Following the oil price rises of late 1973 and early 1974, several organizations predicted massive accumulations of international reserves by the major oil exporting countries by the end of 1980. 1 In the event, their balance of payments surpluses on current account and their reserve accumulations have been substantially lower than expected. 2 One reason is that world economic growth has been slower than assumed. But almost certainly the major source of error concerned the absorptive capacity of the oil exporting nations. The speed with which highly ambitious development strategies could be formulated and implemented in the oil exporting countries during the period after 1973 was not anticipated by many commentators in the early days of the oil crisis. Largely as a result of these factors, the balance of payments surplus on current account of the major oil exporting countries 3 declined from $68 billion in 1974 to $35 billion in 1977; and it has been projected to decline further, to around $10 billion in 1978.4 Further, this surplus is now concentrated among five countries of relatively low absorptive capacity—Saudi Arabia, Kuwait, the United Arab Emirates, Qatar, and the Socialist People’s Libyan Arab Jamahiriya. The current account position of the other seven oil exporting countries as a group was expected to move into deficit in 1978.
The major purpose of this paper is to begin analyzing the fiscal policy experience of the major oil exporting countries during the period 1972–78. Data are assembled on the major fiscal aggregates, distinguishing high- and low-absorption countries. 5 Where the data permit, attention is focused on the domestic, rather than the overall, budget balance in order to evaluate the impact of fiscal policy on domestic liquidity and prices.
The major findings of the paper can be summarized briefly:
(1) The growth of government expenditures in the major oil exporting countries over the period 1972–78 has, by any standards, been spectacular; in both absolute and percentage terms, it has been more rapid than the growth in their revenues.
(2) The rate of growth of government expenditures has been no less rapid in the low-absorption countries than in the high-absorption countries.
(3) The ratio of government expenditure to gross domestic product (GDP) in the major oil exporting countries jumped from 27 per cent in 1972–73 to 40 per cent in 1975, and it has continued to rise. The ratio of government revenue to GDP, however, has been on a declining trend since 1975, reflecting the changing composition of GDP, with an increase in the share of the comparatively lightly taxed non-oil sector in total GDP. As a result, the overall budget surplus was reduced to less than 1 per cent of combined GDP in 1977; and a substantial budget deficit (equal to $14.5 billion, or 4 per cent of forecast GDP) was projected for 1978.
(4) The high-absorption countries as a group have been incurring overall budget deficits since 1975; the low-absorption countries as a group continue to run overall budget surpluses, although the magnitude of these surpluses has been sharply reduced.
(5) During the period since 1972, there has been a close relationship, for the oil exporting countries taken together, between the overall budget surplus/deficit and the overall balance of payments surplus/deficit.
(6) In the oil exporting countries, fiscal policy is the primary determinant of domestic liquidity and aggregate domestic demand. It must be the primary instrument of demand management. For the six oil exporting countries for which data are available, there exists a close relationship between domestic budget deficits, domestic liquidity expansion, and inflation. For the same six countries, it appears that sharp reductions in the rate of growth of the domestic budget deficit have been closely associated with the restoration of domestic financial stability; earlier attempts to control inflationary pressures through subsidies and price controls, while highly expansionary fiscal policies were pursued at the same time, do not appear to have been successful.
(7) For the five oil exporting countries for which adequate data are readily available, preliminary tests indicate a stable relationship between the demand for real liquid balances and real non-oil GDP.
The plan of the paper is as follows. Section I outlines a framework for considering the impact of fiscal policy on monetary expansion in an oil exporting country. Section II provides a profile of fiscal policy in the oil exporting countries over the period 1972–78 and relates fiscal policy to domestic liquidity, price, and balance of payments developments. Section III discusses briefly policy approaches adopted by the oil exporting countries to achieve rapid growth within a framework of domestic financial stability and considers some of the major issues involved in the formulation and implementation of appropriate fiscal and monetary policies. An agenda for further research is provided in Section IV.
