the most tightly restricted groups are finished goods and food and agricultural products. The items which were liberalized were largely noncompeting inputs for locally manufactured items, and domestic production remains insulated from international competition. vii. Tariffs. Tariff rates and dispersion were reduced sharply. The average rate fell from 612 to 30Z and the standard deviatior was halved between 1985 and 1988. There remain twenty-five tariff rates between 0 and 200Z. The large number of rates is symptomatic of attempts to fine-tune protection by distinguishing, for example, among the rate of .12 which applies to 110 commodities, 22--applied to 112 items, and 5Z--assigned to 254 tariff positions. viii. The average tariff for raw materials and capital equipment is 212, compared to 512 for consumer goods. Tariffs tend to be higher for commodities subject to licensing requirements than for the freely importablc commodities, capturing some of the rents from recipients of import licenses. ix. Tariff receipts are reduced by extensive exemptior.3, including Plan Vallejc, Aneean Group, and ALADI, as well as commodity-specific and importer-specific exemptions. In 1984, when the trade-weighted average tariff was 21Z, actual collections (without surcharges) were estimated at 1OZ. This makes it difficult to know or to control the fiscal or protectionist effects of the tariff system. x. Tariff surcharges were introduced in 1985 at a uniform rate of lOZ and increased tu 18Z in 198/, and are less widely exempted than customs duties. They have become a significant part of trade taxes. In the first quarter of 1987, revenue from surcharges approximately equalled that from the customs duty. xi. The function of tariffs is unclear. Import licenses are the binding 4-onstraint on many imports, raising protection above the levels provided by tariffs a'one. Estimates of effective protection revealed no significant correlation between tariffs and nominal protection calculated on the basis of domestic and international price comparisons, nor between the effective protection implicit in the tariff rates and effective protection as measured by price comparisons. xii. The Government is replacing the present schedule of 25 tariff rates with five rates. The rates would be chosen from ranges according to whether a good is produced or producible within Colombia, and the level of national value added. The proposal would be an improvement on the present system, but would lead to high and variable protection. To the extent that licenses restrict imports of final goods, it would cause input costs to fall while the price of final goods was unchanged, and could increase effective protection. A better sequence would be to replace import licenses with moderate, uniform tariffs and then to reduce the tariffs over time. Export Promotion xiii. The present structure of export incentives has been in effect more than twenty years. The three major elements, codified under Decreto  - iii - Ley 444 in 1967, are the Plan Vallejo system, which provides exporters with access to duty-free imported inputs, the CERT export subsidy, and subsidized PROEXPO credit. xiv. Plan Vallejo. The Government has made several improvements in the procedures and requirements for Plan Vallejo, most significantly, permitting exporters to use Plan Vallejo for imported inputs without regard to whether there is local production. Plan Vallejo enjoys an international reputation Ps a good model for providing exporters with the benefits of free-trade status, and the Government's improvements in the plan have led to increased participation. xv. CERT. Though the CERT was designed to rebate indirect taxes to exporters, there is little connection between actual taxes and the rebate which is calculated as a percentage of f.o.b. value of an export. It appears the CERT has over-compensated exporters and is more analogous to an export subsidy. The use of CERT to promote exports is declining: the range of CERT rates was reduced from 15-30Z in 1984 to 3-12U in 1988. xvi. PROEXPO Credit. PROEXPO offers 33 distinct lines of credit, with interest rates between 4Z (for capital goods) and the DTF (fixed-term deposit) rate, currently 35Z. Terms vary from six months for working capital to ten years for capital goods. PROEXPO has reduced the subsidy in most of its credit lines. PROEXPO's interest rate for most credit was 192 during 1980-83 and 22% in 1984-86 while commercial banks' rates averaged 44Z. Recently, PROEXPO has increased its financing of fixed investment, offering rates of 18Z for terms of up to eight years, while raisix,g the rates for pre- and post-shipment credit to the DTF rate. Exporters still receive more favorable credit terms from PROEXPO than from commercial sources. xvii. For the export promotion measures fully to offset the bias in favor of selling at home, they would have to be more than five times as large as they are: nominal intarest rates on PROEXPO credit would have to be negative and CERT payments would have to average more than 6OZ. Furt..er, while the averages would have to be increased by a factor of five, the Government would have to fine-tune the mix of incentives by sector and by products within sectors. Conclusions and Recommendations xviii. The reform program has not fundamentally altered the inward bias of productive incentives. By limiting the pressure of international competition on domestic industry, protection. has probably contributed to the disappointing productivity performance of Colombian industry. A preliminary assessment of industrial efficiency change in Colombia indicates that total factor productivity growth has been low or negative in many industries, particularly in some which are highly protected by import licenses (e.g., textiles). xix. While the Government's prudent debt and macroeconomic management has brought an impressive export recovery in recent years, it appears to have come primarily from petrole-um and coal and from increased utilization of existing capacity. Maintaining recent growth will require investment in export-oriented sectors. The low productivity and competitiveness of the  - iv - industrial sector suggest that growth has been achieved at a high cost of domestic resources. Continued growth without excessive foreign borrowing or reduced consumption could be encouraged by efficiency gains from a more open trade regime. xX. The Government can move toward a more outward-oriented trade regime by extending the reform program it began in 1984. Further measures should use the exchange rate as the principal instrument of protection and export promotion, replace import licenses with moderate tariffs, and eliminate export