Abstract
This Selected Issues paper and Statistical Appendix analyzes economic developments in Colombia during 1996–99. Output growth slowed sharply in 1996 and early 1997, but subsequently rebounded owing to stronger exports, a temporary boom in world coffee prices, and an easing of credit policy. Despite efforts at addressing the fiscal imbalances, the nonfinancial public sector deficit widened further to more than 4 percent of GDP in 1997. Monetary policy during 1996 and most of 1997 was geared toward stimulating domestic demand.
IV. The experience of Capital Controls in Colombia, 1991-9821
51. In recent years, Colombia has experienced large capital inflows, averaging about 3.5 percent of GDP per year and increasing from 0.1 percent of GDP in 1990 to nearly 6 percent in 1997. These flows facilitated a significant acceleration of domestic demand during the first half of the decade, and have also played an important role in financing the widening current account deficit (Figure 1). However, they also exerted upward pressure on the real exchange rate, posing a threat to competitiveness. These concerns prompted the introduction of a set of policies to curb the destabilizing effects of the capital inflows, including the imposition of controls. This chapter discusses Colombia’s experience with such controls.
Colombia: Balance of Payments
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.Colombia: Balance of Payments
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.Colombia: Balance of Payments
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.A. Background: 1991-93
52. Colombia embarked on an ambitious structural reform program in the early 1990s, which included a comprehensive liberalization of the exchange and trade system, the dismantling of interest rate controls, and financial sector reform which allowed full foreign ownership of banks. In addition, a new financing strategy was defined, aimed at diversifying the country’s financing sources, with emphasis on domestic financing for the public sector and foreign direct investment for the private sector. Foreign direct investment was promoted by a new law introduced in 1991 (Box 1).
53. The implementation of these reforms, together with a tightening of credit conditions and a reduction in the rate of crawl of the currency aimed at lowering inflation, contributed to a large increase in private capital inflows beginning in 1991. The overall balance of payments position was also boosted by a significant improvement in the external current account balance which was partly attributable to steps taken to lower the fiscal deficit. In the event, the external current account improved from near balance in 1989 to a surplus that amounted to 5.6 percent of GDP in 1991, and net international reserves rose from 5.5 to 8.9 months of imports of goods and services. To limit the growth of liquidity resulting from the buildup of foreign reserves the monetary authorities took a number of steps during 1991, which included (i) aggressive open market operations in the form of sales of central bank securities; (ii) the imposition of a temporary 100 percent reserve requirement on most new private deposits with the financial system; and (iii) the issuance of non-interest bearing certificates denominated in U.S. dollars in exchange for most foreign exchange receipts.
Foreign Direct Investment in Colombia
Foreign direct investment in Colombia increased sharply in the second half of the 19th century, mainly related to the extraction of gold and textile production. In the early 1930s, such flows became subject to registration. In the 1960s, in the context of the import substitution strategy pursued at that time, increasing concerns about foreign participation in the process of industrialization led to the introduction of regulatory measures on foreign direct investment. This created a negative investment climate which was aggravated in the early 1970s when foreign participation was banned in important sectors (infrastructure, electricity, transportation and communication, public services, health and financial services) and all other areas where it was expected to seriously damage the competitiveness of domestically owned firms. In addition, the repatriation of profits was controlled, foreign companies were restricted to short-term credits in the domestic market, and local equity funding was prohibited.
The situation changed in the 1980s when access to international financial markets became increasingly restricted for sustained financing of the country’s external current account deficit. With the domestic financial market not sufficiently developed to finance growth, Colombia was prompted to change its external financing policy. As a result, in 1987 foreign participation was permitted in areas previously restricted, except in the financial sector, and access to domestic credit was granted to foreign owned firms.
Barriers to foreign investment were further removed in 1991 with the introduction of a law that (i) gave foreign and domestic companies equal access to domestic credit and offered new foreign investments broadly the same tax treatment as domestic investments; (ii) allowed foreign investment in most sectors; and (iii) liberalized restrictions on profit repatriation. In October 1992 all limitations on annual transfers of profits were abolished. The impact of the new law contributed to a substantial increase in foreign direct investment. Net foreign investment rose from 1 percent of GDP in 1991 to 5.1 percent of GDP in 1997, bringing its share in total net capital flows to about 90 percent. In recent years, privatization has accounted for a major part of the foreign direct investment in Colombia.
54. The large scale sterilization resulted in a substantial weakening in the quasi-fiscal position of the central bank (from near equilibrium in 1990 to a deficit of 0.8 percent of GDP in 1991).22 By end-1991, the increasing quasi-fiscal losses by the central bank and sustained capital inflows led the authorities to modify their underlying policies. The peso was revalued by 3 percent, restrictions on capital outflows were eased further, and import liberalization was accelerated. Nevertheless, large-scale capital inflows continued through 1993. Faced with limited effectiveness of existing instruments to stem the inflows, the authorities adopted capital controls in September 1993.
