Monetary Policy and Internal and External Balance
  • 1 https://isni.org/isni/0000000404811396, International Monetary Fund

THIS PAPER is concerned with ways in which monetary instruments, implemented by a central bank acting on its own, might contribute to the objectives of internal and external balance within a framework of highly integrated money and capital markets.1

Abstract

THIS PAPER is concerned with ways in which monetary instruments, implemented by a central bank acting on its own, might contribute to the objectives of internal and external balance within a framework of highly integrated money and capital markets.1

Introduction

THIS PAPER is concerned with ways in which monetary instruments, implemented by a central bank acting on its own, might contribute to the objectives of internal and external balance within a framework of highly integrated money and capital markets.1

Four familiar situations involve deviations from both internal and external balance. These are (1) a balance of payments surplus together with a domestic inflation, (2) a balance of payments surplus together with a domestic recession, (3) a balance of payments deficit together with a domestic inflation, and (4) a balance of payments deficit together with a domestic recession. Consider now the use of traditional monetary instruments (discount policy, open market operations, reserve/liquidity requirements) to deal with these situations. In situations (2) and (3), monetary policies would generally contribute to the correction of both internal and external imbalance.2 An expansionary monetary policy in (2), for example, will at the same time reduce the surplus and create more employment; a contractionary monetary policy in (3), on the other hand, will reduce both the deficit and the rate of inflation.3

Situations (1) and (4) are more complicated in that they create familiar dilemmas and challenges for monetary policy.4 It is useful, therefore, to review and evaluate the monetary policy options in these conditions and to examine ways in which these dilemmas can be, and have been, decided. The discussion will confine itself largely to situations resulting from the existence of both unwanted surpluses5 and inflation—i.e., situation (1).

It is possible to distinguish three kinds of monetary policy that might be pursued in the conditions assumed. First, priority could be attached to the objective of internal balance (controlling the rate of inflation). This involves controlling the cash base and/or use of the cash base (e.g., cash reserve requirements) by traditional techniques in such a way as to restrict the growth of liquidity. There is now an extensive literature on the difficulties involved in trying to follow this kind of policy. Basically, the discussion has centered around two problems.6 One deals with the extent to which the central banks in the industrial countries in continental Europe, faced with this situation, were equipped with adequate policy instruments to enable them to control the rate of growth of liquidity. There were, for example, the familiar constraints on the use of open market operations;7 in addition, other available policy instruments (e.g., cash/liquidity reserve requirements, credit ceilings, and rediscount policies) tended in varying degree, depending on the country, to be inadequate to enable complete or partial neutralization of the effects of changes in the cash base owing to external surpluses. What this means is that inflation tended to be imported simply because the effects of the external surpluses on domestic liquidity could not be offset. In other words, the result of having inadequate instruments is a situation of involuntary imported inflation. Another problem concerns the possibility that this kind of policy will in the end be self-defeating, in the sense that as long as domestic liquidity is being restricted, the shortage in liquidity will tend to be met from external sources. This amounts to saying that it is impossible for countries to follow independent interest rate policies; and any attempt to do this is bound in the end to be frustrated by inflows of capital. This proposition tends to hold in a world of perfect capital mobility but, as long as we depart from this unrealistic assumption, independent monetary policies are possible—although, of course, such a policy will tend to be reflected in large movements in short-term capital flows. In other words, we suppose that central banks are able to exercise control over the money supply.8 To sum up, then, while a policy of controlling liquidity may not be feasible because instruments are inadequate, it need not be self-defeating (assuming the instruments are adequate) as long as capital mobility is imperfect.

Second, priority could, alternatively, be attached to eliminating or reducing the external surplus by allowing the growth of liquidity to respond to the surplus. The dilemma created by this policy is that while it may reduce the surplus it will tend to aggravate the rate of inflation. In effect, the choice involves opting for importing inflation as a result of a voluntary decision, as distinct from importing inflation involuntarily because policy instruments are inadequate.

A third kind of policy involves attenuating the dilemma by pursuing monetary policies that while they control the rate of inflation by restricting the growth of liquidity will at the same time limit or discourage the inflow of capital. In an extreme form, this policy could reconcile internal/external objectives in respect of direction by encouraging a sufficient shift out of domestic assets and into foreign assets.9 Policies aimed at promoting both internal and external balance are of three types: first, the use of credit ceilings both to minimize the interest rate effect and to discourage expansion of domestic assets by the banking system; second, the maintenance of an interest rate differential while at the same time implementing monetary policies directed at the banking system and aimed at discouraging the inflow of capital; third, a policy on the structure of interest rates, which involves keeping short-term rates low, to discourage any inflow of capital, and long-term rates high, to discourage domestic expenditure. Each of these policies will now be discussed in somewhat more detail.

