Article

The Export Promotion System and the Growth of Minor Exports in Colombia

Author(s):
International Monetary Fund. Research Dept.
Published Date:
January 1973
Share
  • ShareShare
Show Summary Details

TRADITIONALLY, COFFEE HAS PROVIDED the bulk of foreign exchange receipts for Colombia. The problems of this dependence upon a single commodity were perhaps most clearly apparent during the period 1957–60, when the international market for Colombian coffee collapsed. World prices for Colombian coffee fell from an annual average of 74 U. S. cents a pound (nearly a record figure) in 1956 to 64 U. S. cents a pound in 1957 and to 45 U. S. cents a pound in 1960. Although this price decline of almost 40 per cent was partially offset by increases in volume, it resulted in a shrinkage of the value of coffee exports by about one third, from a peak of $487 million in 1955 to $332 million in 1960 (Table 1). Given the great importance of coffee in the commodity composition of Colombia’s exports, this price fall also resulted in a substantial decline in overall receipts from commodity exports, although not to such a great extent. Since 1960, however, the growth of so-called minor, or nontraditional, exports has provided some cushion against shortfalls in coffee receipts.1 Over the past decade, minor exports have grown at an average annual rate of 14.5 per cent, and, as a result, their share in total commodity export earnings increased from 11.3 per cent in 1960 to an estimated 34.7 per cent in 1971. In the same timespan, the share of coffee fell from 71.5 per cent to 57.8 per cent.

Table 1.Colombia: Structure of Exports, on a Customs Basis
CoffeeOilOtherTotalCoffeeOilOtherTotal
Million U. S. dollarsPer cent
19054.67.211.839.061.0100.0
192565.518.083.578.421.6100.0
1945103.922.214.4140.574.015.810.2100.0
1955487.461.547.8596.781.710.38.0100.0
1960332.280.052.4464.671.517.211.3100.0
1965343.988.2107.0539.163.816.419.8100.0
1967322.461.2126.3509.963.212.024.8100.0
1968351.436.3170.5558.263.06.530.5100.0
1969343.956.7206.9607.556.69.334.1100.0
1970458.862.5202.3723.663.48.628.0100.0
1971399.051.4239.6690.057.87.534.7100.0
Source: Banco de la República.
Source: Banco de la República.

In this study, an attempt is made to explain the growth of minor exports during the past 20 years in the light of Colombia’s export promotion policies. At first, the promotion of minor exports was undertaken on a trial-and-error basis in which the definition of minor exports often varied and policies sometimes changed from one year to the next. However, in recent years the promotion of minor exports has been managed in a more systematic way, using a comprehensive package of policies that included the exchange rate, fiscal incentives, import-export schemes, special credit facilities, and regional trade agreements and other institutional arrangements. Accordingly, it is hypothesized in this study that the strong growth of minor exports in recent years has been due in large measure to these policies.

As a background to the main body of the study, Section I sets out in some detail a description of the various components of the export promotion system and explains its evolution to the present day. In this part of the study, a measure of the effective exchange rate is developed that seeks to capture the effects of the most important export promotion policies. Then Section II discusses the growth of minor exports in more detail and formulates a simple statistical model for the purpose of measuring the sensitivity of minor export growth to variations in the effective exchange rate, as defined in Section I. The section also considers the relative importance of the various components of the export promotion system. Section III indirectly measures the resource allocation effects of the export promotion policies followed in Colombia through the fiscal cost of the export promotion system. This cost is analyzed in terms of the revenues forgone in the central government budget. Finally, Section IV discusses conclusions and policy implications for the future.

I. The Export Promotion System

During the postwar period, the export promotion system in Colombia has developed in an ad hoc fashion. At first, export incentives were established primarily through the exchange system via multiple rates. Then, beginning in 1960, certain fiscal incentives were created in the form of income tax exemptions. These benefits were amplified in 1967 by a system of tax credit certificates. Along with the establishment of fiscal incentives, special treatment was given to the exports of manufactured goods under the so-called Vallejo Plan, in the form of exemptions from prior licensing, advance import deposits, and customs duties on imported inputs. Special credit facilities in the form of export prefinancing, which were free of exchange risk, were made available during the early 1960s for nontraditional exports. While each of these instruments was in operation throughout the 1960s, only in the latter part of the decade were effective coordination and widespread use achieved.

In the remainder of this section, each of these policy instruments is discussed separately. Then an attempt is made to measure their joint significance and impact in calculating an effective exchange rate for minor exports. Finally, some observations are made about regional trade agreements and other institutional arrangements that have benefited minor exports.

The exchange rate

Exchange rate policy throughout the postwar period has been governed strongly by the changing fortunes of coffee. As a result, the exchange rate that the Colombian authorities have deemed appropriate for the coffee sector has not always been adequate from the point of view of the promotion of minor exports. This situation has led in practice to the use of multiple exchange rates, either explicitly in terms of specific rates or implicitly in terms of export taxes, at many times during the postwar period for the purpose of achieving different economic objectives. The dominance of coffee helps to explain not only the structure of the system but also its inflexibility over time. It is this latter feature that has often produced both exchange rates that were biased against minor exports and the need for elaborate administrative controls on imports. In Table 2, a chronology of changes in the basic exchange rate for most nontraditional exports attempts to summarize the postwar history of exchange rate policy for the promotion of minor exports. Table 3 shows the basic exchange rate applicable to most nontraditional exports for the period 1948–71.

Table 2.Colombia: A Postwar Chronology of Changes in the Basic Exchange Rate for Most Nontraditional Exports(Quotations in Colombian pesos)
Time PeriodType of Exchange RateBeginning/Final

Quotation1
June 1948–Mar. 1951Fluctuating “certificate” rate21.75/2.50
Mar. 1951–July 1952Fixed “official” rate32.50/2.50
Aug. 1952–May 1955Free “export voucher” rate43.50 5/4.00
May 1955–Nov. 1956Free rate64.00/6.62
Dec. 1956–June 1957Free “certificate” rate76.86/5.96
June 1957–Mar. 1958Fluctuating “certificate” rate85.96/6.10
Mar. 1958–Jan. 1959Fixed rate96.10/8.12
Jan. 1959–June 1965Fluctuating “preferential” rate 108.12/18.75
July 1965–Mar. 1967Fixed rate 1113.50/13.50
Mar. 1967–PresentFluctuating “certificate” rate1213.50/22.32 13
Sources: International Monetary Fund, Annual Report on Exchange Restrictions and International Financial Statistics, various issues.

End of month (buying) through 1959; thereafter, average monthly rate.

This rate was introduced on June 1, 1948 and was applied to all exports except coffee, petroleum, bananas, and raw hides.

This rate was introduced on March 20, 1951 and was applied to all exports.

This rate was introduced on August 1, 1952 and was applied to exports of most agricultural products (excluding coffee and bananas), salt, tanned hides, leather manufactures, designated textiles, cement, beer, sugar, sulfur, tobacco, nonprecious metals, nonmetallic mineral substances, and gold manufactures. Gold was also included during the period August 1952–August 1953, after which a special gold market rate was introduced.

During the period August-December 1952, no direct quotations are available for this rate; however, it is known that the premium in this market was about Col$l above the official rate of Col$2.50 per US$1.

This rate was introduced on May 13, 1955 and was applied to all exports.

This rate was introduced on November 30, 1956 and was applied to all exports.

This rate was introduced on June 17, 1957 and was applied to all exports; in addition, all exports were subject to a tax of 15 per cent, which was reduced to 2 per cent on September 25, 1957 for all exports except coffee, bananas, and precious metals. The rate of 2 per cent was reduced to ⅛ of 1 per cent on March 7, 1961 and was eliminated altogether on October 24, 1962.

This rate was introduced on March 26, 1958 and was applied to all exports before export taxes.

This rate was introduced on January 16, 1959 and was based on the average level of the buying rate in the free market for the preceding week. Initially, it was applied to all exports except coffee, bananas, raw hides, and precious metals. After August 19, 1959, exports of manufactured goods with an import content of more than 40 per cent were also excluded; on March 7, 1961, this limit was raised to 50 per cent. From January 11, 1963 until October 25, 1964, this rate was pegged at Col$9.98 per US$1.

This rate was introduced on June 30, 1965; after August 21, 1966, it also applied to all exports of manufactured goods.

This rate was introduced on March 22, 1967 and was applied to all exports.

September 1972.

Sources: International Monetary Fund, Annual Report on Exchange Restrictions and International Financial Statistics, various issues.

End of month (buying) through 1959; thereafter, average monthly rate.

This rate was introduced on June 1, 1948 and was applied to all exports except coffee, petroleum, bananas, and raw hides.

This rate was introduced on March 20, 1951 and was applied to all exports.

This rate was introduced on August 1, 1952 and was applied to exports of most agricultural products (excluding coffee and bananas), salt, tanned hides, leather manufactures, designated textiles, cement, beer, sugar, sulfur, tobacco, nonprecious metals, nonmetallic mineral substances, and gold manufactures. Gold was also included during the period August 1952–August 1953, after which a special gold market rate was introduced.

During the period August-December 1952, no direct quotations are available for this rate; however, it is known that the premium in this market was about Col$l above the official rate of Col$2.50 per US$1.

This rate was introduced on May 13, 1955 and was applied to all exports.

This rate was introduced on November 30, 1956 and was applied to all exports.

This rate was introduced on June 17, 1957 and was applied to all exports; in addition, all exports were subject to a tax of 15 per cent, which was reduced to 2 per cent on September 25, 1957 for all exports except coffee, bananas, and precious metals. The rate of 2 per cent was reduced to ⅛ of 1 per cent on March 7, 1961 and was eliminated altogether on October 24, 1962.

This rate was introduced on March 26, 1958 and was applied to all exports before export taxes.

This rate was introduced on January 16, 1959 and was based on the average level of the buying rate in the free market for the preceding week. Initially, it was applied to all exports except coffee, bananas, raw hides, and precious metals. After August 19, 1959, exports of manufactured goods with an import content of more than 40 per cent were also excluded; on March 7, 1961, this limit was raised to 50 per cent. From January 11, 1963 until October 25, 1964, this rate was pegged at Col$9.98 per US$1.

This rate was introduced on June 30, 1965; after August 21, 1966, it also applied to all exports of manufactured goods.

This rate was introduced on March 22, 1967 and was applied to all exports.

September 1972.

Table 3.Colombia: Basic Exchange Rate Applicable to Most Nontraditional Exports, 1948–711(In Colombian pesos per U.S. dollar)
YearRateYearRateYearRate
19482.3619565.09196410.41
19493.0219576.18196514.76
19503.1119586.32196613.50
19512.5319597.64196714.51
19522.9219606.90196816.30
19533.4719618.28196917.35
19543.4919629.11197018.49
19554.15196310.00197120.00
Sources: Banco de la República; International Monetary Fund, International Financial Statistics, various issues.

Annual average of end-of-month quotation for 1948–58 and annual average of monthly average quotation for 1959–70 (buying).

Sources: Banco de la República; International Monetary Fund, International Financial Statistics, various issues.

Annual average of end-of-month quotation for 1948–58 and annual average of monthly average quotation for 1959–70 (buying).

Colombia’s policies for promoting minor exports during the postwar period began on June 9, 1948, when Decree No. 1949 was issued to establish a free market “certificate” exchange rate expressly for that purpose.2 During the last seven months of 1948, this free rate fluctuated around an average level of Col$2.738 per US$1, in comparison with the “basic” export rate of Col$1.745. A new par value for the peso was established on December 16, 1948 at Col$1.95 per US$1, and until this rate was changed on March 20, 1951, the free market certificate continued to operate. The maximum monthly quotation recorded during this time was Col$3.447 at the end of March 1950. During this first experience in promoting minor exports, nontraditional exports were defined to include all products except coffee, raw hides, bananas, and petroleum. For purposes of the exchange rate system, this definition prevailed, in general, until December 26, 1962; at certain times during this period, precious metals were also included in the category of major exports. After this date, minor exports were defined as they are known today, that is, all exports except coffee, petroleum, and raw hides.

From March 20, 1951 until August 1, 1952, receipts from minor exports were converted at the basic export rate of Col$2.50 per US$1. This adjustment in the exchange system represented a significant appreciation in the minor export rate from the average level of the preceding two years. However, in August 1952, a second export promotion device was introduced into the exchange regime in the form of an “export voucher” system. Under this arrangement, upon surrender of foreign exchange, certain minor exports 3 were granted export vouchers, which gave the exporter the right to import, in an amount equivalent to the value of the exchange surrendered, specified goods included in the prohibited list. These vouchers were negotiable and were sold at a premium in the free market. On average, as this system operated from August 1952 until May 1955, the premium for these export vouchers was about Col$1.00 above the basic export rate of Col$2.50 per US$1.

