Andrew D. Crockett and Duncan M. Ripley
The increase in oil prices that occurred at the beginning of 1974 transformed the world balance of payments situation. The major oil exporting countries developed current account surpluses that were likely to persist for many years; the rest of the world experienced unprecedented current account deficits. Such a situation presents dangers, coming at a time when the par value system has broken down, and there is no agreement yet on what should replace it. If all the oil importing countries seek to eliminate their current account deficits in the short run, disruption of trade and prolonged recession could result. This is because there is little that can be done to reduce the overall deficit of the oil importing nations. Attempts by individual countries to improve their balance of payments—whether by deflation, currency depreciation, or trade and payments restrictions—can only pass the problem on to others. The existence of accepted guidelines for the management of floating rates mitigates these dangers considerably. However, for effective implementation, the guidelines require some more precise notion of what constitutes a desirable medium-term balance in the current account.
In these circumstances, what is needed is a tacit or explicit agreement among the major oil importing countries to pursue consistent external policies. This means that each country must be prepared to accept its share of the overall current account deficit, and not try to eliminate it by means that simply export the problem to its trading partners. The difficulty is to decide on the appropriate way to share this deficit. The discussion that follows suggests that there is no unique “right” distribution; even if there were, it would probably change through time. Many different factors must be taken into account in forming any view on how best to share the deficit among the oil importing countries.
Balancing the accounts
Since, in the absence of reserve movements, the capital account must balance the current account, some forecasting of future capital flows is necessary to arrive at any view on how current account deficits should be allocated. A matching of the capital inflow with the current account deficit can be brought about by assessing how capital will flow, and then by taking measures to adjust the current account; by assessing where capital ought to flow, and then taking measures to adjust both the capital and current account; or by a combination of these two approaches.
Unfortunately, there is little information on where capital might be expected to flow in the absence of official action. Past experience is of limited use in making projections since the sums involved now are much larger, and the countries controlling these flows have had little opportunity to demonstrate their investment preferences. It is likely that these preferences will be strongly affected by both security and liquidity considerations, but just how various investments will be assessed in terms of risk is not clear. In addition, it is probable that these investors may prefer initially to hold liquid assets, but the returns on these assets are particularly sensitive to national monetary policies.
Because of these difficulties, the approach followed here is to determine where capital might be expected to flow on the basis of efficiency criteria, and to use this as a guide for allocating current account deficits among countries. Those who feel that the pattern of capital flows should not be guided by official action may view this approach as a short-term response to a lack of adequate information; others may view it as a more normative approach, since capital movements are affected by national monetary policies in any case, and by a number of market influences that may need to be counterbalanced if an optimal allocation of investment is to be achieved.
Although in principle the problem concerns all countries, only the allocation of the oil deficit among developed countries is dealt with here. This is partly to keep the analysis within manageable bounds, and partly because noneconomic considerations are likely to play a more important role in the case of many developing countries. Furthermore, the rather sweeping assumptions which are necessary to derive results are likely to be less unrealistic if applied to a relatively homogeneous group of countries, such as the developed, oil importing nations.
The analysis applies only to the additional balance of payments flows resulting from the oil price increases. It is therefore implicitly assumed that pre-existing objectives in the balance of payments field, whether or not strictly optimal from an efficiency point of view, can be accepted as “given” in the present analysis.
Effects on savings and investment
The inability of many oil exporting countries to spend their higher receipts in the short term or medium term will result in a large increase in the savings of these countries, which will of necessity be reflected by a fall in savings or an increase in investment in the rest of the world. However, it is by no means certain that this change in the distribution of world savings and investment will be achieved without affecting levels of employment and economic activity. It is well known that a sudden increase in thrift can generate a circular contraction in incomes if investment is not perfectly responsive to the greater availability of savings. Governments will presumably wish to resist this tendency, and our discussion assumes that policy responses are geared to maintaining a full employment level of output in each country.