subsidies. xxi. Macroeconomic Co-ordination. One clear lesson from trade reforms in aev.ral countries is the linkage between trace reform and macroeconomic policy. The Colombian Government's macroeconomic management has provided a competitive exchange rate to encourage exports and efficient import substitution. Further adjustments would encourage expansion of the traded goods sector. xxii. Import Restrictions. The Government should continue to reduce the coverage of import licenses, retaining them only to protect public health and safety. If the Government desires to offer moderate protection beyond that provided by the exchange rate, it should rely on tariffs to maintain a link between domestic and international prices. To reduce anti-export bias and improve the competitive performance of industry, tariffs should be low and uniform, in the range of 10% to 20Z. Exemptions should be retained only for Plan Vallejo and Andean Pact imports. To provide uniform treatment, the Government should continue its program of improving the customs administration. xxiii. Export Promotion. The earmarking of import taxes to finance PROEXPO should be replaced by appropriations under the national budget. The subsbdy provided through PROEXPO credit should be eliminated, with credit offered only at interest rates compatible with those which prevail for domestic economic activity. In line with these reforms, CERT should be eliminated.  COLOMBIA: COMMERCIAL POLICY SURVEY I. BACKGROUND 1. In 1984 the Colombian Government began an adjustment program with the objectives of facilitating export-led growth and encouraging Px diversification. The principal elements of the program live beer.: devaluation and maintenance of a competitive exchange r tel ; elimination of selected import and export restrictions; reduction o- the average tariff and the dispersion among rates; and changes in export incentives. During the program, exports have grown and the foreign exchange budget for imports has been increased. Exports have been diversified, largely through major investments by the state petroleum and coa: companies, ECOPETROL and CARBOCOL. 2. There is an extensive base of literature and consultants' studies on the Colombian trade system. and reforms. This paper has three purposes: (a) To review the literature; (b) To summarize the accomplishments of the program; and (c) To consider further reforms. A separate Industrial Sector Report assesses the sector's efficiency and competitiveness, and recomn'enl:s policy changes to improve performance. 3. This report contains five sections. Following this introd.ctoTy section, there are sections on each of the three main elements of the program: the management of the exchange rate; i.rcort rsstrictions; ar,n export promotion. A concluding section contains a discussion of poss1l _t directions for ,ntinued adjustment. A. Economic Overview 4. Table 1.1 shows macroeconomic data on real output. absorptior., an,` income for 1980-87. During 1981-85, domestic absorption grew mure rapid.y than domestic output, resulting in deficits in the curLent account whikh averaged 5.2Z of GDP during 1980-84 and peaked at 7.cZ in 1982. Imports were reduced and the deficit was eliminated in 1986 By c. boom iTi ct.2 exports (Table 1.2). The export recovery was maintained in 1987 by large petroleum shipments. Real GDP growth during 1980-87 averaged 3.1i, w th mining and construction above average (19.4Z and 5.8Z respectively) and services (2.92), manufacturing (2.6Z), and agriculture (2.4Z) below the average. Population growth a',eraged 2.0Z, so real per capita output gre,w 1.0Z per year. The shares of mining and construction in GDP each increased, while the shares of services, agriculture, and manufacturilng declined. For manufacturing, which remained at 22Z of GDP, this ontinues a trend going back at least to 1965 in which its share in GDP has been unchanged. This compares to Korea, where the s,.are grew from 18Z to 30Z, Mexico (21Z to 26Z), Brazil (26Z to 28Z), Turkey (16Z to 251), and Jamaiea (171 to 222). (World Bank 1988, pp. 226-7). 1/ The exchange rate is defined as the number of Colombian pesos per ur.- of foreign currency.  - 2- TABLE 1.1: Colombian Output, Absorption and Income, 1980-87. (Billions of Pesos, at 1975 Prices) Avg. Growth 1980 1981 1982 1983 1984 1985 1986 1987E 1980-87 GDP 526 538 543 551 570 588 618 651 Grewth Rate 2.32 0.92 1.6Z 3.4Z 3.1? 5.12 5.42 3.12 Sectoral Output Agriculture 119 123 121 124 126 128 133 140 2.4Z Manufacturing 118 115 113 114 121 125 133 141 2.62 Construction 18 19 20 22 24 26 26 26 5.8? Mining 7 7 7 8 10 14 21 23 '9.4? Serv'ces 261 272 280 283 288 293 304 319 2.92 DOMESTIC ABSORPTION 542 569 584 583 59u 589 608 633 2.2Z Private Cons. 385 396 402 404 415 423 436 452 2.32 Public Cons. 54 56 59 59 61 64 65 68 3.2? Gross Fixed Inv. 88 94 96 97 99 94 101 107 2.8Z Change in Stocks 15 23 27 23 15 9 7 7 CURRENT ACCOUNT Resource Balance 17 32 41 31 20 2 -9 -17 Imports of GNFS 101 106 115 104 100 93 96 101 Exports of GNFS 84 74 73 73 80 92 105 118 Source: Revista del Banco de la Republica. E = estimate. 5. The 1980's were unfavorable for most Latin American countries, and Barandiaran (1988) finds that Colombia fared better than any of the other eleven countries in his study of economic adjustment among Latin America's indebted cotuntries. During 1983-87, Colombia maintained a GDP grcwth rate of 3.0?, which was approached only by Brazil's 2.82. In terms of per capita income, only Colombia and Brazil grew between 1980 and 1987. The other ten countries had lower income at the end of the period than at the beginning. At one extreme, Bolivia's had fallen by nearly 30?, Venezuela's by 18?, and Argentina's by 17?, while at the other, Chile's fell by 2?. 6. Tables 1.2, 1.3, and 1.4 display recent data2 on Colombian 2/ Data are not always consistent. For example, for some purposes, sugar and molasses are considered agricultural exiorts (Table 1.3) and in other studies they are classified as manufactured exports (Table 1.4). The definitions and measurements of 'traditional' or "minor' exports also differ among studies. In general, the differences are small and in this report there is no attempt to reconcile them.  -3- TABLE 1.2: Exports of Goods and Services, 1981-87. (US$, hillions) 1980 1981 1982 1983 1984 1985 1986 1987 GOODS Coffee 2208 1507 1515 1443 1735 1712 2742 1633 Petroleum 100 34 213 434 445 410 619 1341 Coal 10 9 14 17 38 121 201 263 Ferronickel 46 62 55 48 76 Gold 310 239 169 177 245 365 359 380 Minor Exports 1668 1608 1371 1030 1099 1125 1363 1556 Subtotal 4296 3397 3282 3147 3624 3788 5332 5249 SERVICES Financial 494 647 510 280 121 106 153 N/A Nonfinancial 1451 1485 1503 903 983 962 1211 N/A Subtotal 1945 21,32 2013 1183 1104 1068 1364 N/A TOTAL 6241 5529 529: 4330 4728 4856 6696 N/A Source: Revista del Banco de la Republica (March 1987, p. v), (July 1988, Sep. Rev. 729, p. v), other issues. TABLE 1.3: Principal Agricultural and Livestock Exports, 19P0-85. (US$, Millions) 1980 1981 1982 1983 1984 1985 Sugar and Molasses 192 88 59 76 40 40 Cotton 82 93 27 23 48 59 Tobacco 26 19 21 21 20 22 Rice, Beans 19 14 1 2 7 11 Bananas 94 122 151 148 198 156 Shellfish 22 18 23 23 26 27 Beef, Live Animals 27 54 46 31 11 7 Flowers 97 109 111 121 129 132 Other 41 35 23 4 10 8 TOTAL 599 552 462 449 490 464 Source: Revista del Banco de la Republica (January, 1987) p. iv.  -4- exports. Table 1.5 contains data on the percentage of Colombian industrial output which is exported. These tables show: o Since the beginning of the orogram in 1984, minor exports have increased. However, through 1986, they had not returned to the levels achieved in 1980-81 (Table 1.2). This lield both for agricultural (Table 1.3) and manufactured exports (Table 1.4). In constant dollars, applying the percentages implicit in Table 1.2 to the data on real exports in Table 1.1, real minor exports in 1986 were 52 below 1980. O Significant export diversification has been achieved principally in petroleum and coal exports, and this diversification is independent of the adjustment program. Chemicals, ferronickel, bananas, shellfish, and flowers are the other exports in which the data show an expansion. O With 6.32 of output destined for export markets (Table 1.5) in 1986, the non-petroleum industrial sector was less export-oriented than in 1980. The sector exported 7.5Z of its output in its best years, 1975 and 1980. During the 1970's, both wrod and furniture and non-metal minerals had years in which over 20Z of output was exported, compared to 14.62 and 10.22, respectively, in 1986. The only subsector where export orientation near its peak is chemicals and rubber products. 7. Export growth was more robust over 1965-74 than during 1975-85. This is due to the reversal of exports during 1978-83. Data from Garcia Garcia (1987) show that manufacturing exports grew at double-digit rates before falling during 1978-83 (Table 1.6). Using three definitions of the manufacturing sector--depending upon whether it includes all manufacturing activities (Manufacturing I), excludes agricultural-based manufacturing (Manufacturing II), or excludes agricultural and petroleum refining--he found growth rates of 131 to 162 during 1965-74, compared to -112 to -52 during 1978-83. B. Evolution of Trade Policy 8. Garcia Garcia (1988) finds that import licenses have been used more frequently than not since the 1930's to control the quantity and composition of imports. Coffee booms have allowed short-lived import booms followed by sudden reversals in the face of foreign exchange shortages. The government has used large devaluations, multiple exchange rates, prior deposits, and frequent changes in tariffs, licenses, and prohibited lists to ration foreign exchange in times of shortages. Between the early 1950's and the 1984 adjustment program, there were thzee liberalizations and three subsequent reveraals. 9. A 1954 import liberalization was reversed beginning in 1955. In 1957 the Superintendencia Nacional de Importaciones was created to administer import restrictions and Law 1 of 1959 allowed the Government to set new priorities for import licenses. It gave special preferences to the mineral industries and to machinery and equipment covered by foreign capital. Prior deposits were used extensively, and by 1962 all imports except 15 items required licenses. 10. A second liberalization was begun in September, 1965, with the introduction of multiple exchange rates. Imports were classified as preferential or as intermediate. Preferential imports received an exchange  TABLE 1.4: Colombian Industrial Exports, 1980-86. (US$, Millions) 1980 1981 1982 1983 1984 1985 1986 Food, Beverages. Tobacco Sugar and Molasses 192 88 59 76 40 40 48 Coffee Extracts 11 39 18 35 37 46 73 Other Foods 23 25 25 24 33 29 30 Beverages and Tobacco 2 3 4 4 4 4 4 Subtotal 227 154 10' 139 113 119 15" Textiles, Leather Textiles 139 112 90 70 77 77 9i Garments 114 114 1i1 59 39 50 77 Leather 24 29 32 25 23 39 52 Shoes 1. 15 20 8 7 11 20 Subtotal 290 270 272 162 146 176 243 Wood and Wood Products Wood 12 14 17 10 6 13 15 Non-metal Furniture 3 5 4 2 2 4 8 Subtotal 15 18 21 11 8 17 23 Paper and Publishing Paper, Paper Products 32 45 31 23 36 25 32 Publishing 40 47 44 34 36 48 58 Subtotal 72 92 75 57 71 73 90 Chemicals Industrial Chemicals 66 60 64 74 73 92 1u7 Other Chemicals 28 33 31 30 32 35 36 Petroleum, Coal Prds. 7 6 5 3 4 5 10 'hubber Products 6 6 3 3 4 5 6 Plastics 11 12 9 13 11 10 11 Subtotal 117 115 112 123 125 146 170 Non-metal Products Ceramics 8 13 6 3 3 4 5 Glass 17 15 11 8 8 7 7 Cement 35 30 33 19 17 22 25 Others 11 14 13 4 7 4 4 Subtotal 71 72 63 34 35 38 41 Basic Metals Iron and Steel 1 1 3 48 66 56 52 Ferronickel (2) (46) (62) (55) (48) Non-Ferrous Metals 2 1 1 1 1 1 1 Subtotal 4 2 4 49 67 57 53 Machinery and Equipment Metal Products 46 53 56 29 22 23 27 Nonelectric Machinery 39 38 37 19 12 18 20 Electrical Machinery 16 26 28 11 8 16 15 Transportation Eqpt. 27 32 18 11 9 8 26 Scientific Equipment 10 11 8 6 5 6 7 Subtotal 138 160 147 75 56 71 94 Other 14 19 19 7 3 6 8 TOTAL 949 903 818 656 624 702 875 Source: Banco de la Republica (1987a).  - 6 - TABLE 1.5: Percentage of Industrial Output Exported, 1975-86. Average 1975 1976-9 1980 1981 1982 1983 1984 1985 1986 TOTAL 8.62 8.02 8.52 7.12 7.52 6.9Z 7.02 9.22 9.62 Subtotal--Non-Petroleum 7.5 7.1 7.5 6.1 6.3 5.0 4.8 6.0 6.3 Consumer Non-Durables 8.1 6.6 7.2 5.2 5.9 4.9 4.4 5.5 5.1 Food Products 5.8 3.8 5.5 3.5 4.7 4.7 3.9 5.1 4.3 Beverages .0 .0 .0 .1 .1 .1 .1 .1 .2 Tobacco .1 .5 .2 .5 .6 .3 .3 .2 .2 Textiles, Clothing 17.4 14.4 15.5 12.5 12.9 9.0 8.9 10.8 11.0 Durables, Intermediates 10.1 10.6 10.4 9.6 9.8 10.5 11.7 15.3 16.4 Non-petroleum 6.9 8.4 7.9 6.9 6.5 5.4 6.0 7.3 8.2 Wood and Furniture 6.3 15.7 9.6 12.3 13.9 6.3 5.0 10.8 14.6 Paper and Printing 3.6 7.8 6.8 8.6 6.5 5.2 6.7 7.6 8.2 Chemicals and Rubber 6.5 6.8 7.4 5.7 5.6 6.3 6.6 7.7 9.0 Nonmetal Minerals 18.1 20.4 19.2 12.8 11.0 6.4 8.3 9.6 10.2 Basic Metals 5.0 4.7 3.5 3.9 4.3 2.3 2.7 3.3 3.1 Refir.ed Petroleum 32.6 28.1 31.8 30.3 34.6 46.9 52.6 72.4 69.7 Capital Goods 5.3 7.0 6.7 7.5 6.9 3.5 2.1 3.6 4.9 Machinery, Equipment 7.6 10.0 9.4 10.1 10.9 5.2 3.2 5.6 5.8 Trans. Equipment 2.7 3.5 3.6 4.4 2.4 1.6 1.1 1.2 3.8 Other Industries 8.4 14.1 16.7 17.8 13.6 6.9 5.4 6.8 7.2 Source: World Bank Staff calculations based on data in Tables A.1 and A.2 in the Appendix. rate 332 below the intermediate goods. Prior deposits were to be eliminated gradually for the preferential imports. Subsequently, large numbers of noncompeting imports were shifted from the prior license to the free list, and tariffs were increased to soften the demand for imports. 