Mr. Morgan, economist in the Fiscal Analysis Division of the Fiscal Affairs Department, is a graduate of La Trobe University, Australia. He received his master’s degree and doctorate from the London School of Economics. Before joining the Fund, he was an economist at the Australian Treasury.
The estimates of the reserve increase between 1973 and 1980 ranged from $180 billion to more than $600 billion. A discussion of these estimates may be found in Rimmer de Vries, “The Build-Up of OPEC Funds,” World Financial Markets (Morgan Guaranty Trust Company, New York, September 23, 1974), pp. 1–10; “Survey: The Gulf,” Economist, Vol. 255 (May 17, 1975), p. 27; and Donald A. Wells, Saudi Arabian Development Strategy (Washington, 1976), p. 2.
The increase in international reserves of the major oil exporting countries between the end of 1973 and the end of 1977 is estimated to be in the range of $60–95 million. The former estimate derives from official estimates appearing in the Fund’s monthly publication, International Financial Statistics. The latter estimate is taken from the Fund’s Annual Report 1978 (Washington, 1978), p. 19, where it is noted that the dividing line between capital movements and reserve asset changes is uncertain for some oil exporting countries.
The 12 major oil exporting countries are Algeria, Indonesia, Iran, Iraq, Kuwait, Nigeria, Oman, Qatar, Saudi Arabia, the Socialist People’s Libyan Arab Jamahiriya, the United Arab Emirates, and Venezuela. This classification is consistent with International Financial Statistics.
See Bank of England, Quarterly Bulletin, Vol. 19 (March 1979), pp. 28–29.
Where this distinction is made in the remainder of the paper, low-absorption countries are considered to be the five mentioned previously, which have relatively undiversified economies as evidenced by small non-oil sectors, shortages of natural resources other than oil, and, typically, small populations.
Government imports will, however, involve domestic expenditure in the form of internal handling and transportation costs.
The domestic deficit data in this paper could not be adjusted to take account of these transactions. This is analogous to the problems raised by government transfer payments. Obviously, recipients of these transfer payments may use them to finance imports. This problem assumes considerable importance when the relationship between transfer payments and imports changes, as may occur when severe port capacity constraints are eased.
Given the limited role played by domestic nonbank borrowing in the economies considered here, the terms are used interchangeably for the remainder of the paper. The central bank often acts as a conduit for government investments from abroad. In oil exporting countries, such income is usually not recorded as central bank profits in the budget.
See Appendix I.
This approach can be complemented by an analysis of the supply of, and the demand for, real resources. The total supply of goods and services to the domestic economy is equal to the sum of GDP and imports minus exports. If non-oil GDP is defined to include the domestic activities of oil companies (such as construction, transport, and communications) and non-oil exports are insignificant, then aggregate domestic supply is closely approximated by non-oil GDP plus imports. Aggregate domestic demand (domestic absorption) equals government purchases of goods and services plus private sector expenditure. The obvious point to note from this simple and traditional framework is that the oil sector (or, more precisely, oil exports) makes no direct contribution to domestic supply; its indirect contribution is via imports.
Derived from International Financial Statistics.
In projecting U.S. dollar magnitudes for 1978, it has been assumed that the most recent exchange rates available at the time of writing (August 1978) between the U.S. dollar and the currencies of individual oil exporting countries would prevail for the remainder of the year.
There are a number of important differences between the balance of payments data from which oil export data are derived and central government fiscal data from which oil revenue data are derived. Balance of payments statistics are on an accrual basis, whereas budget data are more often on a cash basis. In some cases, fiscal years did not correspond to calendar years, necessitating difficult, and sometimes arbitrary, adjustments to the fiscal data. The budgetary systems of some oil exporting countries are highly fragmented, which creates difficulties in the derivation of comprehensive, consolidated central government data. Additionally, nonfinancial public enterprises engaged in oil production and marketing may retain some oil revenue, and state or local government shares of oil revenues will sometimes be excluded from central government data. Details regarding the coverage of the fiscal data are available on request from the author.
