Currency Boards for Developing Countries: A Handbook
by Steve H.
Hanke and Kurt Schuler
[The original version of this study was published by ICS
Press,
In the beginning God created sterling and the franc.
On the second day He created the currency board and, Lo,
money was well managed.
On the third day God decided that man should have free will
and so He created the budget deficit.
On the fourth day, however, God looked upon His work and
was dissatisfied. It was not enough.
So, on the fifth day God created the central bank to
validate the sins of man.
On the sixth day God completed His work by creating man and
giving him dominion over all God's creatures.
Then, while God rested on the seventh day, man created
inflation and the balance-of-payments problem.-Peter B. Kenen (1978:13)
Contents
About the authors
1. The case for currency boards
Currency board versus central bank
Currency boards versus dollarization, multinational central
banks, and free banking
Currency boards and wider economic reforms
Outline of this study
2. The case against typical central banks
The functions of money
Stability and credibility
Credibility and exchange rates
Convertibility and foreign-exchange controls
Central banking and deficit finance
Political independence: an unattainable goal
Inadequate staff
Flexibility: a problem even in theory
3. Currency boards, central banks, and the money supply
Money supply in a currency board system
Money supply in a central banking system
Central banks sometimes mistaken for currency boards
A brief history and assessment of currency boards
Currency boards and currency board-like systems today
4. How to establish a currency board
How to convert a central bank into a currency board
The alternative: a parallel currency approach
How to establish the currency board as the issuer of a
parallel currency
How to choose a reserve currency
How to calculate the initial foreign reserves
How to obtain the initial foreign reserves
5. How to operate and protect a currency board
How to operate a currency board
How to protect a currency board
How to change the reserve currency, if necessary
6. Objections to currency boards
No lender of last resort
Does size matter?
Fixed versus floating exchange rates
Deflation
The inflation tax
The cost of reserves
Colonialism
The worst case
7. Conclusion
Appendix: A model currency board constitution
Bibliography
Notes
Figures
Figure 1.1. A typical currency board versus a typical
central bank
Figure 3.1. Balance sheets
Figure 3.2. Money supply increase in a currency board
system
Figure 3.3. Money supply increase in a currency board
system, initial stage
Figure 3.4. Money supply increase in a currency board
system, intermediate stage
Figure 3.5. Money supply increase in a currency board
system, final stage
Figure 3.6. Money supply decrease in a currency board
system
Figure 3.7. Money supply decrease in a currency board
system, initial stage
Figure 3.8. Money supply decrease in a currency board
system, intermediate stage
Figure 3.9. Money supply decrease in a currency board
system, final stage
Figure 3.10. Money supply increase in a central banking
system with a floating exchange rate
Figure 3.11. Money supply decrease in a central banking
system with a floating exchange rate
Figure 3.12. Central bank and currency board balance sheets
Figure 4.1. Currency boards and currency board-like systems
today
Figure 7.1. Summary of proposals
ABOUT THE
AUTHORS
Steve H. Hanke is Professor of Applied Economics at The
Johns Hopkins University in
Kurt Schuler is a senior economist at the Joint Economic
Committee of the U.S. Congress. (This study, written long before he began
working at the Joint Economic Committee, does not necessarily reflect the views
of the Committee.) He has worked as a consultant on monetary matters in Africa,
Asia, Eastern Europe, and
1. THE CASE
FOR CURRENCY BOARDS
Since the final breakup of the Bretton Woods monetary
system in 1973, the gap that once existed between the currencies of developed
and developing countries has widened into a gulf. During the Bretton Woods
system and the "classical," pre-1914 gold standard, and to a lesser
extent also between the world wars, developing countries and developing
countries alike generally had sound currencies. A sound currency is one that is
stable, credible, and fully convertible. Stability means that current annual
inflation is relatively low, usually in single digits. Credibility means that
the issuer creates confidence that it will keep future inflation low. Full
convertibility means that the currency can buy domestic and foreign goods and
services, including buying foreign currencies at market rates without
restriction.
Today, most developed countries still have currencies that
are sound, if not as stable as under the gold standard. Most developing
countries, in contrast, have unsound currencies (Schuler 1996). The loss of
sound currencies is connected to the rise of central banking in developing
countries. During most of period before the final breakup of the Bretton Woods
system, most developing countries did not have modern-style central banks.
Instead, they had competitive issue of notes by privately owned commercial
banks (in Latin America, East Asia, and self-governing British colonies),
monopoly issue of notes by a privately owned commercial bank (in the colonies
of European powers other than Britain) or currency boards (in most
non-self-governing British colonies and some independent nations) (Conant 1969
[1927], Schuler 1992a, b). Central banking is a recent arrival in most
developing countries. It did not become widespread in Latin America until the 1920s,
and did not become widespread in Africa, the Middle East, and
The high inflation and limited convertibility
characteristic of unsound currencies have hindered the economic development of
developing countries. Unsound currencies discourage domestic and foreign
investment alike. A number of developing countries that had net inflows of
capital under colonial rule and during the first years of independence, when
they had sound currencies, became net exporters of capital after their
currencies became unsound, as foreign capital dried up and domestic capital
illegally fled to avoid confiscation by means of inflation. Consequently,
economic development slowed, stagnated, and in some cases even reversed.
Historical experience suggests that to attract renewed capital investment and
encourage renewed economic development, developing countries need sound
currencies. As we will discuss in more detail later, historical experience also
suggests that central banking has little likelihood of providing sound
currencies soon in most developing countries.
Of rival systems to central banking, the currency board
system was especially widespread: it has existed in more than 70 countries.
Currency board systems continue to exist in some countries, most notably
Currency board versus central bank
Central banking is familiar to most people, at least on a
practical level, as the monetary system of their country. A central bank is a
monetary authority that has discretionary monopoly control of the supply of the
reserves of commercial banks. Usually this implies a monopoly of the supply of
notes (paper currency) and coins. (1)
Discretionary control means the ability to choose a monetary policy at will, at
least partly unconstrained by rules. Reserves mean the medium used to settle
payments. Often, a monetary system uses two types of media to settle payments:
one type that is used primarily domestically, such as notes issued by the
monetary authority, and foreign reserves (ultimate reserves) used in
international trade, such as gold, foreign bonds, or notes issued by a foreign
central bank. Commercial banks, in the broad sense that we will use the term,
are all banks other than the central bank, including cooperative banks,
investment banks, and savings banks.
Though the currency board system has enjoyed a revival of
interest recently, it remains unfamiliar to many people. A currency board is a
monetary institution that issues notes and coins (and, in some cases, deposits)
fully backed by a foreign "reserve" currency and fully convertible
into the reserve currency at a fixed rate and on demand. The reserve currency
is a convertible foreign currency or a commodity chosen for its expected
stability. The country that issues the reserve currency is called the reserve
country. (If the reserve currency is a commodity, the country that has the
currency board is itself considered the reserve country.)
As reserves, a currency board holds low-risk,
interest-earning securities and other assets payable in the reserve currency. A
currency board holds reserves equal to 100 per cent or slightly more of its
notes and coins in circulation, as set by law. The simplest type of currency board
accepts no deposits and issues no securities; if a currency board does accept
deposits or issue securities, they too must be backed 100 per cent or slightly
more by assets payable in the reserve currency. A currency board earns profits
from the difference between the return on the reserve-currency securities it
holds and the expense of maintaining its notes and coins in circulation. It
remits to the government (or to its owner, if not the government) profits
beyond what it needs to pay its expenses and to maintain its reserves at the
level set by law. A currency board does not have discretionary control of the
quantity of notes, coins, and deposits it supplies. Market forces determine the
quantity of notes, coins, and deposits it supplies, and hence the overall money
supply in a currency board system.
A currency board is only a part of the monetary system in
any country that has commercial banks and other financial institutions. The
currency board system comprises the currency board, commercial banks, and other
financial institutions. It also comprises certain rules of behavior by them and
the government concerning exchange rates, convertibility, government finance,
and so on.
Table 1.1 lists differences between a typical currency
board and a typical central bank. This section briefly explains each
difference. Later chapters discuss the differences in detail. We emphasize that
the descriptions are accurate for typical actual currency boards and central
banks, past and present. The description of a currency board does not describe
a theoretically ideal currency board, nor does it describe an exceptionally
good actual currency board. It describes atypical actual currency board,
although the actual performance of currency boards has been close to the ideal
they have been established to strive for (see chapter 3). Similarly, the
description of a central bank does not describe a theoretically ideal central
bank. Nor does it describe an exceptionally good actual central bank such as
the U.S. Federal Reserve System or the Monetary Authority of Singapore. It
describes a typical actual central bank. The description fits most central
banks, especially those in developing countries, which are a substantial
majority of central banks in existence today.
|
Figure 1.1. A typical
currency board versus a typical central bank |
|
|
Typical currency board |
Typical central bank |
|
Usually supplies notes and coins
only |
Supplies notes, coins, and
deposits |
|
Fixed exchange rate with reserve
currency |
Pegged or floating exchange rate
|
|
Foreign reserves of 100 per cent
|
Variable foreign reserves |
|
Full convertibility |
Limited convertibility |
|
Rule-bound monetary policy |
Discretionary monetary policy |
|
Not a lender of last resort |
Lender of last resort |
|
Does not regulate commercial
banks |
Often regulates commercial banks
|
|
Transparent |
Opaque |
|
Protected from political
pressure |
Politicized |
|
High credibility |
Low credibility |
|
Earns seigniorage only from
interest |
Earns seigniorage from interest
and inflation |
|
Cannot create inflation |
Can create inflation |
|
Cannot finance spending by
domestic government |
Can finance spending by domestic
government |
|
Requires no
"preconditions" for monetary reform |
Requires
"preconditions" for monetary reform |
|
Rapid monetary reform |
Slow monetary reform |
|
Small staff |
Large staff |
|
Note: The characteristics listed
are those of a typical actual currency board or central bank, especially one
in a developing country, not those of a theoretically ideal or exceptionally
good currency board or central bank. |
|
To begin at the top of the list in Table 1.1, a typical
currency board usually supplies notes and coins only, whereas a typical central
bank also supplies deposits. Some past currency boards have accepted deposits;
today, the Hong Kong Monetary Authority and some currency board-like monetary
authorities issue securities. The deposits of a typical currency board are
subject to the same foreign reserve requirement as its notes and coins. To
simplify exposition of the currency board system, this study usually discusses
currency boards as if they issue notes and coins only. The additional
complications that result from deposits and securities are minor and do not
significantly change the analysis if they are interchangeable for their holders
and subject to a uniform foreign reserve requirement for the currency board.
Notes and coins issued by the currency board or central
bank and deposits of commercial banks at the currency board or central bank
constitute the monetary base. The monetary base counts as reserves when held by
commercial banks, but not when held by the public. Deposits of the public at
commercial banks and notes and coins held by the public constitute the money
supply. (2)
Notes and coins in circulation, whether held by the public or by commercial banks,
constitute cash.
Deposits at the central bank are the main form of reserves
for commercial banks in a typical central banking system. In a typical currency
board system, commercial banks hold no deposits at the currency board; instead,
reserve-currency assets are their main form of reserves. In a currency board
system and a central banking system alike, commercial banks hold "vault
cash"--notes and coins of the currency board or central bank--to satisfy
their depositors' requests to convert deposits into notes and coins.
A typical currency board maintains a truly fixed exchange
rate with the reserve currency.The exchange rate is permanent, or at most can
be altered only in emergencies. The exchange rate may be written into the
constitution that describes the legal obligations of the currency board. The
record of currency boards in maintaining fixed exchange rates has been
excellent (see chapter 3). A typical central bank, in contrast, maintains a
pegged or floating exchange rate rather than a truly fixed rate. A pegged
exchange rate is constant for the time being in terms of a reserve currency,
but carries no credible long-term guarantee of remaining at its current rate. A
floating exchange rate is not maintained constant in terms of any reserve
currency. The exchange rate maintained by a central bank is typically not
written into law, and can be altered at the will of the central bank or the
government. When a typical central bank suffers heavy political or speculative
pressure to devalue the currency, it devalues. As we shall see later, allegedly
fixed exchange rates maintained by central banks have in reality typically been
pegged exchange rates.
As reserve assets against its liabilities (its notes and
coins in circulation), a typical currency board holds securities in the reserve
currency; it may also hold bank deposits and a small amount of notes in the
reserve currency. It holds foreign reserves of 100 per cent or slightly more of
its note, coin, and deposit liabilities, as set by law. Many currency boards
have held a maximum of 105 or 110 per cent foreign reserves to have a margin of
protection in case the reserve-currency securities they held lost value. A
typical central bank, in contrast, has variable foreign reserves: it is not
required to maintain any fixed, binding ratio of foreign reserves to
liabilities. Even where a minimum ratio exists, a typical central bank can hold
any ratio in excess of that. For example, a central bank required to hold at
least 20 per cent foreign reserves may hold 30, 130, or even 330 per cent
foreign reserves. A typical central bank also holds domestic-currency assets,
which a typical currency board does not.
A typical currency board has full convertibility of its
currency: it exchanges its notes and coins for reserve currency at its stated
fixed exchange rate without limit. Anybody who has reserve currency can
exchange it for currency board notes and coins at the fixed rate, and anybody
who has currency board notes and coins can exchange them for reserve currency
at the fixed rate. However, a currency board does not guarantee that deposits
at commercial banks are convertible into currency board notes and coins.
Commercial banks are responsible for holding enough notes and coins as vault
cash to satisfy their contractual obligations to their depositors to convert
deposits into notes and coins on demand. If the government imposes no minimum
reserve requirement on commercial banks, commercial banks may hold any quantity
(or ratio of reserves to liabilities) that they think is prudent; they are not
required to hold 100 per cent foreign reserves like the currency board, nor 100
per cent currency board notes and coins against deposits. (In other words, M0
is backed 100 per cent by foreign reserves in a currency board system, but broader
measures of the money supply such as M1, M2, and M3 are not.) Deposits at
commercial banks denominated in the currency board currency are fully
convertible at a fixed exchange rate into currency board notes and coins, and
currency board notes and coins are fully convertible at a fixed exchange rate
into the reserve currency. As for foreign currencies other than the reserve
currency, the currency board has no direct role in determining exchange rates
with them. Commercial banks trade them at market-determined exchange rates,
which may be fixed, pegged, or floating against the reserve currency and hence
against the currency board currency.
A typical central bank, in contrast, has limited
convertibility of its currency. Central banks in almost all developed countries
and in some developing countries have fully convertible currencies, but most
central banks in developing countries have partly convertible or inconvertible
currencies. They restrict or forbid certain transactions, particularly
purchases of foreign securities or real estate (see the International Monetary
Fund's Annual Report on Exchange Arrangements and Exchange Restrictions).
A typical currency board has a rule-bound monetary policy.
A currency board is not allowed to alter the exchange rate, except perhaps in
emergencies. (Chapter 5 proposes rules for defining such emergencies.) Nor is a
currency board allowed to alter its reserve ratio or the regulations affecting
commercial banks. A currency board merely exchanges its notes and coins for reserve
currency at a fixed rate in such quantities as commercial banks and the public
demand. When the demand for money changes, the role of a currency board is
passive. Market forces alone determine the money supply through a
self-adjusting process described in chapter 3.
A typical central bank, in contrast, has a partly or
completely discretionary monetary policy. A central bank can alter at will, or
with the approval of the government, the exchange rate, its ratio of foreign
reserves, or the regulations affecting commercial banks. It is not subject to
strict rules like a typical currency board.
A typical currency board is not a lender of last resort,
that is, it does not lend to commercial banks or other firms to help them avoid
bankruptcy. Commercial banks in a currency board system must rely on
alternatives to a lender of last resort (see chapter 6). A typical central
bank, in contrast, is a lender of last resort.
A typical currency board does not regulate commercial
banks. Banking regulations in a currency board system are usually few, and are
enforced by the ministry of finance or an office of bank regulation. A typical
central bank, in contrast, often regulates commercial banks. Perhaps the most
common form of regulation is imposing reserve requirements on commercial banks.
The required reserves, which are held in the form of deposits at the central
bank and vault cash, typically exceed the prudential reserves that commercial
banks would hold if no reserve requirements existed.
The activities of a typical currency board are transparent,
because a currency board is a very simple institution. It is merely a sort of
warehouse for reserve-currency securities that back its notes and coins in
circulation. The activities of a typical central bank are opaque. A central
bank is not a warehouse; it is a speculating institution whose effectiveness
partly depends on the ability to act secretly sometimes. (3)
Because a typical currency board is rule-bound and
transparent, it is protected from political pressure. It is protected by
implicit rules of political behavior, or, better yet, by an explicit
constitution such as that of the Appendix. A typical central bank is
politicized. Some central banks, such as the U.S. Federal Reserve System and
the European Central Bank, are politically independent in the sense that their
governors, once appointed, have sole control of the monetary base and cannot be
fired by the executive branch of the government or the legislature during their
fixed terms of office, and that the executive branch or legislature cannot
dictate lending by the central bank. Even the most politically independent
central banks sometimes yield to strong political pressure, though.
A typical currency board has high credibility. Its 100 per
cent foreign reserve requirement, rule-bound monetary policy, transparency, and
protection from political pressure enable it to maintain full convertibility
and a fixed exchange rate with the reserve currency. An appropriately chosen
reserve currency will be stable; therefore, the currency issued by the currency
board will be stable. A typical central bank, in contrast, has low credibility.
A few exceptionally good central banks, which exist mainly in developed
countries, have high credibility, but the majority do not. Because a typical
central bank has discretion in monetary policy, is opaque, and is politicized,
it has the means and the incentive to break promises about the exchange rate or
inflation whenever it wishes.
Readers may ask how credible a typical currency board can
be if, as has been the case with most currency boards, its reserve currency is
issued by a central bank. Might not the central bank of the reserve country
create monetary instability in a country with a currency board, for example? No
central bank has a perfect record of combating inflation. The central bank of
the reserve country may export instability and inflation to the currency board
country through the currency board.
We reply that one must think in terms of relative
credibility. The reserve currency, if a currency issued by a central bank
rather than a commodity, should be issued by an exceptionally good central
bank, such as the U.S. Federal Reserve System, the European Central Bank, or
the Bank of Japan. Those three central banks are not perfect, but they have
much more credibility than most other central banks now have or will have in
the foreseeable future. The U.S. dollar, the euro, and the Japanese yen have
exceptionally good records and good prospects for future stability, whereas
most currencies issued by other central banks have bad records and bad
prospects for future stability. For example, among developing countries with
more than 1 million people, 90 percent had their currencies depreciate against
the dollar from 1970 to 1993 (Schuler 1996: 33). A currency board transmits the
relative credibility of the reserve country's central bank to the currency
board country, whereas no such effect occurs if the country continues with a
typical central bank. A currency board can "import" the monetary
policy of an exceptionally good central bank by means of a fixed exchange rate
to the currency issued by that central bank.
A typical currency board earns seigniorage (income from
issue) only from interest. The currency board earns interest from its holdings
of reserve-currency securities (its main assets), yet pays no interest on its
notes and coins (its liabilities). Gross seigniorage is the income from issuing
notes and coins. It can be explicit interest income or implicit income in the
form of goods acquired by spending money. Net seigniorage (profit) is gross
seigniorage minus the cost of putting and maintaining notes and coins in
circulation.
A typical central bank also earns interest on its holdings
of securities, which include domestic as well as foreign securities. It earns
seigniorage on its notes and coins in circulationand on the deposits that
commercial banks hold with it. The deposits, like notes and coins, usually pay
no interest. But a more important source of seigniorage for a typical central
bank is inflation. To define inflation precisely, it is a general increase in
nominal prices, typically caused by an increase in the nominal money supply
that is not the result of increased voluntary saving. A typical currency board
cannot create inflation because it does not control the ultimate reserves of
the monetary system. For instance, the currency board system of
A typical currency board cannot finance spending by the
domestic government or domestic state enterprises because it is not allowed to
lend to them. A typical central bank finances spending by the domestic
government, whether to a relatively small extent (as in the
A typical currency board requires no
"preconditions" for monetary reform. Contrary to claims that have
been made by studies from the International Monetary Fund (Baliño, Enoch and
others 1997: 18; Enoch and Gulde 1997: 26-7), government finances, state
enterprises, or trade need not be already reformed before the currency board
can begin to issue a sound currency (Hanke 1999). A typical central bank cannot
issue a sound currency unless the "fiscal precondition" exists, that
is, the government no longer needs to finance budget deficits by means of
inflation. Once governments start to depend on central banks for financing
deficits, they usually have trouble stopping.
A typical currency board is conducive to rapid monetary
reform. It can be established quickly and fulfill its purpose quickly. A
typical central bank is a hindrance to rapid monetary reform.
Finally, a typical currency board needs only a small staff
of a few persons who perform routine functions that are easily learned. A
typical central bank needs a large staff trained in the intricacies of monetary
theory and policy. The People's Bank of China employs 150,000 people; the
Central Bank of the
Currency boards versus dollarization, multinational central
banks, and free bankingWhen the original
edition of this study appeared in 1994, currency boards were the only method
being widely discussed for unifying monetary policy in developing countries
with the generally much better monetary policy of the countries issuing the
major international currencies. Since 1999, dollarization and multinational
central banks have also received attention as methods of unifying monetary
policy. Another monetary system, free banking, has enjoyed a revival of
academic interest even though it does not exist anywhere today.
Dollarization occurs when residents of a country
extensively use the U.S. dollar or another foreign currency alongside or
instead of the domestic currency. Unofficial dollarization occurs when
individuals hold foreign-currency bank deposits or notes (paper money) to
protect against high inflation in the domestic currency. Unofficial
dollarization has existed in many countries for years. Dollarization has been
in the news recently because of interest in officialdollarization, which occurs
when a government adopts foreign currency as the predominant or exclusive legal
tender. For many years the largest economy to be officially dollarized was
The main difference between dollarization and an orthodox
currency board is that under dollarization no distinct national currency
exists. Unless arrangements can be made with the country whose currency it
uses, a dollarized country loses the seigniorage it would earn with a currency
board. Dollarization also deprives a country of whatever political benefits
exist from having a distinct national currency. On the other hand, if the
practical choice is not between an orthodox currency board and dollarization,
but between a currency board-like system and dollarization, dollarization is
likely to be more transparent, credible, and beneficial for the economy because
it is harder to change. The greater transparency and credibility of
dollarization influenced
The establishment of the European Central Bank to issue the
euro in 1999 spurred interest among Eastern European and Balkan countries in
the possibility of joining the European Central Bank as a method of unifying
their monetary policy with that of the euro zone. The European Central Bank is
one of four multinational central banks today. The others are the
Banque Centrale des Etats de l'Afrique de l'Ouest (Bank of
West African States) and the Banque des Etats de l'Afrique Centrale (Bank of
Central African States), which issue the CFA franc in a total of 14 countries,
mainly former French colonies; and the East Caribbean Central Bank, which
issues the East Caribbean dollar in six former and two current British
colonies. (5)
The East Caribbean Central Bank originated as a multinational currency board.
With a multinational central bank, the exchange rate is
pegged (as with the central banks issuing the CFA franc and the East Caribbean
Central Bank) or floating (as with the European Central Bank) rather than fixed
as is the case with a currency board. Exchange controls may exist, as they do
for all current multinational central banks except the European Central Bank.
Free banking is a system of competitive issue of notes and
other liabilities with minimal regulation. In particular, a completely free
banking system has no central bank, no lender of last resort, no reserve
requirements, and no legal restrictions on bank portfolios, interest rates, or
branching. Free banking systems have existed in nearly 60 countries in the
1800s and early 1900s. In general, they were relatively stable and successful
in preserving the fixed exchange rates to gold or silver that almost all
maintained (Dowd 1992). Currency boards or central banks replaced free banking
systems because of intellectual and political trends toward monopolizing issue
of notes and coins in a government body, and no free banking systems exist
today. In the last 25 years, however, interest in free banking among economists
has revived because of dissatisfaction with the performance of central banks
and, recently, the possibility that electronic money will make notes and coins
obsolete, enabling banks to offer full-fledged rivals to government-issued
currencies (Selgin and White 1994, Rahn 1999).