11. This liberalization took place with a fixed exchange rate, amidst a sharp inflation, and with declining coffee prices. A balance of payments crisis forced its reversal by the end of 1966. In 1967 there was virtually no free list, import deposits were increased, and new exchange controls were imposed in Decreto Ley 444--the basic trade legislation which has governed for over twenty years. This law also introduced the framework for  TABLE 1.6: Export Growth Rates, 1965-85. (Based on Exports Measured in Current US$) 1965-74 1975-85 1978-83 1965-85 Non-Coffee Agriculture (2, 3) 8.5Z -0.62 -4.3? 5.32 Manufacturing I (9-11, 13-25; 14.1Z 0.5Z -5.4Z 6.6Z Manufacturing II (16-25) 13.3? 1.1? -5.OZ 6.7Z Manufacturing III (16-19, 21-25) 15.6Z 3.2? -10.6? 8.1? Goods (1-25) 5.8Z 1.9? -1.5Z 4.6Z Goods and Services 5.5Z 3.0Z -1.4Z 4.4? Source: Garcia Garcia (1987) Table 5. Numbers in parentheses refer to economic sectors as defined by DANE and listed in, e. g., Table 3.11. export incentives intended to offset the pervasive anti-export bias: o The Plan Vallejo system which provides exporters with duty-free access to imported inputs. o An export subsidy--ostensibly to compensate for indirect taxes- -originally named CAT (Certificado de Abono Tributario) and, currently, CERT (Certificado de Reembolso Tributario). It provided a uniform 15Z subsidy until 1974 when differentiated rates were introduced. o Subsidized credit for exporters under several specialized programs for working capital, fixed capital, and postshipment finance. 12. Between 1967 and 1974, tariffs were used more and licenses, less, to control foreign trade. A crawling peg was managed to reduce the quantity of foreign exchange demanded. The pace of liberalization quickened after 1975, and, after coffee prices surged in 1977-8, was intensified further in 1979-81. When the coffee boom subsided, aggregate demand was maintained through higher public expenditures, domestic credit creation, and foreign borrowing. The index of the real exchange rate fell from 129 in 1975 to 97 in 1982.3 Balance of payments problems ensued in 1982 when the current account deficit reached 7.4Z of GDP. The liberalization was reversed, and by 1984 only 23 items remained on the unrestricted list. 13. 'Gradualism" is often-cited as a hallmark of Colombian economic policy. For example, in the Letter of Development Policy for the Trade Policy and Export Diversification Loan, the Minister of Finance wrote that "The Colombian development model has followed three fundamental principles during the past 20 years: pragmatism, gradualism, self-discipline. The country has demonstrated repeatedly that it is capable, when necessary and 3/ A fall in the index implies a currency appreciation.  - 8 - appropriate, of undertaking its own macroeconomic policy adjustment., World Bank (1985), p. 43. 14. While this may be true in other aspects of economic policy, Colombian trade policy has been more like a roller coaster than a gradual, long-term improvement. Liberalizations may have been gradual. but their reversals have been abrupt. Figure 1 displays an "index of liberalization" for Colombia for the period 1950-1983, developed in connection with the World Bank's study of the timing and sequencing of trade liberalization policies in nineteen countries (Papageorgiou, Michaely, and Choksi, 1986). The index, a subjective ranking derived by Garcia Garcia (1988) and based on a broad view of all determinants of the openness of an economy, shows that while the three liberalizations were four, eight, and fourteen years from trough to peak, the reversals were four, one, and two years from peak to trough. Further, the net effect of the three liberalizations is nil, and the trading system in 1983 stood virtually where it was thirty-four years earlier. Most of the nineteen countries in this research showed a gradual, upward trend with occasional reversals, but with the overall index of openness higher at the end of the study period than at the beginning. Only four--Peru, the Philippines, Sri Lanka, and Singapore--showed less sustained progress over the periods they were studied, ending with the inde.x at a lower value than at the beginning. In the case of Singapore, it hed begun with a considerably higher level of openness and, even though the openness had been reduced over the period of the study, Singapore would st4.l be classified as an outward-looking economy. C. Present Adjustment Program :. The Government's present program was begun in 1984, and the or o-. _pai objectives were 'to revitalize growth and employment generation OVC.' the medium-term, to strengthen the incentives system towards the Z,.,ti4on of a comprehensive export oriented development strategy and to r-lress the balance of payments deterioration." (World Bank 1985, p. 42). he early stages, the program emphasized macroeconomic adjustments. The nominal exchange rate was managed to maintain the real effective exchange r-.te inr.ex at the level which prevailed in the mid-1970's, and the fiscal ":t,s t was reduced from about 7Z of GDP in 1981-83 to 4.5Z in 1985. 1>. Changes in the trade regime were supported by the Trade Policy and Export Diversification Loan of $300 million in May, 1985, and the Trade and agricultural Policy Loan for $250 million in March, 1986. The Government met its commitments under the loan agreements, and the loans were fully disbursed. First Phase 17. The first phase of the reform program (see Annex VI, World Bank 1985) was designed to promote free trade status for export production. The policy measures were in two major areas, export incentives and import restrictions. '. Althougi Garcia Garcia's study ended with 1983, by many measures, the trough of the reversal of the 1967-81 liberalization did not occur until 1984 (see, e. g., Table 3.1 below). If 1984 is the trough, then the index w:ould be equal to or below its beginning (1950) level.  FIGURE 1: TRADE LIBERALIZATION INDEX X 950-I 983 20 '' 1 I, o '6 19 so 9 55 *960 9 55 I 9) 19 75 1980 is. Export Incentives. The Government rationalized export incentives, broadened access for first-time and infrequent exporters. reduced the administrntive costs associated with the export incentives, and reduced the scope of export restrictions. Specific measures included: o Reduction in the dispersion of CERT rates across products, and payment of CERTs on the basis of domestic value-added. O Automatic access to imported inputs vithout regard to vhether there is domestic production. O Broader access to duty-exemptions for exporters. O Improvements in export financing. O Reform of many procedures involving import licenses and registrations. O Reduction in the number of commodities subject to export restrictions. 