Current Developments in Monetary and Financial Law, Vol. 2

Chapter 18 Central Bank Responsibility for Exchange Rate Policy and Implementation

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

There is no simple, unique answer to the question of what is/what should be the responsibility of the central bank for exchange rate policy and implementation. Rather, there is a range of practices, determined largely by the interplay of several factors in a given country: the legal provisions governing international economic relationships; the degree of central bank independence; and the choice of exchange rate regime. In many respects, these elements are the result of deliberate decisions on the part of governments. However, inherent in this issue are some fundamental economic linkages. The exchange rate is an asset price; it is the foreign currency price of a unit of the domestic currency. As such, it is linked to monetary policy in inescapable ways. Thus the implementation of monetary policy and exchange rate policy cannot be separated, and the central bank cannot be excluded from exchange rate policy.

Legal Provisions

One straightforward place to start thinking about this topic is which governmental entity has legal title to the country’s foreign exchange reserves. Those reserves are whatever foreign-currency-denominated assets are owned by the public sector. Various alternative arrangements are possible and are in use. It may be that ownership is concentrated in the hands of one public sector entity. That may be the central bank. It may be the executive branch, typically controlled by the Finance Ministry or Treasury but basically on the balance sheet of the executive branch. Or, ownership may be split between two or more public sector entities.

Table 1 shows the current situation in the United States. Ownership of foreign exchange reserves is split between the Federal Reserve and the Treasury, about evenly. Holdings of the Federal Reserve are in the System Open Market Account (SOMA). Those of the Treasury are in the Exchange Stabilization Fund (ESF). The latter is an account specifically created to hold such reserves. The creation of such an account is a common way to structure foreign exchange holdings. As you can see, the United States now limits its holdings to euro and yen; but many countries hold a greater number of foreign currencies in their reserves, and in the past the United States has held others. The fact that the holdings are equal in the two accounts, for each of the currencies, is a relatively new development that was deliberately brought about by transactions between the two.

Table 1.Foreign Currency Holdings of U.S. Monetary Authorities Based on Currency Exchange Rates(Changes in balance by source)

as of






as of

March 31,

System Open Market Account (SOMA)
Japanese yen9,221.51.0(51.1)9,171.4
U.S. Treasury Exchange Stabilization Fund (ESF)
Japanese yen9,221.51.0(51.1)9,171.4

Ownership of reserves creates a presumption that some actions related to foreign exchange rate policy could be undertaken. Ownership also likely reflects how decision-making authority has been allocated and how it has been executed in the past. In the case of the United States, ownership accurately reflects some degree of shared responsibility for exchange rate policy between the Fed and the Treasury.

Of primary importance, however, is the allocation of authority to decide upon the exchange rate regime that will be put in place. Before specific policy steps and implementation can come into play, there has to be a decision of where in the spectrum of possible regimes the country will choose to operate. At one extreme are fixed-rate regimes, of which there are more than one kind; at the other extreme, a freely floating exchange rate regime. And there are many possibilities in between. I will discuss below how this decision bears on the conduct of monetary policy and thus on the interaction of monetary policy and exchange rate policy. My point here is that this decision of regime is important in determining the role of the central bank in exchange rate policy and that this decision is inherently a political one. Thus, it is lodged with governments.

Central bank officials are generally consulted, particularly at times when a decision to make a change is under consideration in the context of some kind of crisis. Indeed, in some cases central banks might have the authority on their own to modify the regime from one kind of mixed regime (somewhere between fixed and floating) to another. But the choice of regime requires the kind of political authority that resides within governments and therefore not solely the central bank. Some regimes require active cooperation between two or more countries. That kind of international commitment is really only possible at the level of government. The choice of exchange rate regime can be imbedded in legislation. Alternatively, it can be an executive branch prerogative, delegated to a particular official in the executive branch or the head of government.