The degree to which the additional savings by the major oil exporting countries has to be offset by deliberate policy action on the part of other countries depends to a considerable extent on whether one takes a Keynesian or a monetarist view of income determination. Under an extreme monetarist view, a current account deficit which was precisely offset by capital inflows would not affect aggregate demand, since it would have no effect on money balances. There might, of course, be a change in the distribution of expenditure, since the shift in the world savings schedule would be expected to reduce interest rates and thus encourage capital accumulation at the expense of current consumption. Under more Keynesian assumptions the reduction in real income in the oil importing countries, caused by the deterioration in their terms of trade, would be a more potent factor in reducing current consumption than the greater availability of finance would be in increasing investment. Thus additional measures would have to be taken to preserve full employment equilibrium.
There is, however, no very satisfactory way of establishing just how the distribution of expenditure between investment and consumption will be affected by the present situation. Most evidence suggests that the availability of savings is not a particularly important short-term constraint on investment in developed countries, so that private investment would not increase much in the short term in response to the availability of savings from the oil exporting countries. However, capital formation may well increase because there is substantial need for structural changes in many economies, particularly for investment in energy conservation and in alternative sources of energy production. Furthermore, governments may undertake investment projects on their own as a means of maintaining full employment.
Rather than make necessarily tentative assumptions about the precise nature of policy responses and about the relative sensitivities of savings and investment to interest rates and real income, our analysis proceeds in two stages. First, it is assumed that the additional savings of oil exporting countries are fully reflected in new investment in the rest of the world, and optimal allocation schemes based on this assumption are derived. Second, the assumption of no additional investment is made so that maintenance of full employment in the oil importing countries requires a large reduction in savings brought about through a reduction in their average propensity to save; how this reduction in savings might be allocated is examined. In principle, the results of these approaches can be combined to reflect what is perceived to be the most likely or appropriate response of savings and investment.
The distribution of new investment
The key question in this connection is how to distribute a given increase in global investment among countries. Clearly the optimum distribution of a given volume of investment from a global perspective is one which equalizes the expected marginal efficiency of capital in all employments. No reliable figures are available, however, on the marginal efficiency of capital in different countries, or on how this would change with additional investment. Incremental capital-output ratios are frequently not reliable, and in any case are not relevant in this context since they reflect increases in output resulting from capital formation in conjunction with changes in other factors of production.
To arrive at an optimal distribution of this additional investment, therefore, a number of simplifying assumptions are needed. One such assumption would be that the marginal productivity of capital was the same in all developed countries. This assumption could be justified on the grounds that substantial discrepancies in capital productivity tend to be eliminated by differential rates of capital accumulation and by capital movements across national frontiers.
A second and perhaps more important assumption is needed about how the marginal product of capital will change in response to additions to the capital stock. The most straightforward assumption would be that, at any one time, proportionate additions to the capital stock in different countries would result in equal reductions in the marginal efficiency of investment. In other words, equality in the rate of return on investment would be preserved between countries when the capital stock of each country was increased by the same proportion.
Investment and capital stock
If both of the foregoing assumptions are accepted, it follows that the additional capital from the oil exporting countries should flow to other countries in proportion to the existing capital stock of these countries. Direct estimates of capital stock could be used in this context, but in general these estimates are of rather doubtful reliability. Without undue violence to the facts, however, it might be assumed that a country’s gross national product (GNP) is proportionate to its capital stock. To the extent that the value of a given stock of capital is the discounted value of its product, it could even be argued that GNP was a better measure of capital value than a more direct estimate. It would then follow that countries’ optimal shares in the additional investment would be in proportion to their GNP (see column 1 in the table).
These assumptions concerning capital productivity imply that, all other things being equal, the rates of capital accumulation ought already to be the same in all countries. Since rates of accumulation over time have not been equal, we may consider to what extent these differences suggest that the stated assumptions about the productivity of investment are inaccurate, or reflect different rates of growth of other factors of production, for example, the labor force.