Like a currency board free banking typically involves a
fixed exchange rate and no exchange controls. But rather than being determined
by the reserve country, monetary policy is determined by competition among
banks. Banks are at liberty to offer consumers any type of currency. The
exchange rate can be fixed, pegged, or floating. If fixed or pegged, the anchor
currency can be a commodity or a foreign currency. The preferences of consumers
decide what kind of currency circulates and therefore what kind of monetary
policy succeeds in competition. Also, unlike the case under a currency board,
under free banking net seigniorage, like other forms of profit, tends to be
competed away to zero and passed along to consumers. Because free banks have no
monopoly of issuing notes and coins, they tend to issue them up to the point
where the profit is zero. This does not mean that free banking is inherently
inflationary; the commitment to a fixed exchange rate that almost all free
banking systems have had tends to prevent inflation.
Currency boards and wider economic reforms
Until the revival of interest in currency boards, almost all
economists simply assumed that most countries should have central banks.
Preventing central banks from misbehaving has been a key problem in economic
reform packages such as those implemented in
The alternative approach achieves stabilization by means of
deregulation of prices and trade and by means of a currency board with a fixed
exchange rate. The conventional approach tries to achieve stabilization by
means of liberalization of prices and trade, and by means of a central bank
with a pegged exchange rate. The means appear similar, but have quite different
effects.
The alternative approach advocates deregulation of prices
and trade. It permits the government to abolish price controls as quickly as is
politically feasible. The conventional approach advocates liberalization rather
than deregulation of prices and trade. It abolishes many price controls, trade
quotas, and tariffs, but retains others. In the conventional approach, full
deregulation of prices and trade is not possible because the central bank
allows the government and state enterprises to operate with soft budget
constraints. Soft budget constraints mean that because economic agents receive
subsidies, they can spend more than the sum of their earnings and their
unsubsidized borrowing.
In the alternative approach, the means for imposing the
hard budget constraints essential to macroeconomic stabilization is a currency
board with a fixed exchange rate. A typical currency board tends to end soft
budget constraints and impose their opposite, hard budget constraints, because
it cannot create inflation to finance the government budget deficit. A currency
board does not lend to the domestic government or to domestic state
enterprises, whether directly, or through domestic commercial banks, or through
the government budget. The hard budget constraint that a currency board tends
to impose on the government limits the nominal and real amount of subsidies
that the government can grant to state enterprises. That creates pressure on
the government to restructure or close unprofitable state enterprises.
Restructuring or closing state enterprises hardens their budget constraints. In
the alternative approach, centralized price controls are unnecessary, from a
strictly economic standpoint. The determination of prices can be left to market
forces within the framework of hard budget constraints, because state
enterprises that are unprofitable will tend to be closed or restructured by the
government.
In the conventional approach, in contrast, the means for
trying to impose hard budget constraints is a central bank whose currency is
subject to devaluation. (In many countries, the president or cabinet rather
than the central bank has legal authority over exchange-rte decisions.)
Devaluation softens budget constraints by allowing the central bank to continue
to finance the government budget deficit with no ultimate nominal limit. When
lending by the central bank increases the monetary base so much that the
central bank loses large amounts of foreign reserves at the existing pegged
exchange rate, the central bank can devalue the currency, re-pegging at a new
rate that holds until it devalues.
Both the alternative approach and the conventional approach
agree that microeconomic restructuring can begin before macroeconomic
stabilization, but differ concerning the relationship between microeconomic
restructuring and macroeconomic stabilization. The alternative approach tends
to impose hard budget constraints rapidly. In the alternative approach, the
government tends to finance itself by levying taxes and by borrowing at
positive real interest rates. It may continue to subsidize some state
enterprises, so that their budget constraints remain soft, but it is forced to
choose which unprofitable state enterprises it will continue to subsidize and
tends to harden the budget constraints of the rest by restructuring or closing
them. State commercial banks tend to increase interest rates to positive real
levels and cease being conduits for subsidies to state enterprises, because in
the alternative approach no central bank exists as a lender of last resort to
subsidize state commercial banks or state enterprises. Commercial banks tend to
become monitors of the performance of enterprises, which in turn tends to make
enterprises more market-oriented, for example by limited wage increases to
levels justified by the productivity of workers.
The alternative approach has no preconditions, fiscal or
otherwise; rather, by tending to impose hard budget constraints, it solves the
fiscal preconditions for economic reform and simultaneously introduces a sound
currency immediately. It tends to achieve macroeconomic stability more rapidly
and allows other economic reforms to proceed sooner than in the conventional
approach. The experience of the recent currency board-like systems and of some
orthodox currency board systems that implemented economic reforms after the
Second World War indicate that the alternative approach can be successful. The
conventional approach, in contrast, often fails to harden budget constraints
rapidly because it relies on a central bank. As a result, nagging inflation
prevents it from achieving its full economic potential.
Most countries experiencing economic problems need reforms
in addition to monetary stabilization. The details of a good package of
economic reforms differ widely across time and space, but the need to keep
inflation low and impose hard budget constraints is a recurrent theme. A
currency board is not an panacea, and the main advocates of currency boards
have never claimed that it is. By tending to impose hard budget constraints,
however, a currency board creates pressure for other economic reforms and
increases the prospect that they will be successful. In economic policy there are
rarely any certainties; rather, there are higher and lower chances for success.
A typical currency board provides higher chances for success than a typical
central bank.
Outline of this study
This study explains how a currency board system works and
how to establish a currency board. Chapter 2 states general theoretical and
practical arguments against using a typical central bank as a means of trying
to provide a sound currency.
Chapter 3 explains how the money supply is determined in a
currency board system, and contrasts it with how the money supply is determined
in a central banking system. Chapter 4 describes two approaches to establishing
a currency board. One approach is to convert the central bank into a currency
board. The other approach is to establish the currency board as the issuer of a
parallel currency to the central bank currency. (A parallel currency is one
that circulates extensively alongside another currency. The parallel currency
can have a fixed, pegged, or floating exchange rate with the other currency,
and can circulate legally or illegally. (6)) If
the currency board is the issuer of a parallel currency, its currency will
circulate officially alongside the central bank currency at the market rate of
exchange, much as foreign currency now does unofficially in many countries.
A currency board issues notes and coins backed 100 per cent
by foreign reserves in a reserve currency such as the U.S. dollar, the euro, or
the Japanese yen. A currency board exchanges its currency for the reserve
currency at a truly fixed exchange rate. The currency board can obtain its
initial foreign reserves in several ways. Chapter 4 describes how to calculate
the appropriate size of the initial foreign reserves of the currency board and
how to obtain the initial foreign reserves.
Chapter 5 explains how to operate a currency board. It also
contains ideas about ways to protect the currency board from political
pressure, in particular from pressure to convert it into a central bank.
Furthermore, it suggests how a currency board system can deal with a reserve
currency that becomes unstable. Chapter 6 responds to objections to the
currency board system. Chapter 7 summarizes our proposals. The Appendix
contains a model constitution for a currency board.
2. THE CASE
AGAINST TYPICAL CENTRAL BANKS
There are a number of general theoretical and practical
arguments against using a typical central bank to provide a sound currency. By
reviewing them carefully, we can untangle some confusions that have arisen
because economists and politicians have assumed that a central bank is the only
means by which to achieve macroeconomic stabilization. Many problems of
stabilization that occur in a typical central banking system do not occur in a
typical currency board system.
The functions of money
Consideration of the performance of a monetary system must
begin with consideration of the functions of money. Money functions as a medium
of exchange, store of value, and unit of account. A sound currency fulfills all
three functions satisfactorily and enables participants in a market economy to
make decentralized exchanges efficiently. Without a sound currency,
decentralized exchange cannot reach its full potential. A currency that suffers
high inflation reduces the efficiency of decentralized exchange and hence of a
market economy.
Many currencies issued by central banks in developing
countries are at present unsound, fulfilling none of the three functions of
money satisfactorily. They are inadequate media of exchange: the outside world
refuses to accept them. That impedes foreign investment and trade, and hence
the growth of national economies. They are also avoided domestically in cases
where people can illegally use sound foreign currencies easily.
Many currencies issued by central banks in developing
countries are unreliable stores of value: high inflation makes their value
unpredictable. As a result, people in countries with such currencies save by
hoarding commodities or relatively stable foreign currencies, which retain
value better than the domestic currency. Such hoarding is rational because of
the rapid depreciation of domestic currencies, but costly compared to being
able to use a sound domestic currency. Hoarding of commodities diverts goods
from use by their ultimate consumers into the stockpiles of enterprises and
middlemen. Hoarded bricks, for instance, could be used to build apartments
instead of piling up at construction sites.
In many countries with unsound currencies, extensive
holding of foreign currency notes is widespread. Researchers from the Federal
Reserve System have estimated that 55 to 70 percent of U.S. dollar notes are
held outside the
Many currencies issued by central banks in developing
countries are unsatisfactory units of account: high inflation distorts the
structure of prices and makes economic calculation very difficult. Real
(inflation-adjusted) prices fluctuate widely because sellers have difficulty
estimating the effect of inflation, not because of underlying real changes in
supply and demand. The competitiveness of export products in world markets
consequently experiences large fluctuations caused entirely by inflation, making
it difficult for domestic industries to plan production for export markets and
to earn foreign currency.
Stability and credibility
To understand more precisely how sound domestic currencies
would benefit the economies of countries that now lack them, let us consider
the connection between each of the qualities of a sound currency enumerated in
the previous chapter--full convertibility, credibility, and stability--and the
corresponding functions of money--as a medium of exchange, a store of value,
and a unit of account, respectively.
A currency's usefulness as a unit of account depends on its
stability. Stability means thatcurrent annual inflation is relatively low,
usually in single digits. We will combine discussion of stability with
discussion of credibility. Credibility means that the issuer creates confidence
that it will keep future inflation low. The concept of credibility in a sense
compresses expectations of the future stability of the currency into the
present. A currency's usefulness as a store of value depends on its
credibility.
Central banks everywhere have difficulty achieving
credibility. The difficulty is especially severe in developing countries,
because central banks there experience very strong political pressure for
inflation. For example, in the last 15 years
Lack of credibility typically contributes to high inflation
and high real interest rates. For example, workers base their demands for wages
on the expectation that the central bank will depreciate the domestic currency.
State enterprises and government ministries incur deficits because they expect
that the government will rescue them by pressuring or passing laws ordering the
central bank to finance the deficits. Such behavior creates momentum for
continued inflation.
In a typical central banking system, high real interest
rates often accompany high inflation. To compensate investors for perceived
exchange risk--the risk of using a currency prone to devaluation--real interest
rates for borrowers without access to subsidized credit typically must be high.
High real interest rates stifle economic activity by increasing the cost of
borrowing. Alternatively, real interest rates can be kept low by imposing
controls on interest rates and convertibility, but then shortages of credit
occur because demand exceeds supply at the controlled interest rates, which do
not take into account the exchange risk. Where currencies are sound, real
interest rates in truly unrestricted markets for credit are typically positive
but low, which encourages savings yet avoid stifling economic growth.
In contrast to a typical central bank, a typical currency
board has high credibility. A typical currency board system tends to have low
inflation because its reserve currency has low inflation. Also, because a fixed
exchange rate with the reserve currency prevents depreciation, a typical
currency board system tends to have interest rates similar to the levels
prevailing in the reserve currency, plus an allowance for political risk (the
risk that private property rights will not be enforced or that the government
will seize private property), taxes, and transactions fees. (7)Exchange
risk tends to be absent.
Credibility and exchange rates
The degree of credibility of a monetary system affects the
type of exchange-rate arrangements that are sustainable for it. An
exchange-rate arrangement that is not credible tends not to be sustainable
because it imposes high costs on the economy.
Along the continuum of exchange-rates from most flexible to
most rigid, there are three basic types. To reiterate, a pegged exchange rate
is constant for the time being in terms of a reserve currency, but carries no
credible long-term guarantee of remaining at its current rate. A pegged
exchange rate should not be confused with a fixed exchange rate. A fixed
exchange rate is permanent, or, if the monetary authority can alter it in
emergencies, the monetary authority must follow well-defined rules known in
advance by the public. A floating exchange rate is not maintained constant in
terms of any reserve currency.(8)
The main difference between a fixed exchange rate and a
pegged exchange rate is credibility. A typical central bank cannot maintain a
truly fixed exchange rate because it has low credibility. A declaration by a
typical central bank that it is maintaining a fixed exchange rate is not
credible because the monetary rule of a fixed exchange rate conflicts with
discretionary monetary policy. The only period when a large number of central
banks have maintained truly fixed exchange rates was during the
"classical" gold standard (1880 or before to 1914). At the time,
central banks were still in the minority among the monetary systems of the
world, and many central banks were owned privately rather than by governments
(Conant 1969 [1927]). Those conditions, which induced central banks to maintain
fixed exchange rates, no longer apply.
Since 1914, almost all central banks that have claimed to
maintain "fixed" exchange rates, including exceptionally good central
banks, have in reality maintained pegged exchange rates. Almost all central
banks devalued their currencies against gold or silver during the First World
War, the Great Depression, the Second World War, and the breakup of the Bretton
Woods system in the early 1970s. Most central banks with pegged exchange rates
have also devalued their currencies individually at other times (for a history,
see Yeager 1976: 295-610).
The main purpose of a pegged exchange rate is to make hard
budget constraints more credible. A hard budget constraint means that an
economic agent cannot spend more than it earns plus what it can borrow in
unsubsidized credit markets. The opposite of a hard budget constraint is a soft
budget constraint, where an economic agent receives subsidies and therefore can
spend more than it earns and borrows in unsubsidized credit markets. In
practice, though, pegged rates in developing countries are typically sinking
rates, involving devaluation or foreign-exchange controls to accommodate soft
budget constraints. The bias toward devaluation creates perverse consequences.
As compensation for the perceived risk of devaluation of a pegged exchange
rate, lenders and investors demand higher real rates of interest than would
exist with a truly fixed exchange rate (Walters 1990: 14-15). It may take years
for a typical central bank to achieve substantial credibility for a pegged
exchange rate. In the meantime, high real interest rates to defend a pegged
exchange rate create high costs that are particularly painful to
capital-intensive industries and to export industries that compete against
foreign counterparts based in countries with more credible monetary
authorities. Alternatively, a central bank may keep real interest rates low by
imposing interest rate controls and restricting convertibility, but that causes
a shortage of credit.
High real interest rates are one perverse consequence of
pegged exchange rates; frequent high real exchange rates are another. The real
exchange rate is the ratio of the prices of traded goods to the prices of
nontraded goods. Traded goods are exports and imports. Nontraded goods are
goods such as land and labor, which for various reasons cannot easily be moved
from one country to another. There are several ways to calculate price ratios
of traded to nontraded goods, but the important thing is that all are intended
to show whether exchange rates are systematically affecting the competitiveness
of export industries. Overvalued real exchange rates mean that export
industries are relatively uncompetitive, whereas undervalued real exchange
rates mean that export industries are relatively competitive. (9)
Overvalued real exchange rates in countries with pegged
rates can result from sudden increases in foreign investment or, more commonly,
from domestic inflation. A typical central bank maintaining a pegged exchange
rate tends to inflate, which causes the monetary base to increase faster than
its foreign reserves. Consequently, the domestic price level increases. Imports
become less expensive and exports decrease because they become more expensive
in world markets. The central bank loses foreign reserves as people buy more
imports and as currency speculators bet that the loss of foreign reserves will
induce the central bank to depreciate the currency. To avoid losing more
foreign reserves and to revive exports, the central bank imposes
foreign-exchange controls. Alternatively, the central bank devalues the
currency, re-pegging it at an exchange rate sufficiently undervalued that for a
while the central bank can safely continue to inflate. An undervalued exchange
rate temporarily revives exports by making them suddenly inexpensive and chokes
imports by making them suddenly expensive, but as inflation continues and the
exchange rate later becomes overvalued, the opposite effects occur.
For an economy with a typical central bank maintaining a
pegged exchange rate, therefore, a pegged rate alternately tends to depress and
overstimulate economic activity. Export industries often experience high real
interest rates and high costs for the nontraded goods that they use to help
produce traded goods. Rather than increase their productivity to remain
competitive, export industries often take the easier course of pressuring the
central bank to devalue the currency. Political pressure to devalue the
currency also comes from government ministries and state enterprises, who desire
their soft budget constraints to continue. The incentives facing a typical
central bank are such that it usually benefits more by devaluing the currency
than by maintaining the existing exchange rate. The clash between the long-term
goals and the short-term incentives of central banks is so pervasive that it
has a name: "time consistency" (or "time inconsistency";
see Kydland and Prescott 1977). The Southeast Asian currency crisis of 1997-8
is recent example of time inconsistency in central bank policy.
Even central banks that are better than average have had
difficulty maintaining a "hard" peg, which allows no systematic
depreciation. The difficulty of maintaining a hard peg has induced some central
banks to experiment with a crawling peg.(10)A
crawling peg is a limited appreciation or depreciation of the currency
according to a schedule. For example, a central bank may promise not to
depreciate its currency against the U.S. dollar by more than 20 per cent a
year. With a decelerating crawling peg, depreciation slows year by year and
perhaps eventually stops. A decelerating crawl is implicitly a promise by the
central bank to reduce the growth of the monetary base.
A possible advantage of a crawling peg compared to a hard
peg is that a crawling peg may be less costly to the economy because it does
not try to achieve credibility immediately; instead, the crawl can decelerate
to a hard peg when the central bank seems to have enough credibility to
maintain a hard peg without high real interest rates or extensive
foreign-exchange controls. Another way in which a crawling peg may be less
costly to the economy than a hard peg is that if expectations of inflation
pervade behavior and long-term contracts, a crawling peg reduces the shifts of
real wealth that occur with a suddenly imposed hard peg.
The main disadvantage of a crawling peg is that it can
accelerate, rather than decelerate to a hard peg. Countries that have tried
crawling pegs have generally had higher inflation than other countries in their
regions that have maintained harder pegs punctuated by occasional devaluations
(Connolly 1985). A crawling peg does not change the governance or the
incentives of the central bank, so it is little more credible than a hard peg.
In addition, countries with very high inflation tend to have no long-term
contracts in domestic currency because the domestic currency is not a reliable
unit of account, so a sudden, credible end to inflation would not cause big
shifts of real wealth from debtors to creditors. Consequently, a crawling peg
has no significant advantage for a country with very high inflation (say, more
than 100 per cent a year) compared to a hard peg. For a country with moderately
high inflation (say, 20 to 100 per cent a year), we think that whatever
advantages a crawling peg may have can be duplicated by means of a parallel
currency (see chapter 5) or by de-indexing indexed wages and prices.
Since most central banks, including those that are better
than average, have lacked the credibility to maintain fixed or pegged exchange
rates, many economists have advocated floating exchange rates for them. An
advantage of a floating exchange rate compared to a pegged rate is that a
floating rate requires little credibility, because the central bank makes no
promise concerning the exchange rate. The central bank need not worry that lack
of credibility will cause it to lose foreign reserves supporting the exchange
rate, because it need not support the exchange rate at all.
Another advantage of a floating exchange rate is that in
theory, full convertibility can be allowed. There need be no fear that full
convertibility will create the perverse consequences that can occur with a
pegged exchange rate. In practice, though, most countries with floating
exchange rates impose foreign-exchange controls.
A floating exchange rate can be "clean" or
"dirty." In a clean float, the central bank does not try to influence
the exchange rate systematically. It is possible to for a central bank to hold
zero foreign reserves if it allows a clean float. The Reserve Bank of
Even a clean floating exchange rate has disadvantages, the
main one being that it does not counteract political pressure for inflation
from interest groups and ambitious politicians who favor soft budget
constraints and short-term considerations. Since the early 1980s, some central
banks in developing countries have successfully combined floating exchange
rates with low inflation, low real interest rates, and full convertibility, but
almost no central banks in developing countries have done so over long periods
(Collier and Joshi 1989: 103).
Another disadvantage of a floating exchange rate is that
real interest rates can be high, as they tend to be with a pegged exchange
rate, if high inflation appears likely. A typical central bank maintaining a
floating exchange rate may encounter difficulties like those of a typical
central bank maintaining a pegged exchange rate. A low inflation rate is a
"price-level peg" similar to an exchange-rate peg. Although a central
bank needs little credibility to operate a floating exchange rate, it needs
substantial credibility to combine low inflation with low real interest rates
and full convertibility. If its credibility is low, real interest rates will be
high, although probably not so high as with a pegged exchange rate that has low
credibility. (Speculative pressure on the central bank tends to be less with a floating
exchange rate than with a pegged exchange rate, because the central bank has no
obligation to maintain a particular exchange rate. With a floating rate,
speculators typically engage more in offsetting speculation against each other
and less in speculation against the central bank than with a pegged rate.)
Yet another disadvantage of a floating exchange rate is
that it requires the central bank to target something other than the nominal
exchange rate. However, other targets have problems of definition, control, and
appropriateness. Suppose a central bank decides to target the money supply. The
central bank must define which measure of the money supply it wishes to
control--the monetary base, or broader measures that include components beyond
its direct control, such as deposits at commercial banks. The central bank must
try to control the measure it has chosen, setting targets and achieving them,
revising them for leads and lags in economic activity and for other factors,
such as changes in banking technology and people's habits of holding money that
can change the money supply unpredictably. The central bank must also
frequently evaluate whether the target it has chosen is appropriate, or whether
another target would be more appropriate for encouraging economic growth and
price stability. Problems of defining and achieving appropriate targets caused
most central banks in developed countries to abandon strict adherence to money
supply targets after experimenting with money supply targets in the 1980s. However,
price levels, interest rates, real growth rates, and nominal income also have
problems as targets (see Lindsey and Wallich 1987). (11)
The nominal exchange rate is the easiest target to define, control, and
evaluate.
Because a typical currency board has high credibility, it
can maintain a truly fixed exchange rate. A fixed exchange rate maintained by a
currency board avoids the disadvantages of a pegged or floating exchange rate maintained
by a typical central bank. A typical currency board need never worry that its
foreign reserves are inadequate, because they are equal to 100 per cent or
slightly more of its notes and coins in circulation. Because the exchange rate
is fixed, lenders and investors tend not to demand high real interest rates as
compensation for a perceived risk of depreciation against the reserve currency.
A well-chosen reserve currency will be stable, credible, and fully convertible.
(Chapter 4 discusses how to choose a reserve currency.) Leaving aside the issue
of political risk and considering exchange risk only, a typical currency board
system therefore does not experience the high real interest rates present in a
typical central banking system with a pegged exchange rate and full
convertibility, and present even in some central banking systems with floating
exchange rates.
Convertibility and foreign-exchange controls
A currency's usefulness as a medium of exchange depends on
its convertibility. Convertibility means that the currency can buy domestic and
foreign goods and services, including foreign currencies. Without a convertible
currency, people cannot easily make the decentralized exchanges using money
that make a market economy work efficiently.
Convertibility has three gradations corresponding to the
extent to which a government allows a currency to function as a medium of
exchange. The most basic type of convertibility is cash convertibility--the
ability to exchange a unit (say, a dollar) of bank deposits for a unit of notes
and coins on demand. Cash convertibility is so much taken for granted in
developed countries that it is seldom discussed, except when bank runs occur.
Nevertheless, it does not exist in a few developing countries, such as the
remaining centrally planned economies.