I

Many countries in continental Europe (including Belgium, Denmark, Finland, France, the Netherlands, Spain, Sweden, and Switzerland) have made active use of ceilings on bank credit as a monetary instrument.10 The motives were mixed and varied in the different countries. The principal rationale was that direct methods of control over bank credit not only were more effective than the indirect methods of control but also made their impact more quickly. Another motive was to minimize the fluctuations in interest rates, fluctuations that were considered disruptive to capital markets and the management of the public debt. Of more direct interest to this paper is the view that resort to credit ceilings can help to moderate the conflict between internal and external balance (e.g., inflation and a balance of payments surplus).11 This view rests on the assumption that credit ceilings, as against an interest rate policy (usually identified with changes in the discount rate), may have the same restrictive effect on inflation but at the same time provoke less net capital inflow.

Suppose that the discount rate is unchanged; that credit ceilings are successful;12 that banks’ assets are made up of cash, government securities, foreign assets, and credit; and that ceilings apply to the last category. Obviously, if no expansion in credit is permitted, repatriation of foreign assets could be profitable only if the interest rate on government securities were higher than on foreign assets. Normally, on the Continent, this would entail purchasing government securities from the central bank, since there is only a narrow market in securities. In addition, frequently, holdings of government securities are small, so that there is little scope for expansion; in any case, the interest rate on securities tends to be lower than on advances. Hence, there may well be some discouragement to repatriation. For the same reasons there will be little incentive to borrow in foreign markets and to invest domestically; finally, there may be some incentive to divert a passive inflow13 of funds into foreign markets.

With discount rates unchanged and deposit and advance rates likely to remain unchanged, if some limited expansion in credit is permitted there will be less incentive for the banks to finance the expansion by borrowing overseas. Also, with bank and money market rates kept down in this way, there will be less incentive for foreigners to place their funds in domestic markets. On the other hand, of course, the private sector, denied funds from the banking sector, will have an incentive to seek funds overseas.

II

Forward market intervention represents one familiar method by which short-term capital movements may be influenced and independent monetary policies maintained. There are two types of forward market intervention. One involves intervention in the open market to influence the forward premium or discount, this rate being available to all participants in the forward market. This is the method that has been used extensively in the United Kingdom, the United States, and Canada. Another involves setting a premium or discount in a “closed” market, which is available only to the banking system. This is the device more commonly used on the Continent, and we shall limit our comments to this form of intervention. In this form of intervention the central bank sells spot dollars to the banks and buys them back at rates sufficiently favorable to weaken the incentive to invest in domestic assets.14 Suppose, for example, that the domestic interest rate is higher than the overseas rate; in these conditions the central bank can offer a forward discount on the domestic currency that will discourage the repatriation of foreign assets and will encourage the investment of excess liquidity in overseas markets. The central bank would thus bear the cost of purchasing dollars forward at a more expensive rate than the rate at which it sold the dollars. The size of the forward discount can be made to depend on the interest rate differential, and, since it is a special rate set for the banks, it need not bear any close relation to the forward premium/discount that prevails in the free market.15 Indeed, in conditions of external surplus, to the extent that the forward rate reflects speculators’ expected future spot rate, any anticipated revaluation of the domestic currency will tend to be reflected in the free market in a premium instead of a discount on the domestic currency.16

A number of European central banks have engaged in swap operations with the objective of reducing domestic bank liquidity and their own foreign reserves17 (excluding, of course, forward commitments). Germany began using swap operations early in 1959 and continued until late in 1962 when they were discontinued.18 Swaps were reinstated during 1964 and 196519 and again late in November 1967 and were used vigorously during August and September 1968, when there was a heavy inflow of foreign exchange. On August 29, for example, the central bank lowered its discount on forward dollars from 4 per cent to 3 per cent, and the next day this discount was further reduced to 2.5 per cent. As a result of these changes the covered differential swung sharply in favor of investments in the Euro-dollar market.20 In Italy, at the end of 1959 swap facilities at par became available, subject to a ceiling for each bank. Without this facility sales of forward dollars in the open market would have resulted in a forward discount on the dollar and hence would have discouraged investment in foreign assets.21 Since the end of 1965 the facility has been available only to banks with a net debit position vis-à-vis foreigners. The Dutch, Swiss, and French authorities have also at times offered swap facilities to their banks at preferential rates.

Special reserve requirements on gross or net foreign liabilities represent a second device by which capital inflows through the banks might be discouraged and capital exports encouraged. There are many possible variations to this theme. Reserve requirements may be set in relation to all nonresident deposits, including deposits in domestic currency; alternatively, they may be limited to foreign currency deposits held by nonresidents (and perhaps even by residents). Reserve requirements may be “waived” to the extent that foreign liabilities are offset by holdings of foreign assets (in other words, the reserve requirement is on net liabilities). Correct policy may be further obscured because loans in foreign currency may be placed in foreign markets (to nonresidents) or extended to residents.22 Some loans in domestic currency may be made to nonresidents.