Once the exchange rate regime has been determined, there follows a need to formulate and implement policy on a day-to-day basis. For some regime choices, as will be discussed below, that policy is essentially fully implied by the regime choice. For others, there are significant decisions to be made over time. Responsibility for those decisions is a complex matter. The political authorities may wish to reserve that responsibility for themselves and may do so in some legal sense. However, the economic links between monetary policy and exchange rate policy imply that if the central bank has any control over and responsibility for monetary policy, it will inevitably have an impact on the exchange rate and therefore some role in exchange rate policy.

Central Bank Independence

The issue of central bank independence is framed in terms of monetary policy. If a central bank is empowered to make the decisions that govern its instruments of monetary policy without interference from other branches of government, then it is independent.

Different political systems imply some variation in what it means to be independent of other government authorities. In the United States, with the executive and legislative branches separate, independence is most importantly independence from the executive branch. The Federal Reserve System was created by Congress and could be changed by Congressional legislation at any time. However, both legally and in practice, the decisions of Federal Reserve officials are not subject to review or overturn by officials of the executive branch. In a parliamentary system, the line between executive and legislative is less sharp. However, independence implies that the cabinet in the normal course of governing does not have authority over monetary policy. Clearly, the parliament could legislate changes in the central bank structure and function if it chose.

In the past, we have observed a range of experience across countries. Prior to 1997, the Bank of England was an example of a central bank that did not have the final authority for monetary policy. The Chancellor of the Exchequer, i.e., the finance minister, made the decisions and was the official responsible to parliament for them. Over time, various arrangements were used for securing input from the central bank for these decisions, and the central bank clearly had an advisory role. It also had a major role in implementation of monetary policy and exchange rate policy. But if the central bank lacks final authority with respect to monetary policy, then its role with respect to exchange rate policy will be limited as well.

Examples of countries that have functioned with substantial independence throughout most of the post-World War II period include the Bundesbank, the Swiss National Bank, and the Federal Reserve. Independence is more importantly a question of practice rather than of legal provision. As a consequence, it is to some extent a question of degree. Many central banks have operated in a middle range where they exercised some independence but were at times subject to the influence of government officials.

Over the past decade or so, there has been a significant shift toward empowering central banks with greater independence. The creation of the European Central Bank (ECB) is perhaps the most extreme example. That institution was created by treaty among the countries participating in the European Union and was given by that treaty total independence from the governments of those countries with respect to monetary policy. Because of evidence that the inflation outcome in countries tends to be improved when the central bank is independent, moves to grant such independence have been common in recent years. Examples of countries that have changed central bank legislation to do this include (in addition to those in Europe) Japan and Mexico.

Because central bank independence is a question of practice and of degree, it can be divided into elements. A common distinction is between independence with respect to the setting of the ultimate goal of monetary policy and independence with respect to decisions about particular policy actions. The legislation establishing the central bank, perhaps subsequently modified, typically states or implies the objectives for the central bank in order to define its responsibilities. But that provision can be quite general or can be multifaceted. In that case, considerable discretion remains within the central bank to define for itself the specific objectives of monetary policy. The more narrowly stated the goal or mandate of the central bank by legislation, the less independence in this element of monetary policy it has. In the extreme, the goal can be stated so narrowly that the central bank effectively loses all independence with respect to monetary policy and exchange rate policy.

In those cases where the goal is stated more generally, even including a specific inflation target over some horizon, central bank authorities can retain independence with respect to settings of policy instruments. Implementation of policy measures generally results in changes to short-term interest rates (either directly or indirectly). The change in these interest rates in turn have an impact on the exchange rate (discussed further below). It is this linkage that unavoidably involves the central bank in exchange rate policy.

Exchange Rate Regime

As noted above, the choice of exchange rate regime is a political step, the ultimate responsibility for which rests with governments. The nature of the link between the central bank and exchange rate policy depends upon the exchange rate regime. Regimes range from fixed, of which there are several variations, to freely floating.