B. The Imposition of Capital Controls
55. The capital controls took the form of a deposit requirement; initially 47 percent of external disbursements were to be deposited in an unremunerated account at the central bank for 18 months. In an effort to target short-term inflows, the deposit requirement was imposed only on debts up to 18 months’ maturity, while trade credits of up to 6 months, capital inflows related to privatization and concessions, loans contracted by Colombians to finance investments in other countries, and financing of capital goods were exempt.
56. Subsequently, the terms of the deposit requirement have been adjusted on several occasions to reflect changes in external and domestic conditions (Table 1). These changes have affected mainly the percentage of the loan amounts subject to the deposit; the length of the deposits; and the maturity of the loans subject to the requirement. Furthermore, all external borrowing plans, including for short-term loans (less than a year), had to be registered at the central bank prior to disbursement. The terms and conditions of the deposit requirement were applied to the date of the loan registration regardless of any subsequent changes to the requirement prior to disbursement. With no penalty for the cancellation of registered loans, the system provided an incentive to maintain a stock of registered borrowing plans to avoid the effects of a future tightening of the deposit requirement.
Summary of Capital Controls in Colombia: 1993-98
Percentage of the loan amount to be deposited at the central bank.
Trade credits for longer maturity than allowed were considered regular external financing and were subject to the same terms and conditions of regular deposit requirement, unless they are separately specified.
Prepayment of debt prior to maturity was permitted subject to authorization of the central bank. In addition, the external borrowing registration requirement was modified from prior to disbursement to the time of disbursement.
Summary of Capital Controls in Colombia: 1993-98
| In Effect | Maximum Maturity Subject to Restriction (months) | Period after which the Prepayment was Allowed (months) | Deposit Requirement | Trade Credits Subject to Deposit Requirement (Months) 2/ | |
|---|---|---|---|---|---|
| Period (months) | Value 1/ (in percent) | ||||
| September 1993 to March 15, 1994 | 18 | 18 | 12 | 47 | Import and export financing over six months. |
| March 16, 1994 to August 15, 1994 | 36 | 36 | Less than 12 | 93 | Import and export financing over six months. |
| Between 12 and 18 | 64 | ||||
| Over 24 | 50 | ||||
| August 16, 1994 to February 20, 1996 | 60 | 60 | Minimum 42.8 to maximum 140 depending on the maturity, higher for shorter maturity | Import financing over four months but less than six months subject to 30 percent deposit Export financing over six months. | |
| August 1995 to February 20, 1996 | Import and export financing over 6 months. | ||||
| February 21, 1996 to March 18, 1996 | 48 | 48 | Minimum 10 to maximum 85 depending on the maturity, higher for shorter maturity | Export financing over three months subject to 25 percent deposit for 48 months. Import financing over six months. | |
| March 19, 1996 to January, 1997 | 36 | 36 | 18 | 50 | Export financing over four months subject to 15 percent deposit for 36 months. Import financing over six months. |
| March 12, 1997 to May 21, 1997 | 60 | 60 | 18 | 50 | No changes |
| May 22, 1997 to January 30, 1998 3/ | all 3/ | 18 | 30 in pesos 3/ | Export financing over four months subject to 15 percent deposit in U.S. dollars for 18 months. Import financing over six months. | |
| January 31, 1998-September 17, 1998 | all | 12 | 25 in pesos | No changes | |
| September 18, 1998 | all | 6 | 10 in pesos | Export financing over four months subject to 10 percent deposit in U.S. dollars for 36 months. Import financing over six months no longer subject to deposit requirement. | |
Percentage of the loan amount to be deposited at the central bank.
Trade credits for longer maturity than allowed were considered regular external financing and were subject to the same terms and conditions of regular deposit requirement, unless they are separately specified.
Prepayment of debt prior to maturity was permitted subject to authorization of the central bank. In addition, the external borrowing registration requirement was modified from prior to disbursement to the time of disbursement.