On May 13, 1955, the export voucher system was discontinued and was replaced by a free market for minor exports that existed, except for a period of nine months, under various names until mid-1965. During these ten years the world market for coffee, which had remained rather buoyant for the postwar period up to the mid-1950s, with average annual prices reaching a peak of almost 80 U. S. cents a pound, weakened considerably. With the fall in world prices, which proceeded without a major break until 1964, Colombia underwent a series of exchange crises. As a result, the Colombian authorities experimented with a number of exchange systems during that time. At first, a free market for minor exports was created that was modified by the imposition of a 15 per cent export tax beginning on June 17, 1957. Initially, the 15 per cent tax rate applied to all exports, but it was reduced to 2 per cent on September 25, 1957 for all exports except coffee, bananas, and precious metals and was maintained until 1962. Then, on March 27, 1958 a fixed buying rate of Col$6.10 per US$1 was applied to all exports, but this rate was maintained only until the end of the year. On January 16, 1959 a “preferential” rate based on the average level of the buying rate in the free market for the preceding week was applied to minor exports, while a fixed “certificate” rate of Col$6.10 per US$1 was applied to major exports. At first, the fixed certificate rate applied only to coffee, bananas, and precious metals; however, after August 19, 1959, manufactured goods with an import content of 40 per cent or more were also treated at this rate. Because of administrative difficulties involved in controlling the import content of exports of manufactured goods, on March 7, 1961 this limit was raised to 50 per cent.

The free market for minor exports was continued, with periodic upward adjustments in the major export rate, until January 11, 1963, at which time the free market rate was pegged at Col$9.98 per US$1. On October 25, 1964, the Bank of the Republic withdrew from the market and the minor export rate was freed once again, until mid-1965. During this period, the rate increased, reaching a peak of Col$18.75 per US$1 at the end of June. On June 30, 1965, however, the minor export rate was fixed at Col$13.50 per US$1, and this rate remained in effect until the major exchange reform of March 1967, when the present exchange system was inaugurated. Shortly before this reform, on August 21, 1966, proceeds from all manufactured exports, regardless of their import content, were to be negotiated at the minor export rate of Col$ 13.50 per US$1 rather than at the then ruling major export rate of Col$9.00 per US$1.

The exchange system that has been in operation since early 1967 has been characterized by a unified rate for all exports except petroleum, adjusted for taxes on coffee exports and for subsidies on minor exports. This so-called certificate market exchange rate has been adjusted periodically, often at daily intervals, mainly to reflect movements in domestic costs. This practice has meant, in effect, an average annual depreciation of the Colombian peso of 7.3 per cent over the period 1967–71.4

Fiscal incentives

Fiscal incentives for minor exports were introduced in 1960 by Law No. 81, which took effect in 1961. Among other things, this Law made export profits exempt from income taxes, under the assumption that such profits were as much as 40 per cent of the gross value of exports and by allowing presumed export profits to be deducted from other profits for income tax purposes. This Law excluded from its benefits not only coffee and petroleum but also bananas and raw hides.

The original intention of the export incentives contained in Law No. 81 of 1960 was to stimulate the use of excess productive capacity; however, it soon became recognized as a powerful incentive for promoting nontraditional exports. With an average income tax rate of about 36 per cent, the fiscal subsidy to nontraditional exports implied in the Law was equivalent to 14.4 per cent of the gross value of exports.

While this subsidy provided strong incentives for exports, it created certain distortions, as the system was used to reduce or even to eliminate income tax liabilities arising from activities entirely unrelated to the export business; also, the higher the income tax bracket of the exporter, the larger the subsidy that was offered. Thus, a number of large corporations, such as commercial banks, became engaged in the export business, with the result that the fiscal subsidy to nontraditional exports tended to benefit primarily large corporations.

The existence of these distortions was recognized at the time of the exchange reform of March 1967, and the income tax exemption established by Law No. 81 of 1960 as an export incentive was replaced by Decree-Law No. 444 of 1967 (Art. 165) with a tax credit certificate (Certificado de Abono Tributario—CAT). Since then, CATs have been issued in the amount of 15 per cent of the value of exports to all exports excluding only coffee, petroleum, petroleum derivatives, and raw hides. The CATs, which are freely negotiable, can be used to pay at face value income, sales, and import taxes one year after they are issued. An important additional feature of the CATs is that they are tax exempt. The Government retained the right to change the 15 per cent rate of the CAT or the term for accepting it in payment of taxes.

The afore-mentioned distortion in the operation of Law No. 81 was, in part, corrected through the system of CATs. Whereas under the system established by Law No. 81 of 1960 the exporter needed to have substantial profits from other activities before he could benefit from the subsidy, the system of CATs provided a significant cash incentive to any exporter, regardless of his tax status. Even for a firm without any tax liability, CATs could be sold to other taxpayers. Indeed, substantial amounts of CATs have been bought and sold in the Bogotà and Medellín stock exchanges. After such transactions were initiated in October 1967, 17 per cent of the CATs issued in the last quarter of that year were sold, with this percentage ranging between 60 and 95 in succeeding years. The CAT system, however, still retains a certain bias, as its tax-exempt status provides a larger subsidy to the exporter whose income tax liability (i.e., marginal tax rate) is larger. 5

In order to increase the subsidy to nontraditional exports involved in the CATs, in 1970 the authorities reduced the CAT redemption period from 12 months to 9 months and late in 1971 further reduced this period to 3 months for exports of processed goods and to 6 months for all other nontraditional exports. As shown by the corresponding implementing regulation, the boundary between these two concepts is somewhat arbitrary, as is also evident from the fact that textiles and several chemical products are classified for the purposes of this regulation as nonprocessed products.

Although the export subsidy established by Law No. 81 of 1960 can be easily quantified as 14.4 per cent of the gross value of nontraditional exports, the quantification of the effective subsidy provided by CATs is somewhat more complicated. This is so because CATs provide a different subsidy depending upon whether the exporter retains them until the time of redemption or sells them immediately on the stock exchange. In the Appendix, an attempt is made to estimate the effective subsidy provided by CATs. In practice, this subsidy has ranged from 13.7 per cent to 18.3 per cent of the value of exports, with the value of the subsidy in recent years in excess of 15 per cent. The fiscal cost implied in this subsidy is discussed in a later section.

Import-export schemes

Law No. 1 of 1959 (Art. 55) established an import-export scheme that became known as the Vallejo Plan, the objective of which was to promote the utilization of excess capacity in the manufacturing sector. Just as with the presumptive income tax exemption under Law No. 81, it became evident that the Vallejo Plan was a powerful device for promoting nontraditional exports, and it was defined as such in the exchange reform of March 1967. The Vallejo Plan’s principal provisions are the following: raw materials, intermediate products, and capital equipment to be used for manufacturing products destined for foreign markets may be imported exempt from prior license, advance import deposit, and customs duties. The conditions under which access to the benefits of the Vallejo Plan can be gained are quite stringent. A contract between the exporter and the Government’s Institute of Foreign Trade (Incomex) is required, under which the exporter must demonstrate that imports are being financed as prescribed by the corresponding regulations, submit an exchange surrender guarantee as well as a guarantee covering twice the amount of customs duties that he would have to pay otherwise, commit himself to carry separate accounting to register all transactions under these contracts, and undertake to provide Incomex with ad hoc progress reports on the way his contract is being carried out.

In spite of these requirements, the Vallejo Plan is very attractive. First of all, by not requiring import licenses it reduces—on the basis of Incomex’s normal administrative process—the time required to obtain an import permit from about 20 days to 2 or 3 days. Because imports have been available in Colombia free of restrictions for only a few months since 1956, this feature is quite important. Second, the Vallejo Plan has a substantial financial value to the exporter because of the exemption from advance import deposits and customs duties. This financial value, or subsidy effect, of the Vallejo Plan has been estimated at between 10 and 12 per cent of the gross value of nontraditional exports (see the Appendix). The corresponding cost to the Government of the Vallejo Plan subsidy through the exemption from customs duties is discussed later.

There have been numerous complaints in Colombia about the shortcomings of the Vallejo Plan related to the complexity of the required contract, a factor that makes the system unavailable to most medium-sized and small industrialists. Another alleged shortcoming of this scheme is that, because of the inherent difficulties in making firm contracts to export to new markets, and thus to provide the exchange surrender guarantee that is required, the Vallejo Plan is not available to exporters who are trying to penetrate new markets. Also, exporters benefiting from the Vallejo Plan receive a double subsidy resulting from the combined effect of CATs with the Vallejo Plan, which is not available to minor exports of nonmanufactured goods.

In response to these complaints and shortcomings, the Colombian authorities in 1967 created two alternative import-export schemes: the Vallejo Jr. Plan and the “drawback.” The Vallejo Jr. Plan offers the same incentives as the Vallejo Plan—that is, no prior licensing requirement and exemption from advance import deposits and customs duties—but on an ex post basis. After an exporter has completed an export, he can claim the benefits of the Vallejo Jr. Plan for his next import. The “drawback” is a system for the partial refunding of customs duties, with the amount refunded depending on the value-added component of the product exported. Decree-Law No. 444, which created the drawback, established that the refund would only be made effective one year after the date of import.

Of the three import-export schemes, only the Vallejo Plan is used widely. The drawback has not yet become operational, since the Government has not issued the required implementing regulations. Moreover, the Vallejo Jr. Plan has not attracted many users, partly because of lack of knowledge and partly because it implies that more information must be provided than some exporters are prepared to furnish. Over the past decade, trade under the Vallejo Plan has expanded rapidly. Imports grew from less than US$1 million in 1962 to US$26 million in 1971, and exports jumped from less than US$1 million before 1962 to more than US$65 million in 1971 (Table 4). The extent of the growth in these exports since 1967 that is partly the result of the CAT subsidy is difficult to measure.

Table 4.Colombia: Value of Exports and Imports Under the Vallejo Plan, 1960–71(In millions of U. S. dollars)
YearExportsImportsImports as Percentage

of Exports
19600.10.1100.0
19610.20.2100.0
19620.80.225.0
19635.82.237.9
196412.95.139.5
196526.19.837.5
196645.912.026.1
196741.917.040.6
196852.017.934.4
196961.513.722.3
197058.124.742.5
197164.926.040.1
Sources: Incomex, Anàlisis Sobre el Desarrollo de los Sistemas Especiales de Importación y Exportación, 1971; Banco de la República (figures for 1971).
Sources: Incomex, Anàlisis Sobre el Desarrollo de los Sistemas Especiales de Importación y Exportación, 1971; Banco de la República (figures for 1971).

The three import-export schemes described earlier, especially the Vallejo Plan, have often been criticized on the ground that they do not promote efficient resource allocation. It is contended that these schemes are biased toward industries with a high import component, and that they tend to maximize the import intensity of a certain group of exports, thereby reducing the domestic value added and the net exchange earnings produced.6 Subject to important constraints because of the lack of data to demonstrate this point fully, Urdinola and Mallon 7 have produced quantitative evidence that at least for wood, paper, chemicals, food products, clothing, and miscellaneous manufactures (corresponding to group 32 under the standard UN two-digit classification of industry by branches) the high import content of exports is due to the relatively greater import intensity of the particular commodities exported under the Vallejo Plan rather than to the substitution of imported for domestic inputs. Those industrial groups account for more than one third of nontraditional exports under the Vallejo Plan.

Special credit facilities

The principal system of special credit facilities for nontraditional exports used by Colombian exporters is reflected in the technique of reintegros anticipados, or advance exchange surrender. Under this system, potential exporters borrow in a foreign currency, for periods ranging from a few months to more than one year, and repay with their eventual export proceeds. Prior to January 1973, the exchange risk to the exporter resulting from the peso devaluation in the period when the loan was outstanding was absorbed by the Bank of the Republic, as repayment of the loan was made at the exchange rate prevailing on the day of advance surrender. This system is used almost exclusively by exporters who are already established in their markets to obtain financing for working capital up to the full value of the export sale contract, at interest rates that are substantially below those offered by domestic commercial banks. The principal source of this kind of financing is a line of credit obtained by Colombian commercial banks with their correspondents abroad. Sometimes this type of credit is also extended directly by foreign commercial banks with branches in Colombia and, for Colombian affiliates of foreign firms, is obtained through the head offices abroad.