19. I1wort Restrictions. Import restrictions vere red-aced by moving items from the prohibited list to the prior license list and from the prior  - 10 - license list to the free list. With respect to tariffs, the Government reduced dispersion and imposed a surcharge and an 82 'temporary' import duty. These were projected to raise the average rate of tariffs collected from 122 to 19Z, and it was expected that they could be eliminated during the second phase of reforms. Second Phase 20. The second phase of the program, begun in 1986, continued to emphasize macroeconomic adjustment. The major elements of the program were reduction of the fiscal deficit from 4.5I of GDP in 1985 to below two percent in 1956; further rationalization of the public investment program; expansiorn of credit to the private sector; and maintenance of the real exchange rate on par with the level which prevailed in 1975. This phase also included reforms in agricultural policies and agricultural and general trade mneasures. 21. The trade reforms generally extended the earlier program. Additional items were shifted from the prohibited to the prior license list and others were moved from prior license to free. Nearlv all export prohib:tions for economic purposes were eliminated. Three reforms were introduced specifically to benefit the agricultural sector: o Procedures were simplified for agricultural exporters to benefit from export incentives. o Import restrictions (tariffs and licensing requirements) for agricultural inputs--including fertilizers, pesticides, and capital equipment--were eased. o Private imports (within global quotas) were permitted for some commodities previously restricted to the agricultural marketing institute (IDEMA). 22. The following three sections contain descriptions of the principal elements of the trade system--the exchange rate, the import restrictions, and export promotion--after four years of reform. The final section discusses options ior continued reform.  - 11 - II. THE EXCHANGE RATE AND MARKET INCENTIVES 23. In 1984 the Government devalued the peso and announced it would use a crawling peg to maintain the real exchange rate at the level which prevailed in December, 1975. It was felt this was a period when the external accounts were satisfactory and there were sufficient incentives for exports. The Government reached the target by the end of 1985, and, by adjusting the nominal exchange rate daily with a view to relative inflation between Colombia and eighteen trading partners, there has been a further real depreciation of approximately 15Z. (See Line 1, Table 2.1). The real exchange rate index measures movements in nominal exchange rates and the wholesale price indices of Colombia's trading partners divided by Colombia's wholesale price index. The index reached a peak of 155 (where 1983 = 100) during December, 1986, and averaged 150 in the subsequent 15 months through March, 1988. The Government's present goal is to maintain the index at the December, 1986 level. 24. Exporters have cited the depreciation as a major factor behind the continuing export recovery. Several studies (e. g., Rodriguez 1987 or Edwards 1985--which includes a survey of the earlier literature) have found that "... a strong level of economic activity in the rest of the world, more specifically in the industrial countries, is crucial in order for Colombia's noncoffee exports to grow. From a policy perspective, however, the most important finding from these estimations is the significant effect of the real exchange rate on the behavior of noncoffee exports.' (Edwards 1985, p. 179). 25. The reform program has not affected all sectors uniformly. Table 2.1 shows ratios of selected end-of-year price indices prepared by Banco de la Republica. These ratios--the domestic terms of trade--indicate the changes in the relative prices among sectors and, within sectors, between the domestic and export markets. As prices increase in one market relative to another, incentives are offered to transfer productive resources from the markets with declining prices into those with inLreasing prices. 26. Over the adjustment program, the domestic terms of trade have shifted in favor of an outward orientation. However, the changes--except in the manufacturing sector--are not as great as would be indicated by the increase in the real exchange rate index. While the reform program has benefited the tradable goods sectors with an index of the real exchange rate more favorable than it was in 1975, the terms of trade between tradables and nontradables are uniformly less favorable than they were in 1975. Further, the price ratios were, overall, most favorable at the end of 1985 and, despite continued depreciation in 1986 and 1987, the price ratios have not continued to shift significantly to favor a more outwardly- oriented economy. 27. During 1976-77, Colombia enjoyed a coffee boom ar.d the peso became overvalued. From 1975 through 1982, the real exchange rate index fell from 129 to 97, and changes in the prices of traded goods relative to nontradables favored the production of nontradables. Lines 5 and 6 in Table 2.1 show that the prices of imports and noncoffee exports, relative to nontraded goods, fell by 35Z resulting in increased demands for _.nports  - 12 - TABLE 2.1: Relative Price Indices, 1975-87. (1983 = 100) 1976- 1975 1981 1982 1983 1984 1985 1986 1987 AVG. 1 Real Exchange Rate Index 129 110 97 100 107 137 149 148 (Pesos/Foreign Exchange) 2 Terms of Trade 87 116 92 100 92 104 105 92 (Export Prices/Import Prices) 3 Traded/Nontraded 139 121 91 300 109 131 117 116 4 Exports/Nontraded 130 130 86 100 104 139 114 110 5 Imports/Nontraded 148 112 96 100 114 123 120 122 6 Noncoffee Exports/Nontraded 137 112 88 100 106 115 106 116 7 Noncoffee Traded/Nontraded 145 112 93 100 111 121 115 120 8 Mfg. Exports/Nontraded 131 106 97 100 99 113 140 145 9 Mfg. Exports/Mfg. Nontraded 116 100 97 100 92 103 126 128 10 Mfg. Imports/Mfg. Nontraded 122 106 96 100 109 114 109 107 11 Chemical Exports/Nontraded 142 107 96 100 109 124 117 109 12 Chem. Exports/Chem. Nontraded 106 100 94 100 109 119 116 iC4 13 Chem. Imports/Chem. Nontraded 130 102 92 100 107 113 120 116 14 Imports/Consumer Goods 171 125 96 100 113 123 122 124 15 Noncoffee Exports/Cons. Goods 159 126 89 100 105 115 108 118 16 Imported Inputs/Cons. Goods 196 139 97 100 117 129 136 156 Source: World Bank staff calculations based on end-of-year price indices from Revista del Banco de la Republica; IMF, International Financial Statistics. and decreased supplies of exports. In 1981-83, imports exceeded exports by an average of six percent of GDP (see Table 1.1), compared to 1976-80 when exports exceeded imports by an average of three percent of GDP. 28. The ratio of the prices of traded to nontraded goods (Line 3) moved with the real exchange rate index (Line 1), through 1985. In 1986 and 1987, the traded/nontraded price ratio (a trade-weighted average of the exports/nontraded ratio, Line 4, and the imports/nontraded ratio, Line 5) declined even while the real exchange rate index increased. Both the rise of the real exchange rate index and the decline of the traded/nontraded price ratio appear to be related to the coffee boom, which increased the demand for, and prices of, nontradables relative to tradables. 