Summary of Capital Controls in Colombia: 1993-98
| In Effect | Maximum Maturity Subject to Restriction (months) | Period after which the Prepayment was Allowed (months) | Deposit Requirement | Trade Credits Subject to Deposit Requirement (Months) 2/ | |
|---|---|---|---|---|---|
| Period (months) | Value 1/ (in percent) | ||||
| September 1993 to March 15, 1994 | 18 | 18 | 12 | 47 | Import and export financing over six months. |
| March 16, 1994 to August 15, 1994 | 36 | 36 | Less than 12 | 93 | Import and export financing over six months. |
| Between 12 and 18 | 64 | ||||
| Over 24 | 50 | ||||
| August 16, 1994 to February 20, 1996 | 60 | 60 | Minimum 42.8 to maximum 140 depending on the maturity, higher for shorter maturity | Import financing over four months but less than six months subject to 30 percent deposit Export financing over six months. | |
| August 1995 to February 20, 1996 | Import and export financing over 6 months. | ||||
| February 21, 1996 to March 18, 1996 | 48 | 48 | Minimum 10 to maximum 85 depending on the maturity, higher for shorter maturity | Export financing over three months subject to 25 percent deposit for 48 months. Import financing over six months. | |
| March 19, 1996 to January, 1997 | 36 | 36 | 18 | 50 | Export financing over four months subject to 15 percent deposit for 36 months. Import financing over six months. |
| March 12, 1997 to May 21, 1997 | 60 | 60 | 18 | 50 | No changes |
| May 22, 1997 to January 30, 1998 3/ | all 3/ | 18 | 30 in pesos 3/ | Export financing over four months subject to 15 percent deposit in U.S. dollars for 18 months. Import financing over six months. | |
| January 31, 1998-September 17, 1998 | all | 12 | 25 in pesos | No changes | |
| September 18, 1998 | all | 6 | 10 in pesos | Export financing over four months subject to 10 percent deposit in U.S. dollars for 36 months. Import financing over six months no longer subject to deposit requirement. | |
Percentage of the loan amount to be deposited at the central bank.
Trade credits for longer maturity than allowed were considered regular external financing and were subject to the same terms and conditions of regular deposit requirement, unless they are separately specified.
Prepayment of debt prior to maturity was permitted subject to authorization of the central bank. In addition, the external borrowing registration requirement was modified from prior to disbursement to the time of disbursement.
57. The evolution of Colombia’s capital controls since their introduction in 1993 has undergone four phases: (i) the period through early 1996 when the terms and conditions of the deposit requirement were tightened in response to upward pressure on the real exchange rate; (ii) during the remainder of 1996, when the terms of the deposit requirement were eased as part of an effort to lower domestic interest rates; (iii) during the first half of 1997, when the terms of the deposit requirement were tightened to help forestall a further appreciation of the real exchange rate; and (iv) during the remainder of 1997 and 1998 when the deposit requirement was successively relaxed in the face of recurrent episodes of strong private capital outflows.
58. Despite the imposition of the deposit requirement, private capital inflows remained strong, increasing from 4.1 percent of GDP in 1993 to 6.5 percent of GDP in 1996 (Figure 2). The increase was accounted for in part by debt-creating flows but also by a significant growth in foreign direct investment, which reached 3.7 percent of GDP in 1996, compared with 1.3 percent in 1993. Net debt-creating flows by the private sector remained strong but mostly stable at 3.2 percent of GDP on average during the period 1993-96, compared with 1 percent in 1992, with a notable lengthening in the maturity structure. The share of net short-term flows in net debt creating flows declined from 60 percent in 1993 to less than 20 percent in 1996. Accordingly, the maturity structure of the private debt stock changed; the share of medium- and long-term debt rose to 70 percent of the total debt stock in 1996, from 40 percent in 1993 (Figure 3).
Colombia: Net Capital Inflows
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.1/ Does not include changes in net foreign assets.2/ Includes net foreign direct investment.3/ In 1997 excludes changes in net foreign assets related to privatization proceeds kept abroad.Colombia: Net Capital Inflows
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.1/ Does not include changes in net foreign assets.2/ Includes net foreign direct investment.3/ In 1997 excludes changes in net foreign assets related to privatization proceeds kept abroad.Colombia: Net Capital Inflows
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.1/ Does not include changes in net foreign assets.2/ Includes net foreign direct investment.3/ In 1997 excludes changes in net foreign assets related to privatization proceeds kept abroad.
Colombia: Debt Stock
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.Colombia: Debt Stock
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.Colombia: Debt Stock
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.59. As part of the authorities’ efforts to contain the downward exchange rate pressures which began after mid-1997, the deposit requirement was relaxed in January 1998 and again in September; the retention period was reduced, the deposit ratio was lowered and import financing over six months became exempted. Despite these measures, preliminary figures through September 1998 indicate that the trend of net private capital outflows persisted.
C. Effectiveness of Capital Controls
60. There are three questions to be addressed in assessing the effectiveness of capital controls in Colombia: (i) have they been effective in reducing the total volume of net capital inflows; (ii) did they contribute to reducing the relative amount of short-term capital; and (iii) how did they affect savings and investment.