This system of export prefinancing has been available throughout the past decade, but its use became widespread only after 1964, when the uncertainties associated with exchange rate policies were removed. This is due to the fact that, until January 1973, when the regulations were changed, in this type of external financing, the final peso liquidation was made at the exchange rate prevailing on the date of the definitive exchange surrender. Any exporter using this system was entitled to cancel his advance exchange surrender by returning to the Bank of the Republic the amount of pesos equivalent to his foreign borrowing, at the same exchange rate that prevailed on the date of making the advance surrender. Thus, exchange risk existed only if the export order, on account of which advance exchange had been surrendered, was canceled and the Colombian peso had appreciated in the exchange market in the meantime. Under the exchange system introduced in March 1967, this risk was virtually eliminated, so that widespread use has been made of this system. As a result, more than two thirds of the exchange surrendered to the Bank of the Republic in 1970 and 1971 for exports was in the form of reintegros anticipados (Table 5). The actual proportion of advance to total exchange surrender may be even higher than shown by the Bank’s statistics, because these are based on quarterly data and show the position of advance surrender at the end of the quarter without accounting for advance surrender that has been canceled within the quarter. In May 1972, a certain element of exchange risk was introduced into the system by establishing that when export orders were canceled the final liquidation of the contract would be made at the exchange rate prevailing at that time, rather than at the time of the original exchange surrender.

Table 5.Colombia: Advance and Normal Exchange Surrender for Nontraditional Exports, 1966–71(In millions of U. S. dollars)
YearAdvance

Exchange

Surrender
Normal

Exchange

Surrender
Total

Exchange

Surrender
Advance Exchange

Surrender as a

Percentage of Total
196633.162.695.734.6
196766.750.4117.157.0
196899.678.8178.455.8
1969116.590.7207.256.2
1970160.276.8237.067.6
1971204.293.6297.868.6
Source: Banco de la República.
Source: Banco de la República.

The export subsidy provided by this type of export prefinancing was, as already indicated, equivalent to the differential in interest rates existing between Colombian and foreign banks. After taking into account a service commission of 1½ per cent charged by local commercial banks, the spread between these interest rates is estimated to range between 4 and 8 per cent (see the Appendix). This system has no fiscal cost, but it has been largely responsible for a substantial increase in the short-term foreign liabilities of the Colombian banking system, the outstanding balance of which was estimated at about $194 million, on a net basis, at the end of 1971.

In an attempt to replace some of this foreign financing with domestic resources, in March 1967 the Government established the Export Promotion Fund (Proexpo), which was funded through an import surcharge of 1½ per cent to provide financing at low interest rates for working capital, as well as for expenses related to promotion and advertisement abroad, and other export-related activities.8 The growth of foreign export prefinancing, however, continued unabated through the first quarter of 1972, when Proexpo was provided with a special rediscount facility for the equivalent of US$30 million at the Bank of the Republic and was authorized to reduce its interest rates in an effort to make them competitive with those available abroad.9

The effective exchange rate for nontraditional exports

The term “effective exchange rate” for nontraditional exports is used in this study in its broadest sense, including any official financial measure that affects positively or negatively the peso income of the Colombian exporter. Defined in this fashion, any attempt to quantify the effective exchange rate for nontraditional exports in Colombia during the postwar period has to take into account not only the taxes that were applied to exports over part of the period but also the financial value of the system of incentives described in the preceding parts of Section I. The detailed calculations of the financial value of each of these incentives are presented in the Appendix. Table 6 summarizes the results of those calculations, for the years 1960–71, by showing the spread between the actual certificate market exchange rate and the effective exchange rate for both manufactured and other non-traditional exports, as well as the structure of the effective exchange rate. In 1971, for example, the average certificate market exchange rate was Col$20.00 per US$1, whereas the effective exchange rate for most nontraditional exports was Col$23.89 per US$1, that is, the effective rate was about 20 per cent higher than the actual exchange rate. However, the effective exchange rate applied to manufactured exports covered by the Vallejo Plan was even higher, at Col$26.17. Prior to 1961, the existence of export taxes and the absence of any export incentives meant that the effective return to exporters was less than the actual exchange rate. In 1961, this spread became positive, and between that year and 1964, the effective exchange rate for most non-traditional exports was about 14 per cent higher than the certificate market exchange rate. From 1965 to 1971, the average spread between these two rates increased to about 18 per cent, with an increasing trend in the last three years that originated mainly in the progressive reductions in the redemption term of CATs. Since the elimination of export taxes in 1962 and the start of export prefinancing in 1964, the structure of the effective exchange rate has remained unchanged, with only small variations in the relative importance of the different components.

Table 6.Colombia: Annual Averages of the Effective Exchange Rate for Nontraditional Exports, 1960–71
196019611962196319641965196619671968196919701971
Columbian pesos per U.S. dollar
Effective exchange rate for manufactured exports6.7610.3411.3612.3612.9118.8317.2519.1220.9621.9223.7626.17
Vallejo Plan0.910.950.920.931.551.521.551.681.832.002.28
Effective exchange rate for other nontraditional exports6.769.4310.4111.4411.9817.2815.7317.5719.3120.0921.7623.89
Basic exchange rate6.908.289.1110.0010.4114.7613.5014.5116.3017.3518.4920.00
Taxes–0.14–0.04–0.01
Tax credit certificates1.191.311.441.502.131.942.482.462.482.763.08
Special credit facilities0.070.390.290.580.550.260.510.81
Percentage distribution
Effective exchange rate for manufactured exports100100100100100100100100100100100100
Vallejo Plan98778988889
Effective exchange rate for other nontraditional exports1009192939392919292929291
Basic exchange rate1028080818179787677807876
Taxes–2
Tax credit certificates1112121111111312111212
Special credit facilities12233123
Source: Table 19 (in the Appendix).
Source: Table 19 (in the Appendix).

Regional trade and other institutional arrangements

Colombia’s efforts at regional integration and the promotion of regional and subregional free trade areas appear to have been successful. Table 7 presents Colombia’s nontraditional exports by regional groups and shows a rapid expansion of Colombian exports to the regional groups with which it is associated. This expansion has resulted in the fact that the overall share of exports to the Andean Group, the Latin American Free Trade Association (LAFTA), and the Caribbean Free Trade Association (Carifta) in total nontraditional exports has grown from about 20 per cent in 1968 to nearly 30 per cent in 1971. Within these groups, the Andean Group emerges as the most dynamic, with Colombian exports more than trebling over a four-year period.

Table 7.Colombia: Minor Export Registrations by Regional Groups, 1968–711(In millions of U. S. dollars)
1968196919701971
Total, world154.2203.6203.4233.8
Total, regional groups29.544.552.870.0
Andean Group15.928.337.949.4
Bolivia0.30.50.71.1
Chile3.53.86.88.5
Ecuador6.59.19.012.7
Peru5.614.921.427.1
Other LAFTA 2 members13.014.413.518.4
Carifta30.61.81.42.2
Regional groups as percentage of world total19.121.926.029.9
Source: Incomex.

Includes regional groups of which Colombia is a member and Carifta.

Latin American Free Trade Association (LAFTA).

Caribbean Free Trade Association (Carifta).

Source: Incomex.

Includes regional groups of which Colombia is a member and Carifta.

Latin American Free Trade Association (LAFTA).

Caribbean Free Trade Association (Carifta).

In addition to their efforts to promote nontraditional exports by means of financial incentives, the Colombian authorities have also made a number of institutional changes aimed at advancing the objectives of export promotion. The most important among these were the establishment of Proexpo; the creation of an export insurance mechanism; membership in the Andean Group and LAFTA, and efforts to associate with Carifta; the establishment of free ports within Colombia; the simplification of administrative procedures for foreign trade; and the improvement of port and other infrastructural facilities.

In addition to credit operations, Proexpo engages in other export promotion activities, such as providing exporters with information on foreign markets and technical assistance regarding transport, packing, advertising, etc. Proexpo also organizes export courses for managers and is responsible for the activities of commercial attachés stationed abroad. Finally, it is charged with organizing fairs in Colombia and abroad and with administering the recently created export insurance, which provides protection against political and other noncommercial risks.

Proexpo has given a broad interpretation to its responsibilities. It has undertaken the task of instilling an “export mentality” into the traditionally inward-looking Colombian entrepreneurs. It seeks to achieve this objective by several means that range from simple billboards (“exporting is the best business in Colombia”) to preparing four-year export plans that are drawn up in participation with the private sector.

The Colombian authorities have also established free ports at San Andrés, Barranquilla, and Cartagena. Although the experience with these free ports is mixed and as yet inconclusive, there are plans for setting up one or two additional free ports in the near future.

An area where noticeable institutional improvement has taken place, especially since 1967, is in the administration of exports. Currently, an exporter has to deal with three government institutions, whereas before 1967 this could involve as many as two dozen government agencies. Further improvement in this area is believed possible, and plans are being considered for creating a foreign trade ministry that would eliminate two of the three remaining agencies involved in the export process.

Finally, the objectives of export promotion have been given important consideration in public investment programs. As a result, the capacity of the four main seaports—Buenaventura, Barranquilla, Cartagena, and Santa Marta—has been greatly enlarged, and there are programs, for which financing from international lending agencies has already been obtained, to expand them further. These expansion programs, which relate not only to dock capacity but also to cargo-handling facilities, are coordinated with highway investments to improve access to the seaports. The most important highway plans are the project for the Cali-Buenaventura highway and the program to complete paved roads connecting Bogotà and Medellín with the ports located on the Caribbean coast.

II. The Growth of Minor Exports in Colombia

Minor exports in Colombia are currently defined to include all exports except coffee, petroleum, and raw hides. As noted in the previous section, this definition has varied from time to time for certain legal purposes; however, for the discussion of economic effects in this section, it will be applied uniformly over time. Defined in this way, minor exports can be measured on three different bases that correspond to separate stages in the export process. First, one can use registrations of minor exports, which correspond to the exporter’s advance declaration of future sales to Incomex, on the basis of which export licenses are obtained. Second, one can measure minor exports on the basis of exchange surrender, which arises from the delivery of foreign exchange to the Bank of the Republic by the exporter in advance of, or simultaneous with, the export sale. Finally, one can measure minor exports in more conventional terms, on a customs basis, which correspond to the actual shipment of goods in international trade. These data are reported, with some lag, by the National Statistics Office (Departamento Administrativo Nacional de Estadística-DANE).

The value of minor exports

The measurement of minor exports according to each of the above-mentioned sources is presented in Table 8 for the period 1960–71. The figures shown by the three measures have generally moved together, although in the period since 1967 the value of minor exports measured by exchange surrender has grown faster than the values measured by registrations and by customs flows. This widening divergence reflects predominantly the widespread use of export prefinancing, which was encouraged by the exchange reform measures in 1967, reviewed in Section I. What is not reflected in any of these measures is contraband or unrecorded border trade in minor exports; for purposes of balance of payments calculations, this category of trade in minor exports has been estimated at an average of about US$33 million over the past five years. The most important unrecorded item of export trade is probably cattle sold illegally to Venezuela. However, only data on recorded minor exports are used in this study. Furthermore, both exchange surrender and customs data are used in the analysis because of their availability on a monthly basis since the early 1960s.

Table 8.Colombia: Three Bases for Measuring Nontraditional Exports, 1960–71(In millions of U.S. dollars)
YearRegistrationsExchange

Surrendered
Customs

Basis
196032.930.352.3
196144.836.858.4
196252.654.070.8
196348.645.266.5
196471.273.278.9
1965113.498.8107.1
1966105.795.7107.7
1967119.2117.1126.3
1968154.2178.4170.5
1969203.6207.2206.9
1970203.4237.0202.3
1971233.8297.8239.6
Sources: Incomex; Departamento Administrativo Nacional de Estadística (DANE); Banco de la República.
Sources: Incomex; Departamento Administrativo Nacional de Estadística (DANE); Banco de la República.

In the context of Colombia’s balance of payments, the growth of minor exports has been an important development because it has reduced the country’s overwhelming dependence on receipts from coffee exports. For example, since the early 1950s Colombia’s trade account has revealed large swings that have been closely associated with changes in the international price of Colombian coffee. These prices have fluctuated within a broad range, from a record of nearly 80 U.S. cents a pound in 1954 to about 40 U. S. cents in 1963.10 Over time, however, the importance of changes in world coffee prices to Colombia’s balance of payments situation has declined because of the increasing share of minor exports in total export earnings. In the four-year period 1950–53, coffee exports (on a customs basis) accounted for 78 per cent of total commodity exports, while minor exports accounted for only 7 per cent (Table 9). By the period 1966–69, coffee’s share had declined to 62 per cent, while that of minor exports had increased to 28 per cent. In 1971, it is estimated that the share of minor exports had risen to nearly 35 per cent, while the share of coffee exports in total exports was down to 58 per cent. Also, in 1971, for the first time in modern Colombian history, the growth of minor exports (on an exchange surrender basis) more than offset the decline in coffee exchange earnings that resulted from the sharp fall in world coffee prices that year.