Data assembled by the U.S. Council on International Economic Policy derive average government revenue per barrel of oil, taking account of production costs, royalty and tax arrangements, as well as government equity participation. These data are fully consistent with the percentages cited above. See International Economic Report of the President (Washington, January 1977), pp. 174–76.
Decisions were also required on the appropriate rate of depletion of oil reserves. This paper assumes that, for the oil exporters as a group, the rate of oil production is, by and large, determined by the rest of the world’s requirements. It is also assumed that oil prices charged are, in the main, determined centrally (by the Organization of Petroleum Exporting Countries (OPEC)). Thus, this paper regards oil revenues as largely exogenous and focuses on the deployment of given oil revenues in pursuit of other objectives.
The government expenditure data include net lending—that is, lending undertaken for purposes of public policy rather than for management of government liquidity or for earning a return. For a discussion of the distinction between net lending and financing transactions, see the Fund’s, A Manual on Government Finance Statistics (Draft, Washington, June 1974), p. 130.
The projected large jump in the government expenditure/GDP ratio for the low-absorption countries in 1978 reflects, in part, low GDP growth (which, in turn, reflects projected developments in the oil sector).
Again, this estimate does not rely on unrealistic budget estimates of expenditure. Aggregate government expenditures in nominal terms were projected to increase by 20 per cent in 1978 over 1977 expenditures, and this is easily the lowest growth rate in the period since 1972. The sharp decline in the surplus reflects primarily the large impact of stagnant oil exports on government revenues and GDP growth.
Everett E. Hagen, The Economics of Development (Homewood, Illinois, 1975), pp. 430–31.
For a discussion of absorptive capacity with special reference to the oil exporting countries, see Farid Abolfathi, and others, The OPEC Market to 1985 (Lexington, Massachusetts, 1977), Chapter 2, pp. 23–32.
Data are drawn from International Financial Statistics. Oman, Qatar, and the United Arab Emirates are excluded from Section II.4 owing to unavailability of official price data.
Consumer price data are considered to provide the least unreliable indication of inflationary trends in the major oil exporting countries, but they remain subject to a number of serious deficiencies. Expenditure weights are often very outdated, and coverage is sometimes confined to relatively small segments of the population. Key items in the index have been heavily subsidized in many cases. In the oil exporting countries where widespread price controls have been in force, movements in official price indices may provide little guide to the extent of underlying inflationary pressures. These important reservations need to be kept in mind in evaluating the price data in Chart 3.
These are: Iran, Iraq, Kuwait, Nigeria, Saudi Arabia, and Venezuela.
These aggregate percentage changes are calculated in a manner analogous to those appearing in Chart 3—namely, percentage changes for individual countries are calculated in terms of domestic currency, and these changes are weighted by GDP expressed in dollars.
The pattern of movements in domestic liquidity and prices for these six countries is virtually identical to that in Chart 3, reflecting the fact that this subgroup accounts for almost 70 per cent of the weights employed there.
It should be noted that a significant proportion of balance of payments capital account transactions of the government would be included below the government budget surplus/deficit line.
Oman and Qatar are excluded from both series, since official data on their reserve movements are unavailable.
The divergence in 1977 may reflect, in part, increased foreign borrowing to bolster international reserves. Data from various sources that are assembled in the July 31, 1978 (Vol. 7) issue of the Fund’s fortnightly newsletter, the IMF Survey, indicate higher foreign borrowing in 1977 by several oil exporting countries, including Iran, the United Arab Emirates, and Venezuela. (See, in particular, pages 228, 231, and 232.)
Fund staff estimates indicate that the average annual growth rate in non-oil GDP in the 12 major oil exporting countries over the period 1974–76 was around 12 percent, and that it declined to 9–10 percent in 1977–78. See Annual Report 1978, op. cit., p. 14.