The second type of convertibility is commodity
convertibility--the ability to buy domestic goods and services. It too is so
much taken for granted in developed countries that it is seldom discussed.
Nevertheless, it too does not exist in some developing countries. In countries
with commodity convertibility, all that is usually required to buy domestic
goods and services is cash or credit to pay a domestic seller; domestic trade
is little restricted compared to a centrally planned economy. The exchange of
goods and services is much more extensive, rapid, and efficient where commodity
convertibility exists than where it does not.
The third type of convertibility is foreign-exchange
convertibility--the ability to buy foreign goods and services, including
foreign currencies. If no restrictions exist on buying foreign goods and
services, including foreign currencies, at market rates of exchange, a currency
is said to have full foreign-exchange convertibility. A currency with cash, commodity,
and full foreign-exchange convertibility has full convertibility.
Foreign-exchange convertibility almost always implies cash and commodity
convertibility, so henceforth full convertibility will be synonymous with
unrestricted foreign-exchange. The currencies of most developed countries are
fully convertible, but the currencies of most developing countries are partly
convertible or inconvertible. For example, many currencies are convertible for
most current-account purchases, in which residents use domestic currency to buy
foreign goods and services for import, but inconvertible for many
capital-account purchases, in which residents use domestic currency to buy
foreign financial assets such as foreign currencies and securities, and certain
nonfinancial assets such as real estate. (12)
Restrictions on capital-account transactions are called capital controls.
Current-account convertibility exposes domestic producers
to foreign competition if trade quotas and tariffs are low. Current-account
convertibility tends to introduce into the domestic economy the structure of
prices that prevails in world markets. World prices are signals that help
people determine which areas of production to specialize in. By specializing in
the goods they produce most efficiently, then trading those goods for other
goods, wealth increases globally.
Current-account convertibility is helpful for foreign
trade, but is insufficient for attracting substantial foreign investment; for
that, capital-account convertibility is necessary. Unless foreigners can
repatriate some profits, they will usually be reluctant to make large
investments.
Almost universal agreement exists that cash and commodity
convertibility are desirable in most countries immediately. However, many
economists have advised developing countries to delay full foreign-exchange
convertibility (for example, Greene and Isard 1991: 12-13, 16; McKinnon 1991:
156; Williamson 1991: 379). One argument against immediate full convertibility
is that it would worsen capital flight, that is, domestic investment would
leave on a large scale. Another argument, inconsistent with the first, is that
immediate full convertibility would allow excessive foreign investment and would
thereby make export goods uncompetitive. Foreign investment increases the
prices of land, labor, and other nontraded goods. Prices of exported goods
would then increase because they are made partly from nontraded goods. A large,
sudden appreciation of the real exchange rate could make exports uncompetitive,
causing a depression (as Ronald McKinnon argues in Hanke and Walters 1991:
187-9). A third argument against immediate full convertibility is that it would
create problems of moral hazard.
The arguments against full convertibility, although perhaps
valid for a typical central banking system, are not applicable to a typical
currency board system. Immediate full convertibility in a typical currency
board system is credible, so it tends to encourage a net inflow of capital
rather than capital flight. Although foreign investment increases the prices of
labor, land, and other nontraded goods, but if the investment is used
productively, the new higher prices reflect the increased productivity of
nontraded goods (Schmieding 1992: 196). The experience of currency board
systems has been that foreign investment does not cause large, sudden
appreciations of the real exchange rate that make exports uncompetitive and
cause depressions. International capital movements, like domestic interregional
capital movements, tend to be self-correcting if they overreact to
opportunities for arbitrage. Immediate full convertibility tends not to create
problems of moral hazard in a typical currency board system because a typical
currency board is not a lender of last resort. The unhappy experience of
The historical experience of currency boards supports the
view that problems with full convertibility in a typical central banking system
are caused by central banking rather than by full convertibility. The currency
board devised by John Maynard Keynes that existed in Northern Russian from 1918
to 1920 maintained full convertibility from the day it opened, even though
The experience of these currency board systems, and the
contrast with the typical experience of central banking systems, is part of a
pattern. The International Monetary Fund'sAnnual Report on Exchange
Arrangements and Exchange Restrictions, published since 1950, describes
restrictions on convertibility in its member countries and some of their colonies.
Never have the majority of central banking systems reporting to it had full
convertibility. Almost all developed countries with central banks now have full
convertibility, but only a minority of developing countries do. Among currency
board and currency board-like systems, only that of
Central banking and deficit finance
The lack of stability, lack of credibility, and restricted
convertibility of a currency issued by a typical central bank results from its
subordination to financing government budget deficits. In recent years,
government budget deficits have been typical of developed and developing
countries alike.
Most developed countries have fully convertible currencies
and well-developed financial markets, so their governments can finance budget
deficits by issuing debt that financial markets hold voluntarily. For example,
until it reduced its government budget deficits to meet the targets for the
single European currency,
Most developing countries are not like
Because financial markets are backward in many developing
countries, the capacity of the markets to absorb government debt, even when
compelled to do so, is small. That leaves inflation as the main way of
financing government budget deficits. Central banks in developing countries
have little scope for discretionary monetary policy because they must create
inflation sufficient to finance budget deficits. The subordination of central
banks and banking systems in developing countries to the demands of government
finance keeps their financial systems backward, overregulated
("repressed"), and incapable of mobilizing savings efficiently to
encourage economic growth (Fry 1988: 13-16).
The main argument that economists make for preferring a
central bank to a currency board is that a central bank has more flexibility.
On theoretical grounds the advantages of central bank flexibility are doubtful,
as we will see. But even if the theoretical case for central bank flexibility
were correct, a typical central bank in a developing country has little flexibility
because it is subordinate to government deficit finance.
A typical currency board cannot finance spending by the
domestic government or by domestic state enterprises because it cannot lend to
them. It cannot finance domestic government budget deficits, so it avoids
subordinating the monetary system to deficit finance.
Political independence: an unattainable goal
To help central banks to resist political pressure to
finance government budget deficits, the International Monetary Fund and many
economists have suggested that they be politically independent (for example,
Camdessus 1992: 342). Political independence for a central bank means that its
governors, once appointed, have sole control of the monetary base and cannot be
fired by the executive or legislative branches of government during their fixed
terms of office. It also usually means that the central bank finances itself
from seigniorage, rather than depending on the administration or legislature
for funds, and that they cannot dictate lending by it. For developed countries,
the more politically independent the central bank, the lower inflation tends to
be (Alesina 1989).
However, it has been impossible to fully protect central
banks from political pressure even in developed countries with long traditions
of representative government and separation of powers among the branches of
government. Developing countries that are still trying to achieve fully
representative government and that have no indigenous tradition of separation
of powers have found the task even more difficult. Furthermore, political
independence does not seem to result in lower inflation for central banks in
developing countries (Cukierman and others 1992: 369-76).
A typical currency board, in contrast, can be protected
from political pressure, as past currency boards have been. Chapter 5 proposes
safeguards to provide maximum protection from political pressure for a currency
board.
Inadequate staff
Besides the difficulty of protecting central banks in
developing countries from political pressure, there is also the problem of
finding sufficient qualified staff. The staff of many central banks,
particularly those in former centrally planned economies, lack thorough
training in the art of central banking, and in many cases it is doubtful that
they could administer an appropriate monetary policy even if they were
protected from political pressure. After decades of investing time and money in
training staff, the results in developing countries have been no better than if
they had simply adopted the monetary policy of a major international currency.
A typical currency board needs only a small staff. The
tasks that the staff of a currency board performs are simple and do not require
as much skill as the tasks performed by the staff of a central bank.
Flexibility: a problem even in theory
A typical central bank in a developing country faces many
practical obstacles to issuing a sound currency. There are also strong
arguments against discretionary monetary policy, and hence against a typical
central bank, on theoretical grounds. Their gist is that even a central bank
that has the flexibility to act according to its best judgment, rather than in
response to political pressure, is likely on average to destabilize the
economy.
Theoretical arguments against discretionary monetary policy
come from three sources. The monetarist school of economic thought emphasizes
that long and variable lags can make the effects of discretionary monetary
policy unpredictable. Unless the central bank knows approximately how long a
lag exists between changes in the monetary base and changes in prices, for
example, it can destabilize rather than stabilize the economy by trying to
influence economic activity. Hence there is reason to think that discretionary
policy on average will have worse results than rule-bound policy, even though
in particular cases discretionary policy may get lucky and do better than
rule-bound policy (Friedman 1948, 1960; Laidler 1982: 25-34, 153-63, 187-92; Meltzer
1992).
The rational expectations school of economic thought
emphasizes that whatever systematic discretionary policy a central bank can
administer, people can anticipate and counteract. In fact, many people, such as
bankers and traders, can make profits by correctly anticipating the policy of
the central bank. Their actions create problems for the central bank because
much of the effectiveness of its policies depends on surprise. If people
correctly anticipate that the central bank will create higher inflation, they
adjust prices and interest rates accordingly, and inflation has no temporary
stimulative effect on the economy. The rational expectations school claims that
to the extent that a central bank has a systematic policy, people tend to
anticipate the policy even if the central bank tries to hide it. The only type
of central bank policy that is consistently effective, then, is one of random
surprises. On average, random surprises destabilize the economy because they
create unwelcome uncertainty. Monetary rules are the best attainable policy
given the ability of people to anticipate and counteract the policies of the
central bank. Discretionary monetary policy can easily result in higher
inflation and lower economic growth than rule-bound policy (Barro and Gordon
1983, Kydland and Prescott 1977, Persson and Tabellini 1990: 19-33).
A related criticism of discretionary monetary policy, made
by the "Austrian" school of economic thought, is that discretionary
monetary policy is a type of central planning. Central banking has the same
disadvantages as, say, central planning of agricultural production. Central
planning suppresses price and quantity signals that convey information to
people who have the skill to interpret them correctly. In the monetary system,
among the most important signals are changes in reserves. Changes in the
balance of payments or in the public's holdings of notes and coins cause
changes in reserves and, through them, in the money supply, interest rates, and
income. Discretionary monetary policy, to be worthy of the name, must try to
fight markets as they adjust toward a new set of market-clearing (equilibrium)
prices. By doing so, it usually makes adjustment more prolonged and costly.
Luck may occasionally enable discretionary monetary policy to have beneficial
effects. On average, though, discretionary monetary policy is harmful, because
a central bank has no way of knowing in advance, or often even in retrospect,
which discretionary policies are beneficial and which are harmful (Selgin
1988b: 85-125).
The monetarist school and Austrian school emphasize aspects
of the central bank's ignorance, whereas the rational expectations school
emphasizes the central bank's inability to surprise people with systematic
policy. Although none of the three schools has ever considered in detail a
currency board as a monetary rule, the arguments of all three schools lend
support to the case for currency boards, since currency boards are rule-bound
and have no discretion in monetary policy.
Advocates of discretionary monetary policy consider it
undesirable for governments to renounce discretionary monetary policy as a
possible tool for making real adjustments in national economies. As an example
of the benefits that discretionary monetary policy can bring, they appeal to
cases in which nominal wages are too high for full employment, yet rigid, so
that some people who wish to work cannot find jobs. A central bank, they say,
can increase employment by creating inflation. Inflation can reduce real wages
and make the domestic currency depreciate against foreign currencies, causing a
temporary increase in exports.
When considering the validity of these examples, one must
ask why nominal wages are rigid. The reasons are usually that people expect
that the central bank will inflate, and that laws give workers who already have
jobs special privileges protecting them from competition by unemployed workers.
If the central bank and special privileges for already employed workers were
abolished, greater incentives would exist for nominal wages to become flexible.
Central banking is the problem rather than the solution in the example. Nominal
wages have been flexible in
A currency board cannot administer a discretionary monetary
policy, so its policy is inflexible. Rather than being a disadvantage, though,
the inflexibility of a currency board tends to protect an economy from the
destabilizing effects of discretionary monetary policy and tends to force wages
and prices to be flexible. The case for establishing a currency board or other
monetary rule is much the same as the case for establishing a constitution that
limits the powers of government. Constraints external to day-to-day politics
are necessary to prevent most governments from abridging political freedoms. It
is possible for a government to agree to bind itself to constitutional
constraints, and to create institutions that successfully enforce the
constraints, even though the government would not behave in the same manner if
no constitutional constraints existed. Constitutions are desirable because even
a government that respects political freedoms at one time may later abridge
them unless a constitution establishes them as durable features of political
life. The experience of constitutions is far from perfect, but it does support
the claim that a constitution can be more effective at constraining the powers
of government than no constitution.
A currency board is a form of monetary constitution that
prevents the domestic government from abridging economic freedoms by levying a
high inflation tax not desired by the public. Because a currency board cannot
finance budget deficits of the domestic government, the currency board system
establishes an implicit low-inflation fiscal constitution. A clever, very
determined government can probably find ways to subvert a currency board, just
as it can subvert a political constitution, but safeguards can secure a new
currency board maximum protection from such scheming and should be sufficient
to deter a less than very determined government. (13)
3. CURRENCY
BOARDS, CENTRAL BANKS, AND THE MONEY SUPPLY
Because a typical central bank in a developing country has
great difficulty providing a sound currency, developing countries should
consider a different approach to providing a sound currency, namely,
establishing a currency board. We will show how the money supply is determined
in a currency board system, how to establish a currency board, and how to
operate the currency board and protect it from political pressure to convert it
into a central bank.
Money supply in a currency board system
A typical currency board system relies entirely on market
forces to determine the amount of notes and coins that the currency board
supplies. Market forces also determine other components of the supply of money
in the broad sense (in the examples here, the public's deposits at commercial
banks) by processes described below.
In a currency board system and a central banking system
alike, commercial banks are entrepreneurs of credit. A commercial bank cannot
for long lend more to borrowers than depositors wish to lend to the bank, in
the form of deposits held instead of spent. If a commercial bank lends
excessively, the borrowers spend the excess, for instance by writing cheques.
In the payments system, more funds flow out of the bank than flow into the
bank. To prevent the outflow from bankrupting it, a commercial bank holds
reserves. The loans of commercial banks are limited by their need to maintain
sufficient reserves to enable depositors to convert deposits into reserves on
demand and to withstand outflows of reserves through the payments system.
A typical currency board has no active role in determining
the monetary base. A fixed exchange rate with the reserve currency and the
requirement that the currency board hold 100 per cent foreign reserves prevent
it from increasing or decreasing the monetary base at its own discretion. Nor
does a typical currency board influence the relationship between the monetary
base and the money supply by imposing reserve ratios or otherwise regulating
commercial banks. The money supply in a typical currency board system,
therefore, is determined entirely by market forces. A typical central bank, in
contrast, can at its discretion increase or decrease the monetary base. It can
lend to commercial banks, creating reserves for them, even if its foreign
reserves are decreasing. More reserves tend to enable commercial banks to make
more loans, which they do by creating deposits for borrowers. The money supply then
increases. Decreasing the monetary base tends to have the opposite effect.
Besides changing the monetary base, a typical central bank can also influence
the supply of commercial bank loans by changing the reserve requirements for
commercial banks.
Despite the inability of a typical currency board to create
reserves for commercial banks at its own discretion, the money supply in a
typical currency board system is quite elastic (responsive) to changes in
demand, because the system can acquire foreign reserves. The elasticity of the
money supply in a currency board system is one reason that
The ultimate reserves in a currency board system are the
monetary base of the reserve currency. The only way to acquire the ultimate
reserves is to obtain them from the reserve country. (14)
In its simplest form, that requires achieving a current-account surplus. Making
certain assumptions about a currency board system, changes in the
current-account balance begin a sequence of events that change the money supply
in the same direction. A current-account surplus ultimately increases the money
supply, whereas a current-account deficit ultimately decreases the money
supply. (The overall balance of payments is the gain or loss of reserves in a
period. It consists of the current-account balance--trade in goods and
services--plus the capital-account balance--investment and gifts.) The
assumptions that underlie this simplified description of the money supply
process in a currency board system are:
1. Deposits of commercial banks are convertible into
currency board notes and coins (cash) at a fixed rate.
2. The currency board is the only domestic issuer of notes
and coins.
3. Commercial banks' ratio of reserves to deposits is
constant.
4. The public's ratio of currency board notes and coins to
commercial bank deposits to is constant.
5. Income and the money supply move in the same direction.
6. There is no international branch banking between the
currency board country and the reserve country.
7. Changes in the balance of payments occur only in the
current account; the capital account does not change.
8. No binding minimum reserve ratios or other special bank
regulations exist.
9. People do not hold reserve currency or use the reserve currency
in domestic transactions. (15)
To show how money is supplied in a currency board system,
and how the process differs under central banking, we will use a combination of
balance sheets and flow diagrams. The flow diagrams (Figures 3.2, 3.6, 3.10,
and 3.11) depict sequences of events, whereas the balance sheets depict
conditions at particular stages for the relevant agents. (16)
Figure 3.1 shows simplified balance sheets for a typical
currency board as well as the combined balance sheet of commercial banks and
the combined balance sheet of members of the public. (The public is anyone
other than the currency board and commercial banks.) Recall that in a balance
sheet, by definition, assets = liabilities + net worth. The money supply is the
left-hand (asset) side of the public's balance sheet.
|
Figure 3.1. Balance
sheetsCurrency board |
|
|
Assets |
Liabilities |
|
Foreign reserves |
Notes in circulation Deposits of commercial banks (optional) Net worth |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves) Loans to public |
Deposits of public Net worth |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes Bank deposits |
Bank loans Net worth |
|
Monetary base = notes in
circulation in currency board's balance sheet Money supply = currency board notes + bank deposits in
public's balance sheet Commercial banks' ratio of reserves to deposits =
currency board notes ÷ deposits of public in commercial banks' balance sheet Public's ratio of currency to deposits = currency board
notes ÷ bank deposits in public's balance sheet |
|
Figure 3.2 is a flow diagram that illustrates an initial
situation where the current-account balance is zero: the value of exports
equals the value of imports. Now assume that a current-account surplus occurs.
As the surplus works its way through the monetary system, the currency board
plays an explicit (though passive) role in the sequence only at the stage
labeled "demand for goods in general, including currency board notes and
coins, increases." The system is self-adjusting, and it eventually
achieves a new equilibrium, that is, the current-account balance returns to
zero and the relevant markets clear.
To illustrate the relationship between commercial banks and
the currency board in the sequence of events illustrated in Figure 3.2, the
balance sheets that follow use hypothetical numbers. The "initial
stage" of the balance sheets (Figure 3.3) is the first stage of Figure
3.2, a situation where the current-account balance is zero and the relevant
markets clear--an equilibrium. For the sake of simplicity, assume that net
worth in the balance sheet of the currency board and stockholders' equity in
the balance sheets of commercial banks are zero. Assume further that commercial
banks desire a reserve ratio of 2 percent, and that the public desires a
currency-to-deposit ratio of 10 percent. Initially, 600 units of currency board
notes and coins are in circulation. (The numbers in the examples that follow
were chosen because they result in convenient whole-number solutions. Numbers
in bold represent changes from the previous stage.)
Now let there be a current-account surplus of 12 units, in
the form of foreign currency that the domestic public deposits in domestic
commercial banks. Since, by assumption, commercial banks hold all reserves in
the form of currency board notes and coins, the banks exchange the foreign
currency for currency board notes and coins. The assets and liabilities of the
currency board become 12 units more than in the initial stage, reserves of
commercial banks become 12 units more than in the initial stage, and deposits
of the public at commercial banks become 12 units more than in the initial
stage. Furthermore, the money supply is 12 units more than in the initial
stage. This is the intermediate stage (Figure 3.4).
Notice that commercial banks have a reserve ratio of
approximately 2.23 percent in the intermediate stage, which is more than their
desired, initial ratio of 2 percent. Notice also that the public has a
currency-to-deposit ratio of approximately 9.98 percent in the intermediate
stage, which is less than its desired, initial ratio of 10 percent. Hence the
monetary system is in disequilibrium (the relevant markets do not clear).
Commercial banks therefore increase their loans, and the public increases its
holdings of cash, to restore the ratios of the initial stage. In the final
stage, they do so, achieving a new equilibrium, with the money supply now 110
units more than it was in the initial stage (see Figure 3.5).
|
Figure 3.2. Money supply
increase in a currency board system 1. Current-account balance is zero (exports equal
imports)--equilibrium (Figure 3.3) 2. Domestic demand for imported goods decreases or
foreign demand for currency board country's goods increases 3. Current-account surplus (exports exceed imports) 4. Reserves of commercial banks increase 5. Loans by commercial banks increase (Figure 3.4) 6. Interest rates decrease 7. Income increases 8. Demand for goods in general, including currency board
notes and coins, increases 9. Prices of domestic goods increase 10. Domestic demand for foreign goods increases or
foreign demand for currency board country's goods decreases 11. Current-account balance returns to zero--new
equilibrium (Figure 3.5) |
|
Figure 3.3. Money supply
increase in a currency board system, initial stage Currency board |
|
|
Assets |
Liabilities |
|
Foreign reserves 600 |
Notes in circulation 600 Net worth 0 |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves)
100 Loans to public 4900 |
Deposits of public 5000 Net worth 0 |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes 500 Bank deposits 5000 |
Bank loans 4900 Net worth 600 |
|
Monetary base = 600 Money supply = 5500 Commercial banks' ratio of reserves to deposits =
100:5000 = 2% (equilibrium) Public's ratio of currency to deposits = 500:5000 = 10%
(equilibrium) |
|
|
Figure 3.4. Money supply
increase in a currency board system, intermediate stage Currency board |
|
|
Assets |
Liabilities |
|
Foreign reserves +12 612 |
Notes in circulation +12
612 Net worth 0 |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves) +12
112 Loans to public 4900 |
Deposits of public +12 5012
Net worth 0 |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes 500 Bank deposits +12 5012 |
Bank loans 4900 Net worth +12 612 |
|
Monetary base = 612 (expansion =
12 over initial stage) Money supply = 5512 (expansion = 12 over initial stage) Commercial banks' ratio of reserves to deposits =
112:5012 = approximately 2.23% (disequilibrium) Public's ratio of currency to deposits = 500:5012 =
approximately 9.98% (disequilibrium) |
|
|
Figure 3.5. Money supply
increase in a currency board system, final stage Currency board |
|
|
Assets |
Liabilities |
|
Foreign reserves 612 |
Notes in circulation 612 Net worth 0 |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves) -10
102 Loans to public +98 4998 |
Deposits of public +88
5100 Net worth 0 |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes +10
510 Bank deposits +88 5100 |
Bank loans +98 4998 Net worth 612 |
|
Monetary base = 612 (expansion =
12 over initial stage, no change over intermediate stage) Money supply = 5610 (expansion = 110 over initial stage,
98 over intermediate stage) Commercial banks' ratio of reserves to deposits =
102:5100 = 2% (equilibrium) Public's ratio of currency to deposits = 510:5100 = 10%
(equilibrium) |
|
As the balance sheets illustrate, efforts by commercial
banks to reattain their desired reserve ratio, and by the public to reattain
its desired ratio of currency to deposits, increase the money supply. Their
efforts cause changes in interest rates, prices, and incomes that move the
currency board system to a new equilibrium when a current-account surplus
occurs. The foregoing example in a sense collapses the effects of those
relationships into the reserve ratio of commercial banks and the public's ratio
of currency to deposits. The currency board responds passively by virtue of its
100 per cent reserve ratio, its fixed exchange rate with the reserve currency,
and its inability to change the reserve ratio of commercial banks by imposing
reserve requirements on them.
When a current-account deficit occurs in a currency board
system, the money supply process works as in Figure 3.6. As in the case of a
current-account surplus, the monetary system is initially in equilibrium, and
the balance sheets of the currency board, commercial banks, and the public are
initially as in Figure 3.7, which is the same as Figure 3.3.
Now let there be a current-account deficit of 12 units.