29. Coffee was about half of Colombia's recorded exports during 1986-87, and its unit value rose by 602 between the end of 1984 and the end of 1985, from $1.43 per pound to $2.31 in the New York market, then fell in  - 13 - December, 1986, to $1.33 and further to $1.24 at the end of 1987. The volume of coffee exports increased by 18Z between 1985 and 1986, and by another four percent in 1987. 30. Line 6 shows that, overall, the pricer of noncoffee exports, relative to nontraded goods, have not changed as much as the real exchange rate index. This, however, masks differences amor.g sectors: between 1983 and 1987, manufactured exports' prices increased by 452 (Line 9) relative to nontraded; noncombustible primary materials increased by 9Z; while noncoffee agriculture and petroleum export prices both declined. The first two groups account for 45% of the index and the latter two for 48Z. By contrast, the prices of imports increased an average of 22Z (line 5) relative to nontraded gooAs with chemicals and manufactures rising 21Z and 32Z and transportation equipment by 17Z. These three groups represent 802 of the imported price index. The gain in incentives for noncoffee tradables (Line 7) owes more to the gain in incentives for import substitution (Line 5) than the incentives for exporting (line 6), though the difference between the two sectors is small. 31. Lines 8-10 and 11-13 show the effects of the reform program on the manufacturing and chemicals industries. For each industry, the first ratios (Lines 8 and 11) display the re;tive prices for its export products compared to the prices of all nonttadables. The next two ratios show the prices within the industry for export production relative to nontradables (Lines 9 and 12) and import prices relative to nontradables (Lines 10 and 13). For both industries, incentives have increased for production of tradables, though for chemicals the gain is not as great as the change in the real exchange rate index. 32. Lines 14 and 15 show the changes in prices of tradable goods- -imports and noncoffee exports--relative to consumer goods, while Line 16 shows imported inputs relative to consumer goods. Relative incentives have shifted away from consumer goods toward tradables, and, further, the costs of imported inputs have risen relative to the prices of the consumer goods they are used to produce. 33. Assessment. The Government's continued management of the exchange rate to maintain a competitive environment for exporters has been a significant achievement of the reform program. For a trade reform progtm to increase and diversify exports, it must change the relative prices of traded and nontraded goods. The fundamental mechanism for giving incentives to shift resources from production of nontradables to tradables is the exchange rate. The exchange rate affects the balances of the external accounts as well as internal macroeconomic variables: inflation, employment, and productive incentives. Exchange rate management must consider the effects on all these policy targets. 34. During the reform program, exchange rate management has been consistent with the goal of maintaining a competitive rate. The index rose from 100 in 1983 to 150 during 1986, and has been maintained at about that level. However, given the existing tariffs, there remains excess demand for foreign exchange which necessitates continued use of import licenses to ration it. Further, while the exchange rate index in 1987 stood 15Z above the 1975 level, the domestic terms of trade show that the prices of manufactured exports relative to nontraded goods are 152 below those which  - 14 - prevailed in 1975. Similarly the incentives for import substitution are not as great as they were in 1975, with the ratio of import to nontraded prices down by 15Z. 35. Most of the gains in pric_, of tradables relative to nontradables was in the first two years of the program and, except for manufacturing exports, there has been little change since the end of 1985. Exports have continued to grow, but this is consistent with normal lags between changes in the real exchange rate and changes in performance. (See, e. g., Edwards, 1985 for estimates of the lags). Real nontraditional exports in 1987 were still net as great as they were in 1980, and the manufacturing sector exported a lower share of its output in 1986 than it did in 1975. 36. There are at least three possible explanations why the continuing depreciation in 1986 and 1987 was not translated into the relative prices of tradables &nd nontradables: (1) the indices may not be accurate for the purposes they are being applied; (2) the underlying structure of the economy may have changed, and with it the equilibrium exchange rate; (3) the internal prices of importables may be controlled not by the exchange rate but by the licensing constraints. 37. The Colombian Government maintains an extensive series of price indices, based on patterns of production, consumption, and trade in 1969. It is always difficult to separate goods into distinctly "traded' and "nontraded" categories and to ensure, over time, that the goods sampled accurately portray the characteristics of the larger aggregates. Detailed knowledge of the goods in the sample would be necessary to assure that there is no sampling bias. However, the nontraded/traded price ratio did more closely track the exchange rate index during 1975-85. 