Consider now the differences between a reserve requirement on the banks’ gross foreign liabilities (defined as liabilities in foreign currency and domestic currency to nonresidents) and a reserve requirement on the banks’ net foreign liabilities (where assets are defined as lending in foreign currency to nonresidents). It is useful to distinguish four types of potential inflow in response to higher domestic interest rates. First, the banks may actively seek foreign funds to convert into domestic currency and to expand their lending to residents (Case 1). Second, nonresidents may take the initiative and place funds in the domestic banking system, i.e., passive inflow (Case 2). These liabilities may be in foreign or domestic currency. Third, the nonbanking sector may experience a net inflow of capital, which would normally show up as an increase in liabilities to residents (Case 3). Fourth, the banks may have the incentive to repatriate their existing foreign assets and to expand their lending to residents in domestic currency (Case 4).

We want now to evaluate the effects of the two types of reserve requirement (gross and net) on the likely net inflow of funds. Suppose that the purchase of a domestic asset entails a loss of bank cash (liquidity) equivalent to the purchase of a foreign asset. That is to say, in evaluating the choice between investing domestically or in a foreign market it is assumed that there is no “return flow” accruing from the purchase of a domestic asset. This assumption may be justified on one of two grounds. One is that the domestic asset is purchased from the central bank; the other is that the banking system is sufficiently large that an individual bank does not assume in its calculations that the acquisition of a domestic asset (e.g., a loan to a resident) will result in significant “return flow” (i.e., it disregards the possibility of a multiple expansion of credit). In a monopolistic banking system, where our assumption obviously does not hold, it can easily be demonstrated that, given an accrual of liquidity as a result of an external surplus, the foreign interest rate would have to be significantly higher than the domestic interest rate to offset the multiple expansion of domestic assets. We also disregard the costs of forward cover and the profit margin to the banking system. Suppose now that Rd, Rf = the domestic and foreign interest rates, respectively, on bank assets; Rb = the interest rate at which banks may borrow overseas; x = the potential inflow of funds to the banking system; rf = the reserve requirements on gross or net foreign liabilities; rd = the reserve requirement on resident liabilities.

Case 1

Gross and net reserve requirements. The cost of seeking funds in overseas markets = xRb; the return on a domestic investment = x(1 - rf) Rd23 In equilibrium, at the margin, when the cost of borrowing is equal to the return on domestic investment, RbRd=1rf. It is clear that rf can be set at such a level that any potential gain from borrowing to switch into domestic assets, owing to differential interest rates, may be neutralized. For example, if rf is as high as 0.2, there is no incentive to switch as long as the borrowing rate is below 80 per cent of the domestic interest rate. The existence of a gross or a net reserve requirement will discourage borrowing to invest domestically.

Case 2

Gross reserve requirements. In this case there is a passive inflow of funds, which we assume that the banks accept. The choice for the banks is between a domestic and a foreign investment. The existence of a reserve requirement on gross liabilities reduces the return on the accrual of funds, but it does not affect the choice between a domestic and a foreign investment, which will be determined by the interest rate differential.

Net reserve requirements. Now, xRf = the return on the foreign investment (since there is no reserve requirement when there is an increase in both foreign liabilities and foreign assets); x(1 - rf) Rd = the return on the domestic investment. Again in equilibrium, RfRd=1rf, so that the reserve requirement can be such as to neutralize the interest rate differential. It is clear, then, that in this case a reserve requirement on net liabilities will provide an additional incentive to place the incoming funds overseas.

Case 3

Since gross foreign liabilities are not affected in this case, a reserve requirement on gross liabilities cannot influence incentives in any way. With a reserve requirement on net liabilities, x(1 - rd) Rd = the return on the domestic investment; x(1 - rd) (1 + rf) Rf = (roughly) the return on the foreign investment (remembering that to the extent that a domestic deposit is invested overseas net foreign liabilities and, hence, reserve requirements will fall). In equilibrium, RdRf=1+rf. There is again a clear additional incentive in the case of reserve requirements on net liabilities to place the funds overseas.

Case 4

Again, gross liabilities are not affected by the repatriation of foreign assets, so that a reserve requirement on gross liabilities will not affect the incentive to repatriate. On the other hand, repatriation increases net liabilities and, hence, reserve requirements. In equilibrium in this case, as in Case 2, RfRd=1rf. The results of these four cases are summarized in the following tabulation.

article image

The results demonstrate that reserve requirements on net liabilities are more comprehensive and provide a stronger inducement to divert newly acquired funds into foreign assets and a greater discouragement to the repatriation of existing foreign assets.

Three additional points are worth making on this analysis. First, the distinction between Cases 1 and 2 centered around who took the initiative, the banks or the nonresidents. Of course, to the extent that banks actively seek out foreign depositors the distinction becomes blurred. Also, if banks want to discourage foreigners from depositing funds they can presumably reduce the interest rate that they will pay on foreign deposits. Second, the options could also be evaluated in terms of a choice between foreign and domestic borrowing by the banks, given the returns available. In other words, the banks will try to minimize the cost of obtaining funds. If the reserve requirement is higher on foreign borrowing than on domestic borrowing, they will have some incentive to borrow domestically, even at higher domestic interest rates. Third, effective credit ceilings (and disregarding the possibilities of investing in securities) generate the same incentive effects as reserve requirements on net foreign liabilities. With credit ceilings, if an accrual of liquidity is not invested overseas, it will be subject to a de facto marginal reserve requirement of 100 per cent. In this case the effective return on domestic assets is zero, so that any return on foreign assets will be sufficient incentive to place or to retain funds overseas.

