Journal Issue

Angola: Selected Issues and Statistical Appendix

International Monetary Fund
Published Date:
September 2003
  • ShareShare
Show Summary Details

III. International Reserve Adequacy in Angola24

Executive Summary

47. Vulnerability to fluctuations in the world market price for oil and high dependence on imports make reserve adequacy an important factor for stable economic development and management in Angola. This note reviews the holdings of gross international reserves by Angola and compares them to reserve holdings in a number of African countries which have exchange rate regimes similar to Angola’s.25 The main findings are the following:

  • When facing possible balance of payments shocks affecting the goods markets (i.e., the commodity export and import markets), Angola has been consistently less insulated than similar countries in the region. In particular, Angola’s reserve coverage in terms of months of imports of goods has been one third of the average levels recorded in other African countries with similar exchange rate regimes.
  • Angola’s current reserve position, equivalent to 1.1 month of imports by end-December 2002, is particularly worrisome given the economy’s high dependence on imported consumption, intermediate and capital goods. These high import penetration ratios have resulted in inelastic production structures which heighten Angola’s vulnerability to balance of payments shocks, and may impair the economy’s ability to reallocate resources within and across the tradable and nontradable sectors in the event of an external shock.
  • When facing possible shocks to capital markets, Angola has been consistently less insulated than its peers according to all measures of reserve coverage assessed using estimates for monetary aggregates, external debt service and/or short-term external debt stocks.
  • Given Angola’s difficult macroeconomic environment, characterized by high inflation and large fiscal deficits, the possibility of borrowing from abroad might be more costly when reserves are low, despite the existence of substantial oil resources.

A. Reserve Adequacy: General Considerations

48. As a result of numerous currency and financial crises across the world, a number of theoretical and empirical papers have focused on assessing the economic factors behind those crises. In general, the analysis has concluded that the roots of recent crises lay with the following: (i) inconsistencies between “fundamentals” and fiscal and/or monetary policies (the so-called “first generation” models such as Krugman (1979)); (ii) self-fulfilling expectations about a country’s unsustainable balance of payments position (the “second generation” approach such as in Obstfeld (1986)); and (iii) contagion from crises elsewhere, along with possible investors’ herding behaviors (“third generation” literature (Calvo, 1995)).

49. An idea implicit in the theoretical and empirical research has been that balance of payments crises are triggered by macroeconomic imbalances which can be ultimately quantified in terms of liquidity, solvency, and external competitiveness indicators for the country under stress. In particular, central banks’ international reserve holdings have been usually viewed as an adequate liquidity indicator, thus measuring the country’s ability to back up their short-term external liabilities and needs when facing shocks. Reserve holdings—and their fluctuation over time—are also viewed as an instrument that could be used to minimize the economy’s adjustment costs (in terms of output losses and unemployment) in the upshot of a balance of payments shock. In addition, the analysis has yielded an extensive discussion over the “optimal level” of international reserve holdings and ways to estimate those optimal reserve targets.

50. As suggested by some authors,26 the choice of an “optimal” level of reserve holdings should take into account a number of factors including: (i) the economy’s vulnerability to balance of payments to shocks; (ii) the consequences of running out of reserves; (iii) the opportunity costs of holding reserves; and (iv) the speed with which a country is prepared to adjust its balance of payments when the need arises.

51. The policy prescriptions from the literature are rather straightforward. First, greater vulnerability to balance of payments shocks requires a bigger reserve accumulation. Second, the worse the economic consequences from international reserves depletion, the higher should be the country’s target for the reserve stock. Third, higher opportunity costs of holding reserves call for a lower reserves target or norm. Fourth, faster adjustment to shocks permits economizing in reserves. The caveat, however, is that economic adjustment could be more or less costly in terms of output losses and unemployment depending on: (i) the composition of an import cutback needed to restore the external balance (for example, abruptly cutting essential imports of intermediate or capital goods could be very painful for a country, so a higher level of reserves may help avoid such a costly adjustment); (ii) the short-run elasticity of exports; and (iii) the country’s ability to borrow from abroad to finance its current account deficits.