Foreigners only accept payment in foreign currency. By assumption, the currency
board has all the foreign currency in the domestic monetary system. Domestic
residents obtain reserve currency by converting 12 units of deposits at
commercial banks into currency board notes and coins, then converting the
currency board notes and coins into foreign currency. The assets and
liabilities of the currency board become 12 units less than in the initial
stage, reserves of commercial banks become 12 units less than in the initial
stage, and deposits of the public at commercial banks become 12 units less than
in the initial stage. Furthermore, the money supply is 12 units less than in
the initial stage. This is the intermediate stage (see Figure 3.8).
Notice that commercial banks have a reserve ratio of
approximately 1.76 percent in the intermediate stage, which is less than their
desired, initial ratio of 2 percent. Notice also that the public has a
deposit-to-cash ratio of approximately 10.02 percent in the intermediate stage,
which is more than its desired, initial ratio of 10 percent. Hence the monetary
system is in disequilibrium. Commercial banks therefore decrease their loans,
and the public decreases its holdings of cash, to restore the ratios of the
initial stage. In the final stage, they do so, achieving a new equilibrium,
with the money supply now 110 units less than it was in the initial stage (see
Figure 3.9).
|
Figure 3.6. Money supply
decrease in a currency board system 1. Current-account balance is zero (exports equal
imports)--equilibrium (Figure 3.7) 2. Domestic demand for imported goods increases or
foreign demand for currency board country's goods decreases 3. Current-account deficit (exports are less than
imports) 4. Reserves of commercial banks decrease 5. Loans by commercial banks decrease (Figure 3.8) 6. Interest rates increase 7. Income decreases 8. Demand for goods in general, including currency board
notes and coins, decreases 9. Prices of domestic goods decrease 10. Domestic demand for foreign goods decreases or
foreign demand for currency board country's goods increases 11. Current-account balance returns to zero--new
equilibrium (Figure 3.9) |
|
Figure 3.7. Money supply
decrease in a currency board system, initial stage Currency board |
|
|
Assets |
Liabilities |
|
Foreign reserves 600 |
Notes in circulation 600 Net worth 0 |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves)
100 Loans to public 4900 |
Deposits of public 5000 Net worth 0 |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes 500 Bank deposits 5000 |
Bank loans 4900 Net worth 600 |
|
Monetary base = 600 Money supply = 5500 Commercial banks' ratio of reserves to deposits =
100:5000 = 2% (equilibrium) Public's ratio of currency to deposits = 500:5000 = 10%
(equilibrium) |
|
|
Figure 3.8. Money supply
decrease in a currency board system, intermediate stage Currency board |
|
|
Assets |
Liabilities |
|
Foreign reserves -12 588 |
Notes in circulation -12
588 Net worth 0 |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves) -12
88 Loans to public 4900 |
Deposits of public -12
4988 Net worth 0 |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes 500 Bank deposits -12 4988 |
Bank loans 4900 Net worth -12 588 |
|
Monetary base = 588 (contraction
= 12 over initial stage) Money supply = 5488 (contraction = 12 over initial stage)
Commercial banks' ratio of reserves to deposits = 88:4988
= approximately 1.76% (disequilibrium) Public's ratio of currency to deposits = 500:4988 =
approximately 10.02% (disequilibrium) |
|
|
Figure 3.9. Money supply
decrease in a currency board system, final stage Currency board |
|
|
Assets |
Liabilities |
|
Foreign reserves 588 |
Notes in circulation 588 Net worth 0 |
|
Commercial banks |
|
|
Assets |
Liabilities |
|
Currency board notes (reserves) +10
98 Loans to public -98 4802 |
Deposits of public -88
4900 Net worth 0 |
|
Public |
|
|
Assets |
Liabilities |
|
Currency board notes -10
490 Bank deposits -88 4900 |
Bank loans -98 4802 Net worth 588 |
|
Monetary base = 588 (contraction
= 12 over initial stage, no change over intermediate stage) Money supply = 5390 (contraction = 110 over initial
stage, 98 over intermediate stage) Commercial banks' ratio of reserves to deposits = 98:4900
= 2% (equilibrium) Public's ratio of currency to deposits = 490:4900 = 10%
(equilibrium) |
|
As in the case of a current-account surplus, efforts by
banks to reattain their desired deposit-to-reserve ratio, and by the public to
reattain its desired ratio of currency to deposits, reduce the money supply and
move the currency board system to a new equilibrium when a current-account
deficit occurs.
We made some simplifying assumptions earlier. If we discard
them, the picture becomes too complex to analyze easily. However, the many
additional factors that can complicate the analysis should not obscure the main
point: market forces determine and limit expansion of the money supply in the
currency board system. As long as it is more profitable to invest funds in the
currency board country than elsewhere (after making allowances for differences
in rates of inflation, exchange risk, political risk, and transaction fees),
commercial banks in the currency board system tend to increase their loans. They
can do so because foreign investment tends to occur, bringing additional
foreign reserves to the currency board system. Eventually commercial banks
expand their loans in the currency board system to such an extent that making
further loans there is less profitable than investing the funds abroad. At that
point, commercial banks hold the supply of loans constant in the currency board
system, and the money supply ceases to increase.
Market forces rather than discretionary action by the
currency board cause the money supply to adjust to the current-account balance.
The monetary system is self-adjusting. The currency board is passive in
response to changes in demand for notes and coins; it merely supplies whatever
quantity is demanded at the fixed exchange rate with the reserve currency.
Because the exchange rate is fixed, arbitrage occurs through changes in the
money supply, interest rates, and the current-account balance, rather than
through the exchange rate. In that respect, the currency board system is like the
"classical" gold standard or gold-exchange standard practiced by many
central banks before the First World War, which had truly fixed exchange rates.
Arbitrage is the key to changes in the money supply of a
currency board system. It works by making prices in the reserve country an
"anchor" for nominal prices in the currency board country. (17)
The currency board maintains a fixed exchange rate with the reserve currency,
but it controls no other nominal or real price in the economy. Instead,
arbitrage determines those other prices. Arbitrage also occurs with a floating
exchange rate, but exchange risk creates additional costs that tend to make
arbitrage less efficient than it is with fixed exchange rates. (18)
Suppose the U.S. dollar is the reserve currency; then
An exception to the foregoing remarks about arbitrage is
that consumer price indexes can diverge between the currency board country and
the reserve country. Divergence can occur because consumer price indexes
contain many nontraded goods, such as rent and local services, whereas
wholesale price indexes mainly contain traded goods, such as foodstuffs,
minerals, and manufactures. Ultimately, prices of domestically produced traded
goods reflect the cost of both the nontraded goods (particularly rent and
wages) and the traded goods used to produce them. In a currency board system,
prices of nontraded goods can persistently increase faster than prices of
traded goods, although indirect arbitrage through traded goods tends to limit
increases in the prices of nontraded goods to the extent justified by increases
in productivity. For example, suppose that the productivity of labor in export
industries increases 4 per cent a year in the currency board country and zero
percent a year in the
The experience of currency board countries confirms both
the effectiveness of arbitrage with the reserve country for prices of traded
goods and the possibility of divergence in the rates of inflation for prices of
wages, rents, and other nontraded goods. In Hong Kong, interest rates and the
prices of exported goods have closely followed their counterparts in the
For the sake of clarity, the foregoing exposition of how the
money supply adjusts to the balance of payments in a currency board system made
some simplifying assumptions. Real conditions are never so simple. It is in
fact common in a currency board system for the money supply to change in the
opposite direction from the current-account balance. One factor that can loosen
or break the link between the money supply and the current-account balance is
foreign investment, which is part of the capital-account balance. Foreign
investment can offset or exceed current-account deficits, resulting in a gain
of reserves. Hong Kong had current-account deficits for decades at a time, yet
its money supply increased because
Money supply in a central banking system
A typical currency board cannot administer a discretionary
monetary policy because a fixed exchange rate with the reserve currency and the
requirement that the currency board hold 100 per cent foreign reserves allow it
no discretion. The currency board does not alter the exchange rate, nor does it
actively control the monetary base or regulate commercial banks. Its influence
on real economic activity is passive: it provides a sound currency for economic
agents to use as they wish. The stated purposes of a typical central bank, in
contrast, are to stabilize the price level and influence real economic activity
by controlling such instruments of monetary policy as the monetary base,
reserve requirements, and interest rates charged by commercial banks. Unlike a
typical currency board, a typical central bank can administer a discretionary
monetary policy, and unlike commercial banks, its decisions are not necessarily
guided by considerations of economic profit and loss. A typical central bank is
bound neither by strict rules concerning its behavior nor by the discipline of
profit and loss.
To reiterate, a central bank is a monetary authority that
has discretionary monopoly control of the supply of the reserves of commercial
banks, and usually also a monopoly of the supply of notes and coins. A typical
central bank performs other functions besides supplying reserves of commercial
banks plus its notes and coins in circulation (which are the components of the
monetary base in a typical central banking system). A typical central bank
regulates commercial banks, acts as their lender of last resort, gives economic
advice to the government, and perhaps helps operate the payments system. Those
functions are secondary to its role in supplying the monetary base, though.
Usually, only the central bank controls the monetary base, whereas other
government bodies can and often do perform the secondary functions. For
example, in the United States, only the Federal Reserve System supplies Fed
funds (deposit reserves of commercial banks) and notes, but it shares
regulatory powers with the Treasury Department, powers as a lender of last
resort with government deposit insurance agencies, (19)duties
as an economic advisor with several other government bodies, and the operation
of the payments system with private organizations. The discussion that follows
omits consideration of the secondary functions of a typical central bank and
concentrates on how a typical central bank supplies the monetary base, and how
its actions affect economic activity.
In a typical currency board system, the starting point in
the sequence of events in the example of an increase in the money supply
(Figure 3.2) was a decrease in the demand for imported goods in the currency
board country or an increase in foreign demand for the currency board country's
goods. Changes in demand for goods originate in the market, as a result of
changes in people's desires. In a typical central banking system, in contrast,
the starting point can be a decision by the central bank to increase the
monetary base; perhaps it does so to finance deficit spending by the
government. That is not a decision that originates in the market.
Diagrammatically, the simplified sequence of events for an
unexpected increase in the monetary base in a typical central banking system is
as in Figure 3.10. We consider only the case of an unexpected increase to avoid
complications of the type emphasized by the "rational expectations"
school of economists. The discussion of the money supply process here assumes
that the central bank maintains a floating exchange rate. A floating exchange
rate allows the greatest discretionary control of the monetary base, and hence
differs most from the fixed exchange rate maintained by a typical currency
board. Since a pegged exchange rate maintained by a typical central bank tends
not to persist, periodic devaluations of a pegged rate offer almost as much
long-run discretionary power to increase the nominal monetary base as would
exist if the central bank operated a floating exchange rate.
To show more clearly the contrast with a typical currency
board system, assume that there are no lags: nominal prices adjust very
quickly, leaving real prices unchanged. The only effect of the central bank's
action is that the domestic currency depreciates against foreign currency. If
instead, more realistically, some nominal prices are "sticky," the
central bank's action has real effects on the economy. In the sequence in
Figure 3.10, the likely effect of the central bank's action is that the real
exchange rate will depreciate, that is, prices of nontraded goods will decrease
compared to the prices of traded goods. That will cause a temporary increase in
exports, because nominal wages and prices will take some time to increase in
response to the depreciation of the exchange rate. In the meantime, real wages
and prices will be lower than before. (If the nominal exchange rate were pegged
rather than floating, the likely immediate effect of the central bank's action
would be the opposite: the real exchange rate would temporarily appreciate, causing
a decrease in exports. Eventually, though, a typical central bank would
probably devalue, and the monetary system would then attain a new, though
probably transitory, equilibrium.)
The sequence of events for an unexpected decrease in the
monetary base in a typical central banking system with a floating exchange rate
is as in Figure 3.11. Again, the figure omits consideration of expectations and
lags, and assumes that nominal prices adjust very quickly, leaving real prices
unchanged. The only effect of the central bank's decision is that the domestic
currency appreciates against foreign currency. If some nominal prices are
"sticky," the likely effect of the central bank's actions is that the
real exchange rate will appreciate, that is, prices of nontraded goods will
increase compared to prices of traded goods. That will cause a temporary
decrease in exports, because nominal wages and prices will take some time to
decrease in response to the depreciation of the exchange rate. In the meantime,
real wages and prices will be higher than before. (If the nominal exchange rate
were pegged rather than floating, the likely immediate effect of the central
banks' action would be the opposite: the real exchange rate would temporarily
depreciate, causing an increase in exports. Eventually, though, the real
exchange rate would appreciate and the monetary system would then attain a new,
though probably transitory, equilibrium.)
|
Figure 3.10. Money supply
increase in a central banking system with a floating exchange rate 1. Equilibrium--say, 1 unit of domestic currency = US$1 2. Unexpected decision by central bank to increase
monetary base (say, by lending to government) 3. Reserves of commercial banks increase 4. Loans by commercial banks increase 5. Exchange rate of domestic currency depreciates to 1.05
units = US$1--new equilibrium |
|
Figure 3.11. Money supply
decrease in a central banking system with a floating exchange rate 1. Equilibrium--say, 1 unit of domestic currency = US$1 2. Unexpected decision by central bank to decrease
monetary base (say, by selling assets) 3. Reserves of commercial banks decrease 4. Loans by commercial banks decrease 5. Exchange rate of domestic currency appreciates to 0.95
units = US$1--new equilibrium |
Central banks sometimes
mistaken for currency boards
A few central banking systems are sometimes mistaken for
currency board systems. Among them are the monetary systems of
After
The CFA franc, the common currency of 14 African countries,
has a pegged exchange rate with the French franc. The central banks that issue
the CFA franc are only required to hold a minimum of 20 percent foreign
reserves, not 100 percent like currency boards. This feature caused them
trouble in 1993 and 1994, when the central banks first suspended convertibility
into the French franc and then devalued the CFA franc from 50 per French franc
to 100 per French franc.
Other central banking systems sometimes mistaken for
currency board systems hold 100 per cent reserves only against some
liabilities. A central bank required to hold 100 per cent foreign reserves
against notes and coins in circulation, but not against deposits, is not a
currency board. To illustrate the difference, compare the balance sheets in
Figure 3.12, which use hypothetical numbers. For simplicity, assume that net
worth is zero. Assume for the moment that the currency board, like a minority
of past currency boards, accepts deposits from commercial banks. Assuming that
the currency board accepts deposits makes comparison of the balance sheets
simpler. The currency board holds 100 per cent foreign reserves against its
deposits and against its notes and coins in circulation. For a given amount of
total liabilities, the foreign reserves of the currency board are the same
whether the liabilities are held as notes and coins or as deposits. The
currency board holds 100 per cent foreign reserves against all liabilities, no
matter what form the liabilities take. The reasons that most currency boards
have issued notes and coins only, and have not accepted deposits, originate
from historical peculiarities, not from inherent differences in notes and coins
versus deposits as liabilities.
|
Figure 3.12. Central bank and
currency board balance sheets Central bank |
|
|
Assets |
Liabilities |
|
Foreign securities 150 Domestic securities 600 TOTAL 750 |
Notes and coins 150 Deposits of commercial banks 600 Net worth 0 TOTAL 750 |
|
Central bank |
|
|
Assets |
Liabilities |
|
Foreign securities 750 Domestic securities 0 TOTAL 750 |
Notes and coins 150 Deposits of commercial banks 600 Net worth 0 TOTAL 750 |
The foreign securities (foreign reserves) of the central
bank in Figure 3.12 are equal to its notes and coins in circulation, and its
total securities (foreign and domestic) are equal to its notes and coins in
circulation plus the deposits of commercial banks with it. That does not make
the central bank a currency board, however. The liabilities of the central
bank, like those of a currency board that accepts deposits, include deposits of
commercial banks as well as notes and coins in circulation. The different types
of liabilities are interchangeable: for instance, a one-unit deposit at the
central bank or the currency board can be converted into a one-unit note issued
by the central bank or the currency board, respectively.
For the purpose of the example, what matters is not the
ratio of foreign securities to notes and coins, but the ratio of foreign
securities to all liabilities. To illustrate the point, suppose that all
deposits at the central bank and the currency board are converted into notes
and coins. The total liabilities of the central bank and the currency board
remain unchanged at 750 units. The central bank will have 150 units of foreign
securities and 750 units of notes and coins in circulation, making a reserve
ratio of 20 per cent. The currency board, in contrast, will have 750 units of
foreign securities and 750 units of notes and coins in circulation, making a
reserve ratio of 100 per cent. To become a currency board, the central bank of
Figure 3.12 would have to increase its foreign securities to 750 units and sell
its domestic securities, or maintain its foreign securities at 150 units and
cease to accept deposits from commercial banks.
In a currency board system, only the liabilities of the
currency board must be backed 100 per cent by foreign reserves. Few currency
board countries have imposed reserve requirements on commercial banks.
Commercial banks in currency board systems have often held reserves as low as a
few per cent of deposits; contrary to a misconception (Congdon 1985: 95), it is
not necessary for them to hold 100 per cent foreign reserves. Nor is the
currency board system like the Chicago Plan proposed by economists at the
We stress the importance of formal institutional protection
for countries that wish to establish currency boards today. Many past currency
boards, especially in British colonies, have lacked formal legal protection
from changes in their operating rules. They have had much informal protection,
though. Most have been managed by British or British-trained civil servants who
absorbed a long tradition of financial rectitude. Most British colonial
currency boards have had fixed exchange rates with the pound sterling, and the
British government would have fired colonial officials who tried to devalue
colonial currency against sterling. For independent countries, the lack of
informal protection makes it all the more important to establish formal legal
protection such as chapter 5 and the Appendix suggest. A central bank that
mimics currency board-type rules but can stop doing so at any time is unlikely
to obey the rules for long, hence it will tend to have low credibility.
A brief history and assessment of currency boards
It is now time to summarize briefly the historical record
of currency boards and indicate the extent to which it supports the case for
currency boards. The summary is based on a synthesis of many studies. (20)
More than 70 countries have had currency boards (for a
list, see Hanke, Jonung, and Schuler 1993: Appendix C). The first currency
board was established in 1849 in the British Indian Ocean colony of
Currency board systems have typically been successful in
encouraging foreign investment. With currency boards, many countries have taken
the decisive step from primitive monetary conditions to modern monetary systems
that include sophisticated banking and foreign-exchange services. Inflation in
currency board systems has typically been low, which has encouraged the use of
modern currency in transactions. Economic growth has typically been
satisfactory, and in some cases spectacular. Trade in export goods that have
remained characteristic of certain countries originated during the years of the
currency board system. Export of cocoa and peanuts in West Africa, rubber and
tin in
Currency board systems have typically been stable. All
currency boards have successfully maintained fixed exchange rates and full
convertibility into their reserve currencies, although in the 1970s some
currency boards changed from sterling to more stable reserve currencies. (21)
Even during the Great Depression, all currency boards then existing maintained
fixed exchange rates and full convertibility, unlike almost all central banks
then existing. The oldest remaining currency board, in the Falkland Islands,
has maintained a fixed exchange rate of
Bank failures have been minor in orthodox currency board
systems. Fixed exchange rates with a reserve currency have encouraged foreign
commercial banks, especially those based in the reserve country, to establish
branches. Their multinational branch networks have enabled them to diversify
risk. Domestic banks have had to be strong to survive competition from the
foreign banks. Only some small commercial banks have failed in orthodox
currency board systems, and the losses they have inflicted on depositors have
been tiny. The recent currency board-like systems, which inherited banking
systems made fragile from bad monetary policy under central banking, have
suffered significant bank failures; however, the failures and their effects
have been no worse than those in neighboring central banking systems
The full convertibility inherent in the currency board
system has resulted in capital flight from some currency board systems during
periods of economic or political uncertainty. However, capital flight in
currency board systems seems to have been small compared capital flight in
central banking systems where the threat of foreign-exchange controls has
existed. Currency board systems have experienced severe shocks to their real
exchange rates, but so have central banking systems. Central banking appears on
average to have been no more successful than the currency board system in
alleviating shocks to real exchange rates or other real prices in the economy,
although no systematic study of evidence on the subject exists. An important
element encouraging economic growth in currency board systems has been that
full convertibility has encouraged foreign investors to take advantage of
opportunities for profit during economic downturns. Foreign investment has
helped alleviate shocks to the real exchange rate and other real prices in
currency board systems.
Despite the economic success of currency board systems,
national governments converted most currency boards into central banks in the
late 1950s and the 1960s. Some governments were influenced by theoretical
arguments that a central bank could promote economic growth better than a
currency board. The arguments seemed compelling at the time, but now appear
wrong. More important than theoretical arguments against the currency board
system were political factors. Newly independent countries established central
banks because of the association of the currency board system with colonial rule,
and because older, more established countries had central banks. A central bank
was a symbol of independence, like a national flag. Politicians in some newly
independent countries may also have understood the political advantages of
politicized central banks.
The results of central banking in former currency board
systems have been lackluster. On average, inflation has been higher and
economic growth has been lower in central banking systems than in the currency
board systems they replaced, lower than in their former reserve countries, and
lower than in the remaining currency board systems (Schuler 1992b: 202-3). Most
central banks that have replaced currency boards have restricted the
convertibility of their currencies.
Overall, then, the historical record of currency boards and
currency board systems has been good, as measured according to several of the
most important criteria that economists use. The characteristics we have
described as typical of a currency board really have been typical. The actual performance
of currency boards has been close to the ideal they have been established to
strive for, namely, to maintain full convertibility into the reserve currency
at a fixed exchange rate according to strict rules of procedure. Historical
experience supports the claim that currency boards, if established according to
similarly strict rules in the manner described in the next two chapters, have a
high probability of success.
Currency boards and currency board-like systems today
Orthodox currency boards exist today (2000) in the Cayman
Islands, the Falkland Islands, the Faroe Islands, and
We have described the differences between a historically
typical, orthodox currency board and a historically typical central bank. In
recent years a number of countries have reformed their central banks or have
established new monetary authorities, giving them some but not all
characteristics of orthodox currency boards. Countries that have established
such currency board-like systems in the 1990s are
Like orthodox currency boards, the currency board-like
systems have a constant exchange rate with the anchor currency.
The currency board-like systems differ from orthodox
currency boards with respect to their reserve ratios and their power to act as
lenders of last resort. None of the currency board-like systems have a maximum
reserve ratio. For an orthodox currency board, if it is allowed to accumulate
foreign reserves exceeding 100 percent of the monetary base, the amount of the
surplus has a definite upper limit, which historically has usually been 10
percent. The purpose of the surplus reserves is to guarantee that reserves are
always at least 100 percent by providing a cushion against losses in the
securities the currency board invests in. An orthodox currency board is not
allowed to use the surplus in a discretionary manner, and all profits beyond
those necessary to maintain the small surplus go to the government. Most of the
currency board-like systems, in contrast, are allowed to accumulate surplus
reserves indefinitely from their profits (though in practice there is political
pressure to contribute some reserves to the general government budget). The currency
board-like systems are also allowed to use their surplus reserves in a
discretionary manner to act as lenders of last resort to commercial banks. In
some cases they can also use their main reserves in the same manner: the Banco
Central de la República Argentina has a minimum reserve ratio of only 66
percent, though in practice it holds approximately 100 percent foreign
reserves, while the Bulgarian National Bank is explicitly allowed by law to act
as a lender of last resort in case of a crisis affecting the banking system as
a whole.
We have been pleasantly surprised by the good performance
of the currency board-like monetary authorities, which have acted more like
currency boards than like central banks, and have provided sound currencies. In
all cases, inflation has fallen dramatically and economic growth has been
positive--quite a contrast with the experience of those countries under central
banking. We see a potential problem with the currency board-like systems,
though, because they are a compromise between the discipline of orthodox
currency boards and the desire of governments to retain some discretion in
monetary policy, particularly the ability to have a lender of last resort.