Special reserve requirements on liabilities to nonresidents (irrespective of currency) have been used extensively since 1957 in Germany, and under legislation introduced in July 1969 they may be set as high as 100 per cent.24 In 1957 these reserve requirements were set at a higher rate on foreign than on domestic (resident) liabilities.25 From May 1961 reserve requirements on foreign liabilities were waived when these were offset by certain foreign assets.26 The offset formula was somewhat complicated in that the waiver applied first to offsets against foreign sight liabilities on which reserve requirements were highest and then against foreign time liabilities; this meant that the biggest gain applied to the offset against sight liabilities, the size of the gain depending on the reserve requirement rate. The offset right was suspended early in 1967. In December 1968 the minimum reserve ratio on the increase in the banks’ gross foreign liability was set at 100 per cent.27 In recent years this minimum ratio has been changed on a number of occasions.

In Switzerland early in 1964 something like a 100 per cent reserve requirement was introduced on increases in net nonresident liabilities in Swiss francs. The banks were, in effect, given the option to deposit 100 per cent reserve requirements on increases in these liabilities with the central bank or to purchase foreign currencies. Since no interest was paid on frozen deposits with the central bank and some return could be earned on foreign currencies, the banks had a strong incentive to opt for the second alternative. The 100 per cent reserve requirement was lifted in October 1966. A reserve requirement of 100 per cent was reintroduced in August 1971 under a new agreement that permits the imposition of a reserve requirement against the increase in net foreign positions. In Denmark a 100 per cent reserve requirement (which could be met by deposits of cash or bonds) was in force for a time for increases in the net foreign liability position beyond the January 1965 level.28 Also, a certain percentage of any increase in domestic deposit liabilities had to be absorbed either in deposits or bonds with the central bank or in the purchase of foreign assets. In Austria the central bank has powers to impose reserve requirements against both the level and the growth of net foreign currency liabilities. In Norway, too, the central bank has powers to impose special reserve requirements against increases in foreign liabilities.

A third device for discouraging capital inflows and reducing liquidity is to direct the banks to increase (reduce) their net foreign assets (net liability) positions.29 This direct control contrasts with the indirect method of control involved in swap operations and special reserve requirements. Italy began using this method in 1960 when the banks, which had a large net foreign liability position, were experiencing large increases in their liquidity position as a result of an external surplus. The banks were instructed to liquidate their net foreign liabilities denominated in foreign currencies.30 The banks met this requirement by building up their foreign assets rather than by reducing their foreign currency liabilities. Directives to the banks, adapted to changing conditions, have continued to be issued from time to time since then. In the Netherlands, too, the device has been used since 1964, when, to reinforce a policy of monetary restraint, a ceiling was set on the net foreign liability position (including the position in guilders) of the banks.31 The French banks were under direction until the end of 1966 to maintain a balanced external position in spot and forward combined. This did not act as a serious deterrent to capital inflows, since it was possible for banks to meet the requirement by borrowing spot dollars, converting the dollars into domestic currency for lending, and buying the dollars forward. (Hence, a spot liability is matched by a forward asset.) Since 1969 the French authorities have used directives to the banking system as a device for influencing capital movements through this sector.

A fourth device for restricting capital inflows is to prohibit the payment of interest to nonresidents on certain time or demand deposits. Germany,32 Switzerland, France, and the United Kingdom have all used this measure, and in Sweden nonresident deposits with banks normally do not bear interest. Of course, in periods of exchange rate uncertainty nonresidents will continue to have the incentive to place their funds in deposits denominated in a currency that is expected to revalue. Also, since some of the inflows stopped by this device might find their way into the domestic market securities, Germany and Switzerland have found it necessary to take measures to discourage investment by nonresidents in their domestic markets.

III

A third type of monetary policy, intended to mitigate the conflict between internal and external balance, involves manipulating the term structure of interest rates. In conditions of inflation and a balance of payments surplus this entails keeping short-term rates low to discourage net inflows and keeping long-term rates high to moderate inflation. The only country where this general type of policy has been tried is the United States, when in the early 1960s, faced with unemployment and a deficit in the balance of payments, an attempt was made to maintain long-term rates relatively low and to raise short-term rates (a policy commonly referred to as “operation twist”). The Treasury and the Federal Reserve System together reduced the maturity of the existing debt by selling short-term securities and buying long-term ones. In addition, maximum rates payable by commercial banks on their time and savings deposits were raised.