One possibility is that a high investment country may have deliberately accepted a lower marginal productivity of capital, and have oriented its domestic policies so as to increase the rate of capital formation and maximize growth. In a riskless situation it should be noted that this result could come about only if there were imperfections in the world capital market, or if certain countries pursued policies specifically designed to encourage investment. Differences in the physical productivity of capital in the various countries could be considered as “given,” since they reflect social or political considerations which are not taken account of in a purely economic calculation. In any case, there is no reason for additional investment to be directed disproportionately toward these countries. Under certain assumptions the existing discrepancies in marginal rates of return between countries would be maintained if the additional investment continued to be distributed in proportion to the existing capital stock.
Alternatively, a relatively high rate of capital accumulation in a particular country may indicate that its marginal efficiency of capital schedule declines less rapidly than that of other countries with additional units of investments. If so, this should be reflected in the distribution of the additional investment, and this country should receive a larger share of the additional investment than other countries. Under some circumstances such a scheme might be based on shares in net capital formation in the various countries rather than on shares in GNP. In the absence of consistent and reliable information on capital depreciation, gross investment can be used as a proxy for net capital formation (shares in gross domestic investment are shown in column 2 in the table).
Before a satisfactory distribution scheme for allocating additional investment can be proposed, it would be useful to understand what lies behind existing variations in the rate of capital accumulation. However, in the absence of further information on how best to combine these considerations, a heuristic distribution could be constructed as an average of the distribution based on proportionate additions to capital stock (proxied by GNP) and the distribution based on proportionate additions to increases in capital stock (proxied by gross investment).
Reductions in savings
The opposite extreme from assuming that the whole of the oil exporting countries’ increased savings is translated into additional investment in the rest of the world, is to assume that there is no response of investment. Although it is probable that investment will respond to a large extent in the medium term, the immediate increase is likely to be quite small.
If investment in oil importing countries does not respond to an increase in savings by oil exporting countries, and if there is not to be a fall in employment, the net inflow of capital from the oil exporting countries must be matched by a large reduction in the average propensity to save, and thus the level of savings, in the capital receiving (oil importing) countries. The fall in real incomes in the oil importing countries will itself contribute to a reduction in desired saving, and changes in financial market conditions may also affect the savings propensity of the private sector. In the short run, however, it seems unlikely that the reduction in domestic savings in the oil importing countries will be sufficient to offset the demand reducing effects of the oil situation, and, therefore, further government intervention will be necessary in the form of a change in the budgetary position of the public sector.
If the increase in saving by the major oil exporting countries has to be matched by a reduction in saving on the part of the rest of the world, the problem for the adjustment process becomes one of how such a reduction should be shared in order to minimize the welfare loss to the group of oil importing countries. In a perfect market situation, one might expect the fall in real income in oil importing countries, combined with the fall in the interest rate (reflecting the decline in the interest sensitivity of global savings), to squeeze out marginal savers. The allocation of the required reduction in saving among these countries would then be to some extent a function of different rates of time preference between savers in different countries.
However, given the goal of maintaining full employment, and assuming that in the short run savings may not be very sensitive to the interest rate, it seems unrealistic to expect that purely market forces will bring about the most efficient allocation of the required reduction in savings in oil importing countries. Much of the reduction in savings may have to be achieved by government action. Nevertheless, by means of some not-too-unrealistic assumptions, it is possible to get some useful answers as to how a reduction in savings might best be shared among the oil importing countries.
A two-step approach
The approach followed here is in two stages. First, it considers the reduction in savings in countries due to the fall in real income generated by the increased real cost of oil. Second, it allocates the additional reduction in savings, which will result from direct government intervention, according to an optimal sharing scheme.
|GNP as per|
cent of total
as per cent
as per cent of
|Exports to oil-|
as per cent
in deficit in
share in deficit
|Federal Republic of Germany||9.7||11.4||11.0||11.0||13.7||11.5||5,750|
|14 Industrial Countries||93.0||92.9||90.9||92.6||94.9||92.7||46,350|
|Other Developed Countries||7.0||7.1||9.1||7.4||5.1||7.3||3,650|
|All Developed Countries||100.0||100.0||100.0||100.0||100.0||100.0||50,0001|
Recent developments suggest that the total figure may be high for the medium term.