Of the fixed regimes, a multilateral currency union is among the most binding and absolute. It involves a decision by two or more countries to share the same currency. In 1999, we saw the initiative of the European Economic and Monetary Union (EMU) move to the third stage. Eleven members of the European Union now share one currency, the euro. One can draw an analogy with the 50 U.S. states. They form a currency union and share the U.S. dollar. However, EMU is unique in that the 11 remain independent sovereign states and previously had distinct currencies. Within the currency union, the commitment is so tight that in fact there are no exchange rates and no exchange rate policy. With respect to other currencies, together the 11 face the range of choices from fixed to floating. They have only one central bank, the European System of Central Banks, but many governments.

Currency union can also be unilateral. This results when a given country decides for itself to use the currency issued by another country and its central bank as the domestic currency in circulation. Today we frequently hear reference to dollarization, but in principle other currencies could be chosen for such use. Dollarized countries include Panama and, in a recent move, Ecuador. Liechtenstein uses the Swiss franc. For the country choosing to dollarize, domestic currency ceases to exist (although some retain domestic coinage). As a consequence, the country has in essence no exchange rate policy once it has chosen this regime. If a central bank continues to function, it has no monetary policy responsibilities. Both versions of currency union are extreme in that within the currency union, exchange rates and central bank functions have been eliminated.

The regime of a currency board implies a slightly less rigid commitment in that domestic currency continues to exist. Its exchange value, however, is fixed exactly to a particular foreign currency. The country retains a central bank, but its behavior is determined by the “rules” of being a currency board. Those rules state that the currency board, i.e., central bank, stands ready to swap domestic currency for the named foreign currency at the fixed rate. Argentina is an example of a country operating under a currency board regime.

The gold standard is a variation of a fixed exchange rate system in which the domestic currency is pegged not to another currency but to a fixed amount of gold. If other countries also peg their currencies to gold, then the various currencies are in effect fixed with respect to one another. The central bank, and/or the government, is pledged to maintain the value of the currency at the stated gold price. But of course it is possible to announce a change in the peg to gold, and in the history of the gold standard that happened fairly frequently. It is also possible just to announce a suspension of the system and to no longer promise that the currency is worth any stated amount of gold. That happened from time to time as well in history. When the gold standard was being adhered to, it offered some degree of flexibility as it is expensive to present currency to a central bank or government, demand the promised amount of gold, and ship it back to your home country. As a consequence, some flexibility in the implied cross exchange rates was possible under the gold standard.

The term “pegged exchange rate” implies an exchange rate regime in which the monetary authorities announce a promise to maintain the exchange value of the country’s currency with respect to some other country’s currency within a narrow band. They may “peg” their currency to the dollar, as was the case with rates generally under the Bretton Woods System before its breakdown and as some countries choose to do now. During the 1980s and 1990s the Exchange Rate Mechanism of the European Monetary System was a system of simultaneous bilateral commitments to maintain the exchange rates of the participating countries in terms of each other. These commitments defined a matrix of cross rates that was the mutual responsibility of the participating countries. Alternatively, the authorities may peg to a “basket” of currencies, i.e., to maintain the average value of the domestic currency in terms of specified amounts of several other currencies, usually those of important trading partners. One feature of any pegged system is that the pegs, having been set by governments, can be changed by them. Generally the system defines a narrow band of permitted variation that is viewed as consistent with the peg.

Managed floats are exchange rate regimes where some explicit statement has been made about the exchange value of the currency, but it is not fixed. This alternative is intended not only to permit more day-to-day flexibility in the rate, but also to provide some guarantee about its value. A crawling peg system is one in which a path of rates is announced ex ante, rather than a single pegged value, in terms of another currency or a basket. Thus some continuous change is built into the process. Typically, there is a permitted band around the pre-announced path as well. Variations of crawling peg systems have been used for intervals by several emerging market countries.