It could be argued that a reduction in import duties would contribute to an easing of inflationary pressures by leading to an increased flow of imports. This is not plausible for a large number of oil exporting countries during 1975–76, since the flow of imports was constrained not by deficient demand but rather by insufficient port capacity.
Clearly, there is a wide range of sectors (the nontradables sectors) where imports can make little or no direct contribution to improved supply availabilities; these sectors include government, real estate, water, electricity, gas, wholesale and retail trade, and consumer and business services.
In a large number of oil exporting countries, government subsidy expenditures declined in 1976 and 1977 as a result of the fall in international prices. Data on food price declines are provided in the Fund’s Annual Report 1978, op. cit., pp. 15–17.
In several oil exporting countries, the required data over a sufficient time span for estimating even the most rudimentary liquidity demand functions are either nonexistent or subject to severe limitations.
The Pertamina crisis is the most spectacular example of the problems that can arise from decentralized financial authority. Pertamina, Indonesia’s state-owned oil enterprise, entered into extensive foreign borrowing related to its large investment program, which expanded well beyond the confines of its role in the oil sector, and was only partly coordinated under the country’s development plan. In the second half of 1974, large cost overruns on a number of major projects, difficulties and delays in arranging long-term foreign financing, and a reduced flow of oil income led to Pertamina being unable to meet a large volume of short-term debt repayments falling due in 1975 and gave it little prospect of meeting substantial commitments falling due in future years. The Government committed itself to assist Pertamina in meeting its obligations, thus becoming involved in a difficult struggle to overcome severe balance of payments difficulties. This problem has been satisfactorily resolved.
The logarithmic form (semilogarithmic with respect to the rate of inflation) is used essentially for convenience, as elasticities are obtained directly. There is also some empirical evidence that this is the preferred functional form. See Bijan B. Aghevli and Mohsin S. Khan, “Government Deficits and the Inflationary Process in Developing Countries,” Staff Papers, Vol. 25 (September 1978), p. 388.
See William H. White, “Improving the Demand-for-Money Function in Moderate Inflation,” Staff Papers, Vol. 25 (September 1978), pp. 570–73.
The method used here is taken from Bijan B. Aghevli and Mohsin S. Khan, “Credit Policy and the Balance of Payments in Developing Countries” (unpublished, International Monetary Fund, December 27, 1976). It involves the generation of a series of values for π corresponding to alternative values of β ranging from 0.05 to 1.00. This requires the assumption that, at some point, actual and expected inflation rates were equal. The series of π were substituted into equation (10), and one was selected on the basis of the minimum standard error of the estimated equations.
The results for Indonesia and Saudi Arabia need to be treated with even more caution than those for the other countries because of the relatively low number of observations.
The short-run elasticity is given by the coefficient on real non-oil GDP. The long-run elasticity is derived by dividing this coefficient by one minus the coefficient on the lagged liquidity stock.
“If a country is experiencing a rate of economic growth, let us say of 8 per cent a year, in order to keep prices stable you will probably have to increase the quantity of money by something like 15 or 16 per cent a year. The reason for that is that a country at that stage of development is experiencing a spread of monetary institutions, a spread of the money-way of doing things. And therefore with every one per cent increase in income people desire to have a 2 per cent, roughly 1 1/2 to 2 per cent, increase in the real quantity of money, in the quantity of money measured in command over goods and services. Because people are coming to use money increasingly in their transactions, they are coming increasingly to adopt money as a store of value.” Milton Friedman, “Monetary Policy for a Developing Society,” Bank Markazi Iran Bulletin, Vol. 9 (March–April 1971), p. 710.
There has been little success in obtaining suitable proxies to isolate the monetization effect. Some attempts are described in Anand G. Chandavarkar, “Monetization of Developing Economies,” Staff Papers, Vol. 24 (November 1977), pp. 665–721.
Ibid., pp. 694–702.