Recent developments suggest that the total figure may be high for the medium term.
As a first step, it is assumed that full employment output is maintained, and that real income levels in oil importing countries will have been reduced in direct proportion to their higher oil bills. (The relative distribution of the increases in these bills is shown in column 3 in the table.) It seems reasonable to assume, for purposes of exposition, that the first ingredient in the optimal sharing of the required reduction in savings would reflect the initial fall in savings resulting from the reduction in real income.
In principle, it should be possible to compute the marginal propensity to save in individual countries and to multiply the resulting figures by the real income loss resulting from the oil price increase. To keep the present analysis simple, however, it is assumed that the marginal propensity to save is 0.2 in all developed countries. This would mean that the first 20 per cent of the required reduction in saving by oil importing countries as a group would be allocated in proportion to the oil “impact effects,” that is, in proportion to the increase in expenditure for oil imports.
The allocation of the remaining 80 per cent is somewhat more complicated, depending on assumptions about the productivity of savings. First, let us assume that, prior to the oil developments, the marginal return to savings had been more or less equalized across countries. This assumption is not likely to hold precisely, but one may nevertheless note that, before these developments, countries were prepared to accept whatever discrepancies existed in the return on savings without desiring any substantial capital flows to correct them. Next, let us assume that the elasticity of savings with respect to changes in interest rates is the same in all countries. Then, any reduction in savings beyond the 20 per cent of the “impact effect” should be allocated in proportion to existing shares in global savings. (The distribution of gross savings is shown in column 4 in the table, and can be seen to be very similar to that of gross domestic investment, shown in column 2.)
Additional current account effects and dynamic considerations
The optimal distribution schemes mentioned here for capital inflows from the oil exporting countries are based on savings and investment considerations, assuming an “oil surplus” of a given size. However, it needs to be borne in mind that the aggregate oil surplus will presumably decline over time, as the oil producing countries increase their imports and cut back on financial investment abroad. If the future imports of the oil exporters come from the same countries which presently receive petrodollar inflows, there will be no major balance of payments consequences; current account transactions will take the place of capital transactions with no change in the overall balance.
However, the pattern of capital inflows as described in the preceding analysis is not likely to correspond precisely to countries’ prospective shares in the import growth of the oil exporting countries. A better key to the distribution of future imports would be the current exports of the oil importing countries to the oil exporting countries. (Country shares in these exports are given in column 5 in the table.) Countries with relatively large shares in these markets may be expected to have an underlying improvement in their balance of payments, since they will gain more on current account (as these countries increase imports), than they lose on capital account (as these countries reduce capital exports). This implies an appreciation of their currencies vis-à-vis those of countries not benefiting so much from increased absorption in the oil countries.
The consequences of anticipated exchange rate changes have not so far been taken into account, however, although the earlier discussion of optimal current account deficits, reflecting capital flows, in fact implies certain types of exchange rate adjustment. Expectations play no role as long as exchange rate changes are viewed as a once-for-all adjustment; under these circumstances, exchange rate changes need not affect the distribution of savings and investment. Although the income from a given volume of physical investment will have gone up in an appreciating country, the supply price of capital will have changed in roughly the same proportions, leaving percentage yields roughly constant. However, to the extent that further exchange rate changes are anticipated, investment in a country whose currency is expected to appreciate will become more attractive since an exchange rate gain will be added to the marginal real yield. This will lead to larger inflows for countries whose currencies are expected to appreciate (for example, because of their disproportionate share of growth in imports by the members of the Organization of Petroleum Exporting Countries) than could be justified by the savings and investment considerations analyzed earlier. This factor would imply a greater degree of appreciation for these countries, even in the short run, and would bring about equilibrium in the overall balance of payments.
