A. Introduction

105. Using exceptional powers under an economic emergency, the Government of Colombia introduced a special levy on financial transactions at the end of 1998. The receipts from this levy, which was originally designed as temporary, were to fund the exceptional costs associated with the mounting difficulties faced by the financial system. During the following two years, however, the levy changed substantially, including its tax base, rates, earmarking provisions, permanence, and the institutions charged with the collection.

B. Evolution of the Bank Debit Tax

106. The economic emergency decree No. 2331 of August 1998 introduced a special levy on the debit of funds from deposits in financial institutions (debit tax). This levy carried a rate of 0.2 percent and was to last until the end of 1999. A second rate of 0.12 percent was applied to interbank transactions. Exemptions included liquidity operations carried out by the central bank (Banco de la República or BR) and the public deposit insurance agency (Fogafin), and the transfers of tax receipts collected for the treasury by commercial banks.⁸² The resources collected by this levy were earmarked for debt relief to mortgage holders and for the strengthening of financial cooperatives and public and private banks. Collection was entrusted to Fogafin.

107. In March 1999, the Constitutional Court of Colombia extended the 0.2 percent rate to interbank operations (eliminating the preferential 0.12 percent rate), transformed the levy into a tax, and entrusted the collection to the revenue service (DIAN). The court also determined that private banks could not be beneficiaries of the revenues collected by this tax.

108. In July 1999, congress extended the tax through the end of 2000 and earmarked the receipts to finance the reconstruction activities related to the January 1999 earthquake that hit Colombia’s main coffee producing region, known as the Eje Cafetero. Exemptions on interbank credit operations and transactions with government securities were introduced.

109. In December 2000, congress made the tax permanent and eliminated the restrictions on the use of the proceeds.⁸³ The rate was increased to 0.3 percent and exemptions were broadened to include small withdrawals, withdrawals for mortgage payments, and financial transactions of the national, departmental, and municipal governments.

C. International Comparisons⁸⁴

110. In the past decade, a number of Latin American countries have introduced taxes on banking transactions,⁸⁵ which were levied mostly on debit operations. These have tended to be temporary in nature and were introduced to help deal with particularly difficult fiscal situations. Nonetheless, the revenue collected by these taxes did not exceed 8 percent of total tax revenue, with the exception of Ecuador where they accounted for over 26 percent of tax revenue in 1999 (Table 1). Most countries have replaced these taxes with alternative revenue sources.

Table 1.

Comparison of Debit Taxes in Latin America

Sources: SM/00/207; and DIAN data.

Table 1.

Comparison of Debit Taxes in Latin America

			Gross Revenue
	Year	Tax Rate	In Percent of GDP	In Percent of Tax Revenue
Countries where tax is being enforced
Brazil	1994	0.25	1.06	3.6
	1997	0.20	0.80	2.8
	1998	0.20	0.90	3.0
	1999	0.22	0.79	2.6
Colombia	1999	0.20	0.73	7.3
	2000	0.30	0.60	5.3
Countries where tax was discontinued
Argentina	1989	0.70	0.66	4.3
	1990	0.30	0.30	2.0
	1991	1.05	0.91	5.4
	1992	0.60	0.29	1.5
Ecuador	1999	1.00	3.50	26.7
Peru	1990	1.41	0.59	6.4
	1991	0.81	0.46	5.0
Venezuela	1994	0.75	1.30	7.7
	1999	0.50	0.60	4.9

Sources: SM/00/207; and DIAN data.

View Table

Table 1.

Comparison of Debit Taxes in Latin America

			Gross Revenue
	Year	Tax Rate	In Percent of GDP	In Percent of Tax Revenue
Countries where tax is being enforced
Brazil	1994	0.25	1.06	3.6
	1997	0.20	0.80	2.8
	1998	0.20	0.90	3.0
	1999	0.22	0.79	2.6
Colombia	1999	0.20	0.73	7.3
	2000	0.30	0.60	5.3
Countries where tax was discontinued
Argentina	1989	0.70	0.66	4.3
	1990	0.30	0.30	2.0
	1991	1.05	0.91	5.4
	1992	0.60	0.29	1.5
Ecuador	1999	1.00	3.50	26.7
Peru	1990	1.41	0.59	6.4
	1991	0.81	0.46	5.0
Venezuela	1994	0.75	1.30	7.7
	1999	0.50	0.60	4.9

Sources: SM/00/207; and DIAN data.

111. While research on the implications of financial taxes is limited and ongoing, a recent study by FAD found that bank debit taxes were successful in raising revenue in the short term but that the market responses to the tax indicate that there have likely been adverse allocational impacts, in particular, significant financial disintermediation.⁸⁶

D. Preliminary Evaluation of the Tax

112. A rigorous evaluation of the bank debit tax in Colombia is difficult at this time due to a number of reasons. First, the tax is still relatively new and therefore the available information is scarce. Moreover, since its creation, the tax has undergone a series of changes that have altered its nature. And finally, the period of this tax coincides with considerable economic and financial turbulence in Colombia that included a banking crisis, a recession, and substantial exchange rate and interest rate volatility.