61. There has been debate over whether the capital controls have been effective in reducing the total amount of external flows to Colombia. Cárdenas and Barrera (1996) calculated the effect of the capital controls in Colombia by specifying its burden in terms of the opportunity costs incurred by agents holding the unnremunerated deposits. They concluded that total inflows have not been sensitive to the deposit requirement. Ocampo and Mora (1997), on the other hand, using a different measurement of private capital flows showed that they have been effective in reducing the amount of external financing.
62. The deposit requirement seems to have played an important role in lengthening the term structure of inflows to Colombia. The estimation results by Cárdenas and Barrera (1996) as well as by Ocampo and Mora (1997) confirm that the controls have been effective in inducing longer term maturities in the Colombian debt structure. However, reservation may be required in assessing its effectiveness in lengthening the maturity structure. Imposition of the deposit requirement coincided with the introduction of the exchange rate band, which also may have contributed to reducing the short-term flows, making it difficult to isolate the impact of each of these policies on short-term flows.23
63. Records of external debt registers suggest that the effects of capital controls have been diluted by the requirement of debt registration prior to disbursement. The registration system created incentives to anticipate changes in the deposit requirement and secure future “rights” to obtain external financing, as the terms of the deposit requirement were determined at the time of registration rather than at the time of disbursement, as noted above. For example, from late 1993 the maturity of registered loans increased sharply from less than, to greater than 18 months. (Figure 4). However, the increase in the maximum maturity of loans subject to the deposit requirement from 18 to 36 months only took place in March 1994, which would indicate that potential borrowers anticipated the future tightening by the central bank. The same happened in mid-1994 before the terms were tightened further to 60 months; by the time the maturity was lengthened, all borrowing registrations were for over 60 months. Until the requirement of pre-registration was abolished in May 1997, the magnitude and the maturity of the registrations exceeded substantially that of the actual capital inflows.
Colombia: Registers of External Financing
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.Colombia: Registers of External Financing
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.Colombia: Registers of External Financing
Citation: IMF Staff Country Reports 1999, 006; 10.5089/9781451808735.002.A004
Source: Banco de la Republica.64. While the effectiveness of the deposit requirement in controlling the total volume of inflows is uncertain, the data suggest that the deposit requirement has contributed to a shift to alternative financing sources that were exempt from the controls. The surge in foreign direct investment is accounted for in part by the foreign direct investment law adopted in 1991 and, more recently, by large privatization receipts. However, the deposit requirement appears to have provided incentives to switch from debt-creating flows to foreign direct investments which are not subject to this restriction. In some cases this provided a loophole to avoid the deposit requirement.24
65. Finally, on saving and investment, there is a vast literature showing a strong relation between savings and investment in Colombia and external financing in the 1990s. Sánchez et al. (1996) demonstrated a significant substitution of private savings for external financing in the first half of the 1990s. Ocampo and Mora (1997) confirmed the crucial role played by the external financing: large capital inflows substituted, to a large extent, the private savings and encouraged investment by private firms. However, it did not affect the public sector investment which was mostly dependent on tax revenues. The direct impact of the capital controls on savings and investment is subject to further studies in the case of Colombia in light of the uncertain effect of the controls on the total volume of external financing.
References
Cárdenas, Mauricio, and Felipe Barrera, “On the Effectiveness of Capital Controls in Colombia,” mimeo, FEDESARROLLO, 1997.
Ocampo Gaviria, Jose A., and Tovar Mora, Camilo E., “Capital Flows, Savings and Investment in Colombia 1990-96,” 1997, Archivos de Macroeconomía 58, DNP.
Sánchez, F., Murcia G. and Oliva C., “Auge y Colapso del Ahorro Empresarial en Colombia, 1983-1994,” Planeación y Desarrollo, 1996.
Prepared by Keiko Honjo.
This occurred because of the excess in cost per sterilization bond issued over earnings on international reserves.
Some studies have raised doubts about the relevance of discussing the issue of short versus long-term flows. The argument against the short-term flows is based on the assumption that the long-term capital inflows would be more stable because they are mainly guided by longer term fundamentals and hence less sensitive to short-term interest rate differentials. In fact, if capital markets are sufficiently developed, allowing relatively easy transactions of the long-term flows, the long-term capital could be as liquid as the short-term flows. In the case of the United States, for example, there is no classification of flows between short and long.
There have been several cases in the past in which the deposit requirement was circumvented through establishment of a holding company in a tax-heaven country abroad. The external borrowing by such holding companies and their transfers of funds to Colombian domestic companies would be legally recorded as foreign direct investment, exempt from the deposit requirement.