Table 9.Colombia: Value, Annual Growth Rates, and Percentage Shares of Commodity Exports, on a Customs Clearance Basis, 1950–69
Period Averages1950–53

to

1958–61
1958–61

to

1966–69
1950–53

to

1966–69
1950–531954–571958–611962–651966–691950–531966–69
Thousand U.S. dollarsAnnual growth ratesPercentage shares
Total491,461582,124458,191499,335545,814–0.92.20.7100.0100.0
Major exports455,902530,718410,984418,517392,972–1.3–0.6–0.992.872.0
Coffee384,472459,848338,959343,289336,498–1.6–0.1–0.878.361.7
Petroleum71,43070,87072,02575,22856,4740.1–3.1–1.514.510.3
Minor exports35,55951,40647,20780,818152,8423.615.89.57.228.0
Agricultural15,32926,72127,74946,83481,6167.714.511.03.115.0
Cotton15,8309,86719,61316.43.6
Sugar2061,3025,93313,04233.82.4
Bananas9,50221,09314,39113,73222,3565.35.75.51.94.1
Cattle1,0491,6913,6428.10.20.7
Tobacco2,0722,6262,5997,3755,5252.69.96.30.41.0
Other2,7062,7953,6278,23617,4383.721.612.30.63.2
Industrial3,5997,89913,55026,55458,81018.020.219.10.710.8
Fuel oils192,9987,7247,40414,9728.650.32.7
Cardboard boxes1132917,8091.4
Textiles2778254577,23610,8566.548.525.70.12.0
Portland cement4157221,9572,0232,33821.52.311.40.10.4
Pharmaceuticals2433083841,3502,2985.925.215.10.4
Other2,6453,0453,0158,25020,5371.627.213.70.53.8
Mineral11,78911,9359651,5355,605–36.824.5–4.82.41.0
Other4,8324,8514,9435,8956,8110.34.12.21.01.2
Sources: Banco de la República; DANE.
Sources: Banco de la República; DANE.

During the postwar period, the major growth in minor exports has occurred in the decade of the 1960s. In view of the changes in the export promotion system, as detailed in the previous section, this fact is not surprising. From the period 1958–61 to 1966–69, minor exports increased at an average annual rate of nearly 16 per cent, as against an increase of less than 4 per cent in the period 1950–53 to 1958–61 (see Table 9). And within the decade of the 1960s, the growth of minor exports was slightly higher in the second half than in the first. Again, this result is not surprising in view of the important changes introduced under Decree-Law No. 444 of 1967.

The commodity composition and destination of minor exports

As shown in Table 9, the commodity composition of minor exports covers a wide range of agricultural, industrial, and mineral goods. However, in recent years, six commodities have accounted for more than half of this total: bananas, cotton, fuel oils, sugar, textiles, and tobacco. Agricultural commodities have accounted for the bulk of the value of minor exports, but exports of industrial goods have grown at a significantly higher rate during the past two decades. Of the agricultural commodities, bananas, cotton, sugar, and tobacco have been the most important items, although in recent years export of cattle, seafood, and wood have registered significant increases in their shares. For industrial goods, most of the growth prior to 1960 was explained by increases in exports of fuel oil, while in the period since that time, cardboard boxes, textiles, pharmaceuticals, metal machines, and printed materials have been the leading industrial export items. Exports of minerals have been the least important of the three classes of minor exports, with their variation over time dominated by changes in exports of gold and emeralds.

The rapid growth of minor exports in recent years has been accompanied by changes in the relative importance of various foreign markets. Unlike the two major export items, coffee and petroleum, for which the United States has consistently accounted for more than half of foreign sales, minor exports have been distributed among many markets, with the most striking increase taking place in the Latin American region. As shown in Table 10, throughout the decade of the 1960s, Western Europe (especially the European Common Market) and the United States have been the principal markets for Colombian minor exports. However, the relative importance of the European market has declined significantly in the face of a strong growth in minor exports to the LAFTA region (especially the Andean Common Market). Although a detailed breakdown of minor exports on a customs basis is not available after 1969, these trends have continued since that year, as shown in the data for export registrations by country of destination (Table 11). For the three-year period 1969–71, the relative importance of the LAFTA market was even greater than that of the United States, and for the year 1971 alone, it was nearly as important as the Western European market. In general, it appears that whereas the North American and Western European markets have been the strongest centers of demand for the “traditional” minor exports, such as cotton, bananas, and sugar, the Latin American region has been an important market for newer, minor export items, such as textiles, metal products, electrical apparatus, machinery, and chemical products.

Table 10.Colombia: Destination of Minor Exports, on a Customs Basis, 1960–62 and 1967–69
Average
1960–621967–691960–621967–69
Thousand U. S. dollarsPercentage shares
Total60,510167,906100.0100.0
United States14,76944,60424.426.6
Canada7713,4321.32.0
CACM 11,4924,1842.52.5
LAFTA 27,01029,38511.617.5
Chile7483,4211.22.0
Ecuador1,0307,0171.74.2
Peru3,6637,7786.04.6
Bolivia143100.20.2
Venezuela1,2395,7562.03.5
Other3165,1030.53.0
Rest of Western Hemisphere4,60713,7527.68.2
Western Europe28,46362,09147.037.0
Netherlands4,91621,4628.112.8
United Kingdom6,02113,72610.08.2
Germany13,39012,79022.17.6
Other4,13614,1136.88.4
Rest of world3,39810,4585.66.2
Sources: DANE; Banco de la República.

Central American Common Market (CACM).

Latin American Free Trade Association (LAFTA).

Sources: DANE; Banco de la República.

Central American Common Market (CACM).

Latin American Free Trade Association (LAFTA).

Table 11.Colombia: Destination of Minor Exports, Based on Registrations, 1968–71
19681969197019711968196919701971Average,

1969–71
Million U. S. dollarsPercentage shares
Total154.2203.6203.4233.8100.0100.0100.0100.0100.0
United States43.146.350.750.028.022.724.921.423.0
Canada3.14.55.26.82.02.22.62.92.6
CACM 14.75.28.78.43.02.64.33.63.5
LAFTA 228.642.250.766.718.520.724.928.524.7
Chile3.53.86.88.52.31.93.33.62.9
Ecuador6.59.19.012.74.24.54.45.44.8
Peru5.614.921.427.13.67.210.611.69.8
Venezuela6.68.36.37.44.34.13.13.23.5
Other6.46.17.211.04.13.03.54.73.7
Rest of Western Hemisphere11.319.421.424.57.39.510.510.510.2
Western Europe55.070.663.368.035.734.731.129.131.6
Netherlands16.018.211.35.410.48.95.62.35.6
United Kingdom14.913.912.49.69.76.86.14.15.7
Germany8.29.213.218.05.34.56.57.76.2
Other15.929.326.435.010.314.512.915.014.1
Rest of world8.415.43.49.45.57.61.74.04.4
Source: Incomex.

Central American Common Market (CACM).

Latin American Free Trade Association (LAFTA).

Source: Incomex.

Central American Common Market (CACM).

Latin American Free Trade Association (LAFTA).

For most of Colombia’s minor exports, foreign demand can be said to be fairly price elastic, in marked contrast to the demand situation for coffee. Thus, the growth of minor exports has been governed more by conditions of domestic supply than of world demand. Domestic supply conditions tend to be somewhat different for agricultural and industrial exports: for the important agricultural commodities, domestic production is oriented toward the export market, whereas for many industrial goods, the export share of domestic production is very small. For the four principal agricultural commodities—bananas, cotton, tobacco, and sugar—exports accounted for at least 25 per cent of domestic production by volume over the period 1966–70; for bananas this proportion was as high as 43 per cent (Table 12). Moreover, for each of these products except cotton, the share of domestic production exported has remained fairly constant. Finally, for all these commodities gains in export volume since 1964 have been strong, as suggested by the composite export volume index for the four products presented in Table 12. Export volume by this measure in 1970 was nearly three times its level in 1964, whereas unit export values showed a decline of 10 per cent and a maximum variation of 22 per cent over the same period. Most other minor exports, such as cattle and industrial goods, appear to represent a rather small proportion of domestic production, although adequate data for a precise calculation generally are not available.

Table 12.Colombia: Minor Agricultural Exports in Relation to Domestic Production, 1966–70, and Volume and Value Indices, 1964–70
19661967196819691970Average,

1966–70
Thousand metric tons
Export volume
Bananas310.6325.6401.6334.5261.9326.8
Cotton4.530.046.658.860.140.0
Tobacco13.111.98.713.012.711.9
Sugar113.9176.5236.8145.3142.6163.0
Domestic
production
Bananas721.3749.1833.9780.7722.4761.5
Cotton75.496.6120.2125.3127.8109.1
Tobacco42.538.242.042.242.341.4
Sugar537.4596.6663.3708.9672.3635.7
Per cent
Share of
domestic
production
exported
Bananas43.143.548.242.836.342.9
Cotton6.031.138.846.947.036.7
Tobacco30.831.220.730.830.028.7
Sugar21.229.635.720.521.225.6
1964196519661967196819691970
Export volume index 1100.0132.5113.7211.3279.1282.3281.2
Unit export value index2100.088.879.178.289.098.090.6
Source: DANE.

Weighted average of bananas, cotton, tobacco, and sugar with average unit values for 1964–70 as weights.

Weighted average of bananas, cotton, tobacco, and sugar with average volume for 1964–70 as weights.

Source: DANE.

Weighted average of bananas, cotton, tobacco, and sugar with average unit values for 1964–70 as weights.

Weighted average of bananas, cotton, tobacco, and sugar with average volume for 1964–70 as weights.

The sensitivity of minor exports to exchange rate variation

Some theoretical notes

As pointed out at the beginning, the principal intent of this study is to determine the extent to which the remarkable growth of minor exports during the past five to ten years may be explained by the export promotion system that has been developed since 1960. The discussion in Section I has shown that exchange rate policy has been the primary component of this system, supplemented by the use of tax credit certificates, duty-free import schemes, and special credit facilities.

In the analysis that follows, three main factors were considered in determining the effect of the export promotion system on minor exports. First, the effective exchange rate, as calculated in the previous section, was used as a measure of the “true” return in pesos for a dollar’s worth of exports under the export promotion system. However, since this return from selling in the foreign market must be compared with the corresponding return from selling in the domestic market, the effective exchange rate was deflated by an index of wholesale prices in Colombia. This index of wholesale prices, which is considered by the Colombian authorities as the best indicator of changes in domestic costs, in turn was deflated by an index of export prices in the United States to reflect changes in Colombia’s competitiveness vis-à-vis foreign markets. Such a model can be expressed in the following functional form:

Where

Ex =minor exports on a customs basis in U. S. dollars, f.o.b.
ER =the effective exchange rate in Colombian pesos per U.S. dollar
P =ratio of nonfood wholesale prices in Colombia to export price index in the United States

An important assumption implicit in the formulation of this model is that foreign demand for Colombian minor exports is perfectly price elastic, which implies that world export prices are determined exogenously. This assumption seems justified for Colombian minor exports in light of the relative unimportance in world trade of any single minor export item. Whereas Colombia’s coffee exports account for nearly 15 per cent of the value of world trade in this commodity, its minor agricultural exports, such as cotton, sugar, and tobacco, have a share of less than 1 per cent. Only for bananas is this share somewhat larger, for example, 4.3 per cent for the period 1966–70. 11

Obviously, this model cannot represent a complete explanation of the growth of minor exports, especially in the long run. For such an analysis, one would have to consider a number of factors, including weather variation, technological improvement, and changes in domestic consumption patterns. For lack of convenient measures of many of these variables, we have concentrated in this study on the export promotion system, as it seemed most amenable to systematic analysis. It might also be argued that the model should include, as a supply function, some measure of capacity constraint. However, for most of the period under study, it has been assumed that such a constraint did not exist, in view of Decree-Law No. 1 of 1959 and Law No. 81 of 1960, which were designed specifically to promote the utilization of excess productive capacity. Finally, value data for minor exports were used instead of a volume index because no suitable quantum measure or deflator is available for minor exports in Colombia.

In broad outline, the model formulated here is similar to that used in previous studies of minor export growth in Colombia by Sheahan and Clark and by Urdinola and Mallon.12 However, significant differences lie between this work and previous studies in the calculation of the effective exchange rate, which takes account not only of the CATs but also of the duty-free benefits under the Vallejo Plan and of the special credit facilities. In addition, this study has attempted to discriminate between the effect of exchange rate variation for both agricultural and industrial minor exports, and to assess the relative importance of the various components of the export promotion system on the performance of minor exports.