Tax revenues

113. The debit tax in Colombia was successful in raising revenue in the short run, 0.73 percent of GDP in 1999 and 0.60 percent of GDP in 2000 (Table 2), comprising 7.3 percent and 5.3 percent of total tax revenue collected in those years. The yield as a proportion of GDP decreased in 2000 by over 17 percent, despite the fact that the economy recovered from the recession of 1999,⁸⁷ which seems to suggest that economic agents adapted to avoid the tax.

Table 2.

Colombia: Revenue from the Debit Tax

(In billions of pesos)

Sources: DIAN and; Superintendency of Banks.

Table 2.

Colombia: Revenue from the Debit Tax

(In billions of pesos)

			12-month change
Tax revenue	1999	2000	Absolute	Percentage
			(in billions)	(in percentage)
January	26.3	73.9	47.6	180.9
February	96.6	101.6	5.0	5.2
March	82.1	77.1	-5.0	-6.1
April	101.1	71.5	-29.6	-29.3
May	98.7	99.4	0.7	0.7
June	92.4	84.4	-8.0	-8.7
July	118.3	80.0	-38.3	-32.4
August	75.9	95.2	19.3	25.4
September	65.6	79.5	13.9	21.2
October	87.4	98.0	10.6	12.2
November	78.9	77.3	-1.6	-2.1
December	107.3	98.8	-8.5	-7.9
Total	1030.7	1036.7	6.0	0.6
In percent of GDP	0.73	0.60		-17.8

Sources: DIAN and; Superintendency of Banks.

View Table

Table 2.

Colombia: Revenue from the Debit Tax

(In billions of pesos)

			12-month change
Tax revenue	1999	2000	Absolute	Percentage
			(in billions)	(in percentage)
January	26.3	73.9	47.6	180.9
February	96.6	101.6	5.0	5.2
March	82.1	77.1	-5.0	-6.1
April	101.1	71.5	-29.6	-29.3
May	98.7	99.4	0.7	0.7
June	92.4	84.4	-8.0	-8.7
July	118.3	80.0	-38.3	-32.4
August	75.9	95.2	19.3	25.4
September	65.6	79.5	13.9	21.2
October	87.4	98.0	10.6	12.2
November	78.9	77.3	-1.6	-2.1
December	107.3	98.8	-8.5	-7.9
Total	1030.7	1036.7	6.0	0.6
In percent of GDP	0.73	0.60		-17.8

Sources: DIAN and; Superintendency of Banks.

114. The administration of this tax has been facilitated by the sophisticated information technology infrastructure used by financial institutions. This allows for the speedy transfer to the treasury of the weekly collections.

Interbank market

115. The constitutional court ruling of March 1999 raised the rate of the debit tax to 0.2 percent for interbank transactions. Given that most of these are very short term (from 1 to 14 days), the result was a sharp increase in the effective taxation rate of such operations. The BR has estimated that the increase in the rate implied an annual rate of 62 percent for overnight credit operations.⁸⁸

116. The immediate consequence of this measure was that interbank credit virtually dried up. Average interbank lending both in U.S. dollars and Colombian pesos (Figure 1) dropped by about 90 percent from late February 1999 to early May 1999, but rose again in August 1999 after these operations were exempted from the tax.

Colombia: Debit Tax, Money Supply, and Interbank Credit

Citation: IMF Staff Country Reports 2001, 068; 10.5089/9781451808834.002.A007

Source: Banco de la República.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

Colombia: Debit Tax, Money Supply, and Interbank Credit

Citation: IMF Staff Country Reports 2001, 068; 10.5089/9781451808834.002.A007

Source: Banco de la República.

• Download Figure
• Download figure as PowerPoint slide

Colombia: Debit Tax, Money Supply, and Interbank Credit

Citation: IMF Staff Country Reports 2001, 068; 10.5089/9781451808834.002.A007

Source: Banco de la República.

• Download Figure
• Download figure as PowerPoint slide

117. During this period, the BR emerged as a clearinghouse of interbank credit operations since operations with the BR were exempted from the tax. Thus, banks with liquidity needs would carry out repo operation with the BR, while those with excess liquidity would engage in a reverse repos.⁸⁹

Payment system

119. The economic slowdown that began in 1998 brought about a reduction in check clearing activity in Colombia. During 1998, the volume of checks cleared in Bogotá and the regional offices of the BR fell with respect to the previous year, and by the end of that year, the values of these checks were lower than in the corresponding period in 1997 (Table 3).

Table 3.

Colombia: Check Clearing Activity

(In number of checks and millions of CP)

Source: Banco de la República.

Table 3.