Thus far emphasis has been on economic efficiency as a criterion for the allocation of capital inflows. In practice, however, there are imperfections both in capital and exchange markets. Furthermore, financial investment decisions reflect a number of factors other than the physical productivity of capital; these include liquidity and risk considerations. Consequently, the pattern of financial flows is not likely to conform automatically to any of the patterns suggested by the simple models that have been set forth above. Official policies may be used to secure an “optimal” allocation of the overall deficit, but to the extent that this allocation conflicts with forces existing in the capital and exchange markets, attempts to secure this allocation may result in substantial strains being introduced into these markets. Since the avoidance of such strains is a legitimate objective of adjustment policies, it is useful to consider the extent to which existing financial structures and market mechanisms might give rise to a pattern of capital flows different from that expected on efficiency grounds.
One factor that is likely to be important both for economic and noneconomic reasons is the distribution of the impact of the oil price increase on countries’ balance of payments. This impact distribution represents a basic situation which can be modified to a greater or lesser degree through policy actions. It has already been discussed in connection with the optimal distribution of savings and investment. In addition, it will influence the sharing of the deficit that countries are willing to accept.
Another factor that is likely to affect the ability of oil exporting countries to invest their surplus funds is the relative depth of financial markets in capital receiving countries. It is generally thought that the capital markets of London and New York are more broadly based and operate more flexibly than those of other financial centers. This is cited as one reason why a relatively greater proportion of the capital from the oil exporting countries has flowed to the United Kingdom and the United States than to other developed countries. Insofar as this flow is likely to continue, and these funds are absorbed domestically, it could be argued that these two countries are able to bear a somewhat larger share of the overall deficit than would be suggested on the basis of the assumptions put forward earlier in this article. However, to the extent that the initial flows are immediately re-lent to other countries, the size of these inflows need not be important in determining the optimum sharing of the current account deficit.
Servicing and repayment
Another factor which may possibly influence the investment behavior of oil exporting countries is their confidence in the ability of capital receiving countries to ensure the servicing and perhaps ultimate repayment in real resources of their financial claims. This factor, of course, has a political dimension which is beyond the scope of the present piece. But there are also economic considerations bearing on confidence. One might expect investment in countries which have had a strong balance of payments record in recent years backed by large reserve holdings (corrected for compensatory borrowings or other official liabilities to foreign monetary authorities) to be more attractive than investment in countries whose balance of payments and reserve positions have been weak. Thus, the relatively weak reserve positions of Italy, the United Kingdom, and under some circumstances, the United States, might largely offset any advantages which these countries possessed in terms of the superiority of their financial markets.
Finally, it will probably be the case that those countries which are in a relatively favorable position to develop alternative sources of energy will be more attractive outlets for surplus funds, ceteris paribus. There are two reasons for this. In the first place, their capacity to absorb additional financial resources in real investment will be greater; and second, their ability to service debt and ultimately affect a large transfer of real resources will be enhanced. Canada, Norway, and the United Kingdom have larger energy resources relative to their GNPs or to the size of their trading sectors than do most of the other developed countries, and this criterion would argue for a relatively greater flow of investment to these countries than might otherwise be justified.
Basis for further study
The overall distribution of the oil deficit based on our assumptions thus constitutes a “first approximation” which might be modified, to some extent, to take account of two other factors. First, the distribution of the impact effect, representing as it does a given element in an uncertain situation, might be expected to play a direct role in the eventual outcome as well as through its effect on real income and saving. Second, shares in exports to the oil exporting countries will influence balance of payments developments, and may also have exchange rate implications that affect capital flows. To take account of these modifying factors, one might assign a weight of 50 per cent to the “first approximation” and weights of 25 per cent each to the two modifying distributions. This subjective distribution scheme was calculated for the group of developed countries. (The shares of the industrial countries are given in column 6 in the table.)
The increase in the annual current account deficit of the developed countries with the oil exporting countries has been roughly estimated at US$50 billion between 1972 and the second half of 1975 (a distribution of this deficit based on the subjective weighing scheme is given in column 7 in the table.) The deficit allotted to the industrial countries under this approach totals US$46 billion. Although this distribution of the deficit is only one plausible allocation among many, it is presented as a contrast to current account projections, and as a focus for further consideration.