Some empirical results

The theoretical model just outlined was tested empirically with annual data for the years 1948–71 and with quarterly data for the years 1960–71. Minor exports were measured on a customs basis with annual data, and on both a customs basis and an exchange surrender basis with quarterly data. All data were transformed into natural logarithms, and estimates were derived on the basis of ordinary least-squares technique. 13

With annual data, the model was tested with the real effective exchange rate variable as a single regressor, yielding the following results: 14

As revealed in the statistics, the exchange rate variable alone has a high degree of explanatory power. However, the statistical significance of the regression coefficient is likely to be overstated as suggested by the low value of the D-W statistic of serial correlation. Most probably, too, the regression coefficient itself is biased upward because of possible misspecification in the model, arising from the exclusion of other important explanatory variables. At any rate, the size of the coefficient suggests some degree of sensitivity of minor exports to variations in the real effective exchange rate, as measured in this study. 15

In an effort to improve the results of the first test, a second test was tried incorporating a dummy variable for the influence of LAFTA on minor export growth. The role of this regional arrangement was described in Section I. As a measure of this variable, the share of total Colombian exports to the LAFTA region was used. This variable has a value of zero for the years 1948–61 and then rises from 1.6 (per cent) in 1962 to an estimated 10.2 (per cent) in 1971. The results of this test represent a substantial improvement over those of the previous one, as shown below:

With the addition of this variable, the total explanatory power of the regression increases substantially (i.e., higher R2). This result probably gives a more accurate measure of the elasticity of minor exports with respect to the real, effective exchange rate than in equation (1) because of the improvement in the D-W statistic. Notably, this elasticity is greater than unity, which suggests that a positive variation of 1 per cent in the effective exchange rate, corrected for relative price changes, yields a 1.3 per cent change (of the same sign) in minor exports.

An attempt was also made to test ER and P separately as regressors in the model, in an effort to distinguish between the effects of the effective exchange rate alone and the ratio of domestic prices to foreign prices. While these results were marred by problems of multicollinearity, they did show that the effective exchange rate was substantially the more important of the two variables:

The sign of the P variable is negative, as expected, although its statistical significance is somewhat weak. Also, the estimated coefficient of the ER variable indicates that the elasticity of minor exports with respect to the effective exchange rate alone is about 1.2, which is slightly less than the estimate for the real, effective exchange rate shown in equation (2).

As a further refinement, minor exports were also defined as excluding both bananas, on the grounds that this commodity was excluded from the benefits of the export promotion system throughout much of the period under study, and gold, since the highly erratic movements in the export of this commodity were related more to technological or accidental phenomena than to changes in export promotion policies. The elasticity of adjusted minor exports (EXX) is substantially higher than before, as shown in the following results:

One final test was applied to the annual data in an effort to measure the relative importance of the various components of the effective exchange rate on minor exports as presented in Table 4. In this test, each of the components of the effective exchange rate (except that for the Vallejo Plan) was used as an independent variable in the regression analysis to explain variation in minor exports. This is an imperfect test at best; however, it was hoped that this procedure would give some indication of the relative weight of the various elements in the export promotion system on the growth of minor exports. The results were as follows:

Where

ERT =nominal exchange rate applied to minor exports including export taxes
CAT =financial value of fiscal incentives, measured in exchange rate equivalent
SCF =financial value of special credit facilities, measured in exchange rate equivalent
ER =ERT + CAT + SCF
LAFTA =share of total Colombian exports to LAFTA
P =ratio of nonfood wholesale prices in Colombia to export price index in the United States

In this test, the exchange rate itself is overwhelmingly the most important explanatory variable in terms of its statistical significance. It would seem that these results would give some tentative support to the notion that the use of exchange rate adjustment, as distinct from other elements in the export promotion system, has been the most important influence on minor export growth.

The tests just described were also performed with quarterly observations for the years 1960–71. By and large, the results that we have shown were confirmed. However, within the time period chosen, marked differences occurred in the regression analysis using data before and after 1967. This date was selected as a breaking point precisely because of the important changes in the export promotion system introduced early in that year. Important differences also showed up depending on which measure of minor exports was used. The best results were given for minor exports as measured by exchange surrender (ES):

Significantly, one notes a large jump in the value of the regression coefficient of the exchange rate variable from one period to the next, along with an improvement in the degree of statistical significance. This result would suggest a much greater response on the part of minor exports to the operation of the export promotion system since 1967 than under the previous arrangements. Notably, too, regional trade factors, as measured by the LAFTA variable, have an important weight in explaining the growth of minor exports during both periods.

When each of the components of the effective exchange rate was tested separately as an independent variable, the results showed once again the relative importance of the exchange rate variable. The test results for the second period were far superior to those of the first:

Finally, each of the major subgroupings of minor exports, agricultural and industrial, was tested for its sensitivity to movements in the real, effective exchange rate. For manufactured exports, the effective exchange rate (ER*) was defined to include the component for the Vallejo Plan, which was not available for exports of nonmanufactured goods. The analysis was performed with quarterly data, on the basis of exchange surrender, for the entire period for which such a breakdown is available:

As expected, the real, effective exchange rate has a strong degree of explanatory power for both groups of minor exports. Surprisingly, the LAFTA variable is not as significant here as over all, but this failing is explained to some extent by problems of multicollinearity. Given the importance of LAFTA for Colombia’s exports of manufactured goods, one would have expected a much more significant contribution from this variable.

On the basis of these regressions, one can conclude that minor exports have been fairly sensitive to variations in the real, effective exchange rate, especially in the period since the major exchange reforms of 1967. Moreover, the tests that we have applied suggest that the exchange rate itself has been the most important instrument in the export promotion system, as distinct from such other elements in the system as the CATs or special credit facilities. For a more rounded view of the export promotion system, however, the benefits in terms of the growth of minor exports must be seen against the costs of operating the system. The fiscal costs of the system are considered in the following section.

III. The Fiscal Cost of the System of Incentives for Nontraditional Exports

The general case

The need to promote the growth and diversification of exports arises generally from a fundamental disequilibrium in the balance of payments, whether current or projected. Only a few times was a drive for export promotion and diversification initiated solely with domestic objectives, such as employment, in mind. A fundamental disequilibrium in the balance of payments may be corrected most easily by adjusting the exchange rate. In practice, however, several countries have chosen to combine the effect of an exchange rate adjustment with other financial incentives, or have relied entirely on the latter. There are, then, three different, possible combinations of policies: (1) adjusting the exchange rate without introducing other incentives to export; (2) adjusting the exchange rate and introducing other incentives; and (3) introducing other incentives to export without adjusting the exchange rate.

Other policies remaining unchanged, an adjustment in the exchange rate, at the same time that it improves the international price relationship and contributes to the promotion and diversification of exports, will increase the cost of imports—thereby tending to reduce the demand for imported goods and to equilibrate the balance of payments. This result is usually an important effect of currency devaluation, as it will help to accelerate the correction of the disequilibrium in the balance of payments. However, promoting exports by combining exchange rate adjustment with the introduction of other financial incentives usually needs to be complemented, at least in the initial stages, by measures that curtail the demand for imports through other than price effects. Until the export promotion effort begins to bear fruit, the need to reconcile the demand for imports with the availability of foreign exchange under this policy combination will force the authorities of the country concerned to rely on quantitative restrictions and administrative controls. This implies manual direction of the resource allocation process rather than reliance on the market mechanism. In addition to the probability of introducing substantial distortions in resource allocation, this alternative requires the existence of ample domestic resources, usually of a budgetary nature, to finance the cost of the system of financial incentives. In the absence of these resources, the system of financial incentives will require recourse either to central bank credit or to external financing, with the latter, in turn, leading to an increase in the weight of foreign or domestic debt service, or, alternatively, to a postponement of required domestic expenditure.

The third alternative, forgoing any exchange rate adjustment in order to promote exports exclusively with other financial incentives, means magnifying the need to rely on quantitative restrictions to control the demand for imports and to mobilize domestic resources to finance the cost of the incentives.

The Colombian experience

Since 1960, Colombia has chosen the second policy alternative, that is, combining an active exchange rate policy with the introduction of other financial incentives, principally of a fiscal nature, with small variations in the system to gradually improve its effectiveness. Especially since 1967, rather than attempt to reach an exchange rate that would be sufficient to promote nontraditional exports and to control the demand for imports, Colombia has followed a policy of gradual exchange rate adjustment together with large fiscal subsidies—through the CAT, the Vallejo Plan, and other customs duties forgone—and substantial reliance on quantitative restrictions and administrative controls to reconcile the demand for imports with the availability of foreign exchange. Since this policy mix was decided upon, its management has been quite successful, as seen in the preceding sections. Indeed, nontraditional exports, in terms of exchange surrender, expanded at the quite satisfactory annual rate of 21 per cent, on average, from 1960 to 1971. During this same period, the supply of imports increased to levels that permitted an acceleration in the growth of gross domestic product (GDP) in the late 1960s and early 1970s, from an average rate of less than 5 per cent in real terms between 1950 and 1967 to about 6 per cent in 1968–69 and to about 7 per cent between 1970 and 1972. Finally, the net international reserves held by the Bank of the Republic changed from an average negative position of about US$130 million over the first half of the 1960s to a positive amount of approximately US$170 million at the end of 1971, for a cumulative gain of US$300 million. At the end of 1971, this amount was equivalent to about three months’ import payments.

These indicators of success, however, do not reveal the real economic costs of the system. Such costs ideally would be expressed in terms of the direction of resource allocation in the economy. While it is nearly impossible to measure the real effects on resource allocation of the system of financial incentives and import controls, an indirect attempt can be made to quantify these effects in terms of the fiscal costs of the system of export incentives. In this exercise these costs can be divided into direct and indirect components.16 The direct cost of the system is represented by the revenues forgone under the tax exemption of Law No. 81 of 1960, or through the issue of CATs and by the customs duties forgone under the Vallejo Plan, calculated at prevailing exchange rates. These estimates are shown in Table 13. For CATs, the fiscal cost is represented simply by the amount of them that were received by the Ministry of Finance in payment of taxes. For years prior to 1968, the value of the tax exemptions under Law No. 81 of 1960 has been estimated by applying the effective fiscal subsidy of 14.4 per cent, implicit in the provisions of the Law as discussed previously and in the Appendix, Calculation of financial value of fiscal incentives (f), to the gross value of exports converted into Colombian pesos at the annual average exchange rate applicable to minor exports.17 A lag of one year is also allowed, to take account of normal delays in the payment of corporate income taxes.

Table 13.Colombia: Calculation of the Fiscal Cost of Export Incentives, 1960–71 and Projections for 1972
196019611962196319641965196619671968196919701971Projections’

1972
Million Colombian pesos
I.Total fiscal cost of export incentives (II + III)113443982933477307435436268471.2611.6432.825
II.Direct costs (1 + 2 + 3)24366661971861871943795226701,360
1. Income tax exemption (Law 81 of 1960) 12436635478148133
2. Taxes paid with CATs1383314245451,200
3. Customs duties forgone under the Vallejo Plan (VP) (b) × (c)37193854564898125160
a. VP imports0.2 20.2 22.2 25.1 29.8 212.0 217.0 217.9 213.7 224.7 226.0 228.0 2
b. VP imports at effective import exchange rate311194696154251295240458520614
c. Average customs duty rate (per cent)421.020.818.416.115.920.224.421.419.020.121.424.026.0
III.Indirect costs (1 + 2)113203622272866335573564324687399731.465
1. Other customs duties forgone under VP (c) × (d)131713161512273954
a. Effective exchange rate for non-traditional exports56.7610.3411.3612.3612.9118.8317.2519.1220.9621.9223.7626.1729.34
b. Effective import exchange rate36.606.706.828.809.009.8112.8414.7916.4617.5018.5320.0021.94
c. Exchange rate spread (a – b) ÷ (b)0.020.540.670.400.430.9260.340.290.270.250.280.310.34
d. Direct costs of VP (II. 3)37193854564898125160
2. Other customs duties forgone (a) × (b)113203622262836165443404174567129341,411
a. Actual customs duties collection5505925405656596701,6001,1731,5431,8252,5443,0134,150
b. Exchange rate spread (III.1.c.)0.020.540.670.400.430.9260.340.290.270.250.280.310.34
Per cent
IV.Total fiscal cost of fiscal incentives100100100100100100100100100100100100100
1. Direct costs7923181325343145424148
a. Income tax exemption792216102024
b. Taxes paid with CATs2239343343
c. Customs duties forgone under VP12351096885
2. Indirect costs100939177828775666955585952
a. Other customs duties forgone under VP22321222
b. Other customs duties forgone100939177828573636754565750
Sources: Ministry of Finance; Banco de la República.