Discipline and discretion are incompatible elements, and often when they clash
under central banking it is discipline that is sacrificed. Concern that
currency board-like monetary authorities might devalue has led to speculative
attacks against the currency in
|
Figure 4.1. Currency boards
and currency board-like systems today |
||||
|
Country |
Population |
GDP (US$) |
System began |
Exchange rate / remarks |
|
|
37 million |
$374 billion |
1991 |
1 peso = US$1 / Currency
board-like |
|
|
62,000 |
$1.9 billion |
1915 |
|
|
|
320,000 |
$5.4 billion |
1952 |
|
|
|
3.5 million |
$5.8 billion |
1997 |
1 convertible mark = DM1 /
Currency board-like |
|
|
8.2 million |
$34 billion |
1997 |
1 lev = DM1 / Currency
board-like |
|
|
39,000 |
$930 million |
1972 |
Cayman $1 = US$1.20 |
|
|
450,000 |
$530 million |
1949 |
177.72 |
|
|
1.4 million |
$7.8 billion |
1992 |
8 kroons = DM1 / Currency
board-like |
|
|
2,800 |
unavailable |
1899 |
Falklands £1 = |
|
|
41,000 |
$700 million |
1940 |
1 Faroese krone = 1 Danish krone
|
|
|
29,000 |
$500 million |
1927 |
Gibraltar £1 = |
|
|
6.8 million |
$168 billion |
1983 |
|
|
|
3.6 million |
$18 billion |
1994 |
4 litai = US$1* / Currency
board-like |
|
Note: *A few times since 1997, Sources: CIA (1999), Schuler (1992b: Appendix). |
||||
4. HOW TO
ESTABLISH A CURRENCY BOARD
A currency board can be established by converting the
central bank into a currency board or by making the currency board the issuer
of a parallel currency alongside the central bank. Both approaches have been
used by past currency boards. This chapter explains both approaches step by
step.
How to convert a central bank into a currency board
If a government wishes to convert a central bank into a
currency board, how should the government proceed?
Experience with currency boards in places as diverse as
Palestine, Danzig, and the Philippines indicates that administratively,
converting a central bank into a currency board system is simple. The steps for
conversion are as follows.
1. Delegate to other bodies all functions of the central
bank other than supplying the monetary base. For instance, the ministry of
finance can regulate commercial banks and give advice to the president on
monetary affairs. Commercial banks can operate the payments system and provide
mutual deposit insurance protection.
2. Allow a brief period of clean, unrestricted floating
exchange rates for the domestic currency. The exchange rate between the reserve
currency and the domestic currency must be appropriate. An overvalued real
exchange will price exports out of world markets, while an undervalued real
exchange rate will make imports expensive, preventing people from buying
foreign machinery and other goods needed for modernizing national economies.
The best indication of the appropriate real exchange rate is the unrestricted
market rate, which reflects supply and demand. Accordingly, the first step in
fixing an exchange rate for the domestic currency is to allow the exchange rate
to float for a brief period.
When the government allows the domestic currency to float
without restrictions, it should announce its choice of reserve currency and the
date it will fix the exchange rate. The reason for making the announcement is
to reduce uncertainty that otherwise may tend to undervalue the domestic
currency as a store of value. A period not exceeding 90 days should be
sufficient to indicate approximately the appropriate exchange rate. The float
should be a clean one reflecting market forces only, rather than a dirty float
reflecting intervention by the central bank. To promote a clean float, all
existing foreign-exchange regulations should be abolished.
3. Make the actions of the central bank transparent and
predictable. During the period of unrestricted floating, the actions of the
central bank should be transparent and predictable, so that they cause no
destabilizing random surprises to exchange rates. The central bank should be
required to disseminate weekly or even daily reports of its activities and its
balance sheet. Another way of making the actions of the central bank more
transparent and predictable is to require that it hold significant reserves
against further increases in the monetary base. For instance, it can be
required to increase its foreign reserves by 100 per cent of any increase in
the monetary base that occurs after the government announces the monetary
reform.
4. Convert some required reserves of commercial banks
(deposits at the central bank) into currency board notes and coins or into
foreign securities, whichever the commercial banks prefer. Dispose of remaining
reserves. With this step, the deposit liabilities of the central bank will
cease to exist. If the central bank imposes reserve requirements on commercial
banks, not all deposits of commercial banks at the central bank will
necessarily be converted into currency board notes and coins or into foreign
securities. "Excess" reserves can be converted into government bonds
to avoid creating inflation. A later section in this chapter discusses what
quantity of reserves commercial banks may need and what proportion of reserves
is excess.
5. Establish a fixed exchange rate with the reserve
currency. After the deposit liabilities of the central bank cease to exist, all
that will remain are its issues of notes and coins in circulation and its net
worth, on the liability side of the balance sheet, and its foreign reserves and
miscellaneous holdings such as its offices, on the asset side. Its other assets
and liabilities will have been distributed to commercial banks or the
government, or will have been canceled. To convert what remains of the central
bank into a currency board, the government must now establish a fixed exchange
rate with the reserve currency it chose in step 2. Simultaneously, the
government must ensure that reserve assets for currency board notes and coins
in circulation equal 100 per cent.
When the date to establish the fixed exchange rate arrives,
the government should set the rate somewhere within the range of recent market
rates. Setting exchange rates is an art rather than a science; if uncertainty
exists about an appropriate exchange rate, it is best to err on the side of an
apparent slight undervaluation rather than a slight overvaluation compared to
the range of recent market rates, so as to be certain that the exchange rate
does not price exports out of world markets. It is better to start with an
exchange rate that results in competitively priced exports than with an
exchange rate that results in overpriced exports. As Hong Kong's experience in
returning to the currency board system in 1983 illustrates, a range of freedom
exists in setting an exchange rate (see
An alternative way of setting the exchange rate is to set
it immediately, without a period of floating, at an estimate of the
market-clearing rate or at a deliberately undervalued rate. That method seems
disadvantageous because it makes no attempt to gather information from the
market.
Some economists distrust the ability of a float, even a
clean float, to indicate approximately the appropriate exchange rate. They
would prefer to use statistical constructs for setting the exchange rate
(Williamson 1992: 43-4). But there is no reason to think that using statistical
constructs to set the exchange rate would be any more successful than using
statistical constructs to centrally plan other economic activities. Such
measurements often have no direct connection to supply and demand.
6. Ensure that foreign reserves equal 100 per cent of
domestic-currency notes and coins in circulation. The currency board should
begin with foreign reserves equal to 100 per cent of its notes and coins in
circulation. (If the currency board accepts deposits, foreign reserves should
equal 100 per cent of notes and coins in circulation plus deposits.) Allowing
the currency board to operate with a lower reserve ratio might create
possibilities for the board to administer a discretionary monetary policy.
7. Transfer the remaining assets and liabilities of the
central bank to the currency board and open the currency board for business. At
the moment that the government fixes the exchange rate with the reserve
currency, the currency board will replace the central bank as the issuer of
domestic notes and coins and will assume all remaining assets and liabilities
of the central bank. The central bank will cease to exist.
We envision all the steps being completed within 120 days
of the decision to establish a currency board. They could take as little as 30
days. Historical experience has been that monetary reform is usually most
successful at encouraging economic recovery when enacted quickly (see Yeager
and others: 42-3).
The alternative: a parallel currency approach
Attempts to convert a central bank into a currency board
may encounter political opposition. The central bank and state enterprises
dependent on it for subsidized loans often have considerable political power
and may oppose monetary reform. Converting the central bank into a currency
board may suddenly deprive state enterprises and the government of access to
inflationary finance, causing them hardship as they adjust to hard budget
constraints. Is there a way to bypass the potential political obstacles posed
by entrenched political interests and to ease the transition from high
inflation to low inflation?
Establishing a currency board as an issuer of a parallel
currency may do so, although it is not a tactic that has often been used. To
reiterate, a parallel currency is one that circulates extensively alongside
another currency at exchange rates determined by market forces. The parallel
currency can have a fixed, pegged, or floating exchange rate with the other
currency, and can circulate legally or illegally.
Parallel currencies have been common historically. Many
countries have had parallel circulation of gold and silver or silver and copper
coins (bimetallism). Some have also had parallel circulation of, say, gold
notes and deposits with silver notes and deposits. Today, parallel currencies
are common in many border regions; for instance, the U.S. dollar and the
Canadian dollar both circulate along the U.S.-Canadian border, although the
U.S. dollar predominates in the
Parallel currencies are also common in countries that have
had extreme inflation and inconvertible currencies, such as
Parallel currencies have a long history, then, and are part
of the experience of residents of many countries today. What may be new to
readers is the explanation here of how a currency board issuing a parallel
currency can reform a monetary system. In the system of parallel currencies
proposed here, the currency board currency will circulate alongside the central
bank currency as an alternative, officially approved domestic currency. There
will therefore be two competing domestic currencies. Competition from the
currency board will force the central bank either to provide a currency of
comparable high quality to the currency board currency, or to wither away as
people cease to use the central bank currency. (25)
Many economists oppose proposals for legalizing a parallel
currency (Cooper 1991a: 131-2, 1991b: 312-14; Nuti 1991: 54-5; Williamson 1991:
402-4). Their criticisms do not apply to a currency board as a parallel issuer,
though. The main criticism of legalizing a parallel currency is that it would
accelerate inflation in the central bank currency. The central bank currency
continues to be used in many countries mainly because it has the support of the
domestic government, in the form of forced tender laws, which make its use
compulsory. Granting equal legal status to the currency board currency would
supposedly reduce the incentive for holding the central bank currency.
Accordingly, some people would cease holding central bank currency and instead
would hold currency board currency. As demand for the central bank currency
diminished, inflation in the central bank currency would increase (the velocity
of the central bank currency would increase) unless the central bank reduced
the monetary base of the central bank currency. For political reasons, reducing
the monetary base would be difficult. Inflation in the central bank currency
would therefore accelerate, causing a self-reinforcing cycle of further
reductions in demand for the currency board currency. Ultimately the cycle
would end in hyperinflation in the central bank currency, which would deprive
the government of seigniorage from the central bank currency and cause economic
decline.
The criticism is flawed. It assumes that people in
countries with unsound currencies currently use the central bank currency as
their only currency. In reality, those countries typically already have one or
more parallel currencies, such as the U.S. dollar, and many parallel stores of
value besides. Many residents already hold foreign notes at home and some have
foreign-currency deposits abroad. In some countries, people also hold durable
goods such as bricks and soap. Therefore, the choice in such countries is not
between a situation with a single currency and a situation with a parallel
currency, but between two situations with parallel currencies. The only
question that remains to be decided is whether a parallel currency will circulate
illegally alongside the central bank currency, as the U.S. dollar and other
foreign currencies now do, or legally, as the currency board currency will in
this proposal.
It is not certain that the existence of a parallel currency
board currency will reduce demand for the central bank currency, even if the
central bank continues to increase the monetary base of the central bank
currency as rapidly as before. In many cases, legalizing a parallel currency
has had the consequences that critics fear, but that need not always be the
case. By encouraging remonetization of the economy by means of domestic
currency, a sound parallel currency issued by a currency board may actually
increase demand for the central bank currency (Auerbach and others 1992: 19-23;
Rostowski 1992: 95-6; Rostowski and Shapiro 1992: 17, 29). (Ultimately, though,
demand for the central bank currency will diminish toward zero if the central
bank does not use the opportunity provided by an increase in demand to
stabilize the value of its currency.)
Suppose, however, that the existence of the currency board
currency reduces demand for the central bank currency, as critics of a parallel
currency argue it will. The good domestic currency will tend to displace the
bad one in circulation. Inflation in the central bank currency will increase
and seigniorage from the central bank currency will cease. But the currency
board currency will enable the domestic government to recapture some of that
seigniorage as well as seigniorage now lost to foreign central banks. More
important, the currency board currency will greatly reduce the economic
inefficiency caused by an inconvertible currency suffering high inflation. By
using the currency board currency, people will be able to avoid the
disturbances to economic activity that the central bank currency causes because
it is an unsatisfactory medium of exchange, store of value, and unit of account
(Auerbach, Davison, and Rostowski 1992: 11-18; Rostowski 1992: 94-101).
Unlike critics of legalizing a parallel currency, we are
little concerned with the fate of the central bank currency in a parallel
currency approach. If the central bank withers away as an issuer of currency,
fine; it will cease to have an effect on the economy, and the economy will
benefit. Its remaining functions can be assigned to other government bodies,
privatized, or abolished. If competition from the currency board induces the
central bank to change its behavior and the central bank currency continues to
exist as a less inflationary currency than it is now, that is also fine. In
either case, at least one sound domestic currency will exist.
Competition between the central bank currency and the
currency board currency will be greatest if both are allowed by law to serve
identical functions. If the currency board currency is not at first allowed
equal legal tender status with the central bank currency, or if currency board
notes and coins are limited to large denominations, the competition will be
unequal and demand for the central bank currency will be more than it otherwise
would be. To reap the full advantages of the currency board currency as a
parallel currency, the government should not limit competition with the central
bank currency.
How to establish the currency board as the issuer of a
parallel currency
The currency board can be established as the issuer of a
parallel currency according to the following steps, which should take no more
than 60 days. For example, the currency board that existed in North Russian
from 1918 to 1920 was established just eleven weeks after it was first
proposed, despite civil war and reliance on less rapid transportation and
communications than exist today (Hanke and Schuler 1991a).
1. Obtain initial foreign reserves for the currency board.
We describe later how to calculate the quantity of initial foreign reserves
necessary for the currency board and how to obtain the reserves.
2. Make the currency board currency legal tender for
payment of taxes and private debts.The currency board currency should be made a
legally permissible currency in which to pay taxes and private debts. However,
that the currency board currency should not be forced tender for private debts,
that is, people should be allowed to make contracts and payments in the central
bank currency or other currencies if both parties to a contract or payment
wish.
3. Issue currency board currency equal to the initial
foreign reserves. The currency board will have 100 per cent foreign reserves
from the beginning. It will issue no more currency board currency than the
value of its foreign reserves.
4. Put the currency board currency into circulation, for
example, by a distribution to every inhabitant of the country according to a
well-defined plan. The nominal exchange rate between the reserve currency and
the currency board currency can be anything, so long as the real amount of
currency board currency does not exceed the foreign reserves of the currency
board. Suppose that the reserve currency is the U.S. dollar and that the
currency board has $100 million of foreign reserves. The currency board will
issue 100 million units if the exchange rate is one currency board unit per
dollar, or 500 million units if the exchange rate is five currency board
currency per dollar. In either case, the amount of currency board currency
issued equals $100 million. An exchange rate of one to one seems best, because
it will make conversions easiest to calculate.
The currency board will now inaugurate the parallel
currency system by distributing currency board currency representing up to $100
million. The actual distribution can be designed in various ways. An easy
method is to give every resident an equal amount of currency board notes and
coins as a one-time gift. The currency can also be given on a per household
basis, or according to various scales: for example, a certain amount for the
first person in each household, a lesser amount for each additional adult, and
a still lesser amount for each child.
To prevent fraud, the government can take precautions
similar to those used to prevent voting fraud. Residents who receive their
distribution of currency board currency can have their identity documents
stamped or their fingers dipped in indelible ink. The distribution of currency
board currency should occur simultaneously throughout the country over a short
period.
5. Allow the currency board currency to circulate as a
parallel currency to the central bank currency, at an exchange rate determined
by market forces. After the previous steps have been taken, the currency board
currency will circulate alongside the central bank currency. Nobody will be
forced to use the currency board currency. Much of the domestic economy will
use the currency board currency as the medium of exchange, store of value, and
unit of account because the currency board currency will be more stable than
the central bank currency. It will be a matter for individuals to decide
whether they now wish to pay and accept currency board currency or central bank
currency. Commercial banks will also have to decide whether to allow depositors
to convert existing deposits in central bank currency into currency board
currency.
This proposal assumes that there will be an unrestricted
market in exchange, so that people can exchange any amount of central bank
currency for currency board currency or the reverse at the market rate. Hence
it will be no disadvantage, in terms of the function of either type of currency
as a medium of exchange, to be paid in one type of currency or the other.
The central bank will have to make its currency
sufficiently sound to withstand competition from the currency board currency or
it will wither away in importance as an issuer of currency. We suggest that
when the real value of central bank notes and coins diminishes to less than 10
per cent of all domestic currency in circulation, the central bank should be
abolished. By then the political forces supporting the central bank should be
weak.
If currency board currency is introduced by distributing it
to the public, the government will for a time continue to make payments in
central bank currency, because it will have no currency board currency. If the
government wishes to receive currency board currency in tax payments, it must
ensure that accounting rules do not favor central bank currency, for example by
not taking into account the decline in the real value of a given nominal amount
of central bank currency between the time taxes are assessed and paid. If
accounting rules favor the central bank currency, the government may receive no
currency board currency.
If the government uses both currencies in payment for a
while, and if the central bank currency continues to depreciate, the government
will need to devise rules about what combination of currencies it uses. A
simple rule is for small payments to consist only of central bank currency, and
large payments to consist only of currency board currency. The size of payments
considered "large" can decrease to zero as the government accumulates
more currency board currency. Persons receiving payments from the government
will experience little disadvantage from being paid in central bank currency
rather than currency board currency, if the government adjusts its payments to
the market exchange rate of the central bank currency against the currency
board currency. An unrestricted market will exist for exchanging the two types
of currency, so people will be able to exchange their central bank currency
immediately if they wish.
How to chose a reserve currency
What reserve currency should the currency board choose? The
most likely choices for most countries are the U.S. dollar, the euro, or the
Japanese yen, the three key currencies of international trade and finance.
The U.S. dollar is the most widely used currency in
international trade and finance. Raw materials, such as oil, natural gas,
timber, and minerals, are predominantly priced in dollars on world markets. The
dollar also has the advantage that it already is the most widely used
unofficial parallel currency and a popular unit of account in many developing
countries. Choosing the dollar would involve the least change of habits of any
reserve currency in such countries. Historically, the dollar has had low
inflation, high credibility, full convertibility, and low real interest rates.
It has a high likelihood of continuing its exceptionally good historical
performance.
The euro is another possible reserve currency. It is the
currency of Western Europe, the largest trading partner and largest potential
source of new foreign investment for countries in Eastern Europe, the former
Soviet Union, and
The Japanese yen is less widely used in international trade
or as a parallel currency than the dollar or the mark. Even so, Japanese trade
and investment from
A basket of foreign currencies does not seem advantageous
as the reserve currency for the vast majority of potential currency boards. A
basket is less transparent to the public than a single reserve currency, and
thus may not as quickly achieve high credibility for the currency board. A
basket also imposes greater costs on the currency board in terms of management
time and transaction fees. A basket does not eliminate exchange risk with any
single reserve currency; it sacrifices greater potential variability of
exchange rates with each component of the basket for lower variability with the
whole basket. If people are allowed to hold foreign-currency deposits, those
who desire the benefits of lower variability with a basket of currencies can
create their own baskets by holding a combination of currencies or by trading
currency futures and options, as people do in Hong Kong.
We have considered the implications for developing
countries of choosing various reserve currencies. It is also worthwhile to
consider the implications for the reserve country. If the reserve currency is
already widely used as a parallel currency, the demand for reserve-currency
notes will probably decrease because people will exchange reserve-currency
notes for currency board currency notes and coins. On the other hand,
commercial banks will demand more reserve currency, either directly or through
correspondent banks in the reserve country, as a means of settling payments.
The net effect of a currency board currency on demand for the monetary base of
the reserve currency is thus unclear in this case. Whatever the case, the
economies of most developing countries are small compared to the U.S., euro
zone, and Japanese economies, so the initial effect of a currency board
choosing the dollar, euro, or yen as the reserve currency will be
correspondingly small.
Linking the currency board currency to a foreign currency
does not subject a country to foreign political domination. Rather, linking the
currency to an appropriate foreign currency can restore an element of national
pride by giving a country a currency as sound as the reserve currency.
The next chapter suggests what the currency board should do
if the reserve currency becomes unstable.
How to calculate the initial foreign reserves
What quantity of initial foreign reserves will the currency
board need? The answer partly depends on how the board is established. If the
central bank is converted into a currency board, the entire domestic monetary
base will require 100 per cent foreign reserves as backing. If the currency
board is the issuer of a parallel currency, the initial foreign reserves can be
smaller.
Suppose first that the central bank is converted into a
currency board. The foreign reserves that the currency board system needs will
depend on the size of the monetary base and the exchange rate with the reserve
currency.
The monetary base typically comprises currency notes and
coins in circulation (whether held by commercial banks or the public) and
deposits of commercial banks at the central bank. Currency notes and coins in
circulation should be backed 100 per cent by foreign reserves in the currency
board system. As was explained earlier in this chapter, not all currency
reserves of commercial banks need be converted into currency board notes and
coins or foreign securities, although those that are should be backed 100 per
cent by foreign reserves. The usable reserves of commercial banks are those in
excess of minimum required reserves. Therefore, for example, if the reserve
requirement is 20 per cent, a commercial bank with 25 per cent reserves has
only usable reserves of only 5 per cent of the public's deposits with it.
In the currency board system, the government should impose
no reserve requirements. Commercial banks should determine their own reserve
ratios according to their judgments of what is prudent, as has been the case in
most past currency board systems. In modern banking systems with no reserve
requirements, commercial banks have usually held reserves of only a few per
cent of deposits. The primitive condition of banking technology in many
developing countries may at first make it necessary for domestic commercial
banks to maintain higher reserve ratios than banks in modern banking systems
with no legal reserve requirements. To provide domestic commercial banks with
reserves that are more than adequate, suppose that the banks need reserves of
10 per cent of deposits. (These are reserves that they will have after being
restructured, if necessary.) The actual ratio to be used will vary from case to
case.
Since according to this proposal no reserve requirement for
commercial will exist in the currency board system, all the reserves of
commercial banks remaining after monetary reform will be usable reserves.
Allowing all current reserves of commercial banks to become usable reserves in
the currency board system might risk causing renewed inflation, though, because
half or more of current reserves might be excess, not needed by the banks to
ensure convertibility of their deposits into currency board notes and coins.
The excess reserves would become the basis for an increase in bank loans and
hence an increase in prices of domestic goods similar to that indicated in
Figure 3.2. The economy would eventually achieve a new set of higher, market-clearing
prices, but in the meantime inflation would have undesirable effects.
If the central bank is converted into a currency board, the
quantity of initial foreign reserves necessary for the currency board system is
the sum of currency notes and coins in circulation plus 10 per cent of currency
deposits (the somewhat arbitrary but quite adequate figure discussed earlier),
divided by the exchange rate of currency per unit of reserve currency. Deciding
what exchange rate to use in the calculation is difficult in many cases because
no legal unrestricted market for currency exists. Little can be said in general
about the appropriate exchange rate. However, we stress that whatever rate is
used, the calculation will be merely an estimate. The actual amount of reserves
necessary will be determined by the exchange rate attained during the brief
period of floating before the exchange rate with the reserve currency is fixed.
If the currency board is established as the issuer of a
parallel currency, in contrast, no determinate quantity of initial foreign
reserves is necessary, because the currency board will not provide backing for
the monetary base of the central bank currency. All notes and coins of the
currency board in circulation must be backed 100 per cent by foreign reserves,
but the board can begin with whatever initial quantity of foreign reserves it
obtains. The initial quantity of foreign reserves should be sufficient to
indicate to the populace that the currency board is a substantial institution.
Notice that in the parallel currency approach, commercial banks are given none
of the initial foreign reserves. If they wish to offer deposits in currency
board currency, they need to acquire currency board currency notes and coins or
reserve-currency assets as reserves.