B. The Case of Angola

52. This section examines Angola’s gross international reserves holdings vis-à-vis holdings in other African countries with similar exchange rate regimes.27Table III.1 presents a number of indicators usually used for assessing international reserve adequacy for Angola and peer African countries.

Table III.1.Angola: Indicators of International Reserve Adequacy, 1997-2002
A. Ratio of International Reserves to Imports 1/
199719981999200020012002Average 3/
Average Other Countries 2/7.425.774.866.597.467.996.70
B. Ratio of International Reserves to Currency in Circulation
199719981999200020012002Average 3/
Average Other Countries 2/2.131.621.361.922.072.601.95
C. Ratio of International Reserves to Broad Money
199719981999200020012002Average 3/
Average Other Countries 2/0.620.490.400.470.500.540.51
D. Ratio of International Reserves to Scheduled Debt Service
199719981999200020012002Average 3/
Average Other Countries 2/3.623.393.263.574.395.794.09
Average (exc. Rwanda)2.111.710.961.481.882.751.88
E. Ratio of International Reserves to Short Term Debt 4/
199719981999200020012002Average 3/
Average Other Countries 2/8.946.584.2110.2014.8618.3210.58
Average (exc. Algeria)2.161.541.062.883.123.112.20

Imports of goods only.

Includes Algeria, Burundi, Ghana, Guinea, Nigeria, Rwanda and Zambia.

Average are computing without considering 2000. As explained in the text unusual positive term of trade shock induced an important accumulation of reserves in Angola.

All debt with remaining maturity of one year.

Source: International Financial Statistics, Countries’ Databases.

Imports of goods only.

Includes Algeria, Burundi, Ghana, Guinea, Nigeria, Rwanda and Zambia.

Average are computing without considering 2000. As explained in the text unusual positive term of trade shock induced an important accumulation of reserves in Angola.

All debt with remaining maturity of one year.

Source: International Financial Statistics, Countries’ Databases.

Angola is not well shielded against shocks to goods markets

53. A standard measure of a country’s insulation against shocks to the good markets is the ratio of gross international reserves scaled by imports of goods and services. According to this measure, Angola’s reserve coverage has been consistently lower than the average coverage for similar African countries.

54. Angola’s reserves coverage improved significantly after the implementation of a stabilization program in 1997 and 1998. In particular, Angola’s reserves rose from less than 2 months of imports in 1998 (about US$347 million) to 4.6 months (about US$1.2 billion) in 2000. Subsequently, however, large increases in total import levels, combined with a depletion of gross international reserves, resulted in a significant reduction in the reserve import coverage. By end-2002, Angola’s international reserves had reached an all-time low equivalent to one month of imports of goods (about US$375 million).28

55. Although a country’s “optimal” level of reserve coverage might change overtime, practitioners consider a reserve coverage of between 30 to 40 percent of total imports to be generally adequate.29 At Angola’s current import levels (of about US$4 billion), this rule of thumb would mandate reserve holdings ranging between US$1.2 to US$1.6 billion, values which are well above the current BNA’s reserves holdings of 375 million.

56. It should be noted that, in general, most indicators for Angola improved in 2000 due to an important accumulation of reserves, which in turn resulted from an unusual positive shock to the terms of trade. Indeed, the price of oil rose from US$17.6 to US$27.2 a barrel, a 55 percent increase that year.

Angola’s coverage against shocks to the capital account is also limited

57. Another motive for having high international reserve holdings should be to protect a country from financial shocks such as deposit outflows and market contagion, while also permitting debt obligations to be serviced on a timely basis.

58. The ratio of official reserves to currency in circulation reflects the monetary authorities’ ability to withstand pressure on the exchange rate stemming from a flight away from local currency (but not away from banks). As the data in Part B of Table III.1 suggest, Angola’s official reserve coverage of currency in circulation compared less favorably to the average of the sample at the beginning of the sample period, but improved significantly in 2000. Nevertheless, Angola’s situation has significantly deteriorated in the last few years. This deterioration reduced its reserve coverage to below the regional average again. By 2002, while the regional average reached 2.60, the Angolan coverage was only 1.05.