Calculated as 14.4 per cent of the value of minor exports (excluding coffee, petroleum, bananas, and raw hides) of the previous year, converted at basic exchange rate applicable to most nontraditional exports, as given in the text and in the Appendix.

In millions of U.S. dollars.

Effective import exchange rate calculated by dividing total imports in Colombian pesos by total imports in U. S. dollars.

Sources: Ministry of Finance; Banco de la República.

Calculated as 14.4 per cent of the value of minor exports (excluding coffee, petroleum, bananas, and raw hides) of the previous year, converted at basic exchange rate applicable to most nontraditional exports, as given in the text and in the Appendix.

In millions of U.S. dollars.

Effective import exchange rate calculated by dividing total imports in Colombian pesos by total imports in U. S. dollars.

It has been assumed that indirect costs arise from the difference between the nominal and the effective exchange rates for nontraditional exports. For example, in the Vallejo Plan, the indirect costs can be estimated by applying the effective exchange rate instead of the nominal rate for nontraditional exports to imports under the scheme, as was done in estimating the indirect costs. A similar calculation is also made for actual customs duties to derive an estimate of the revenue that is forgone, in general, in using an exchange rate that is more appreciated than the effective exchange rate.

Implicit in these calculations of the indirect fiscal cost of export incentives is the assumption that the level of imports, both under the Vallejo Plan and subject to duties, would be the same at a more depreciated exchange rate as at the rate that actually prevailed. The plausibility of this assumption, in turn, rests on the judgment as to whether or not the restrictive effect on imports of the effective exchange rate for nontraditional exports would be the same as that of the import control system as it actually operates. Unfortunately, it is difficult to reach any conclusions on this question, as there was no significant period in the past decade during which a sufficient proportion of imports was freed from quantitative restrictions and administrative controls so that the elasticity of import demand to changes in the exchange rate could be estimated.18 Nevertheless, despite this difficulty, the foregoing assumption has been accepted with the understanding that the estimated effects may somewhat overstate the actual indirect costs of the export incentive system.

Overall cost and impact of the system of financial incentives

Table 13 shows the results of calculating the fiscal cost of the Colombian system of export incentives under the assumptions that have just been formulated. Two additional points should be considered concerning those results. The first is that no attempt has been made to estimate the economic benefits derived from the export promotion program—in terms of additional income of the export sector and its multiplier effect on the rest of the economy, increased import capacity, better utilization of installed capacities, additional employment opportunities, induced investments, etc.—because they are already built into the system, and their effects, therefore, are already reflected in the Government’s actual fiscal operations. The second point is that the method followed underestimates the actual fiscal cost by not taking into account either the additional income tax revenue that would have accrued to the Government if the CATs had not been tax exempt or the expanded base for applying the sales tax that would have resulted from a higher exchange rate.

The results contained in Table 13 make it apparent that the system of fiscal incentives for export promotion has had a substantial fiscal cost, and that this burden is growing heavier as the export promotion effort succeeds in reaching its goals. The table shows, furthermore, that the greater fiscal cost is related to the indirect component, which implies that the largest fiscal burden of the export promotion system is completely unrelated to the objective of promoting exports. Since 1968, however, after the introduction of CATs, the direct costs of the export incentives have been growing steadily. Because of the shortening of the redemption period of CATs, it is estimated that in 1972 the direct and indirect fiscal costs were nearly equal.

Perhaps the impact of the fiscal cost of export incentives can best be obtained by placing it in the context of the Central Government’s overall cash operations. This is done in Table 14, which shows the Central Government’s cash operations over the past 12 years with estimates for 1972. Except for 1960 and 1966, the Government’s operations showed consistently large deficits that were financed almost exclusively with external development loans, except for a few years when the Government had to rely on central bank financing. The impact of the export incentives on the Government’s fiscal operations can be analyzed from three different perspectives.

Table 14.Colombia: Fiscal Impact of Export Incentives in the Context of the Government’s Overall Cash Operations, 1960–71 and Projections for 1972(In millions of Colombian pesos)
196019611962196319641965196619671968196919701971Projections,

1972
Central government cash operations
Total revenue including CATs2,1322,2152,1483,1073,8824,0076,1396,7888,1959,80112,37414,96116,955
Taxes paid through CAT redemption12436635478148133138331424545930
Total current revenue2,1322,1912,1123,0443,8283,9295,9916,6558,0579,47011,95014,41616,025
Current expenditure1,4121,5151,9142,5092,8783,0104,0704,0914,9026,1987,9199,79611,330
Surplus on current account7206761985359509191.9212,5643,1553,2724,0314,6204,695
Investment expenditure–6831,5321,2431,2961,5631,3301,9022,8283,8664,3445,4426,2827,395
Overall surplus or deficit (–)37–856–1,045–761–613–41119–264–711–1,072–1,411–1,662–2,700
Impact of export incentives
Total cost of export incentives113443982933477307435436268471,2611,6432,825
Surplus or deficit after export incentives48–512–647–468–266319762279–85–225–150–19125
Export incentives as percentage of current revenue116191091912889111118
Export incentives as percentage of current account surplus25120155377939212026313660
Sources: Banco de la República; Fund staff estimates.

Figures for 1961–67 reflect estimate of income tax exemption.

Sources: Banco de la República; Fund staff estimates.

Figures for 1961–67 reflect estimate of income tax exemption.

First of all, had the Government decided to promote exports by eliminating the overvaluation of the peso, its financial position would have been radically different, especially since 1966. Indeed, since that year, instead of recording large deficits, the Central Government’s cash operations would have been practically in equilibrium through 1971, with substantial surpluses in 1966, 1967, and 1972. This would have permitted the Government either to have forgone its heavy and growing dependence on foreign borrowing or to have satisfied a substantial part of the economic and social development needs that it cannot meet under current conditions.

Second, viewed in terms of actual central government revenue, export incentives have been an important factor in accounting for the low buoyancy of the system, despite substantial tax measures in recent years in the form of new taxes, higher tax rates, and administrative improvement. Indeed, as a proportion of GDP, central government revenue increased from 8 per cent in 1960 to only 9.6 per cent in 1971. The latter coefficient would have been 10.7 per cent if the fiscal incentives had accrued to the Treasury instead of being paid out as subsidies to exports and imports.

The fiscal effect of the system of export incentives may also be measured as a proportion of central government revenue (see Table 14). In the period 1961–66, the proportion of the fiscal export incentives in total revenue fluctuated within a wide range of 9 to 19 per cent, reflecting primarily changes in the indirect cost of export incentives. Since 1967, however, this proportion has risen steadily from 9 to 18 per cent, as the weight of direct costs has increased; the high estimate for 1972 reflects principally the increased fiscal cost implicit in the reduction of the redemption period of CATs in late 1971. The conclusion is unavoidable that the proportion of central government revenue that is absorbed by fiscal export incentives will continue to increase, should the nontraditional export promotion effort continue to progress satisfactorily—and recent indicators suggest that it is even gaining momentum.

Finally, when the fiscal cost of the system of export incentives is compared with the Central Government’s current account surplus, the growing cost of the system means that between 1966 and 1971 an average of almost 30 per cent of the Government’s savings available to finance investment expenditure was absorbed by export incentives; moreover, after 1972 one half or more of the Government’s savings will be diverted from investment expenditure to export subsidies.

IV. Conclusions and Policy Implications

Since the beginning of the past decade, Colombia has sustained an impressive increase in nontraditional exports of approximately 15 per cent on an average annual basis. This growth performance raised the share of nontraditional exports from 11 per cent in 1960 to nearly 35 per cent in 1971; preliminary estimates for 1972 indicate that this share was even higher. During this same period, the Colombian authorities gradually developed a comprehensive system of export incentives that was implemented primarily through exchange rate policy but also included fiscal incentives, special import-export arrangements, and special credit facilities. The origin of this system, at least insofar as exchange rate policy is concerned, can be traced back to the early postwar years; however, the system did not really develop fully until the passage of Decree-Law No. 444 in 1967. This Decree-Law established the legal basis for the “adjustable-peg” exchange rate system and the CATs that are in operation today.

This study has attempted to estimate quantitatively the impact of the export promotion system by calculating an effective exchange rate for nontraditional exports that takes into account the financial value of each of the various components of the system. During the period 1961–71, this effective exchange rate for most nontraditional exports on average was about 17 per cent above the nominal exchange rate, although this spread varied as changes were made in the instruments of the export promotion system.

As a result of statistical tests based on regression analysis, the real effective exchange rate was found to be very significant in explaining variation in nontraditional exports with samples of both annual and quarterly data. Moreover, when each component of the real effective exchange rate was considered separately in the regression analysis as an explicit explanatory variable, the nominal exchange rate (adjusted for export taxes) proved to be the most important regressor.

The effects on resource allocation of the export promotion policy were examined in terms of revenues forgone in the central government budget. While the direct costs of the system, especially in the use of CATs, have been growing sharply in recent years along with the rapid growth of nontraditional exports, the heaviest fiscal burden of the system arises from the indirect effects of import duties forgone in applying an exchange rate that overvalued the Colombian peso in relation to the effective exchange rate implicit in the export promotion system. In the past three years, the combined direct and indirect costs of the system have amounted, on average, to about 10 per cent of central government revenues and to about 30 per cent of the current account surplus on a cash basis. In 1972 alone the total fiscal cost of the export promotion system is estimated at about Col$2.8 billion, compared with an estimated overall cash deficit of Col $2.7 billion.

The conclusions of this study raise important questions about the future of the export promotion system in Colombia. In view of the substantial and growing fiscal cost of the system and of the prospects for future growth of minor exports, as suggested in the Government’s recent four-year plan (1972–75),19 clearly some way must be found to relieve the burden of the system on the central government budget so as to free resources for the pursuit of other objectives of economic and social development. The most effective solution to this problem would seem to be increasing reliance in the future on exchange rate adjustment as the principal instrument for export promotion. This recommendation would seem appropriate, especially in view of some of the statistical results given in Section II. By increasing the rate of currency depreciation, the Colombian authorities would be able to reduce the role of the other instruments in the export promotion system without reducing the overall financial value of the system to exporters of nontraditional goods. Of course, an immediate increase in the rate of currency depreciation and elimination of the other instruments of export promotion might not be feasible in the short run, but change could be introduced gradually over time. For example, one way to begin might be to slowly reduce the value of CATs from 15 per cent to 12 per cent to 10 per cent, etc., compensating each adjustment in the financial value of CATs by an increase in the rate of currency depreciation so as to leave the effective peso return to minor exporters unchanged. One convenient byproduct of this policy arrangement would be that, with the increase in exchange rate adjustment, less dependence on direct import-control mechanisms would be needed, as an increased value of the exchange rate would create its own financial restraint on imports.

By relying more and more on exchange rate adjustment in the promotion of minor exports, the Colombian authorities would be able to reduce the present fiscal cost of the export promotion system, without penalizing the growth of minor exports, and to simplify the exchange rate system. The demands of balanced and efficient resource allocation would indicate that both of these objectives should be pursued.

APPENDIX:

Calculation of the Effective Exchange Rate for Minor Exports

The effective exchange rate for minor exports discussed in Section I can be defined in the following way:

ER* = ER (1 + e + f + p + c)

Where

ER* =effective exchange rate for minor exports in Colombian pesos per U. S. dollar
ER =nominal exchange rate for minor exports in Colombian pesos per U. S. dollar
e =export taxes on minor exports in per cent
f =financial value of fiscal incentives under Decree-Law No. 81 of 1960 or under CATs, in per cent
p =financial value of the Vallejo Plan, in per cent
c =financial value of special credit facilities, in per cent

The calculation of each of these parameters, e, f, p, and c, will be considered separately.

Export taxes (e)

The estimate of the parameter e requires no special calculation, as the rates are given directly by law. In 1957, from June 17 until September 25, a rate of 15 per cent was applied to all exports, both major and minor. After September 25, 1957 this rate was reduced to 2 per cent for all exports other than coffee, bananas, and precious metals. On March 7, 1961 the 2 per cent rate was reduced to ⅛ of 1 per cent, and on October 24, 1962, it was eliminated altogether.