38. Alternatively, the equilibrium real exchange rate has changed. The behavior of the domestic terms of trade since 1983 raises questions about the target rate the Government has adopted--initially the index which prevailed in 1975 and, subsequently, that which prevailed in December, 1986. The equilibrium rate at any time is determined by a host of underlying economic fundamentals, and each change in one of the determinants would change the equilibrium. rate. 39. Assuming that the rate which prevailed in 1975 were truly an equilibrium,1 there are many reasons to believe it would have changed. The underlying structure of the Colombian economy has been transformed in a number of ways which are important to the determination of the equilibrium exchange rate. Some of the major changes are the rates of tariff collection and export subsidy, the unemployment rate, the fiscal deficit, the rate of inflation, and, quite likely, the volume and value of illegal 1/ Some would doubt that the 1975 real exchange rate was an equilibrium. There was a recession in the Colombian and the world economies, and during 1975 it was thought that the courtry could have a balance of payments crisis. About 55Z of reimbursible imports were restricted by licensing requirements and some policy-makers were considering an even- more-extensive system of licensing.  - 15 - imports. In addition, the following fundamentals would change the equilibrium real exchange rate: o There has been a marked expansion in Loal and petroleum exports from US$43 million (10Z of goods exports) in 1981 to US$1,604 million (31Z) in 1987. Their increase accounted for 84Z of export growth over this period. These new exports shift the supply curve for foreign exchange outward and reduce the equilibrium rate. O Foreign debt has increased from 212 of GDP to 432, and real interest rates in the world economy are higher. This increases the demand Aor foreign exchange and raises the equilibrium exchange rate. o The volume and composition of unregistered exports has changed. In the 1970's, Colombia was a major exporter of marijuana. In the 1980's marijuana has been replaced by cocaine. The two exports have different levels of local value added and of repatriation of foreign exchange. WhLile measuring illegal exports is problematic, it has been estimated that during 1981-85 they were between 2.6Z and 6.41 of GDP, or ietween 232 and 76Z of the valup of legal exports (Gomez, 1988). The illegal exports tend to reduce cne equilibrium exchange rate, and if the value of illegal exports has increased since 1975, there would be a net downward effect on the equilibrium exchange rate. 40. Finally, in a closed economy in which licenses (rather than tariffs or the exchange rate) are the binding constraint on imports, the internal price of importables is fixed by the quantity of licenses rather than the exchange rate. Within the range in which the licenses are the binding constraint, devaluations would serve to reduce the rents accruing to licenses and would not affect internal prices. During the reform program, in.port licenses were reduced from 991 of domestic manufacturing on 1984 to 822 in 1988 (see Table 3.7) 41. At the current exchange rates and tariffs, there is excess demand for foreign exchange. The "shortage' signifies that the value to importers of an additional unit of foreign exchange is greater thft, Lhe cost to exporters of obtaining it. In order to restore balance so that supply equals demand without Government rationing, the Government could increase the level of tariffs or allow a further depreciation of the exchange rate, or some combination of the two. General tariff increases would be inconsistent with the thrust of the trade reform program and would provide no stimulus for expanding trade, but they would provide an additional source of revenue and, with properly chosen tariffs, allow the Government to rationalize the protective structure. Alternatively, if the Government wishes to expand imports and exports, then further depreciation of the real exchange rate would increase the supDly and decrease the demand for foreign exchange until excess demand is eliminated. This would be consistent with the overall thrust of the reform program to date, and with the Government's goal of continued export expansion and diversification.  - 16 - III. IMPORT RESTRICTIONS 42. The Jurnta Monetaria is responsible for fixing a foreign exchange budget to control the value ot imports. At the exchange rates which have prevailed in Colombia, the demand for imports has couis'stEntly exceeded th- foreign exchange budget. Tariffs and, more significantly, discreti.uL,., licenses have been used to contain imports within tho levels permitted by the exchange budget. This section discusses the recent changes in the import licensing requirements and tariffs and their effects on the structure of protection. A. ImDort Licenses 43. The foreign exchange budget is rationed among importers by .NCOMEX through its im,ort licensing system. Each item in the NABANDINA tariff code belongs to one of three categories: Free (INCOMEX must grant import licenses); prior license (INCOMEX has discretion to refuse licenses); or prohibited (INCOMEX will grant licenses only under unusual circumstances). The Junta de Importaciones determines whether an item is nlaced on the prohibited, previous license, or free list. Presently there are fifty prohibited items (Table 3.1), largely imports which are considered 'luxuries": specialty food products, alcoholic beverages, cigarettes.1 TABLE 3.1: Import Licensing Requirements, 1980-88. (Number of tariff positions) 1980 1981 1982 1983 1984 1985 1986 1987 1938 ProhibiteJ! 0 0 0 0 828 69 56 56 - Prior License 1502 3180 3716 4671 4160 3602 3160 3081 3131 Free 3276 1814 1287 340 23 1359 1826 1905 1939 Total Items 4778 4994 5003 5011 5011 5030 5042 5042 5120 Sources: INCOMEX, Country Economic Memorrndum, Arancel de Aduanas. 44. All registered imports are classified either as reimbursable or non-reimbursable. If an import is reimbursable, the importer receiving a license is assured of foreign exchange from Banco de la Republica. If it is not, lie must supply foreign exchange from his own sources. Those with access to foreign exchange include ECOPETROL, diplomats, foreign dono.s, Plan Vallejo, and foreign capital investment. Even where an importer supplies his own foreign exchange, he still must have a license. l/ Some of these items have local production or close substitutes which receive substantial protection from the import prohibitions. Excise taxes would provide more neutral incentives.  - 17 - TABLE 3.2: Annual Import Licenses by Foreign Exchange Regime. (US$, Millions) REIMBURSABLE IMPORT LICENSES 1981 1982 1983 1984 1985 1986 1987 Frgn. Exch. Budgst 4709 4866 3969 3126 3298 4196 4759 ECOPETROL 241 316 281 186 281 201 125 Plan Vallejo 163 120 91 112 144 189 233 Subtotal 5113 5302 4341 3424 3723 4586 5117 NON-REIMBURSABLE IMPORT LICENSES Petroleum, Mining 671 383 416 322 504 88 142 Temporary Imports 254 309 216 97 296 109 120 Donatio.is 8 8 5 19 17 8 23 Diplomatic 4 4 6 29 7 6 6 Foreign Capital 0 0 0 0 12 8 4 SIEX 19 28 21 38 27 24 25 Other 25 61 26 53 105 45 29 Subtotal 981 793 690 559 966 288 350 TOTAL IMPORT LICENSES 6094 6095 5030 3982 4689 4874 5466 Source: INCOMEX. 