Target zones specify a range around a stated rate, but with bands sufficiently wide that some significant fluctuation is expected and allowed. However, the authorities are committed to keeping the exchange rate from moving beyond the range. Target zones have been suggested as an alternative for limiting the fluctuation of the major currencies in terms of each other, but have not yet been implemented for that purpose.

“Dirty float” is the term applied when there is no single peg, no band, no range, no commitment—the exchange rate is determined in the market each day—yet the authorities are not indifferent to the exchange value of the currency and may at a time of their choosing try to influence the market rate. One form of influence is to make public statements and hope that market participants react to those statements (verbal intervention). Another is for the finance ministry and/or the central bank to sell domestic currency for foreign currency in the market or vice versa in an attempt to alter the market price (market intervention). This regime is the one now used for the dollar, the euro, the yen, and many other currencies.

Logically, there is also the extreme regime of a freely floating exchange rate. Such a rate would be determined in the market at each moment, with the authorities committed never to intervene. But how could a government today make a promise that would bind not only itself but all future governments?

Policy Implementation

A tightly pegged exchange rate regime defines monetary policy. It completely determines what the central bank is to do in response to alternative market developments. The central bank is pledged to maintain convertibility to gold at a stated amount or to keep the market exchange rate equal to the announced pegged rate. Such a system precludes the central bank from using its policy tools to achieve some other goal at the same time. How then do we think about the central bank’s role in this case?

The central bank typically functions as the fiscal agent for the government, but this relationship does not by itself determine the role of the central bank in formulating or implementing exchange rate policy. As fiscal agent, the central bank transacts on behalf of the government. But it can also do so for its own account. The government typically has a domestic-currency-denominated account at the central bank. When the government transacts, this account is debited or credited. But it can also have accounts at commercial banks or other financial institutions. The fiscal agent relationship does not define how decisions are made within the central bank or for exchange rate policy.

In a tightly pegged system, the central bank must use its tools to fulfill the commitments made to those in the market about the exchange rate. The central bank can do this directly by maintaining a pool of foreign exchange reserves and standing ready to transact with market participants on the promised terms. It can do so indirectly by instead exercising control over a very short-term interest rate and changing that rate so as to bring about a market exchange rate consistent with the peg. By bringing about changes in the interest rate, the central bank makes the currency more or less attractive to investors. Market participants trade with each other, not the central bank. The central bank watches the exchange rate and moves interest rates to achieve the exchange rate peg.

The central bank is in an entirely reactive role. It reacts to buying or selling of the currency from whatever source. Usually, it is the institution that has the tools needed to implement the exchange rate policy. But as a consequence of the exchange rate regime, the central bank has no scope for monetary policy directed at domestic objectives and no discretion. It may be at the center of the process, but it has little or no real power.

Capital controls are the one instance when it may be possible for a time to peg or manage an exchange rate and practice a separate monetary policy. This is the extreme exception to the basic framework presented so far. The imposition of capital controls makes certain transactions or investor behavior illegal. As long as they can be enforced, it would be possible to have an exchange rate objective and to implement that with external transactions while at the same time having a domestic objective that is not consistent with that exchange rate. The domestic objective can be pursued using domestic transactions. The capital controls attempt to prevent someone from doing essentially simultaneous transactions to take advantage of a price difference for a given asset, say the domestic currency or a domestic-currency denominated security, in a market within the country and one outside of it. The capital controls prevent price arbitrage. As a consequence, evading them is profitable. Market participants will expend resources to find a way to make those profits. After a time, it is generally the case that evasion begins to undermine the controls, and they become ineffective. Moreover, while they are in place, the capital controls distort prices and contribute to inefficient use of resources.