There are, in general, two difficulties with this approach. First, there is the problem of the feasibility of bringing about any significant change in the term structure by changing the composition of the debt. Second, supposing that the structure is changed along the lines suggested, there is the problem of whether it will achieve the objectives intended. On the first question, while term structure theory can demonstrate that a change in the composition of debt will, in normal conditions, change the term structure of interest rates,33 it has in fact proved extremely difficult to measure the size, if any, of this effect. Indeed, most studies in this field give very conflicting results, so that it is impossible to evaluate the sensitivity of the term structure to changes in the composition of debt.34 A study of “operation twist” by Modigliani and Sutch concluded that the narrowing of the term structure differential at the time was due not so much to the change in the composition of the debt as to the cyclical conditions then prevailing.35 Another related question has to do with whether or not central banks are equipped to implement a policy of this kind in the short run. Most central banks on the Continent cannot carry out open market operations of the type required to change the composition of the debt in the direction required. On the other hand, since markets would be more segmented on the Continent than in the United Kingdom and the United States, conditions would appear to be more favorable to bringing about adequate changes in the structure by changing the maturity of new issues or changing short-term rates sensitive to the discount rate.

The second problem is also serious. Consider a term-structure policy, in conditions of inflation and a balance of payments surplus, that successfully lowers short-term rates and raises long-term rates. If short-term rates affected capital flows predominantly and had very little effect on economic activity, and if long-term rates affected economic activity predominantly (albeit with a long, uncertain lag that may, in the end, prove destabilizing) and capital flows only weakly, the intended effects could be realized. There is, however, no certainty that the two rates do have these significant differential impacts.36 For example, if the short-term rate has a significant effect on inventories of consumer durables and long-term capital movements are sensitive to changes in the long-term rate, this result need not hold. It is very difficult then to be confident that a successful change in the term structure will minimize the capital outflows and at the same time make a significant impact on inflation.

Some Conclusions

We discussed, briefly, three general types of policy that might be invoked to mitigate a conflict between internal and external objectives. The first policy (credit ceilings) proved difficult to evaluate; on the whole, however, we concluded that it might make some contribution to the resolution of the conflict.37 More importance should be attached to the second type of policy. These policies (forward market intervention, special reserve requirements, directives to the banking system, and prohibition of interest payments to foreigners) can be and have proved moderately successful as short-term expedients in coping with situations of conflict.38 Of course, since these measures are directed at the banking system, they do not touch private flows of funds, including those by nonbank intermediaries.39 However, in many countries these are, in varying degrees, constrained by exchange restrictions.

The discussion was limited to a conflict situation involving both a balance of payments surplus and inflation. Much of what was said, however, could be (and has been to some extent) applied in reverse to the other conflict situation involving both a balance of payments deficit and unemployment. The types of policy discussed, while particularly relevant to these situations, must not be thought to be inapplicable to other types of deviation from internal and external balance. Consider, for example, a balance of payments deficit and inflation. Traditional monetary policies will, except where there is perfect capital mobility, make some impact on both the deficit and inflation. When, however, capital markets are highly integrated, very strong monetary policies for domestic inflation would normally be required if the inflow of funds is to be counteracted. A small restrictive effect may, in other words, be accompanied by a very large inflow of funds. In these conditions, central banks may prefer to moderate the inflow of funds by the types of policy discussed and in this way ensure a stronger leverage for any given restrictive policy.40

APPENDIX

Suppose that Mo = money supply, H = base money, B = the external reserves of the central bank, D = domestic assets of the central bank, r is the money-base multiplier, then

(1)ΔMo=rΔH
(2)ΔB+ΔD=ΔH

and

(3)ΔMo=rΔB+rΔD,

where equation (2) identifies the sources of changes in base money.

Suppose that B = -αΔD, i.e., a proportion of an exogenous change in central bank credit is neutralized by an offsetting change in external reserves. If α = 1, the changes in central bank credit are completely neutralized by changes in the reserves of the central bank. This is the case of perfect capital mobility where discretionary monetary policy cannot affect the money supply. If α < 1, then

Δ Mo = -ΔD + rΔD

and

ΔMoΔD=r(1α).

The larger α (the more mobile capital is), the larger the change in domestic assets to realize a given change in the money supply. The types of policy discussed in the text tend to reduce α and hence to increase the leverage of monetary policy, as the concluding comments indicated.

BIBLIOGRAPHY

  • Bank for International Settlements, The Sources and Control of Bank Liquidity, C.B. 352 (mimeographed, Basle, February 1968).

  • Bank for International Settlements, Some Questions Relating to the Structure of Interest Rates, C.B. 354 (mimeographed, Basle, July 1968).

    • Search Google Scholar
    • Export Citation
  • Bank for International Settlements, Regulations and Policies Relating to Euro-Currency Markets, C.B. 358 (mimeographed, Basle, December 6, 1969).

    • Search Google Scholar
    • Export Citation
  • Brehmer, Ekhard,Bundesbank Measures to Ward Off the Speculative Inflow of Foreign Funds in August/September 1968” (unpublished, International Monetary Fund, January 13, 1969).

    • Search Google Scholar
    • Export Citation
  • Brehmer, Ekhard,Official Forward Exchange Operations: The German Experience,Staff Papers, Vol. XI (1964), pp. 389-413.