How to obtain the initial foreign reserves
Once the quantity of initial foreign reserves the currency
board needs has been calculated, how can the currency board obtain the
reserves?
In the conversion approach, the net amount of foreign
reserves the currency board will need will be the gross amount minus existing
foreign reserves. In the parallel currency approach, the net amount of initial
foreign reserves the currency board will need will equal the gross amount,
because the central bank will continue to exist and the currency board will
obtain no reserves from the central bank.
In either approach, if existing foreign reserves are small,
the currency board may need additional foreign reserves to provide backing.
Foreign reserves can be obtained from several sources. The government can lease
or sell state property for fully convertible foreign currency. Equivalently, if
the central bank is converted into a currency board, the government can lease
or sell state property for currency and not reissue the currency. (That would
be like the process described in Figure 3.13, which decreases the monetary base
and the overall supply of money.) A similar technique was used in
Another possible source of reserves is a loan from the
International Monetary Fund, which for example in 1997 lent some of the funds
necessary to bring the reserves of Bulgarian currency board-like system up to
an adequate level. To keep the currency board "pure," its
constitution should forbid it from accepting loans other than a loan for its
initial foreign reserves.
If the currency board borrows from the IMF, the currency
board will eventually have to repay the loan. Even if the currency board has no
other initial reserves than the loan from the IMF, the currency board should
have no difficulty repaying the loan within fifteen to twenty years. Because
the currency board will issue a sound currency, people will tend to exchange
the foreign notes that they now hold, and hold currency board notes and coins
instead. Because the currency board will help the economy to grow, the overall
demand for domestic currency will increase. Consequently, the note and coin
circulation of the currency board will increase from its initial level, its
foreign reserves will increase accordingly, and the seigniorage from the
foreign reserves will increase. Based on historical and current international
comparisons of the relationship between income and circulation of notes and
coins, (26)
the note and coin circulation of the currency board will probably equal at
least 3 per cent of GDP within three years after the currency board opens. If
the net seigniorage the currency board earns is 4 per cent of its note and coin
circulation, the currency board should be able to repay at least 0.12 per cent
of GDP a year to the IMF. These are conservative estimates that many countries
may exceed.
Once the currency board opens, the self-adjusting nature of
the money supply process in a currency board system will enable the money
supply to accommodate changes in demand for money. Experience indicates that the
currency board system has not hindered rapid increases in the money supply that
have been justified by economic growth. Foreign investment has enabled the
money supply in currency board systems to increase consistently despite decades
of current-account deficits.
It is worthwhile to consider a point previously mentioned
that applies only to the conversion approach, not to the parallel currency
approach. Suppose that the impending conversion of the central bank into a
currency board increased the credibility of the currency so much that during
the brief period of floating exchange rates (step 2 of the parallel currency
approach), the exchange rate of the currency appreciated compared to the
previous rate, and the foreign reserves necessary for the currency board system
therefore increased substantially.
In that case, the very success of the monetary reform would
enable the government to acquire the necessary additional reserves. The
appreciation in the exchange rate of the currency would reflect increased confidence
in the economy. Increased confidence will increase the capitalized value
(present value) of assets in the country. Increased confidence in the economy
should enable the government to obtain even a large quantity of initial foreign
reserves for the currency board by taxation or by borrowing.
In practice, lack of initial foreign reserves has not been
a problem for any currency board or currency board-like system. Where the
reserves have not been sufficient at the start, it has been possible to augment
them by borrowing them or securing lines of credit, and the rapid growth in the
monetary base that has occurred as a result of confidence in the currency has
generated the seigniorage to repay borrowed reserves.
5. HOW TO
OPERATE AND PROTECT A CURRENCY BOARD
How to operate a currency board
A typical currency board is simple to operate. Past
currency boards have usually had staffs of ten or fewer people. Past currency
boards have achieved economies by contracting clerical and investment functions
to outside parties; a new currency board can do likewise. The extreme
simplicity of a currency board is one of the advantages of the currency board
system. As has been mentioned, many central banks in developing countries lack
staff with the technical skill to administer monetary policy competently.
The main administrative details of operating a currency
board will be as follows. (27)Constitution:
The Appendix contains a model currency board law that distills features of past
currency board constitutions into a form that should enable the currency board
to operate efficiently.Exchange policy: The currency board will exchange its
notes and coins on demand at a fixed rate into or from the reserve currency at
its offices or agencies. Anybody who has reserve currency will be able to
exchange it for currency board notes and coins at the fixed rate, and anybody
who has currency board notes and coins will be able to exchange them for
reserve currency at the fixed rate. To hold a large supply of reserve-currency
notes would reduce the profits of the currency board, because the board would
be unable to invest those funds in interest-bearing securities. Hence, the
currency board should try to encourage a "wholesale" currency
exchange business with commercial banks and use electronic funds transfer
extensively for payment and acceptance of reserve-currency securities.
Clientele: Although the currency board should encourage a wholesale currency
exchange business with commercial banks, the public should also be allowed to
deal directly with the currency board. Some British colonial currency boards
dealt only with banks, as a way of reducing their need for staff (Greaves 1953:
13). It seems unnecessary and unjust to discriminate against the public so.
Most people will exchange currency through commercial banks in any case, as the
West African Currency Board discovered when it changed from dealing with
commercial banks only to dealing with the public also. Giving the public the
choice of dealing directly with the currency board will place a low limit on
the commission fees that commercial banks charge for exchanging currency board
currency for reserve currency. That will tend to tighten the link between the
currency board currency and the reserve currency, which will make arbitrage
between the currency board country and the reserve country more efficient.
Lower and upper limits to exchanges: To reduce their
handling costs, many currency boards have imposed minimum exchange amounts.
Small British colonial currency boards such as those of
To strengthen confidence, a currency board established
according to the proposal here should impose no minimum. The public will then
know that the currency board is always ready to convert any amount of currency
board currency into reserve currency. There will also, of course, be no upper
limit to the amount of reserve currency or of its own notes and coins in
circulation that the currency board accepts for exchange. No past currency
board except that of
Commission fees: Some currency boards have charged
commission fees of 1/8 per cent to 1 per cent per transaction; the North
Russian currency board, for instance, charged a fee of 1 per cent. Other boards
have charged lower commission fees for large transactions than for small ones.
We recommend that a currency board established according to the proposal here
charge no commission fees for exchanges. The social benefits of not charging
fees greatly exceed the pecuniary benefits to the currency board of charging
fees. Commission fees would loosen the link to the reserve currency, especially
for short-term capital movements, because they would impose high costs relative
to the benefits of arbitrage. A few currency boards, most notably the East
African Currency Board toward the end of its existence, have deliberately
manipulated their commission fees to influence capital movements (Kratz 1966:
246-7). But a currency board is intended to eliminate exchange risk with the
reserve currency, so it is pointless to erect barriers to exchange into and
from the reserve currency. Besides, commission fees would bring little income
to the currency board; it will in any case earn most of its income from
interest on its foreign assets.
Exchanges by the currency board should be exempt from
taxation, to prevent the government from attempting to tax the currency board
out of existence. The currency board should also be exempt from other legal
barriers that might hinder exchanges by it.
Offices: The currency board should have its main office in
the financial center of the country. It should have a few branch offices or
agencies in other large cities, if the country is large. The main office will
do most of the business. The main role of the branch offices or agencies will
be storing and distributing notes and coins. The currency board need not have
actual branches. Instead, one or more commercial banks can be the board's agent
outside the financial center, as the Bank of British West Africa was for the
West African Currency Board. The currency board should also have an office
abroad, in the reserve country or in a safe-haven financial center such as
Management: The currency board should have a small board of
directors to supervise the board's staff. Past currency boards have had three
to eight directors. The powers of the board of directors and of the staff will
be limited; unlike their counterparts in central banks, they will have no
discretionary control of the monetary base. To protect the board of directors from
political pressure to convert the currency board into a central bank, directors
should have staggered terms. Also, some directors could be foreigners,
appointed by foreign commercial banks, or perhaps by the International Monetary
Fund if the IMF lends some of the initial foreign reserves of the currency
board. The next section returns to this proposal.
Staff: The staff of the currency board will perform two
functions: exchanging currency board notes and coins for reserve currency, or
the reverse, and investing the assets of the currency board in low-risk
securities denominated in and payable in the reserve currency. The exchange
work will require only a small number of bank tellers. The investment work will
require some expert financial traders, but since the currency board will follow
rather routine, conservative investment practices, its investment expenses
should be smaller than those of commercial banks with portfolios of similar
size. The assets of the currency board should be held at suitable institutions
abroad, for example with one or more large foreign commercial banks or central
banks.
Past currency boards have had small staffs. The West
African Currency Board, which served
Reserve ratio: The currency board will begin with foreign
reserves equal to 100 per cent of its notes and coins in circulation. In
addition, like most past currency boards, the currency board should accumulate
a reserve fund to ensure that its foreign reserves are never less than 100 per
cent even if its assets lose part of their market value (for example, if
interest rates increase, reducing the principal of fixed-rate securities). Many
currency boards have accumulated a reserve fund of 10 per cent of notes and
coins in circulation (Clauson 1944: 9). They have usually paid all net
seigniorage into the reserve fund until the reserve fund is full. They have
paid all additional net seigniorage to their governments. These rules leave no
opportunity for discretionary monetary policy when the foreign reserves of the
currency board are between 100 per cent and 110 per cent. The currency board
should adopt similar rules.
Composition of reserves: The currency board should hold its
foreign reserves in low-risk assets payable in the reserve currency only. Most
of its foreign reserves will be low-risk, interest-earning securities. It can
also hold some foreign reserves in interest-bearing deposits at reputable
commercial banks in the reserve country, or in reserve-currency notes or
noninterest-earning deposits at the central bank of the reserve country. As
much as possible, the currency board should avoid holding assets that earn no
interest.
The currency board should not hold assets denominated in
domestic currency, because that would open the way to central banking-type
operations. Specifically, the currency board would be able to increase or
decrease the domestic monetary base by changing its holdings of domestic
securities, as a central bank does. Allowing holdings of domestic assets was
one of the steps that led the East African and
Besides opening the way for central banking, holding
domestic assets can be risky, as the experience of the North Russian currency
board shows. The North Russian board held 25 per cent of its reserves in North
Russian government bonds. When the Red Army captured
Limiting the currency board to assets payable in the
reserve currency need not limit the currency board to securities issued in the
reserve country. Many governments and companies to issue securities denominated
in foreign currency in Eurocurrency (offshore) markets. Branches of French
banks in
British colonial currency boards often divided their
foreign reserves into a "liquid reserve" and an "investment
reserve." The liquid reserve consisted of securities payable in less than
two years, and was typically about 30 per cent of total reserves. The
investment reserve consisted of securities payable in more than two years, and
was the rest of total reserves. The investment reserve was equal to the
public's estimated minimum, "hard-core" demand for currency board
notes and coins (Clauson 1944: 8-11). The liquid reserve of the currency board
may need to exceed 30 per cent initially, but later the board should be able to
hold an increasing proportion of assets in the higher-yielding investment
reserve as the economic situation of the currency board country improves.
Expenses: Judging from the experience of past currency
boards, average expenses of the currency board should be no more than 1 per
cent of total assets, and may as little as ½ per cent. The largest expense will
be printing notes and minting coins. Salaries will probably be the next largest
expense. Rent, utilities, and remaining costs will probably be small.
The notes issued by the currency board should be printed
abroad. Printing the notes abroad will prevent the domestic government from
seizing the printing presses and overturning the currency board system by
printing notes unbacked by foreign reserves. The cost of printing notes abroad
may be higher than it would be if they were printed in domestically, but the
extra expense is worthwhile as a type of insurance. The cost of printing notes
is US$26 to $45 per thousand, depending on what design features the notes have
(Berreby 1992; see also Shapiro 1993).
Seigniorage: Unlike securities and many bank deposits,
notes and coins pay no interest. Hence, notes and coins are like an
interest-free loan from people who hold them to the issuer. The currency board
will earn gross seigniorage equal to interest from its holdings of
reserve-currency securities. Its net seigniorage (profit) will be the gross
seigniorage minus the expense of putting and maintaining notes and coins in
circulation. In addition, if notes and coins are destroyed, the net worth of
the currency board will increase, because its liabilities will decrease but its
assets will not.
Suppose the reserve currency of the currency board is the
U.S. dollar. In the currency board system, the only difference between using
currency board notes and coins instead of dollar notes and coins is that the
currency board rather than the U.S. Federal Reserve System will capture the net
seigniorage. The currency board can earn significant net seigniorage. A
portfolio of long-term and short-term securities should earn an average return
of at least 5 per cent a year. The expenses of the currency board should not
exceed 1 per cent a year, and may be as little as ½ per cent a year. Net
seigniorage, then, will probably be at least 4 per cent a year of the average
circulation of the currency board's notes and coins in circulation. (28)
How to protect a currency board
Despite the economic success of the currency board system
earlier in this century, currency boards exist today only in a few countries.
Currency boards elsewhere were converted into central banks because governments
were influenced by incorrect economic criticisms of the currency board system,
the desire to establish central banks as symbols of national independence, and
an understanding of the political advantages of politicized central banks. As
has been mentioned, many currency boards have relied on informal protection
rather than formal legal protection from changes in their operating rules. The
experience of most such currency boards, and of central banking, strongly
suggests that new currency boards should have strong legal protection from
being converted into a central bank. Anxiety that the monetary constitution
embodied in the currency board might be subverted would reduce the willingness
of foreigners to invest, diminishing one of the main advantages of the currency
board system. Therefore, this section proposes ways of strengthening a currency
board as a monetary constitution. The proposals can be summarized as
credibility, commitment, and competition. They are complementary; any can be
implemented separately or with the others.
A currency board can strengthen its credibility by
protecting itself from potential pressure from the government. Previous
chapters explained how the various features of a typical currency board make
credible its commitment to a fixed exchange rate. Since any human institution,
no matter how rule-bound, is administered by people and can be changed by
people, the model currency board constitution of the Appendix includes a
provision that a majority of the board of directors be foreigners. That will
help prevent the government from bending the rules of the currency board. The foreign
directors should be appointed by Western commercial banks, or perhaps by the
International Monetary Fund if the IMF lends some of the initial foreign
reserves of the currency board. The directors appointed by the IMF should not
be IMF officials or officials of IMF member governments, because their
decisions may too easily be influenced by political considerations. Precedents
exist for such an arrangement. For example, only three of the eight directors
of the Libyan Currency Board of the 1950s were Libyans; the rest were Britons,
Frenchmen, Italians, and Egyptians chosen by their respective governments
(Blowers and McLeod 1952: 453). To reduce the political influence of the
domestic government on the domestic directors of the currency board, the domestic
directors could be required not to be government officials and could be
appointed by a trade association of privately owned banks rather than by the
government.
Another way to strengthen the credibility of a currency
board is for it to hold its assets in a safe-haven country such as
Yet another way to strengthen the credibility of a currency
board is for its notes to contain a statement that they are convertible into
the reserve currency at a fixed rate at the board's offices domestically and
abroad. Whether or not notes and coins issued by the currency board contain an
explicit statement of convertibility, they should be considered a type of
contract promising a fixed exchange rate, unlike notes and coins issued by a
typical central bank. Holders of notes and coins should have the right to sue
the currency board for breach of contract in the very unlikely event that it
fails to redeem its notes and coins at the fixed exchange rate on demand.
A currency board can commit itself to buy and sell forward
exchange at the fixed rate with the reserve currency. Some currency boards,
such as that of
A currency board should cease dealing in forward exchange
if the reserve currency approaches the inflation limits discussed in the next
section.
A currency board can be subjected to competition to induce
it to maintain high-quality service. People should be allowed to make
contracts, payments, and deposits in any currency they wish. In particular,
reserve-currency notes should be allowed to circulate alongside the notes of
the currency board, as has been the case in many currency board systems based
on the pound sterling. However, use of reserve-currency notes in the currency board
country will probably be small, for reasons explained in the next section.
(Notice that as a party to contracts and payments, the government need not
accept the reserve currency for tax payments; it can insist on payment in
currency board currency.)
How to change the reserve currency, if necessary
Besides lacking protection from being converted into
central banks, currency boards have had one other defect: they have lacked
well-defined rules for untying their currencies from an unstable reserve currency.
Most currency board currencies were linked to the pound sterling, which was
stable for more than a century until the Second World War. When the currency
boards were established, confidence in sterling was so great that nobody
considered the possibility that sterling would become unstable. After the
Second World War, though, sterling did become unstable. British colonial
currency boards devalued their currencies with sterling against gold and the
U.S. dollar in 1949, 1967, and 1972. Devaluation hurt them by increasing the
cost of many foreign goods that they needed for their economic development,
such as the food that Hong Kong imported from
Changing the reserve currency is beneficial if the existing
reserve currency becomes quite unstable, because otherwise the currency board
system suffers the monetary problems afflicting the reserve country. (However,
freedom to make contracts and payments in other currencies offers some escape
from the problem.) If the currency board has the power to change the reserve
currency, though, the procedure should be carefully specified in its
constitution and should be enacted by the currency board itself, rather than
being a somewhat arbitrary government decision as was the case with the
currency boards that changed reserve currencies in the 1970s.
We suggest that the currency board not be allowed to change
the reserve currency unless annualized inflation in the consumer price index of
the reserve country exceeds the range -5 per cent to 20 per cent for more than
two years, or -10 per cent to 40 per cent for more than six months. These are
inflation rates that historically have caused substantial economic disruption
if exceeded.
(29) If inflation in the reserve country exceeds the specified range,
the currency board should be allowed to devalue or revalue its currency in
terms of the reserve currency by no more than the amount of the inflation rate
in the reserve country for the period just specified (two years or six months).
Alternatively, the currency board should be allowed to choose a new, more
stable reserve currency and set a new fixed exchange rate at the rate then
prevailing between that currency and the original reserve currency. (If gold is
the reserve currency, the currency board country itself will be considered the
reserve country.)
It may also be desirable for the constitution of the
currency board to contain a similar provision allowing the currency board to
reset the exchange rate with the reserve currency if the reserve currency
appreciates or depreciates very rapidly against a basket of foreign currencies
representing other countries important in foreign trade. These provisions may
appear to open a loophole for destabilizing speculation, as occurs with a pegged
exchange rate, but they do not. Destabilizing speculation occurs when the
commitment to an exchange rate is uncertain. The commitment of the currency
board to maintaining the existing exchange rate is certain, provided that the
reserve currency remains within the predetermined range of inflation or
appreciation. Outside the range, the commitment of the currency board to
changing the exchange rate or the reserve currency is certain; hence no
uncertainty exists about the behavior of the currency board, although
uncertainty may exist about the behavior of the reserve currency. In any case,
speculation will not reduce the foreign reserves of the currency board below
100 per cent of its notes and coins in circulation.
We offer the foregoing rules for changing the reserve
currency as suggestions, which are more experimental than the other operating
rules we have discussed. The general point we wish to emphasize is that it is
better to respond to instability in the reserve currency by having well-defined
rules, known in advance to the public, than to respond in the improvised, even
capricious ways that some currency boards and governments have done. (30)
6.
OBJECTIONS TO CURRENCY BOARDS
A currency board system can provide a sound currency and a
framework within which other problems of the monetary system of a typical
developing country can be solved in a way that will be beneficial for its
economy. To investigate whether establishing a currency board has disadvantages
compared to allowing a typical central bank to make monetary policy, let us
consider the main objections to currency boards.
In the 1950s and 1960s, it was claimed that the currency
board system had certain disadvantages compared to central banking. More recent
economic theories and historical investigation have refuted or reduced the
significance of those objections to the currency board system, but since they
continue to be made, and since no widely available refutation exists, this
chapter briefly considers them, as well as more recent objections. (31)
The objections to the currency board system do not apply with full force to a
parallel currency approach. For the sake of argument, though, assume that the
central bank has been converted into a currency board or that the central bank
has withered away because nobody uses its currency.
Most of the objections that economists have made to the
currency board system, in print (for example, Roubini 1998, Williamson 1995)
and in conversation with us, have been purely theoretical. They have neglected
the excellent historical record of currency board systems, which has been summarized
in several studies.
(32) Many of the theoretical objections have had little practical
importance for currency board systems. Many objections to the currency board
system have also neglected the need to compare monetary institutions
systematically. The currency board system and central banking are both integral
wholes. Certain of their features imply certain other features; therefore, one
should not argue as if the advantages of either system are independent of its
disadvantages. For example, the flexibility possible with completely
discretionary monetary policy (if flexibility is really attainable) is
unavoidably connected with the risk of high inflation. It is doubtful whether
that flexibility could be beneficial, even if the political pressures that tend
to frustrate its exercise in practice could be surmounted.
No lender of last resort
Perhaps the most common objection to a currency board system is that it is
susceptible to financial panics because it lacks a lender of last resort. (33)
One possible reply is that the government can be a lender
of last resort even if no central bank exists. The government can lend to
commercial banks; for example, the
A more fundamental reply is that a government-sponsored
lender of last resort creates more problems than it solves. Many central banks
are lenders of last resort not only to commercial banks, but to state
enterprises and to the government. Even if a typical central bank can be
limited to acting as a lender of last resort only to commercial banks, problems
of moral hazard will tend to occur because commercial banks will expect that
the central bank will rescue them when they become illiquid.
Lack of a central bank as a lender of last resort does not
seem to have harmed currency board systems. Failures by commercial banks have
been minor in orthodox currency board systems. No large commercial bank has
ever failed in a currency board system, and losses to depositors from the few
small commercial banks that failed have been tiny (Schuler 1992b: 191-3). Since
the founding of the first currency board in 1849, there have apparently been no
cases in which commercial banks in currency board systems have relied on
central banks as lenders of last resort. For example, British overseas
commercial banks in currency board system apparently have never relied on the
Bank of England as a lender of last resort. Currency board systems have
performed well without lenders of last resort. Even the currency board-like
systems, which have suffered major bank failures and have used the capabilties
of their monetary authorities as lenders of last resort in a limited way, bank
failures have caused fewer problems than in many central banking systems.
Therefore, it seems likely that after an initial restructuring, commercial
banks in countries that establish currency boards can become strong, stable,
and capable of preserving their liquidity without government-sponsored lenders
of last resort.
Two important sources of stability for commercial banks in
currency board systems have been interbank lending markets and international
branch networks. As the commercial banking system develops, a large interbank
lending market is likely to develop. Illiquid banks will borrow from more
liquid ones, as they do in the currency board system of
The risk of financial panics in a currency board system can
also be reduced by private, voluntary deposit insurance. Government deposit
insurance, whether explicit or implicit, is likely to be a burden to taxpayers.
In developing and developed countries alike, it has cost taxpayers billions of
dollars in many recent cases. Competition promotes sound banking, and is the
best guarantee of safety. If compulsory deposit insurance is thought to be a
political necessity, it should be operated by the banks themselves, are the
private, voluntary deposit insurance systems of
Another way to reduce the risk of financial panics is for
commercial banks to include a "notice of withdrawal clause" (option
clause) in their contracts with depositors. The notice of withdrawal clause
would allow a commercial bank to delay for a set period the requests of
depositors to convert notes into currency board notes and coins. In return, the
bank would pay a penalty rate of interest; for example, 3 per cent above the
rate prevailing before it exercised the notice of withdrawal clause. Banks would
be free to offer a notice of withdrawal clause or not, and depositors would be
free to do business with such banks or not (see Dowd 1988, White 1984: 26,
29-30). Notice of withdrawal clauses have precedents; they were widespread
among savings banks in the
Does size matter?