Eventually the various forms of fixed rates, except for those in which the domestic currency ceases to exist, encounter conditions that create a conflict between the exchange rate objective and domestic economic conditions. In some cases, there may be upward pressure on the exchange rate. In order to counter that pressure, the central bank has to expand the domestic money supply and in so doing put in place a monetary policy that is too expansionary for current economic conditions. Alternatively, there may be downward pressure on the exchange rate. In that case, the central bank would need to raise domestic interest rates and to sell foreign currency reserves in exchange for the domestic currency. The tight monetary conditions could damage economic activity. In any event, the available supply of foreign exchange reserves could be exhausted. Because the central bank is the one implementing the exchange rate policy, it is often the central bank that sounds the alarm that reserves are being depleted or that interest rates cannot stay high enough indefinitely. Such a development may require that the country change the peg, i.e., depreciate the currency, modify the system, i.e., widen the bands, or perhaps even abandon a fixed regime and move to a floating regime. The central bank may participate in such decisions, but once again we are essentially talking about a decision with respect to exchange rate regime, a decision in the end made by the government.

Implementation of Floating Rate/Monetary Policy

The role of the central bank is significantly different when the exchange rate regime is that of a loosely managed or floating rate. In this case, the central bank’s responsibilities and daily decisions are directed toward one or more domestic objectives. The exchange rate in this case is one of many financial variables that play a role in the monetary policy process, but it is no longer the central variable.

First, the exchange rate is a channel of monetary policy. It reacts to policy actions taken by the central bank. Because it reacts, and because in turn participants in the economy react to the exchange rate, some of the effects of the monetary policy measure are transmitted through the exchange rate.

Second, the exchange rate is information. By monitoring changes in the exchange rate, central bank officials can anticipate some developments within the economy. In addition, they can infer to some extent market participants’ views on the domestic economy and prospective monetary policy.

Third, the exchange rate is a source of shocks. Because every exchange rate by its nature relates to two economies, not one, the exchange rate is influenced by developments outside the domestic economy and can transmit the effects of those developments into that economy.

In a regime of a floating exchange rate, the central bank can process the information in exchange rate movements and assess the shocks hitting the economy through the exchange rate. It can decide to alter monetary policy in response—not to achieve a particular exchange rate objective, but rather to offset or to reinforce the impact of the exchange rate on domestic variables. The central bank can decide how it perceives the exchange rate having an impact on its final objectives. It still cannot independently choose a domestic objective and a goal for the exchange rate, but it can determine the priorities of reacting to the exchange rate or trying to influence the exchange rate or not as it seeks to achieve its domestic objectives.

What determines the exchange rate in the market? How is it linked to monetary policy? A floating exchange rate responds at each moment to offers to buy or sell domestic currency for foreign currency in the market. What are the incentives behind such offers? An investor is weighing the alternatives open to him to earn the highest possible return on his investment. Everything else being equal, the higher the domestic interest rate is, the more attractive domestic assets in the domestic currency will be to that investor. However, for a given domestic interest rate, the higher the foreign interest rate is, the more attractive a foreign asset denominated in some other currency would be. But the foreign asset, when it matures, will repay the investor in foreign currency. Each investor must choose a single currency in which to make the comparison of which assets yield more return. In order to compare alternatives in more than one currency, the investor has to form a view as to future exchange rates, so that he can determine total returns in one currency. But, of course, future exchange rates cannot be known now with certainty. So this calculation inevitably entails risk. The investor has to make an assessment of how much risk he believes is embedded in each investment strategy and how much risk he is willing to bear. Whatever the outcome of that evaluation, with everything else held constant, higher domestic interest rates tend to attract investors to the domestic currency and cause it to appreciate; higher foreign interest rates tend to do the reverse.

We might ask whether, under a floating rate regime, it is possible to separate the exchange rate from domestic monetary policy. That would permit some other part of the government to retain authority for exchange rate policy. The exchange rate reacts to monetary policy steps, but it also reacts to a variety of other economic events, including the other country’s monetary policy. The authorities may not be indifferent to the exchange rate value that results from these forces. Such was the case for the dollar in 1979 and 1985, for example.