  • Dorrance, Graeme S., and E. Brehmer,Controls on Capital Inflow: Recent Experience of Germany and Switzerland,Staff Papers, Vol. VIII (1961), pp. 427-38.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Garvy, George, The Discount Mechanism in Leading Industrial Countries Since World War II (paper prepared for the Steering Committee for the Fundamental Reappraisal of the Discount Mechanism Appointed by the Board of Governors of the Federal Reserve System).

    • Search Google Scholar
    • Export Citation
  • Gilbert, Milton, and Warren McClam,Domestic Implications of the Evolving International Monetary Mechanism—Domestic and External Equilibrium: European Objectives and Policies,American Economic Association, Papers and Proceedings of the Seventy-seventh Annual Meeting (The American Economic Review, Vol. LV, May 1965), pp. 189-99.

    • Search Google Scholar
    • Export Citation
  • Katz, Samuel I., External Surpluses, Capital Flows, and Credit Policy in the European Economic Community, 1958 to 1967, Princeton Studies in International Finance, No. 22 (Princeton University Press, 1969).

    • Search Google Scholar
    • Export Citation
  • Logue, Ruth, Imported Inflation and the International Adjustment Process, Board of Governors of the Federal Reserve System, Staff Economic Studies (doctoral dissertation submitted to the George Washington University, February 1969).

    • Search Google Scholar
    • Export Citation
  • Malkiel, Burton G.,The Term Structure of Interest Rates,American Economic Association, Papers and Proceedings of the Seventy-sixth Annual Meeting (The American Economic Review, Vol. LIV, May 1964), pp. 532-43.

    • Search Google Scholar
    • Export Citation
  • Mills, Rodney H., Jr.,The Regulation of Short-Term Capital Movements: Western European Techniques in the 1960’s,Board of Governors of the Federal Reserve System, Staff Economic Studies (May 1968).

    • Search Google Scholar
    • Export Citation
  • Modigliani, Franco, and Richard Sutch,Innovations in Interest Rate Policy,American Economic Association, Papers and Proceedings of the Seventy-eighth Annual Meeting (The American Economic Review, Vol. LVI, May 1966), pp. 178-97.

    • Search Google Scholar
    • Export Citation
  • Modigliani, Franco, and Richard Sutch,, “Debt Management and the Term Structure of Interest Rates: An Empirical Analysis of Recent Experience,The Journal of Political Economy (Supplement, August 1967), Vol. 75, pp. 569-89.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Ross, Myron H.,‘Operation Twist’: A Mistaken Policy?The Journal of Political Economy, Vol. LXXIV (April 1966), pp. 195-99.

  • Scott, Ira O., Jr., and Wilson E. Schmidt,Imported Inflation and Monetary Policy,Banca Nazionale del Lavoro, Quarterly Review (December 1964), pp. 390-403.

    • Search Google Scholar
    • Export Citation
  • Yeager, Leland Bennett, International Monetary Relations: Theory, History, and Policy (New York, 1966).

Politique monétaire et équilibre intérieur et extérieur

Résumé

Une balance des paiements excédentaire en période de récession ou une balance des paiements déficitaire en période d’inflation ne manquent pas de poser un dilemme aux responsables de la politique monétaire d’un pays. Cette étude a pour objet d’examiner trois différentes politiques monétaires susceptibles d’atténuer ce dilemme. La première de ces politiques consiste à imposer des plafonds au crédit afin de minimiser l’incidence des taux d’intérêt et de décourager l’augmentation des avoirs intérieurs par le système bancaire. La deuxième vise à maintenir un écart entre les taux d’intérêt tout en mettant en œuvre à l’intention du système bancaire une politique monétaire destinée à décourager les entrées de capitaux. Diverses politiques sont examinées dans le cadre de cette étude. Les banques centrales peuvent fixer une prime spéciale à terme ou décote dans un marché « fermé » exclusivement réservé au système bancaire. Par exemple, si les taux d’intérêt pratiqués dans un pays sont supérieurs à ceux de l’étranger, la banque centrale peut offrir une décote à terme sur la monnaie nationale, dissuadant ainsi le rapatriement d’avoirs extérieurs et encourageant les investisseurs à placer leurs excédents de liquidités sur les places étrangères. Des coefficients de réserves obligatoires peuvent être fixés pour les engagements extérieurs bruts ou nets. Cette étude évalue l’incidence des deux types de réserves obligatoires (brutes et nettes) sur les entrées nettes probables de capitaux. Des directives peuvent être adressées aux banques afin qu’elles modifient la position nette de leurs avoirs extérieurs. Pour que des sorties de capitaux s’effectuent par le système bancaire, les banques peuvent par exemple recevoir l’ordre d’accroître leur position extérieure nette. Les autorités monétaires peuvent également interdire le paiement d’intérêt sur certains dépôts à terme et à vue de non-résidents. Cette étude examine l’application de ces mesures par différents pays. D’une manière générale, tout en contribuant à certains égards à résoudre ce dilemme, ces mesures, encadrement du crédit inclus, n’ont guère d’effet sur les mouvements privés de capitaux y compris sur les sorties de capitaux effectuées par les intermédiaires non bancaires. Les mouvements de capitaux qui sont découragés par le canal du système bancaire peuvent s’effectuer par le secteur non bancaire. Une troisième politique consiste à modifier la structure des taux d’intérêt. En période d’inflation et de balance des paiements excédentaire, par exemple, cette politique implique le maintien de faibles taux d’intérêt à court terme afin de dissuader les entrées nettes de capitaux et parallèlement le maintien de taux d’intérêt élevé à long terme. Cette étude montre que la mise en œuvre de cette politique est particulièrement difficile.