Another objection to the currency board system is that it
is appropriate for small economies that are open (highly dependent on foreign
trade), such as
We could reply that most countries that now have central
banks are small. To repeat,
Even accepting the terms "large,"
"small," "open," and "closed" as meaningful for
monetary policy, experience suggests that the objection has no practical
significance for the currency board system. Currency boards have been
successful in small, open economies such as Hong Kong and large (populous),
closed economies such as
Fixed versus floating exchange rates
Yet another objection to the currency board system is that
a floating exchange rate is best not only for small, open economies, but for
almost all countries, whether large or small. A floating exchange rate
supposedly better enables an economy to adjust to changes in the terms of trade
than does a fixed exchange rate.
We begin by taking the argument on its own terms. We reply
that a fixed exchange rate is preferable for countries with typical central
banks because a credible fixed exchange rate will enforce a durable monetary
and fiscal constitution and will cease to be a subject of political contention.
In particular, a fixed exchange rate will tend to end soft budget constraints.
At present, expectations that a typical central bank will continue to
accommodate soft budget constraints induce a vicious cycle of inflation in many
developing countries. A fixed exchange rate maintained by a currency board, on
the other hand, will stop inflation because the hard budget constraints that
the currency board system tends to impose will induce workers and state
enterprises to limit wage and price increases to competitive levels, and will
prevent the government from acceding to all wage and price increases by
subsidizing all unprofitable state enterprises.
Additionally, a fixed exchange rate will tend to eliminate
exchange risk with the reserve currency. If a country establishes a currency
board using, say, the U.S. dollar as its reserve currency, it will join a
populous and wealthy common currency zone. Trade with countries in the common
currency zone will be easier than it would be with a floating exchange rate
because the fixed exchange rate will tend to eliminate exchange risk in the
prices of goods. People in the common currency zone will be able to make more
exact price calculations for internationally traded goods. That will tend to
enhance economic efficiency by making the lowest-cost producers within the
common currency zone those with the greatest natural advantages, not those
temporarily benefiting from the extreme fluctuations in real exchange rates
common with a pegged exchange rate, and to some extent with a floating exchange
rate. (34)
A fixed exchange rate will also enable entrepreneurs to apply to other problems
talent that, in a monetary system with a floating exchange rate, they would
apply to foreign-currency speculation and hedging. (Exchange risk with
currencies outside the common currency zone will remain, so some wealth and
talent will continue to be applied to foreign-exchange speculation.)
Eliminating exchange risk will encourage foreign investment
in the currency board country, particularly from other countries within the
common currency zone. Investors will know with certainty what exchange rate
they will receive in terms of the reserve currency should they wish to
repatriate profits. A fixed exchange rate will also enable the currency board
country to "piggyback" on the financial markets of other countries in
the common currency zone. Entrepreneurs in the currency board country will be
able to use as points of reference the highly liquid, well-established markets
elsewhere in the zone. Entering financial markets elsewhere in the zone will
become easier. Financial markets in developed countries offer facilities for
interest-rate hedging, foreign-exchange swaps, and other transactions that will
not be available on a similar scale in most developing countries for years.
Easy access to large foreign financial markets, with no exchange risk, will
tend to increase the growth of economies that establish currency boards.
A deeper reply than the foregoing to objections to a fixed
exchange rate is that debate about "fixed" versus floating exchange
rates usually assumes that the monetary authority is a central bank. For that
reason, advocates of floating exchange rates for the major developed countries
(such as Friedman 1988 [1953]: 8-10) correctly contend that the exchange rate
maintained by a central bank cannot be truly fixed, merely pegged. Unlike a
typical central bank, though, a typical currency board can maintain a truly
fixed exchange rate.
(35)
Debate about "fixed" (in reality, pegged) versus
floating exchange rates also usually assumes that everybody in a country uses
the same currency. In currency board systems, though, foreign-currency
deposits, particularly reserve-currency deposits, have been common. In Hong
Kong, foreign-currency deposits exceed Hong Kong dollar deposits (Jao 1992),
and deposits in Japanese yen (a currency that floats against the
Allowing people to hold deposits in foreign currency also
offers a solution to economists' longstanding, inconclusive arguments about
optimum currency areas, that is, the extent to which it is beneficial that a
country should have one currency or multiple currencies, and fixed exchange
rates or floating rates (Fenton and Murray 1992, Kawai 1992, Mundell 1961). (36)Allowing
people to hold deposits in foreign currency will enable them to take advantage
of any benefits of floating currencies by holding deposits in currencies that
float against the currency board currency. Competition among currencies, as
among other goods, is the proper way to determine optimum areas of service (see
White 1989b).
Deflation
Another objection to a currency board is that it is
deflationary in a growing economy. If one makes certain stringent theoretical
assumptions, as the simplified examples of chapter 2 did, an increase in the
demand for currency board notes and coins requires a current-account surplus to
produce additional foreign reserves as backing. As an economy with a currency
board grows, then, it must achieve continual current-account surpluses for the
supply of currency board notes and coins to increase as quickly as the demand.
Continual surpluses are unlikely, implying that in periods of balance or deficit
in the current account, the supply of notes and coins will increase more slowly
than the demand, resulting in deflation. Deflation would not occur if the notes
and coins were liabilities of a typical central bank, which could increase the
supply of notes and coins without acquiring additional foreign reserves. (37)
We reply that theoretical assumptions are so stringent that
they rarely or never apply to actual currency board systems. A developing
country experiencing healthy economic growth, such as most countries with
currency boards have been, typically has a capital-account surplus (foreign
investment) that exceeds its current-account deficit. Furthermore, the
international branch networks typical of commercial banks in a currency board
system reduce the demand for reserves in the currency board country compared to
what it would otherwise be. Commercial banks can pool reserves between the
reserve country and the currency board country. For example, ignoring the
effect of differences in reserve requirements, the overall reserve position of
a commercial bank with branches in Hong Kong and the United States does not
change if customers of the bank in Hong Kong write Hong Kong dollar cheques to
customers of the bank in the United States, which the American customers then
exchange for U.S. dollars. The effect is the same as if customers of the bank
in
There has apparently been only one case of deflation in a
currency board system caused by an increase in demand for notes and coins. It
occurred in
The inflation tax
A somewhat related objection to a currency board is that it
deprives a country of the opportunity to impose an inflation tax of its own
choosing. A currency board will supposedly deprive the domestic government of
revenue precisely when the government most needs revenue. As a corollary, one
may argue that at any time, a country has the sovereign right to change the
rules governing its currency.
We reply that the restraint on inflation that a currency
board tends to impose is an advantage rather than a disadvantage. Many
developing countries are now suffering from the bad effects of a high inflation
tax. Most of their people would prefer a monetary system that drastically
reduces the inflation tax, as shown by the widespread use of foreign currency
and barter in many of those countries.
We also reply that if a government establishes a currency
board as the issuer of a parallel currency, the government can continue for a
time to impose an inflation tax of its own choosing by means of the central
bank currency. In the long term, though, the parallel central bank currency may
vanish from circulation, so the objection reappears. A more fundamental reply,
then, is that a currency board does not try to earn the most seigniorage
possible, but to earn an amount of seigniorage consistent with maintaining a
sound currency. The seigniorage produced by extreme inflation is large in the
short term, then tends to decrease steeply. Abundant experience shows that
seigniorage from extreme inflation is not a reliable source of tax revenue in
the long term. Moreover, extreme inflation hinders economic growth and reduces
the revenue that can be raised from other taxes.
The cost of reserves
Still another objection to a currency board is that
requiring the currency board to hold 100 per cent foreign reserves deprives the
economy of real resources that are available in a central banking system,
because a typical central bank holds much less than 100 per cent foreign
reserves. Economists who investigated this topic in the 1950s claimed that 30
to 50 per cent of the reserves of currency boards were surplus, since there was
a hard core of notes and coins that people would never return to the boards for
conversion into the reserve currency. The hard core corresponded to the
"investment reserve." Surplus reserves are costly, because they could
be used to buy imports, thus increasing the real goods available in the
economy.
We reply that the surplus foreign reserves may not be as
large as was claimed (Birnbaum 1957). But even if they are, consider their
cost. Once spent, they are gone, and earn no interest. Foreign reserves held by
a currency board, in contrast, earn interest because the currency board invests
them. The stream of future interest payments has a capitalized (present-value)
equivalent. The cost of surplus reserves is the difference between the value of
the goods they could buy now and the capitalized equivalent of the interest
that they will earn if invested. Alternatively, it is possible to calculate the
risk-adjusted interest that the surplus foreign reserves would earn if lent
domestically, and to compare it with the risk-adjusted interest from foreign
assets. Only if domestic interest rates are significantly higher than foreign
interest rates for similarly risky investments is a currency board more costly
than central banking in the narrow sense of the cost of holding reserves.
Critics of the currency board system have often failed to
consider that in many currency board systems, the reason that real domestic
interest rates have been higher than real rates in their reserve countries is
that higher rates have reflected higher political risk, higher risk of default
by borrowers because of different property rights, and higher operating costs
for commercial banks. After adjusting for those factors, the rates of return
from domestic investments and foreign investments have been much closer to
equality (Schuler 1992b: 193-5).
Some currency boards have held domestic securities as part
of their assets, partly because they sought a higher return on assets,
unadjusted for risk. New currency boards should not hold domestic assets, such
as domestic government bonds. Holding domestic assets would risk involving the
currency board in domestic politics, for example by purchasing or not
purchasing certain types of domestic securities for political reasons. The more
domestic assets the currency board held, the more it would be subject to
political risk and political pressure from the domestic government. Another
reason that the currency board should not hold domestic assets is that 100 per
cent foreign reserves is a "natural" ratio that is easy to agree
about. If the ratio is 90 per cent, there will be political pressure to
decrease the ratio to 80 per cent, then to 70 per cent, and so on, as with a
few past currency boards. Minimum gold or foreign-exchange reserve ratios
imposed on central banks have tended to be reduced whenever governments deemed
it advisable in the name of temporary expediency. The U.S. Federal Reserve
System, for instance, was originally required to hold a gold reserve ratio of
40 per cent of its notes in circulation; today the ratio is zero. A 100 per
cent ratio for foreign reserves has a psychological appeal shared by no other
ratio.
Colonialism
Another objection to the currency board system is that it
creates a colonial relationship between the currency board country and the
reserve country.
We reply that the currency board system by itself creates
no colonial relationship. Historically, most currency boards have existed in
British colonies, but currency boards have also existed in independent countries,
including
More generally, a fixed exchange rate, or even a pegged
exchange rate, tends to create close economic relationships between countries
adhering to fixed or pegged rates, yet no colonial relationship need be
implied. The gold standard did not make
As for the possibility that the currency board itself could
somehow become a tool of colonialism, chapter 5 proposed ways to protect the
currency board from interference by foreigners and the domestic government
alike. It proposed a role for foreigners as directors of the currency board to
prevent the domestic government from appointing a majority of directors intent
on converting the currency board into a central bank. The proposal that a
majority of the directors of the currency board should be foreigners may seem
to be an insult to national pride, because it imposes an external restraint on
the domestic monetary system. But restraints are typically necessary for a
monetary constitution to be successful, and external restraints are especially
desirable for a country that has a history of lack of self-restraint in
monetary policy.
Anyway, the current monetary systems of most developing
countries are not now objects of national pride; residents shun domestic
currencies in preference to foreign currency. It is difficult to imagine a more
colonialist type of monetary relationship than unofficial yet pervasive use of
foreign currency, which signals the inability of the domestic government to
provide a currency that people wish to hold. A currency board will tend to
reverse currency substitution and restore an element of national pride by
providing a sound domestic currency.
The worst case
A question (not really an objection) is what could happen
to a currency board in the worst case imaginable. The worst case for the
currency board that we can imagine is that the public converts all currency
board currency into reserve currency. We reply that even in that case, little
would happen. The 100 per cent foreign reserves of the currency board would
ensure that it could meet all demands to convert currency board notes and
coins. Instead of having currency board notes and coins, people would have
reserve currency.
(38) If the exchange rate between the currency board currency and the
reserve currency were one to one, it would not even be necessary for shops to
recalculate prices in reserve currency for the benefit of persons spending
reserve currency. Deposits at commercial banks would also be unaffected.
Even in the worst case, then, the currency board would not
disturb the economy as long as the exchange rate is fully credible. (If the
exchange rate is not fully credible because the currency board lacks strong
institutional safeguards, interest rates would rise significantly.) But the
worst case will not happen. Conversion of currency board currency into reserve
currency will begin a sequence of events whose simplest version is sketched in
Figure 3.6. The chain of events is self-correcting, leading to new
market-clearing levels of the nominal money supply, prices, and incomes.
Furthermore, the necessity that payments to the government be made in currency
board currency will create a hard core of demand for currency board currency,
which will limit the amount of currency board notes and coins that people
convert even in the worst case.
Other objections can be made and other questions can be
asked about the currency board system. We think that we have answered the main
objections and questions, however.
7.
CONCLUSION
Central banking in its current form in developing countries
has resulted in unsound currencies, high inflation, and often economic
stagnation. Central banks in most of those countries, even if reformed, are unlikely
to provide sound currencies soon. Therefore, we suggest consideration of the
currency board system. The currency board system, as used in
The currency board system has many advantages for
developing countries. A currency board can be established quickly, it is simple
to operate, it can be a credible monetary authority that will issue a stable,
fully convertible currency, and it tends to encourage international trade and
investment.
There are two ways to establish a currency board: by
converting the central bank into a currency board, or by establishing the
currency board as the issuer of a parallel currency. The conversion approach
tends to impose hard budget constraints immediately. The parallel currency
approach allows a brief transition period during which budget constraints
harden and during which the government can continue to collect some revenue from
inflation in the central bank currency. At the same time, the existence of the
parallel currency board currency will enable much of the economy to protect
itself from high inflation.
Our proposals for monetary reform are detailed and require
governments that implement them to take action on many points. The most
important elements of the proposal are that the currency board be established,
and that the currency board be protected from political pressure to convert it
into a central bank. Appropriate safeguards to protect the currency board can
be devised.
The hard budget constraints that a currency board will tend
to impose will create momentum for further reforms of government finance and
state enterprises. A currency board best promotes economic growth if it is part
of a package of wider reforms; however, a correctly established currency board
is robust enough to survive and help an economy even if political pressure
temporarily delays wider elements of reform. A currency board forces other
monetary and economic reforms to occur because it tends to eliminate the soft
budget constraints that perpetuate the current monetary system.
Monetary reform is a most important step for generating
growth and progress, and a currency board is the most promising means of
achieving durable, beneficial monetary reform in developing countries today.
|
Figure 7.1. Summary of
proposals Converting the central bank into a currency board
(chapter 4) 1. Delegate to other bodies all functions of the central
bank other than supplying the monetary base. 2. Allow a brief period of clean, unrestricted floating
exchange rates for the domestic currency. 3. Make the actions of the central bank transparent and
predictable. 4. Convert some reserves of commercial banks (deposits at
the central bank) into currency board notes and coins or into foreign
securities, whichever the commercial banks prefer. Dispose of remaining
reserves. 5. Establish a fixed exchange rate with the reserve
currency. 6. Ensure that foreign reserves equal 100 per cent of
domestic-currency notes and coins in circulation. 7. Transfer the remaining assets and liabilities of the
central bank to the currency board and open the currency board for business. Establishing the currency board as the issuer of a parallel
currency (chapter 4) 1. Obtain initial foreign reserves for the currency
board. 2. Make the currency board currency legal tender for
payment of taxes and private debts. 3. Issue currency board currency equal to the initial
foreign reserves. 4. Put the currency board currency into circulation, for
example, by a distribution to every resident according to a well-defined
plan. 5. Allow the currency board currency to circulate as a
parallel currency to the central bank currency, at an exchange rate determined
by market forces. |
APPENDIX: A
MODEL CURRENCY BOARD CONSTITUTION
To illustrate the legal foundation necessary for a currency
board to work best, this appendix offers a model constitution for a currency
board. The model constitution has many features adapted from the constitutions
of currency boards in West Africa, Hong Kong, the British Caribbean,
Currency board constitution
1. The Currency Board of Country X is hereby created by the
government of Country Xand the International Monetary Fund in joint
cooperation. The purpose of the Currency Board is to issue notes, coins, and
deposits in Currency Board units, and to maintain them fully convertible at a
fixed exchange rate into a reserve currency as specified in paragraph 6.
2. The Currency Board shall have its legal seat in
3. a. The Currency Board shall be governed by a board of
five directors. Three directors shall be foreign nationals appointed by the
International Monetary Fund. They shall not be employees of the International
Monetary Fund or its member governments. Two directors shall be appointed by
the government of Country X.
b. A quorum shall consist of three members of the board of
directors, including at least one of the directors chosen by the government of
Country X. The board of directors may meet at the Currency Board's legal seat
or in such other locations as it designates. Decisions shall be by majority
vote, except as specified in paragraph 15.
c. The first two directors appointed by the government of
Country X shall serve terms of one and four years. The first three directors
appointed by the International Monetary Fund shall serve terms of two, three,
and five years. Subsequent directors shall serve terms of five years. Directors
may be reappointed once. Should a director resign or die, the appropriate
organization as specified in paragraph 3(a) shall choose a successor to
complete the remainder of the term.
4. The board of directors shall have the power to hire and
fire the Currency Board's staff, and to determine salaries for the staff. The
bylaws of the Currency Board shall determine salaries for the directors.
5. The Currency Board shall issue notes and coins
denominated in currency board units. The notes and coins shall be fully
convertible into the reserve currency. The notes shall be printed outside
Country X. The Currency Board may accept deposits of the reserve currency.
6. a. The reserve currency is the foreign currency or the
commodity to which the currency board currency has a fixed exchange rate.
Initially, the reserve currency shall be Currency Z and the fixed exchange rate
shall be A currency board units per unit of Currency Z.
b. Failure to maintain the fixed exchange rate with the
reserve currency shall make the Currency Board subject to legal action for
breach of contract according to the laws of
7. The Currency Board shall charge no commission for
exchanging Currency Board units for the reserve currency, or the reverse.
8. The Currency Board shall begin business with foreign
reserves equal to at least 100 per cent of its monetary liabilities (notes and
coins in circulation, deposits with it, and so on). It shall hold its foreign
reserves in securities or other forms payable only in the reserve currency. The
Currency Board shall not hold securities issued by the national or local
governments of CountryX, or by enterprises owned by those governments.
9. The Currency Board shall pay all net seigniorage
(profits) into a reserve fund until its net unborrowed reserves equal 110 per
cent of its monetary liabilities. It shall remit to the government of Country X
all net seigniorage beyond that necessary to maintain 110 per cent reserves.
The distribution of net seigniorage shall occur annually.
10. The head office of the Currency Board shall be in City
Y of Country X. The Currency Board may establish branches or appoint agents in
such other cities of Country X as it sees fit. The Currency Board shall also
maintain a branch in
11. The Currency Board shall publish every business day a
summary of its main balance sheet items from the previous business day. It
shall publish a detailed financial statement, attested by the directors,
quarterly or more often. The statement shall appraise the Currency Board's
holdings of securities at their market value. The detailed statement shall be
evaluated by an external auditor annually or more often.
12. The Currency Board may issue notes and coins in such
denominations as it judges to be appropriate.
13. Should the annual change in the consumer price index in
the reserve country fall outside the range -5 per cent to 20 per cent for more
than two years, or -10 per cent to 40 per cent for more than six months, within
sixty days the Currency Board must either:
a. Devalue (if the change in the index is negative) or
revalue (if the change in the index is positive) the Currency Board currency in
terms of the reserve currency by no more than the change in the index during
the period just specified, or
b. choose a new reserve currency and fix the exchange rate
of the Currency Board currency to the new currency at the rate then prevailing
between the new reserve currency and the former reserve currency.
14. If the Currency Board chooses a new reserve currency in
accord with paragraph 13, within one year it must convert all its foreign
reserves into assets payable in the new reserve currency.
15. The Currency Board may not be dissolved nor may its
assets be transferred to a successor organization except by unanimous vote of
the board of directors.
16. Beyond an initial loan of reserves from the
International Monetary Fund, the Currency Board may not accept loans or grants
of reserves from international agencies or foreign governments.
17. Exchanges by the Currency Board shall be exempt from
taxation by the government of Country X.
18. Currency Board currency shall be legal tender for
paying taxes and settling debts in Country X. However, it shall not be forced
tender for contracts between private parties.
BIBLIOGRAPHY
For updated information and links to relevant World Wide
Web sites, see Kurt Schuler's Web site on currency boards, dollarization, and
related topics, <http://www.dollarization.org>. See also Steve Hanke's
Web site, <http://www.forbes.com/hanke>. The Web site of the Bank for
International Settlements, <http://www.bis.org>, has links to the sites
of currency boards and currency board-like monetary authorities.
Alesina, Alberto (1989) "Politics and business cycles
in industrial democracies," Economic Policy: A European Forum, 4, 8,
April: 57-98.
"Analyst" (1953) "Currency and banking in
Auerbach, Paul, Davison, Geoffrey, and Rostowski, Jacek
(1992) "Secondary currencies and high inflation: implications for monetary
theory and policy," Discussion Paper 58, Centre for Economic Performance,
London School of Economics, February.
Tomás Baliño, Charles Enoch, and others (1997)
"Currency Board Arrangements: Issues and Experiences," International
Monetary Fund Occasional Paper 151, August.
Barro, Robert J., and Gordon, David B. (1983) "Rules,
discretion and reputation in a model of monetary policy," Journal of
Monetary Economics, 12, 2: July, 101-21.
Basu, S. K. (1971) A Review of Current Banking Theory and
Practice,
Berreby, David (1992) "The companies that make money
from making money," New York Times, August 23: 10F.
Birnbaum, Eugene A. (1957) "The cost of a foreign
exchange standard or of the use of a foreign currency as the circulating medium,"
International Monetary Fund Staff Papers, 5, 3, February: 477-91.
Blowers, G. A., and McLeod, A. N. (1952) "Currency
unification in
Bordo, Michael D., and Jonung, Lars (1987) The Long-Run
Behavior of Velocity: The International Evidence.
Boughton, James M. (1992) "The CFA franc: zone of
fragile stability in
Bruno, Michael (1991) High Inflation and the Nominal
Anchors of an Open Economy, Essays in International Finance 183, June,
Princeton: International Finance Section, Department of Economics,
Buchanan, James M., and Wagner, Richard E. (1977) Democracy
in Deficit: The Political Legacy of Lord
[Caine, Sydney] (1950) Monetary Systems of the Colonies,
Camdessus, Michel (1992) "Monetary arrangements in
Eastern Europe and the former Soviet Union," unpublished speech delivered
at Bank of Greece colloquium on Monetary Integration in the Eastern European
Economies, Athens, 9 October; excerpted in "Monetary arrangements should
promote price stability, says Camdessus," IMF Survey, 9 November: 341-3.
Caprio Jr., Gerard, Michael Dooley, Danny Leipziger, and
Carl Walsh (1996) "The lender of last resort function under a currency
board : the case of
CIA (1999) U.S. Central Intelligence Agency, CIA World
Factbook, available online at
<http://www.odci.gov/cia/publications/factbook/index.html>.
Chalmers, Robert C. (1893) A History of Currency in the
British Colonies,
Claasen, Emil-Maria (ed.) (1991) Exchange Rate Policies in
Developing and Post-Socialist Countries,
Clauson, G. L. M. (1944) "The British colonial
currency system," Economic Journal, 54, 213, April: 1-25.
Collier, Paul, and Joshi, Vijay (1989) "Exchange-rate
policy in developing countries," Oxford Review of Economic Policy, 5, 3,
Autumn: 94-113.
Collyns, Charles (1983) Alternatives to the Central Bank in
the Developing World, Occasional Paper 20,
Conant, Charles A. (1969 [1927]), A History of Modern Banks
of Issue, 6th ed. rev.,
Connolly, Michael (1985) "On the optimal currency peg
for developing countries," Journal of Development Economics, 18, 2-3,
August: 555-9.