45. Table 3.2 shows data on the annual volume of import licenses. divided between reimbursable and non-reimbursable. Over the period covered by the data, 1981-87, reimbursable imports accounted for 882 of the licenses. Within the non-reimbursable category, most licenses were given for petroleum and mining or for temporary imports. It should be noted that the non-reimbursable exports have not served as a safety valve to avoid tight foreign exchange budgets as there is no correlatior. between reductions in the foreign exchange budget and increases in the non- reimbursable exports. 46. Because there was excess demand for foreign exohange at the rates which prevailed, by 1984 the Government had virtually eliminated unrestricted imports. Less than 12 of the items in the NABANDINA were freely importable, 832 required licenses, and 162 were prohibited. One of the major features of the reform program was the removal of items from the prohibited and prior license lists and expansion of the free list as shown in Table 3.1. The reforms have resulted in a system in which the number of unrestricted and prohibited positions has returned to the level of 1981. In 1981, 3180 items were subject to license, and in 1988, 3181 were either subject to license or prohibited. While this represents progress from the  - 18 - trade regime of 1^84, it may be seen in Table 3.1 that the regime in 1981 was more restrictive than that which prevailed only a year earlier. 47. The foreign exchange budget has grown, and import licenses in 1987 were increased by 37Z (in nominal terms) over the levels of 1984 (Table 3.3). Simultdneously, INCOMEX has improved the pLocedures for granting licenses so that importers are able to plan more effectively. Most decisions are made within two weeks, whereas they have previously required up to six months. 48. in 1987, INCOMEX evaluated 113,000 applications for licenses, approving 84,000 (742) and rejecting 29,000 (262). The demand for imports represented by the applications was US$7,051 million and the licenses granted totaled US$5,466 million (77% of the value requested). These figures cannot be interpreted as the measure of "excess demand' since, on the one hand, unsuccessful firms may re-apply and firms may apply for more imports than they intend to import and, on the other, many who would have imported simply assume a negative determination and do not apply. This factor is particularly strong in the case of consumer goods which, while not prohibited, constitute less than 102 of licenses granted, and INCOMEX's prejudice against them is well known. 49. The foreign exchange budget in 1987 was US$4,759, of which US$2,115 million was applied to free imports leaving US$2,644 for prior licenses. Presumably, the demand for prior licenses was the difference between the total applications and the free licenses granted, or US$4,936 million. Since the foreign exchange available for prior licenses was only 54Z of the value of the applications, priorities had to be set by INCOMEX. Decreto Ley 444 lists the following criteria for evaluating applications (Legis, Regimen de Importaciones, pp. 5-6): o The importance of the good to satisfy consumer needs and to maintain or increase the level of employment. o The contribution to the development and diversification of exports and the effect on the balance of payments. o The net savings of foreign exchange obtained by import substitution. O Whether there is national production in the region where the merchandise is destined and the necessity of protecting the consumer against the abuse of price and quality. o The need for timely imports of merchandise or inputs for the development of local industry in especially depressed regions, even though the imported goods are produced in the country. O Where the application is for goods not produced in the country, their degree of scarcity and grade of necessity. Also, their retail price relative to the cost of importation. o The importer's inventories relative to the sales. o The value of licenses for the same article reqluested by the importer in previous periods.  - 19 - TABLE 3.3: Import Licenses Classified by Use. (USS, Millions) 1981 1982 1983 198S 1985 1986 1987 Consumption 770 1062 576 417 435 643 653 Raw Materials 2210 2080 1877 1862 2213 2053 2503 Capital Goods 2173 2242 1865 1232 1568 1943 2162 Combustibles 902 645 650 432 433 201 124 Others 39 66 61 39 39 35 24 Total 6094 6095 5030 3982 4689 4874 5466 (Percentage) Consumption 132 17% 1l1 10Z 9% 13% 122 Raw Materials 362 34Z 37% 47Z 47Z 42Z 46Z Capital Goods 36Z 37Z 37% 31% 33% 402 40% Combustibles 15% 11% 13Z 11i 9Z 4% 2Z Others lZ 1 1% 1Z 1Z 1Z 0% Source: INCOMEX. 50. In interpreting these criteria, INCOMEX assigns first priority to inputs and consumers' necessities (medicine and food not produced within Colombia), second to capital goods, and residual exchange is used for other purposes. Where INCOMEX believes there is domestic production of sufficient quality and quantity, import licenses are not granted. When there are unanticipated contractions of the foreign exchange budget, INCOMEX is aware of the employment effects of its decisions and tries to minimize unemployment which could be caused by shortages of imported inputs. Consumption goods and capital goods for new investments are reduced before raw materials and intermediate goods. Decisions on individual applications are strongly influenced by a firm's imports in previous years, so that it becomes difficult for a firm to improve its market share against competitors if each is constrained by imported inputs. In this regard, transferring imported inputs from the prior license to the free list could increase domestic competition in the output market. 51. Table 3.3 displays data on import licenses awarded, by end-use, over 1980-87. The data reflect INCOMEX's priorities to award licenses to inputs, consumer necessities, and capital goods. Over the period, there has been a steady decline in imports of combustibles as ECOPETROL's production has increased and the country has imported less petroleum. There has been relatively little variance in the imports of raw materials relative either to capital goods or consumer goods. Consumer and capital goods, the most volatile categories, have been sensitive to the size of the  - 20 - TABLE 3.4: Imports by Use, 1980-87. (USS, Millions) 1980 1981 1982 1983 1984 1985 1986 1987P Consumer Goods Nor.-Durable 307 331 324 302 242 178 185 213 Durable 312 336 367 237 189 155 197 290 Subtotal 620 668 69>1 339 431 333 383 iO2 Raw Materials and Intermediates Agricultural 162 147 190 158