Calculation of financial value of fiscal incentives (f)

The financial value of the fiscal incentives provided under Law No. 81 of 1960 can be calculated in a fairly straightforward fashion, as discussed in the second paragraph of the section, The Colombian experience, by applying a marginal tax rate of 36 per cent20 to the ratio of export profits to gross value of exports assumed in the Law, which is 40 per cent. This calculation yields an effective fiscal subsidy of 14.4 per cent of the gross value of exports.

For CATs, the calculation of the financial value of the fiscal incentive requires the consideration of two cases: one for those exporters who choose to hold the CAT until the end of its redemption period, and the other for those exporters who choose to sell the CAT in the stock exchanges. For those who hold the CAT, the fiscal incentive, f1, can be defined as

Where

CAT =financial value of tax credit certificate as a percentage of export value (i.e., 15 per cent)
t =marginal income tax rate in per cent
n =number of months the CAT is held before redemption
r =prevailing rate of interest, or opportunity cost of holding cash balances

Calculated in this way, the financial value of the CAT in the first case is positively correlated with increases in the tax rate, t, and negatively correlated with increases in the period of redemption, n, and the rate of interest, r.

In the second case, where the exporter sells the CAT in the stock exchanges, the fiscal incentive, f2, can be defined as

f2 = CAT (1–d)

Where

CAT =financial value of tax credit certificate as a percentage of export value (i.e., 15 per cent)
d =discount rate, in per cent, of transactions in CATs of differing maturities

Thus, the financial value to the exporter of the CATs transacted in the stock exchanges is only the 15 per cent basic rate less the discount at which the CATs are sold. The benefit of the income tax exemption does not accrue to the exporter but rather to whoever acquires the CAT in the market.

As a result, the calculation of the effective financial value of CATs, f, is equivalent to a weighted average of the financial value of CATs in the two cases just described, f1 and f2. The weights used are the amounts of CATs sold in the stock exchanges and held by the exporter. This calculation is made in Table 15. During 1967, when sales of CATs in the market were minimal, the effective financial value of CATs substantially exceeded their face value (15 per cent) because of the strong effect of the tax exemption. From the first half of 1968 until the first half of 1970, the combination of a long redempion period with a high proportion of sales in the market reduced the effective financial value of CATs to about 14 per cent. When the authorities began to shorten the redemption period of the CATs in 1970 and 1971, the weighted effective financial value of CATs once again exceeded the face value (15 per cent).

Table 15.Colombia: Calculation of Effective Financial Value of Cats to the Exporter, by Quarters, 1967–71
19671968196919701971
IIIIIIIVIIIIIIIVIIIIIIIVIIIIIIIVIIIIIIIV
Million Colombian pesos
CATs issued37.163.758.977.990.479.084.985.4114.4132.4112.4108.8135.0126.2136.8133.7206.5206.7159.8
Sold in stock exchange10.333.950.261.275.080.187.3128.583.4107.6121.279.693.3110.9126.5150.3133.5
Held by exporter37.163.748.644.040.217.89.95.327.13.929.01.213.846.643.522.880.056.426.3
As percentage of CAT
CATs sold1743557788947697749990636883617383
CATS held1001008357452312624326110373217392717
Average price of CATs sold (%)80.1189.6887.5389.0090.0090.3590.4090.5590.6392.9093.0992.3893.2494.3091.6995.3395.43
As percentage of export value
Effective financial value of CATs
Sold by the exporter12.013.513.113.413.513.513.613.613.613.914.013.914.014.113.814.314.3
Held by the exporter18.318.318.318.318.318.318.318.318.318.318.318.318.318.818.818.818.818.819.5
Weighted average18.318.317.216.215.414.514.113.814.713.714.813.914.415.715.514.915.815.515.2
Sources: Ministry of Finance; Banco de la República; Fund staff estimates.
Sources: Ministry of Finance; Banco de la República; Fund staff estimates.

Financial value of the Vallejo Plan

An exporter using the Vallejo Plan is entitled to automatic import licensing upon request and to exemption from payment of both customs duties and advance import deposits. Algebraically, the financial value arising from these exemptions can be stated as follows:

Where

μ =average imported component of manufactured exports under the Vallejo Plan
δ =average effective import duty, in per cent
π =average effective rate of advance import deposits, in per cent
n =average number of months during which advance import deposits are held at the Bank of the Republic
r =prevailing rate of interest, or the opportunity cost of holding cash balances

As an estimate of μ (the average imported component of manufactured exports under the Vallejo Plan), the average ratio of imports to exports under the Plan during the past ten years (see Table 4) is 0.35. The average effective import duty, δ, was calculated on the basis of annual statistics of imports and import duties, as shown in Table 16. These calculations show that the average effective import duty in Colombia has fluctuated around 21 per cent. The average effective rate of advance import deposits, π, was computed as the weighted average of the deposits actually held by the Bank of the Republic. These calculations are shown in Table 17. The term for these deposits, n, was taken to be about ten months, based on information regarding the actual administrative arrangements for operating the advance import deposit requirement.21 The financial value of the Vallejo Plan arising from the exemption from import duties and advance import deposits, p, is calculated in Table 18. During the period 1960–71, the average financial value of the Vallejo Plan was about 10.5 per cent.

Table 16.Colombia: Calculation of Average Import Duty, 1960–71
Imports,

f.o.b.

(million

U.S.

dollars)
Exchange

Rate

(Colombian

pesos per

U.S. dollar)
Imports,

f.o.b.
Dutiable

Imports1
Duties

Collected
Average

Effective

Import

Duty

(per cent)
Year(million Colombian pesos)
1960496.46.603,2762,62055021.0
1961530.86.703,5562,84559220.8
1962536.96.823,6622,93054018.4
1963497.58.804,3783,50056516.1
1964575.49.005,1794,14365915.9
1965423.59.814,1543,32367020.2
1966638.912.848,2046,5631,600224.43
1967464.314.796,8675,4941,173 421.43
1968615.116.4610,1258,1001,543 419.0
1969647.917.5011,3389,0701,825 420.1
1970802.018.5314,86111,8882,544421.4
1971784.020.0415,71112,5693,013 424.0
Source: Banco de la República.

Dutiable imports in this table are 80 per cent of total imports, f.o.b. A National Planning Department study in 1969 covering the years 1967 and 1968 showed that the imports exempt from customs duties amounted to slightly less than 20 per cent of total imports. Included under those exempt from customs duties were diplomatic imports, imports under the Vallejo Plan, imports by public sector agencies, and imports financed by long-term international credits and grants.

Excludes some Col$300 million collected on automobiles.

Reflects temporary increases in duties in connection with import liberalization program.

Includes customs duties paid with CATs and 3 per cent import surtax.

Source: Banco de la República.

Dutiable imports in this table are 80 per cent of total imports, f.o.b. A National Planning Department study in 1969 covering the years 1967 and 1968 showed that the imports exempt from customs duties amounted to slightly less than 20 per cent of total imports. Included under those exempt from customs duties were diplomatic imports, imports under the Vallejo Plan, imports by public sector agencies, and imports financed by long-term international credits and grants.

Excludes some Col$300 million collected on automobiles.

Reflects temporary increases in duties in connection with import liberalization program.

Includes customs duties paid with CATs and 3 per cent import surtax.

Table 17.Colombia: Calculation of Weighted Average Rate of Advance Import Deposits, Selected Quarters, 1960–71(In millions of Colombian pesos)
1960196119621963196419651966
RateIIIIIVIIIIVIIVIVIIIIVIIIIIIIV
15.46.25.88.39.07.88.58.711.120.020.020.019.014.013.0
53.44.45.09.03.41.61.00.21.9
106.715.124.640.846.531.024.020.017.010.07.0
155.95.24.10.60.4
20144.7155.2153.3145.852.020.410.51.92.5
3071.6104.4168.2294.7334.9319.0239.0173.0116.040.024.0
4040.829.514.80.5
5040.2107.038.614.80.30.1
6539.059.84.521.341.168.080.974.674.037.015.04.01.0
70
758.51.80.30.3
9021.619.916.611.214.2
1006.42.19.5384.364.119.613.52.52.3
11021.611.64.3
120109.3216.6400.9573.1554.7628.4483.0323.0176.041.022.0
130740.7589.8565.150.05.51.21.10.60.5
150142.381.623.00.80.5
170111.168.424.50.70.1
1803.30.7
200188.864.119.80.90.1
Total900.6796.7798.5650.61,072.8747.7818.51,018.41,044.41,072.4803.0551.0332.0106.066.0
Weighted average rate110.9103.6101.978.1107.2103.891.083.680.184.084.481.972.059.453.2
Source: Banco de la República.
19671968196919701971
RateIVIVIIIIIIIVIIIIIIIVIIIIIIIV
115.1117.116.917.919.219.621.325.433.536.639.041.845.145.0
59.726.421.622.019.521.422.826.526.132.429.530.013.126.4
1020.331.736.440.243.546.643.144.064.460.659.769.289.8107.5
15
20
30212.7320.8332.9369.2386.5421.8454.2484.3701.6708.6765.3790.7867.3702.0
40
5012.337.455.1
65
7029.950.264.985.9103.3105.2113.4117.4162.2142.1132.6111.671.035.0
75
90
1008.8102.2338.4467.7
110
120
130494.4617.3642.2653.5644.7647.0700.5758.2964.9925.1967.6830.3501.5217.2
150
170
180
200
Total782.11,163.51,114.91,188.71,216.71,261.61,355.31,455.81,952.71,905.42,002.51,876.51,863.61,655.9
Weighted average rate93.480.888.486.384.883.083.583.780.579.979.780.669.761.9
Source: Banco de la República.
Source: Banco de la República.
Table 18.Colombia: Calculation of Financial Value of the Exemption from Customs Duties and Advance Import Deposits Under the Vallejo Plan (VP), 1960–71(In per cent)
196019611962196319641965196619671968196919701971
1.Average import component0.350.350.350.350.350.350.350.350.350.350.350.35
2.Average effective import duty21.020.818.416.115.920.224.421.419.020.121.424.0
3.Financial value of VP exemption from customs duties (1 × 2)7.357.286.445.645.577.078.547.496.657.047.498.40
4.Weighted average rate of advance import deposits1105.089.899.087.381.483.266.678.185.585.781.980.5
First quarter110.9101.9(88.0)91.0(82.6)(81.5)81.9(63.0)(90.3)88.483.579.7
Second quarter103.6(93.0)(97.0)(88.5)(81.8)(82.8)72.0(73.0)(87.0)86.483.780.6
Third quarter(103.0)(86.0)107.2(86.0)(81.0)84.059.4(83.0)(84.0)84.880.569.7
Fourth quarter(102.5)78.1103.883.680.184.453.293.480.883.079.961.9
5.Annual interest rate (14 per cent) over ten months (nr12)0.11670.11670.11670.11670.11670.11670.11670.11670.11670.11670.11670.1167
6.(4 × 5)12.2510.4811.5510.199.509.717.779.119.9810.009.569.39
First quarter12.9411.8910.2710.629.649.519.567.3510.5410.329.749.30
Second quarter12.0910.8511.3210.339.559.668.408.5210.1510.089.779.41
Third quarter12.0210.0412.5110.049.459.806.939.699.809.909.398.13
Fourth quarter11.969.1112.119.769.359.856.2110.909.439.699.327.22
7.Financial value of exemption from advance import deposits (6 × 1)4.293.674.043.563.323.402.723.193.493.503.352.98
First quarter4.534.163.593.723.373.333.352.573.693.613.413.26
Second quarter4.233.803.963.613.343.382.942.983.553.523.423.29
Third quarter4.213.514.383.513.313.432.433.393.433.463.292.85
Fourth quarter4.193.194.243.413.273.452.173.813.303.393.262.53
8.Total financial value of VP (3 + 7)11.6410.9510.489.208.8910.4711.2610.6810.1410.5410.8411.38
First quarter11.8811.4410.039.368.9410.4011.8910.0610.3410.6510.9011.66
Second quarter11.5811.0810.409.258.9110.4511.4810.4710.2010.5610.9111.69
Third quarter11.5610.7910.829.158.8810.5010.9710.8810.0810.5010.7811.25
Fourth quarter11.5410.4710.689.058.8410.5210.7111.309.9510.4310.7510.93
Sources: Banco de la República; Tables 16 and 17.

Figures in parentheses are interpolations.

Sources: Banco de la República; Tables 16 and 17.

Figures in parentheses are interpolations.