Colombia: Check Clearing Activity

(In number of checks and millions of CP)

					12-month percentage change
	No. of Checks		Value of Checks		No. of Checks		Value of Checks
	1998	1999	1998	1999	1998	1999	1998	1999
January	14,454	11,057	57,950	28,099	-983.22	-23.50	15.96	-0.52
February	14,667	9,976	57,454	26,918	-5.22	-0.32	21.15	-0.53
March	15,827	11,716	60,732	31,062	5.69	-0.26	30.72	-0.49
April	14,791	10,529	58,888	29,136	-10.98	-0.29	3.08	-0.51
May	15,123	10,583	54,293	30,329	-4.21	-0.30	7.89	-0.44
June	14,858	10,027	63,290	29,706	-5.34	-0.33	28.49	-0.53
July	16,023	10,209	67,434	31,211	-8.95	-0.36	14.09	-0.54
August	13,900	10,224	58,600	30,119	-12.23	-0.26	0.13	-0.49
September	13,916	9,924	63,812	30,624	-16.92	-0.29	0.14	-0.52
October	13,404	9,645	63,057	29,559	-19.86	-0.28	0.07	-0.53
November	13,042	10,056	46,004	30,049	-14.76	-0.23	-0.14	-0.35
December	13,209	10,983	37,880	36,874	-26.86	-0.17	-0.44	-0.03

Source: Banco de la República.

View Table

Table 3.

Colombia: Check Clearing Activity

(In number of checks and millions of CP)

					12-month percentage change
	No. of Checks		Value of Checks		No. of Checks		Value of Checks
	1998	1999	1998	1999	1998	1999	1998	1999
January	14,454	11,057	57,950	28,099	-983.22	-23.50	15.96	-0.52
February	14,667	9,976	57,454	26,918	-5.22	-0.32	21.15	-0.53
March	15,827	11,716	60,732	31,062	5.69	-0.26	30.72	-0.49
April	14,791	10,529	58,888	29,136	-10.98	-0.29	3.08	-0.51
May	15,123	10,583	54,293	30,329	-4.21	-0.30	7.89	-0.44
June	14,858	10,027	63,290	29,706	-5.34	-0.33	28.49	-0.53
July	16,023	10,209	67,434	31,211	-8.95	-0.36	14.09	-0.54
August	13,900	10,224	58,600	30,119	-12.23	-0.26	0.13	-0.49
September	13,916	9,924	63,812	30,624	-16.92	-0.29	0.14	-0.52
October	13,404	9,645	63,057	29,559	-19.86	-0.28	0.07	-0.53
November	13,042	10,056	46,004	30,049	-14.76	-0.23	-0.14	-0.35
December	13,209	10,983	37,880	36,874	-26.86	-0.17	-0.44	-0.03

Source: Banco de la República.

120. This phenomenon intensified in late 1998 and throughout 1999 as a result of the deepening of the recession and the introduction of the debit tax. In the second half of 1999, the value of checks cleared was over 50 percent lower than in the corresponding period of 1998.

121. Interviews with Colombian tax officials have revealed a number of ways in which individuals and firms act to avoid the tax, which involve the substitution of checks for other payment instruments that are not subject to the tax. Credit cards are used by economic agents for both receipts and payments, thereby reducing the tax liability to the single net payment at the end of the billing cycle. Other instruments include vouchers that can be used for payments at supermarkets, restaurants, and selected retail stores.⁹¹ More recently, commercial banks have introduced new services aimed at avoiding the debit tax. These include the collection of debt for small and medium-sized firms and the placement of the funds in the banks’ accounts at the central bank. These firms, in turn, make payments to their suppliers that are channeled by means of interbank operations.

E. Conclusion

122. The debit tax in Colombia originated at a time of fiscal stress that coincided with and—to an extent—resulted from the banking problems that emerged during 1998. The authorities explained that a main justification for this particular tax was equity. It was argued that everybody with a bank account would be subject to the tax, including participants in the underground economy, and that the poor do not possess bank accounts, and would therefore not be affected. Yet the notion that this is an equitable tax does not always hold: activities that involve the frequent turnover of working capital, such as retail operations, pay a higher effective rate than other sectors. Moreover, the substitution of check or electronic payments for others payment instruments (such as cash, vouchers, etc.) involve high transaction costs. In addition, this tax clearly will not contribute to the extension of banking services to the poor, but might actually work in the opposite direction.

123. Notwithstanding its contribution to government revenue at a time when fiscal consolidation is paramount, this tax carries considerable risk for the Colombian economy. First, financial disintermediation has resulted as alternative payment instruments not subject to the debit tax have been developed that eliminate or reduce the reliance on bank deposits, as described above. There is also evidence that investors are switching to government securities (where transactions are exempted from the tax) and away from deposits and bonds issued by financial institutions,⁹² ultimately threatening their ability to extend credit.

124. In addition, the considerable progress achieved by Colombia in the modernization of its payment system during the past decade could be reversed by the tax. During this period, a number of state-of-the-art electronic large and small value payment systems were introduced. These systems, which have contributed to greater efficiency and systemic security of the payment system in Colombia, use the liquidity in banks’ accounts at the central bank and in individuals’ accounts at commercial banks. The adverse effects of the debit tax on bank deposits could lead to lesser use of these efficient payment systems.

125. Finally, this tax will likely delay the recovery of the financial system, at a time when it is emerging from three years of heavy losses, since it reduces operating income from payment services. The delayed recovery, in turn, might help prevent the reduction of the high interest rate spreads prevalent in the Colombian economy, with the ensuing negative repercussions on the real sector.