59. To measure a country’s s ability to confront a flight from liquid assets in domestic currency including the banking system, international reserves are scaled by broad money (the sum of money and quasi-money in domestic currency). According to this measure, Angola compared less favorably to other African countries during the period 1997 to 1999, but it significantly improved its position in 1999 and 2000. Nevertheless, this improved coverage was short-lived and Angola’s ratio of international reserves to broad money fell to less than half the average ratio for other similar African countries by 2002.

60. Angola’s relative improvement observed around 2000 mirrored a large reserve accumulation, while broad money (measured in dollar terms) increased proportionally less. Subsequently, while broad money increased by about 60 percent in 2001/2002, official reserves declined steadily from US$732 million at end-2001 to US$375 million by end-2002.

61. Scheduled external debt service (both short- and long-term) also plays a role in evaluating reserves holdings. In a crisis, external debt service would put a strain on reserves, as rolling-over debt and new credits would be difficult. Greater reserve holdings should, in principle, facilitate external debt repayment and rolling-over of old debts.

62. Table III.1, panel D, suggests that Angola’s coverage of official reserves to external debt service has been consistently lower than the mean coverage for comparator African countries during 1997 to 2002. Even in 2000, when Angola’s ratio of international reserves to external debt service reached a peak of 60 percent, this coverage was less than one sixth of the average for other African countries in our sample. By end-2002, Angola’s coverage fell to 19 percent, while coverage in comparable African countries was close to 600 percent.

63. It should be noted that in the case of Angola, since the mid-1990s, most contracted debt has been guaranteed by oil (oil-backed loans). In this sense, the collaterized repayment arrangement of these loans reduces the reserves holdings’ needs and the accumulation of arrears related to this oil-backed loans.

64. Panel E in Table III.1 shows the ratio of official reserves to the stock of short-term external debt for Angola and comparator African countries.30 As with other indicators, 1998 seemed to be a particularly difficult year with reserves covering only 8 percent of the short-term obligations. Towards 2000, the coverage reached a peak of 41 percent due to the large accumulation of reserves. While in 1999 short-term external debt was US$2.4 billion, it increased to US$2.9 billion in 2000. In 2001 and 2002, short-term external liabilities remained at the similar levels (US$2.5 and US$2.1 billion respectively). Nevertheless, the subsequent decline of international reserves, reduced the reserve coverage to 28 percent and 18 percent in 2001 and 2002 respectively.

C. Eventual Benefits of Additional Reserve Holdings

65. Our analysis so far suggests that Angola’s reserves holdings compare less favorably than average holdings in comparator African countries. However, increasing Angola’s reserve holdings should take into account the benefits and costs of such an action.

66. First, reserve holdings imply a cost. For debtor countries like Angola, the opportunity cost of reserves is best measured by the difference between the interest rate paid on external debt and that received on reserve assets. In the last several years, the average interest rate paid on the external debt has been 8.5 percent.31 At the same time, the implicit rate of return on Angola’s official reserve holdings has been around 1.5 percent. Thus, the opportunity cost of holding reserves is about 7 percent.

67. However, recent economic research for Africa and Latin American countries suggests that an increase in reserve holdings benefit a country by lowering the interest rate spreads it faces in the international money market. For instance, some research for South Africa32 suggests that an increase in the ratio of reserves to imports might induce a reduction in the sovereign risk spread between 43 and 88 basis points.33

68. In addition, staff analysis in the context of the Article IV consultation for Uruguay,34 suggests that higher holdings of reserves, ceteris paribus, reduce country risk, increase investor confidence, and thus might reduce interest rates. For example, it is suggested that the elasticity of spread with respect to imports coverage is around minus one half. That is, a one percent increase in the ratio of reserves to months of imports might reduce the spread by 50 basis points. These numbers would, therefore, suggest that additional reserves might help Angola reduce its external financing costs.