La política monetaria y el equilibrio interno y externo

Resumen

Las situaciones que comprendan, o bien un superávit de balanza de pagos junto con una recesión, o un déficit de balanza de pagos junto con inflación, crean dilemas para la política monetaria. En este trabajo se examinan las formas en que pueden aplicarse tres clases de política monetaria para atenuar dichos dilemas. Primero, la utilización de los topes al crédito para minimizar el efecto del tipo de interés y para desalentar la expansión de los activos internos por parte del sistema bancario. Segundo, el mantenimiento de diferencias de tipos de interés al mismo tiempo que se lleva a cabo una política monetaria dirigida al sistema bancario y encaminada a desalentar la entrada de capitales. Aquí se examinan distintas medidas. Los bancos centrales pueden implantar una prima especial a término o un descuento en un mercado “cerrado” ai que solamente tenga acceso el sistema bancario. Por ejemplo, si el tipo de interés interno es más alto que el del extranjero, el banco central puede ofrecer un descuento a término a la moneda nacional, lo cual desalentará la repatriación de activos sobre el exterior y fomentará la inversión del exceso de liquidez en los mercados del exterior. Puede imponerse una obligación especial de reserva sobre los pasivos brutos o netos frente al exterior. En este estudio se evalúan los efectos de las dos clases de obligación de reservas (bruta y neta) sobre la probable entrada neta de fondos. También pueden emitirse directrices al sistema bancario para que varíe su posición neta de activos frente al exterior. Por ejemplo, a fin de que se produzca una salida de capital a través del sistema bancario, pueden dárseles a los bancos instrucciones de que aumenten sus posiciones netas frente al exterior. También puede recurrirse a una prohibición del pago de intereses a los no residentes por ciertos depósitos a plazos y a la vista. Se examina la utilización de dichas medidas en distintos países. En general, aunque dichas medidas, incluidos los topes al crédito, contribuyen a resolver el dilema, no afectan a los flujos privados de fondos, incluidos los de los intermediarios no bancarios. Los flujos cuya circulación se ha desalentado a través de los intermediarios del sistema bancario pueden encontrar un nuevo cauce a través del sector no bancario. Una tercera clase de medidas es la que representa una manipulación de la estructura de plazos de los tipos de interés. Por ejemplo, en condiciones de inflación con un superávit de balanza de pagos habría que mantener bajos los tipos a corto plazo para desalentar las entradas netas, y mantener altos los tipos a largo plazo. Se indica que hay graves dificultades en el cumplimiento de una política de ese género.

*

Mr. Argy, Chief of the Financial Studies Division of the Research Department, is a graduate of the University of Sydney, Australia. He has been a lecturer at the University of Auckland, New Zealand, and a lecturer and senior lecturer at the University of Sydney. He has contributed several articles to economic journals.

1

Other policy mixes to achieve the same ends (e.g., combinations of monetary and fiscal policies and monetary policies combined with wider bands around par values) will not be discussed.

2

The choice between the monetary instruments would have to be determined on the basis of other relevant criteria, e.g., quickness of response, effect on interest rates of government securities, announcement effects, and discriminatory effects on different-sized banks or differently located banks.

3

This is not, of course, to say that these policies will exactly correct the internal/external imbalance. Indeed, in the absence of a second instrument this is most unlikely. The distinction is one between coincidence in internal/external policy in respect cf both direction and degree. Since we are not concerned with exact meeting of the targets, the paper is limited to conflicts, or otherwise, in respect of direction alone. See Yeager [17], p. 92. (Numbers in square brackets refer to items listed in the Bibliography, pp. 524-25.)

4

Situation (1) has been a serious problem for continental Europe since 1958 and has assumed greater importance recently. Situation (4), on the other hand, has been a serious problem at times in the two reserve-currency countries, the United Kingdom and the United States.

5

Put more correctly, we assume that the central bank wants to decelerate its accumulation of reserves. A situation of this kind is, of course, less pressing than one where the external reserves are falling.

6

See Scott and Schmidt [16] and Logue [10].

7

The major constraint being the absence of a developed money market outside the banking system. In some countries, too (e.g., the Netherlands), central banks did not have an adequate portfolio of government securities.

8

See the Appendix on this point.