One tool that is thought by some to offer a way to influence the exchange rate other than through monetary policy is exchange market intervention. Intervention occurs when government authorities enter the market and buy one currency for another; they do this in order to have an impact on the market price. So in a floating rate regime, responsibility for exchange rate policy would seem to reside with that part of the government that has authority over foreign exchange intervention. However, foreign exchange intervention can be done in two ways: sterilized and unsterilized. In the case of unsterilized intervention, interest rates as well as the exchange rate are permitted to move in response to the transaction. In the case of sterilized intervention, a secondary transaction is done in domestic markets to keep interest rates unchanged.

If the intervention is unsterilized, it is in effect monetary policy. As noted above, we would expect monetary policy to be able to influence the exchange rate. If the intervention is sterilized, there is considerable debate as to whether it will have any effect on the exchange rate, or at least any lasting effect on the exchange rate. Much evidence would suggest not.

So the question of exchange rate policy can swing on who has the authority to make decisions about intervention: the central bank or some other part of government, usually the finance ministry. Often, in the case of a floating rate regime, the government will attempt to retain a role for itself by controlling intervention decisions. But if the central bank is independent, it can by itself always undertake the transactions that would transform the intervention from unsterilized to sterilized. So the government may have authority over intervention and may exclude the central bank from participating in setting that policy, but it may end up being sterilized intervention. And sterilized intervention may have little effect.

There have been extended times when the yen/dollar rate has continued to move down despite dollar purchases, which would be thought to strengthen the dollar and raise its value in terms of yen. This happened during much of 1994 and early in 1995. It happened again in 1999. Conversely, the value of the dollar in terms of the yen continued to rise somewhat further in 1998 despite dollar sales. There are episodes that would suggest some effectiveness of sterilized intervention. During mid-1995, dollar purchases occurred at the time of a rising dollar. In early 1999, dollar purchases marked the end of a period of dollar depreciation and the start of a short episode of dollar appreciation. This evidence is very partial and does not attempt to control for other effects on the exchange rate at the time of intervention. We can infer, however, that sterilized intervention is clearly not a dominant, alternative tool that gives officials the ability to control the exchange rate regardless of monetary policy stance or measures in the two countries.

For some intervals, the movement in three-month market interest rates for the United States and for Japan is consistent with the expectation that a widening interest rate differential in favor of dollar rates should contribute to a strengthening of the dollar. Such is the case during the final three quarters of 1995 and on balance during 2000. However, the dollar fell in 1994 while U.S. rates rose relative to Japanese rates. And over much of 1996 through mid-1998, the dollar rose substantially while the interest differential changed little. The exchange rate fluctuations over this period clearly were influenced by other factors.


Central bank responsibility for exchange rate policy and implementation depends crucially on the choice of exchange rate regime. That choice is generally made at a political level, although the central bank may be consulted.

The closer the exchange rate regime is to a tightly pegged alternative, the less scope for discretionary monetary policy there is for the central bank. In such a regime, the exchange rate is at the center of central bank responsibilities, which focus on implementing that exchange rate decision. However, the central bank’s role is reactive. It has little real authority in the determination of either monetary or exchange rate policy.

The closer the exchange rate regime is to a freely floating alternative, the more monetary policy is framed around domestic objectives. In this case, the exchange rate reacts to monetary policy as well as to other economic developments. The central bank can decide in what ways it chooses to incorporate exchange rate movements into its decision making with respect to monetary policy. There may still be a role for other government officials to attempt to set exchange rate policy independently of the central bank through sterilized intervention or by putting in place capital controls. Empirical evidence casts doubt on whether sterilized intervention can have a lasting impact on the exchange rate. Capital controls may work for a time but in most cases introduce distortions and eventually succumb to substantial evasion.

Note: The views in this paper are solely the responsibility of the author and should not be interpreted as reflecting the views of the Board of Governors of the Federal Reserve System or any other member(s) of its staff.

    Other Resources Citing This Publication