69. Further benefits of additional reserve holdings arise from Angola’s oil dependency, which makes the economy highly vulnerable to developments in volatile world oil markets. In a nutshell, 60 percent of Angola’s GDP is generated in the oil sector; oil exports represent 90 percent of total exports; and oil tax revenues have averaged 81 percent of total government revenues during the last five years. Any drastic turbulence in world oil markets could therefore have severe consequences to Angola’s output and employment.

70. Another benefit stemming from a higher reserve norm for Angola relates to the conduct of monetary policy. First, given the relatively high levels of pass-through and dollarization (as commented by Gasha (2003)), any significant increase in the price of the dollar might negatively affect the rate of inflation. Less reserves imply less degrees of freedom for stabilizing the price of foreign currency via foreign exchange interventions. Second, lower levels of reserves might increase the need for large adjustments in the levels of imports. Given the high dependence on essential imported goods, any sizable increase in import prices (or reduction in import supply) would negatively affect aggregate supply and total output.

71. Finally, higher reserve holdings would give the Angolan authorities more flexibility to balance financing needs with macroeconomic adjustment in the event of an external shock. Indeed, the apparent slow speed of adjustment (derived from the high dependence on imports, the large impact of exchange rate movements on prices, and the rather high cost of external borrowing) suggests that if a shock buffets the Angola’s economy, the adjustment process might be particularly difficult. Moreover, given its current weak macroeconomic situation, Angola’s access to international capital markets might not be adequate. In general it is more difficult to borrow when reserves are most needed. Even though the existence of oil resources helps the country access capital markets, these funds might be accessible at higher costs than otherwise the case with larger reserves.


    Bailen, Jose, Juan CarlosJaramillo, andEvanTanner, 1999, “International Reserves Adequacy in Uruguay”, Selected Issues Paper, International Monetary Fund.

    • Search Google Scholar
    • Export Citation

    Calvo, Guillermo, 1995, “Varieties of Capital Market Crises”, Working Paper, University of Maryland.

    Gasha, Jose Giancarlo, 2003, “Angola: A Note on Inflation”, Selected Issues Paper, International Monetary Fund.

    Krugman, Paul, 1979, “A Model of Balance of Payments Crises”, Journal of Money, Credit and Banking 11.

    Obstfeld, Maurice, 1986, “Rational and Self-fulfilling Balance of Payments Crises”, American Economic Review, Vol. 76, No. 1.

    Vocke, Matthias, 2003, “Sovereign Risk Spreads Under Inflation Targeting,” in South Africa: Selected Issues Paper, by Michael Nowak and others. IMF Staff Country Report No. 03/18 (Washington: IMF).

    • Search Google Scholar
    • Export Citation

    Williamson, John, 1988, “Exchange Reserves as Shock Absorbers”, In The Open Economy: Tools for Policy Makers in Developing Countries, ed,R.Dornbusch andF.Helmes.

    • Search Google Scholar
    • Export Citation

Document prepared by Jose Giancarlo Gasha.


Angola’s exchange rate regime has been classified by the IMF as a “managed floating with no preannounced path for the exchange rate.” This classification is published in the monthly issues of the IMF’s International Financial Statistics.


For instance, see Williamson (1988).


The sample comprises countries, which have, according to the latest IFS classification (December, 2001), a managed float with no preannounced path for the exchange rate regime.


Due to data limitations, the import coverage estimates presented for Angola and the rest of African countries is measured against imports of goods only. Measured against goods and services, the import coverage for Angola would fall to 0.6 months by end-2002.


See for instance, Williamson (1988).


Includes all debt with a remaining maturity of one year.


Despite the fact that Angola’s loans are oil backed, the spread versus international interest rates remains wide.


The study pooled data for four inflation-targeting countries: South Africa, New Zealand, Poland and Thailand.


See for instance, Bailen et al (1999).

Other Resources Citing This Publication