9

A slightly different way of formulating the options is as follows. Suppose that a country is in a comfortable reserve position and is experiencing an excessive rate of inflation. Its interest rate is set at a level that is considered appropriate, given the internal situation. Now suppose that the Euro-dollar rate (in response to U. S. conditions) falls. The country may keep its interest rate high and allow an additional inflow of capital, which it would then need to neutralize. It may lower its rate in line with external conditions, but the interest rate might then be wrong for internal conditions. Finally, it may keep its interest rate high and find ways of minimizing the inflow (or find ways of fighting inflation without raising interest rates). These correspond to the three options discussed above. These were precisely the options facing a number of European countries during 1970 and early 1971 as Euro-dollar rates slid and domestic conditions in these countries called for the maintenance of relatively high interest rates. If fiscal policy were sufficiently flexible, one way to deal with this situation would be to go along with the lower rates in the Euro-dollar market by pursuing an expansionary monetary policy but to neutralize the effects on domestic activity through a contractionary fiscal policy.

10

The major exception is Germany. In Italy an analogous scheme is in operation, since bank loans above a certain size are subject to central bank approval.

11

This view is widely held, although it has never been carefully examined. See Logue [10], pp. 92-93; Katz [9], p. 6; Garvy [7], p. 17; Mills [12], pp. 4-5; and the Bank for International Settlements [3], p. 10.

12

Credit ceilings may of course be exceeded, e.g., in France in 1969 and in the Netherlands in 1970. In the Netherlands, credit ceilings may be exceeded but such excesses are subject to a penalty of requiring special deposits to be placed with the central bank.

13

Where nonresidents take the initiative and place funds in the domestic banking system (see later discussion).

14

Or buys dollars forward only to cover existing dollar assets.

15

See, however, the comments on Germany that follow.

16

For example, in 1960 in Germany a large forward discount was offered against the deutsche mark when, simultaneously, a premium was available in the free market.

17

See, on this, Brehmer [4 and 5].

18

Brehmer [5] argues that the major objective of these swap operations in this period was to reduce domestic liquidity. Other monetary instruments, in other words, were not adequate to offset excess domestic liquidity.

19

During the period 1964-65 the facility was confined to investments in U.S. Treasury bills.

20

There are limitations on the extent to which the Bundesbank may set terms for forward dollar sales that are more favorable than those in the free market. Consider an illustration where the premium on the deutsche mark is significantly lower in the closed market than in the open market. German banks may take (and indeed have taken) advantage of this situation in the following way: they borrowed dollars abroad and bought them forward on the free market; they then converted the spot dollars into deutsche mark, swapped the deutsche mark for dollars with the central bank, and sold these dollars at the forward rate in the closed market. In effect the banks would then have matching spot and forward positions, and they could profit from buying dollars forward on the free market and selling them forward in the closed market. The Bundesbank has attempted to discourage these types of operation, which came into evidence in 1968.

21

In Italy the motivation in offering preferential swap facilities was a mixed one. See Mills [12], pp. 23-25.

22

A considerable part of foreign currency loans in Italy, for example, are to domestic residents. A loan in foreign currency to a resident tends to have the same domestic effect as a loan in domestic currency.

23

Of course, Rd will not remain unchanged whatever the size of x. In a macro-model the domestic interest rate would be shown to respond to the size of domestic asset holdings of the banking system. Another qualification is that domestic investment need not be motivated solely by considerations of immediate profitability; good will to customers may also assume some importance.

24

A measure analogous to reserve requirements was also introduced in August 1964. Increases in gross foreign liabilities resulting from foreign credits beyond the average outstanding in the first half of 1964 were penalized by equivalent reductions in the banks’ rediscount quota with the Bundesbank.

25

This differential depended on the form of liability and the size of the bank. See Mills [12], pp. 9-10.

26

Increases in foreign assets taking the form of loans to foreign customers could not be used as an offset.

27

The offset facility was not reintroduced at the time. Prior to the requirement of 100 per cent, reserve requirement rates on resident and nonresident deposits were the same.

28

This applied to increases in the net liability position, not to a reduction in net asset positions.

29

In 1969 the device was widely used not to encourage outflows or to discourage inflows but for opposite reasons. For example, as interest rates in the Euro-dollar market were high, a number of countries (e.g., Belgium, Denmark, France, Italy, and the Netherlands) took measures to protect their reserves by directing banks to reduce their net asset positions.

30

Net foreign liabilities in lire were exempted.

31

This was of limited effectiveness, since the banks as a group had a large net foreign asset position.

32

The prohibition extended to foreign currency deposits as well.

33

This is so, even with a pure expectations theory, provided that expectations are divergent among operators. See Malkiel [11].

37

Of course, there may be reasons for preferring on balance not to use credit ceilings, e.g., the effects on bank competition and the administrative difficulties in implementation.

38

See Brehmer [4 and 5], Mills [12], Katz [9], Logue [10], and Garvy [7].

39

An outstanding recent illustration of this difficulty occurred in Germany during 1970 and early 1971. During this period Germany was pursuing a relatively tight monetary policy. Restrictions on the inflow of funds through the banking system had the effect of encouraging a large volume of borrowings by the nonbank sector from the Euro-dollar market. See, on this, Michael Porter, “Monetary Policy and Capital Flows: The Experience of the Federal Republic of Germany, 1963-70” (unpublished, International Monetary Fund, July 26, 1971).

40

See the Appendix.