Financial value of special credit facilities

Since late 1964 when the exchange risk was removed from the system of advance exchange surrender, Colombian exporters have made extensive recourse to foreign borrowing for the purpose of export prefinance in amounts of up to 100 per cent of the value of their exports. Until recently, when special credit facilities were created at Proexpo, the financial attractiveness of using foreign export prefinancing arose from the large differential in interest rates between Colombian and foreign banks. Colombian banks generally served as financial intermediaries in these operations, charging a service commission of about 1½ per cent. Accordingly, the financial value of these special credit facilities can be defined as follows:

c = αβ[rh – (rf + ρ)]

Where

α =share of total exports prefinanced
β =percentage of export value prefinanced
rh =local interest charge for export prefinance
rf =foreign interest rate for export prefinance
ρ =commission charge of Colombian financial intermediaries

As a measure of the share of total exports prefinanced, α, the share of advance exchange surrender to total exchange surrender was used. The annual statistics for this measure are given in Table 3 for the years 1966–71. Prior to 1966, the value was assumed to be the same as in 1966. The percentage of export value prefinanced was taken to be 100, on the assumption that exporters take advantage of the full limit of the special credit facilities. The local interest charge, rh, as in previous calculations, was assumed to be 14 per cent. As a measure of foreign interest rates, rf, the London rate for Euro-dollars was taken. Finally, the commission charge of Colombian financial intermediaries, p, was assumed to be 1.5 per cent.22Table 19 shows the calculation of the financial value of the special credit facilities.

Table 19.Colombia: Calculation of Financial Value of Special Credit Facilities (SCF), 1964–71(In per cent)
Colombian

Interest

Rate

(1)
Euro-

dollar

London

Rate

(2)
Colombian

Banks’

Commission

(3)
Interest

Rate

Differntial

(4) =

(1)–

(2+3)
Share of

Total

Exports

Prefinanced

(5)
Financial

Value

of SCF

(6) =

(5) x

(4)
19643.501.180.381.950.3460.67
Fourth quarter14.004.701.507.800.3462.70
196514.004.921.507.580.3462.62
First quarter14.004.691.507.810.3462.70
Second quarter14.005.041.507.460.3462.58
Third quarter14.004.731.507.770.3462.69
Fourth quarter14.005.231.507.270.3462.52
196614.006.281.506.220.3462.15
First quarter14.005.531.506.970.3102.16
Second quarter14.005.901.506.600.2711.79
Third quarter14.006.781.505.720.3271.87
Fourth quarter14.006.921.505.580.4622.58
196714.005.541.506.970.5703.97
First quarter14.005.541.506.960.4523.15
Second quarter14.005.071.507.430.4653.45
Third quarter14.005.311.507.190.5313.82
Fourth quarter14.006.221.506.280.7784.89
196814.006.471.506.040.5583.37
First quarter14.005.831.506.670.6074.05
Second quarter14.006.901.505.600.5212.92
Third quarter14.006.241.506.260.5963.73
Fourth quarter14.006.891.505.610.5302.97
196914.009.871.502.630.5621.48
First quarter14.008.241.504.260.5282.25
Second quarter14.009.891.502.610.6001.57
Third quarter14.0011.001.501.500.6100.92
Fourth quarter14.0010.351.502.150.5091.09
197014.008.451.504.060.6762.74
First quarter14.009.221.503.280.7342.41
Second quarter14.008.981.503.520.6492.28
Third quarter14.008.331.504.170.6102.54
Fourth quarter14.007.251.505.250.6963.65
197114.006.571.505.930.6864.07
First quarter14.005.531.506.970.6804.74
Second quarter14.006.791.505.710.6583.76
Third quarter14.007.871.504.630.6723.11
Fourth quarter14.006.101.506.400.7184.60
Sources: Banco de la República; Bank of England, Quarterly Bulletin.
Sources: Banco de la República; Bank of England, Quarterly Bulletin.

The calculation of the effective exchange rate incorporating all the various components just described is presented in Table 20. Results are given on a quarterly and an average annual basis for the period 1960–71.

Table 20.Colombia: Effective Exchange Rate for Nontraditional Exports, 1960–71(In Colombian pesos per U.S. dollar)
A1B2C3D4E5Total,

A+B+

C+D
Total,

A+B+

C+D+E
19606.90–0.146.766.76
First quarter6.79–0.146.656.65
Second quarter6.80–0.146.666.66
Third quarter6.90–0.146.766.76
Fourth quarter7.10–0.146.966.96
19618.28–0.041.190.919.4310.34
First quarter7.53–0.101.080.868.519.37
Second quarter8.21–0.011.180.919.3810.29
Third quarter8.61–0.011.240.939.8410.77
Fourth quarter8.75–0.011.260.9210.0010.92
19629.11–0.011.310.9510.4111.36
First quarter8.78–0.011.260.8810.0210.90
Second quarter8.89–0.011.280.9210.1611.08
Third quarter8.59–0.011.240.939.8210.75
Fourth quarter10.171.461.0911.6312.72
196310.001.440.9211.4412.36
First quarter10.071.450.9411.5212.46
Second quarter9.971.440.9111.4112.32
Third quarter9.971.440.9111.4112.32
Fourth quarter9.971.440.9011.4112.31
196410.411.500.070.9311.9812.91
First quarter9.971.440.8911.4112.30
Second quarter9.971.440.8911.4112.30
Third quarter9.971.440.8911.4112.30
Fourth quarter11.721.690.321.0413.7314.77
196514.762.130.391.5517.2818.83
First quarter13.551.950.371.4115.8717.28
Second quarter16.612.390.431.7419.4321.17
Third quarter15.372.210.411.6117.9919.60
Fourth quarter13.501.940.341.4215.7817.20
196613.501.940.291.5215.7317.25
First quarter13.501.940.291.6115.7317.34
Second quarter13.501.940.241.5515.6817.23
Third quarter13.501.940.251.4815.6917.17
Fourth quarter13.501.940.351.4515.7917.24
196714.512.480.581.5517.5719.12
First quarter13.511.950.431.3615.8917.25
Second quarter14.082.580.491.4717.1518.62
Third quarter14.862.720.571.6218.1519.77
Fourth quarter15.602.680.761.7619.0420.80
196816.302.460.551.6519.3120.96
First quarter15.862.570.641.6419.0720.71
Second quarter16.182.490.471.6519.1420.79
Third quarter16.402.380.611.6519.3921.04
Fourth quarter16.752.360.501.6719.6121.28
196917.352.480.261.8320.0921.92
First quarter16.992.340.381.8119.7121.52
Second quarter17.202.530.271.8220.0021.82
Third quarter17.462.390.161.8320.0121.84
Fourth quarter17.742.630.191.8520.5622.41
197018.492.760.512.0021.7623.76
First quarter18.042.510.431.9720.9822.95
Second quarter18.312.640.421.9721.3723.34
Third quarter18.622.920.472.0122.0124.02
Fourth quarter18.972.940.692.0422.6024.64
197120.003.080.812.2823.8926.17
First quarter19.342.880.922.2623.1425.40
Second quarter19.743.120.742.3123.6025.91
Third quarter20.223.130.632.2723.9826.25
Fourth quarter20.713.150.952.2624.8127.07
Sources: Banco de la República; Tables 1519, inclusive.

A = Basic exchange rate for nontraditional exports.

B = Export taxes in exchange rate equivalent.

C = Financial value of tax credit certificate in exchange rate equivalent.

D = Financial value of special credit facilities in exchange rate equivalent.

E = Financial value of the Vallejo Plan in exchange rate equivalent.

Sources: Banco de la República; Tables 1519, inclusive.

A = Basic exchange rate for nontraditional exports.

B = Export taxes in exchange rate equivalent.

C = Financial value of tax credit certificate in exchange rate equivalent.

D = Financial value of special credit facilities in exchange rate equivalent.

E = Financial value of the Vallejo Plan in exchange rate equivalent.

Mr. Teigeiro, Assistant Chief of the Grancolombian Division of the Western Hemisphere Department, received his undergraduate education at the Central University of Madrid and the University of Villanueva in Havana, Cuba, and did postgraduate work at the University of Villanueva.

Mr. Elson, economist in the Grancolombian Division, was educated at Yale and Columbia Universities, and received his doctorate in economics from the latter institution.

Although the definition of minor exports in Colombia has changed over time, in recent years it has included all exports except coffee, petroleum, and raw hides.

See Banco de la República, XXV Informe Anual del Gerente a la Junta Directiva, July 1, 1947–June 30, 1948 (Bogotà), pp. 60–61.

Although the eligible list of minor exports was defined in terms of specific commodities rather than in terms of all except the four major exports (i.e., coffee, bananas, petroleum, and raw hides), as under the previous system, it appears that, in fact, most of the minor exports were included in the definition, and for this reason, it can be taken as coterminous with the previous one. For a list of the commodities treated under the export voucher system, see the International Monetary Fund, Fourth Annual Report on Exchange Restrictions, 1953, pp. 105–12.

In this study, depreciation or appreciation of a currency is measured by percentage changes in the exchange rate, defined in terms of Colombian pesos per U. S. dollar.

For a further discussion of this and related points, see the Colombian Commission on Tax Reform, Fiscal Reform for Colombia, ed. by Malcolm Gillis (Harvard University Law School, 1971), p. 95.

This point was raised in the context of Colombia’s problems of unemployment by a special mission from the International Labour Office in Towards Full Employment (International Labour Office, Geneva, 1970), p. 273.

Antonio Urdinola and Richard Mallon, Policies to Promote Colombian Exports of Manufactures, Center for International Affairs, Economic Development Reports, No. 75 (Harvard University, September 1967).

Through the end of 1971, the use of these facilities was relatively unimportant. At the end of 1971, total Proexpo credit to the private sector, which includes more than export prefinancing, amounted to about US$16 million, as against outstanding gross short-term liabilities of the commercial banks for export prefinancing of about US$120 million.

As the special credit facilities for export prefinancing through Proexpo have been used only to a small degree, no attempt has been made to take them into account in the calculation of the effective exchange rate in the Appendix.

Average monthly quotation of spot price for Colombian coffee in New York.

Food and Agriculture Organization, Trade Yearbook, 1970, Vol. 24 (Rome, 1971).

John Sheahan and Sara Clark, The Response of Colombian Exports to Variations in Effective Exchange Rates, Center for Economic Development, Research Memorandum No. 11 (Williams College, Williamstown, Massachusetts, June 1967); Antonio Urdinola and Richard Mallon, Policies to Promote Colombian Exports of Manufactures (cited in footnote 7).

The logarithmic transformation of the data implies that the regression coefficients can be interpreted as elasticities.

In reporting the empirical results, the following statistics are used: R2 is the coefficient of determination adjusted for the degrees of freedom; and D-W is the Durbin-Watson statistic of serial correlation. The coefficients in parentheses below the regression coefficients are t-ratios.

In the tests reported, the effective exchange rate is measured exclusive of the component for the Vallejo Plan, as this feature of the export promotion system applies only to manufactured goods.

To simplify the analysis, it is assumed, in the spirit of partial equilibrium analysis, that “other things remain the same,” regardless of changes in the export promotion system. Of course, in actual fact, this is not true. However, tracing out all the direct and indirect effects of changes in the export promotion system on the economy could be attempted only in the context of an elaborate econometric model of a general equilibrium system, which is clearly beyond the scope of this study.

For the purposes of this Law, minor exports excluded coffee, petroleum, bananas, and raw hides.

The least that can be said is that a higher exchange rate than that which actually prevailed in the past would have been needed to eliminate the need for import controls, as the import authority (Incomex) has had to reject, on average, between 10 and 30 per cent of the applications for import licenses that it has received.

See Proexpo, Plan Cuatrienal de Exportaciones, 1972–1975 (Bogotà, December 20, 1971).

This is the predominant marginal income tax rate for private Colombian corporations (sociedades anónimas), as specified in Decree-Law No. 81 of 1960.

When a manufacturer wishes to import raw materials or intermediate products, he is required to make an advance import deposit—at varying rates, depending on the type of merchandise—before he can obtain his import license. The Bank of the Republic holds the deposit until 90 days after the imported merchandise has cleared customs. In effect, this means a period of about nine to ten months: 15 days to one month for obtaining the import license or registration; five to six months as the average time between licensing and customs clearance; and three months for the prescribed retention by the Bank.

Defined in this way, the calculation of the financial value of special credit facilities may be somewhat overstated, as it assumes implicitly that the term of external borrowing for export prefinancing is one year. Although there is no data on which to base a firm conclusion, the average term of this borrowing appears to be probably less than one year. However, any overstatement on this account may be partially or wholly offset by the fact that no estimate was made of the value to the exporter of the absence of exchange risk inherent in the operation of the system, especially since early 1967. Prior to 1967, this freedom from exchange risk was of doubtful value, given the great variations in exchange rate policy.

Other Resources Citing This Publication