Congdon, T. G. (1985) Economic Liberalisation in the Cone
of Latin America, Thames Essays 40,
Cooper, Richard (1991a) "Opening the Soviet
economy," in Merton J. Peck and Thomas J. Richardson (eds) What Is to Be
Done? Proposals for the Soviet Transition to the Market:116-32,
Cooper, Richard (1991b) "Comment," in John
Williamson (ed.) Currency Convertibility in Eastern Europe: 310-14,
Corbo, Vittorio, de Melo, Jaime, and Tybout, James (1986)
"What went wrong with the recent reforms in the Southern Cone,"
Economic Development and Cultural Change, 34, 3, April: 607-40.
Crick, W. F. (ed.) (1965) Commonwealth Banking Systems,
Cukierman, Alex (1992) Central Bank Strategy, Credibility,
and
Cukierman, Alex, Webb, Steven B., and Neyapti, Bilin (1992)
"Measuring the independence of central banks and its effect on policy
outcomes," World Bank Economic Review, 6, 3, September: 353-98.
Culp, Christopher, and Hanke, Steve H. (1992) "The
Hong Kong linked rate mechanism: monetary lessons for economic
development," unpublished paper,
Culp, Christopher L., Hanke, Steve H., and Miller, Merton
(1999) "The case for an Indonesian currency board," Journal of
Applied Corporate Finance, Winter: 57-65.
Dodwell, David (1992) "New institutions offered to
keep growth on track," Financial Times, 8 October: 4.
Dornbusch, Rudiger, Sturzenegger, Frederico, and Wolf,
Holger (1990) "Extreme inflation: dynamics and stabilization,"
Brookings Papers on Economic Activity, 2: 1-64, 77-84.
Dowd, Kevin (1988) "Option clauses and the stability
of a laisser faire monetary system," Journal of Financial Services
Research, 1, 4, December: 319-33.
Dowd, Kevin (ed.) (1992a) The Experience of Free Banking,
Dowd, Kevin (1992b) "Money and the market: what role
for government?" unpublished paper, University of Nottingham, published in
Cato Journal, 12, 3, Winter 1993.
Drake, P. J. (ed.) (1966) Money and Banking in Malaya and
Drake, P. J. (1969) Financial Development in Malaya and
Edo, Michael E. (1975) "Currency and banking
legislation in the
Enoch, Charles, and Anne-Marie Gulde (1997) "Making a
Currency Board Operational," International Monetary Fund Paper on Policy Analysis
and Assessment 97/10, November.
Fenton, Paul, and Murray, John (1992) "Optimum
currency areas: a cautionary tale," unpublished paper, Bank of Canada,
September.
Fischer,
Fisher,
Freris, Andrew (1991) The Financial Markets of Hong Kong,
Friedman,
Friedman,
Friedman,
Friedman,
Fry, Maxwell J. (1988) Money, Interest, and Banking in
Economic Development,
Ghose, T. K. (1987) The Banking System of
Ghosh, Atish, Anne-Marie Gulde, and Holger Wolf (1998)
"Currency Boards: The Ultimate Fix?," International Monetary Fund
Working Paper WP/98/8.
Goldstein, Morris, Goldstein, Morris, Isard, Peter, Masson,
Paul R., and Taylor, Mark P. (1992),Issues in the Evolving International
Monetary System, Occasional Paper 96, Washington: International Monetary Fund.
Goodhart, Charles (1988) The Evolution of Central Banks,
Greaves, Ida C. (1953) Colonial Monetary Conditions,
Greene, Joshua E., and Isard, Peter (1991) Currency
Convertibility and the Transformation of Centrally Planned Economies, Occasional
Paper 81,
Grilli, Vittorio, Masciandro, Donato, and Tabellini, Guido
(1991) "Political and monetary institutions and public financial policies
in the industrial countries," Economic Policy: A European Forum, 13,
October: 341-76, 383-92.
Hanke, Steve H. (1999a) "Some reflections on currency
boards," in Mario I. Blejer and Marko Skreb (eds) Central Banking,
Monetary Policies, and the Implications for Transition Economies: 341-66.
Hanke, Steve H. (1999b) "
Hanke, Steve H. (2000) "Parkinson in Euroland,"
Forbes Global, April 3: 24.
Hanke, Steve H., Jonung, Lars, and Schuler, Kurt (1993)
Russian Currency and Finance: A Currency Board Approach to Reform,
Hanke, Steve H., and Schuler, Kurt (1990) "Keynes and
currency reform: some lessons for
Hanke, Steve H., and Schuler, Kurt (1991) "Keynes's
Russian currency board," in Steve H. Hanke and Alan A. Walters (eds)
Capital Markets and Development: 43-63,
Hanke, Steve H., and Schuler, Kurt (1999) "A monetary
constitution for
Hanke, Steve H., and Walters, Alan A. (eds) (1991)
"Discussion," in Steve H. Hanke and Alan A. Walters (eds) Capital
Markets and Development: 187-202,
Havrylyshyn, Oleh, and Williamson, John (1991) From Soviet
disUnion to Eastern Economic Community? Policy Analyses in International
Economics 35, October,
Hayek, F. A. (ed.) (1975 [1935]) Collectivist Economic
Planning,
Hayek, F. A. (1937) Monetary Nationalism and International
Stability,
Hayek, F. A. (1978) Denationalisation of Money--The
Argument Refined: An Analysis of the Theory and Practice of Concurrent
Currencies, 2nd ed.,
Hazlewood, Arthur (1954b) "The economics of colonial
monetary arrangements," Social and Economic Studies, 3, 3-4, December:
291-315.
Hutton, J. P. (1992) "Real exchange rates," in
Peter Newman, Murray Milgate, and John Eatwell (eds) The New Palgrave
Dictionary of Money and Finance, 3: 303-5, London: Macmillan.
IMF (annual) International Monetary Fund, Annual Report on
Exchange Arrangements and Exchange Restrictions, Washington: International
Monetary Fund.
IMF (monthly) International Monetary Fund, International
Financial Statistics.
Jao, Y. C. (1974) Banking and Currency in Hong Kong,
Jao, Y. C. (1992) "Recent trends in currency
substitution," Asian Monetary Monitor, 16, 4, July-August: 12-21.
Jao, Y. C., and King, F. H. H. (1990) Money in Hong Kong:
Historical Perspective and Contemporary Analysis, Hong Kong: Centre of Asian
Studies,
Jonung, Lars (1979) "Knut Wicksell's norm of price
stabilization and Swedish monetary policy in the 1930's," Journal of
Monetary Economics, 5, 4, October: 459-96.
Jonung, Lars (1984) "Swedish experience under the
classical gold standard 1873-1913," in Michael D. Bordo and Anna J.
Schwartz (eds) A Retrospective on the Classical Gold Standard: 361-99, Chicago:
University of Chicago Press for National Bureau of Economic Research.
Kawai, Masahiro (1992) "Optimum currency areas,"
in Peter Newman, Murray Milgate, and John Eatwell (eds) The New Palgrave
Dictionary of Money and Finance, 3: 78-81, London: Macmillan.
Kenen, Peter (1978) "The role of monetary policy in
developing countries," in Central Bank of Gambia, Research Department, The
Role of Monetary Policy in Developing Counties: Theoretical Perspective and
Institutional Framework: 9-13, Banjul, Gambia: Central Bank of Gambia.
King, Frank H. H. (1957) Money in British East Asia,
Kratz, Joachim W. (1966) "The East African Currency
Board," International Monetary Fund Staff Papers, 13, 2, July: 229-55.
Krugman, Paul R., and Rotemberg, Julio (1992 [1991])
"Speculative attacks on target zones," in Paul R. Krugman, Currencies
and Crises,
Kydland, Finn E., and Prescott, Edward C. (1977)
"Rules rather than discretion: the inconsistency of optimal plans,"
Journal of Political Economy, 85, 3, June: 473-92.
Laidler, David (1982) Monetarist Perspectives,
Lee, Sheng-Yi (1974, 1986) The Monetary and Banking
Development of
Lindsey, David E., and Wallich, Henry C. (1987)
"Monetary policy," in John Eatwell, Murray Milgate, and Peter Newman
(eds) The New Palgrave: A Dictionary of Economics, 3: 508-15, London:
Macmillan.
Loynes, J. B. (1962) The West African Currency Board
1912-1962,
MacDonald, Ronald (1988) Floating Exchange Rates: Theories
and Evidence,
Marx, Karl, and Engels,
Mates, Neven (1992) "Does the government have to clean
bank balance sheets in transitional economies?" Communist Economies and
Economic Transformation, 4, 3: 395-409.
McKinnon, Ronald (1991) The Order of Economic
Liberalization: Financial Control in the Transition to a Market Economy,
Meltzer, Allan (1992) "The Fed at seventy-five,"
in Michael T. Belongia (ed.) Monetary Policy on the Fed's 75th Anniversary:
Proceedings of the Fourteenth Annual Conference of the Federal Reserve Bank of
St Louis: 3-65, Boston: Kluwer.
Mises, Ludwing von (1981 [1932]) Socialism, 2nd ed., trans.
by J. Kahane,
Morgenstern, Oskar (1963) On the Accuracy of Economic
Observations, 2nd ed., Princeton:
Mundell, Robert A. (1961) "A theory of optimum
currency areas," American Economic Review,51, 3, September: 657-65.
Nelson, William Evan (1984) "The imperial
administration of currency and British banking in the Straits Settlements,
1867-1908," unpublished Ph.D. dissertation,
Nevin, Edward (1961) Capital Funds in Underdeveloped
Countries,
Newlyn, W. T., and Rowan, D. C. (1954) Money and Banking in
British Colonial Africa: A Study of the Monetary and Banking Systems of
Nuti, Domenico Mario (1991) "Comment," in John
Williamson (ed.) Currency Convertibility in Eastern Europe: 48-55,
Osband, Kent, and Villanueva, Delano (1992)
"Independent currency authorities: an analytic primer," International
Monetary Fund Working Paper 92/50, July, published inInternational Monetary
Fund Staff Papers, 40, 1, March 1993: 202-16.
Ow, Chwee Huay (1985) "The currency board monetary
system--the case of
Parry, Guillermo E. (ed.) (1997) Currency Boards and
External Shocks: How Much Pain, How Much Gain?,
Persson, Torsten, and Tabellini, Guido (1990) Macroeconomic
Policy, Credibility and Politics,
Pleskovic, Boris, and Sachs, Jeffrey (1992) "Currency
reform in
Porter, Richard, and Ruth Judson (1996) "The Location
of U.S. Currency: How Much Is Abroad?," Federal Reserve Bulletin, v. 82,
no. 10, October, pp. 883-903.
Rahn, Richard (1999) The End of Money and the Struggle for
Financial Privacy, Seattle: Discovery Institute Press.
Revista de análisis económico (1992) special issue on
convertibility and currency substitution, 7, 1, junio.
Rostowski, Jacek (1992) "The benefits of currency
substitution during high inflation and stabilization," Revista de análisis
económico, 7, 1, junio: 91-107.
Rostowski, Jacek, and Shapiro, Judith (1992)
"Secondary currencies in the Russian hyperinflation of 1921-24,"
Discussion Paper 59, Centre for Economic Performance, London School of
Economics, January.
Rothbard, Murray N. (1962) "The case for a 100 percent
gold dollar," in Leland B. Yeager (ed.)In Search of a Monetary
Constitution: 94-136, Cambridge, Massachusetts: Harvard University Press.
Roubini, Nouriel. [1998?]. "The case against currency
boards: debunking 10 myths about the benefits of currency boards."
Available online at <http://www.stern.nyu.edu/~nroubini/asia/CurrencyBoardsRoubini.html>.
Sayers, R. S. (ed.) (1952) Banking in the British
Commonwealth, London: Oxford University Press.
Schmieding, Holger (1992) "Lending stability to
Europe's emerging market economies: on the potential importance of the EC and
the ECU for the transformation process in Central and Eastern Europe,"
Ph.D. dissertation, Christian-Albrechts Universität, Kiel, Germany, published
in Kieler Studien, Tübingen: J. C. B. Mohr (Paul Siebeck).
Schuler, Kurt (1992a) "The World History of Free
Banking," in Kevin Dowd (ed.) The Experience of Free Banking: 7-47,
London, Routledge.
Schuler, Kurt (1992b) "Currency boards,"
unpublished Ph.D. dissertation, George Mason University, available at
<http://www.erols.com/kurrency> or <http://www.dollarization.org>.
Schuler, Kurt (1996) Should Developing Countries Have
Central Banks? Currency Quality and Monetary Systems in 155 Countries, Research
Monograph 52. London: Institute of Economic Affairs.
Schuler, Kurt (1998) "A contingency plan for
dollarizing Hong Kong," HKCER Letters (Hong Kong Center for Economic
Research), 52, September, available at
<http://www.hku.hk/hkcer/articles/v52/kurt.htm>.
Schuler, Kurt, Selgin, George, and Sinkey, Joseph, Jr.
(1991) "Replacing the ruble in Lithuania: real change versus
pseudoreform," Cato Institute Policy Analysis 163, 28 October.
Schwartz, Anna J. (1992a) "Currency boards: their
past, present, and possible future," National Bureau of Economic Research,
published in Carnegie-Rochester Conference Series on Public Policy, December
1993.
Schwartz, Anna J. (1992b) Do Currency Boards Have a
Future?, Occasional Paper 88, London: Institute of Economic Affairs.
Selgin, George A., and Lawrence H. White (1994) "How
Would the Invisible Hand Handle Money?," Journal of Economic Literature
32, 4, December: 1718-49.
Shannon, H. A. (1951) "Evolution of the colonial
sterling exchange standard," International Monetary Fund Staff Papers, 1,
3, April: 334-354.
Shannon, H. A. (1952) "The modern colonial sterling
exchange standard," International Monetary Fund Staff Papers, 2, 2, April:
318-362.
Shapiro, Joshua (1993) "Australia's plastic bills:
durable and hard to counterfeit," New York Times, 7 February: F9.
Simons, Henry (1934) A Positive Program for Laissez Faire:
Some Proposals for a Liberal Economic Policy, Chicago: University of Chicago
Press.
Talley, Samuel H., and Mas, Ignacio (1990) "Deposit
insurance in developing countries," World Bank Working Paper 548,
November.
Tan Pin Neo, Maggie (ed.) (1997), Currency Board System: A Stop-Gap Measure or a
Necessity. Currency Board System Symposium '97, Singapore: Board of
Commissioners of Currency Singapore.
Taylor, Dean (1982) "Official intervention in the
foreign exchange market, or, bet against the central bank," Journal of
Political Economy, 90, 2, April: 356-68.
Vaubel, Roland (1978) Strategies for Currency Unification:
The Economics of Currency Competition and the Case for a European Parallel
Currency, Kieler Studien 156, Tübingen: J. C. B. Mohr (Paul Siebeck).
Walters, Alan A. (1987) "Currency boards," in
Peter Newman, Murray Milgate, and John Eatwell (eds) The New Palgrave: A
Dictionary of Economics, 1: 740-2, London: Macmillan.
Walters, Alan (1990) Sterling in Danger: The Economic
Consequences of Pegged Exchange Rates, London: Fontana/Collins.
Walters, Alan (1992b) "The frontiers of reform: errors
and omissions," AIG World Markets Advisory (AIG Trading Corp.), September:
1, 3-5.
Walters, Alan A., and Hanke, Steve H. (1992) "Currency
boards," in Peter Newman, Murray Milgate, and John Eatwell (eds) The New
Palgrave Dictionary of Money and Finance, 1: 558-61, London: Macmillan.
White, Lawrence H. (1984) Free Banking in Britain: Theory,
Experience and Debate 1800-1845,Cambridge: Cambridge University Press.
White, Lawrence H. (1989a) Competition and Currency: Essays
on Free Banking and Money,New York: New York University Press.
White, Lawrence H. (1989b) "Fix or float? The
international monetary dilemma," in Lawrence H. White, Competition and
Currency: Essays on Free Banking and Money: 137-47, New York: New York
University Press.
White, Lawrence H. (ed.) (1993) Free Banking, 3 v.,
Aldershot, England: Edward Elgar.
Williamson, John (ed.) (1981) Exchange Rate Rules: The
Theory, Performance and Prospects of the Crawling Peg, New York: St. Martin's
Press.
Williamson, John (1991) "The economic opening of
Eastern Europe," in John Williamson (ed.)Currency Convertibility in
Eastern Europe: 363-431, Washington: Institute for International Economics.
Williamson, John (1992) Trade and Payments After Soviet
Disintegration, Policy Analyses in International Economics 37, June,
Washington: Institute for International Economics.
Wlliamson, John (1995) What Role for Currency Boards?,
Washington: Institute for International Economics.
Williamson, John (1996) The Crawling Band as an Exchange
Rate Regime: Lessons from Chile, Colombia, and Israel. Washington: Institute
for International Economics.
World Bank (1992) World Development Report 1992: Development
and the Environment,Oxford: Oxford University Press.
Yeager, Leland B. (1976) International Monetary Relations:
Theory, History, and Policy, New York: Harper and Row.
Yeager, Leland B. (ed.) (1962) In Search of a Monetary
Constitution, Cambridge, Massachusetts: Harvard University Press.
Yeager, Leland (1992) "Monetary constitutions,"
in Peter Newman, Murray Milgate, and John Eatwell (eds) The New Palgrave
Dictionary of Money and Finance, 2: 731-4, London: Macmillan.
Yeager, Leland, and associates (1981) Experiences with
Stopping Inflation, Washington: American Enterprise Institute.
NOTES
1. In
some countries, a government agency other than the central bank issues coins.
Coins are typically a very small proportion of the total money supply, and the
agency issuing coins usually coordinates its policy with that of the central
bank, so it typically has no independent influence on monetary policy.
2. We
make this definition of the money supply for ease of exposition. In practice,
near-monies such as accounts at money market mutual funds may be almost as
liquid and widely accepted in payment as deposits at commercial banks. To avoid
such problems of definition, one can think of commercial banks as symbolizing
all institutions that extend credit widely accepted as means of payment.
3. Even
the most respected central banks have hidden their activities from public
inspection. For example, the economist Oskar Morgenstern (1963: 20-1), who
researched the accuracy of central bank balance sheets, remarked that
Central banks in many countries, the venerable Bank of
England not excepted, have for
decades published deliberately misleading statistics, as,
for example, when part of the gold in their possession is put under "other
assets" and only part is shown as "gold." In democratic Great
Britain before World War II, the Government's "Exchange Equalization
Account" suppressed for a considerable period all statistics about its
gold holdings, although it became clear later that these exceeded the amount of
gold shown to be held by the Bank of England at the time. This list could be
greatly lengthened. If respectable governments falsify information for policy
purposes, if the Bank of England lies and hides or falsifies data, then how can
one expect minor operators in the financial world always to be truthful,
especially when they know that the Bank of England and so many other central
banks are not?
4. For
more on dollarization, see Kurt Schuler's Web site,
<http://www.dollarization.org>.
5. As
of June 2000, the members of the European Central Bank are Austria, Belgium,
Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands,
Portugal, and Spain. The members of the Banque Centrale des Etats de l'Afrique
de l'Ouest (Central Bank of West African States) are Benin, Burkina Faso, Côte
d'Ivoire (Ivory Coast), Guinea-Bissau, Mali, Niger, Senegal, and Togo. The
members of the Banque des Etats de l'Afrique Centrale (Bank of Central African
States) are Cameroon, Central African Republic, Chad, Republic of Congo
(Congo-Brazzaville), Equatorial Guinea, and Gabon. The members of the East
Caribbean Central Bank are Anguilla, Antigua and Barbuda, Dominica, Grenada,
Montserrat, St. Kitts and Nevis, St. Lucia, and St. Vincent and the Grenadines.
Anguilla and Montserrat, which are British colonies, are the only members of
any multinational central bank that are not independent.
6. Some
economists reserve the term "parallel currency" for a currency that
circulates alongside another at a floating exchange rate, often illegally.
7. Very
high real interest rates are usually caused by exchange risk or political risk.
A currency board will tend to eliminate exchange risk with the reserve
currency, but it will not eliminate political risk. Real interest rates will be
high unless property rights are secure.
8. Fixed,
pegged, and floating exchange rates are easiest to distinguish when currencies
are fully convertible. The differences among them become somewhat blurred for
partly convertible or inconvertible currencies, because foreign-exchange
controls often are more important than exchange-rate arrangements in
influencing the true market exchange rates of the currencies. A currency that
has a pegged exchange rate and capital controls may behave like a floating
currency in unrestricted foreign-exchange markets, such as the black market.
9. Several
definitions of the real exchange rate exist; see Hutton (1992).
10. On
crawling pegs, see Williamson (1981). On the related system of crawling bands,
see Williamson (1996), which studies Chile, Colombia, and Israel--all of which
abandoned crawling bands by 1999.
11. An
exceptionally good central bank may be able to keep inflation relatively low by
targeting the price level, a measure of the money supply, or another statistic
(for an example, see Jonung [1979]). However, for a typical central bank such
targets create political pressure for soft budget constraints, because the
standards for evaluating their success are less transparent than for an
exchange rate target.
12. For
more on different types of foreign-exchange convertibility, see Greene and
Isard (1991: 2-3).
13. On
the case for a fiscal constitution, see Buchanan and Wagner (1977). On a
monetary constitution, see Grilli and others (1991) and Yeager (1962, 1992).
14. This
assumes that the ultimate reserves are used only in the currency board system
and the reserve country, not elsewhere. In reality, ultimate reserves of some
currencies are held outside of their country of origin; for example, U.S.
dollar notes are held extensively in Latin America.
15. Only
assumption 1 is necessary for the analysis of currency boards; the other
assumptions can be discarded, but the analysis then becomes much more
complicated. See Ow (1985) and Walters and Hanke (1992) for details.
16. The
account that follows borrows liberally from Greenwood (1981, 1983a).
17. For
a general discussion of the role of a nominal anchor, see Bruno (1991).
18. Quotas
and high tariffs also reduce the efficiency of arbitrage, which is one reason
that the alternative approach to economic reform discussed in chapter 1
recommends deregulation of trade.
19. The
Federal Reserve System will lend to government deposit insurance agencies whose
funds are depleted. It is the ultimate lender of last resort in the American
monetary system.
20. See
Blowers and McLeod (1952), Caine (1950), Chalmers (1893), Clauson (1944), Crick
(1965), Drake (1966, 1969), Edo (1975), Freris (1991), Ghose (1987), Greaves
(1953), Jao (1974), Jao and King (1990), King (1957), Kratz (1966), Lee (1974,
1986) Loynes (1962), Nelson (1984), Newlyn and Rowan (1954), Sayers (1952), and
Shannon (1951, 1952). More recent works that summarize the history of the
currency board system include Schuler (1992b) and Schwartz (1992a).
21. The
East Caribbean Currency Authority in effect devalued the East Caribbean dollar
by about 30 per cent when it switched from sterling to the US dollar as its
reserve currency in 1976. The devaluation was apparently intended to improve
the competitiveness of exports. The assets of the East Caribbean Currency Board
were more than sufficient to have supported the change of reserve currencies
without a devaluation.
22. This
calculates the size of the economy by converting domestic currency into the
U.S. dollar equivalent, not by evaluating purchasing power parities. For
analyses of the extent to which Hong Kong deviates from an orthodox currency
board system, see Culp and Hanke (1992) and Schuler (1998).
23. For
analyses, see Caprio and others (1996) and Hanke (1999a). Hanke and Schuler
(1999) proposed to end uncertainty about Argentina's currency board-like system
by dollarizing. Steve Hanke presented the proposal to Argentine president
Carlos Menem in February 1999.
24. For
recent articles on currency substitution, see Revista de análisis económico
(1992).
25. Typical
monetary analysis, which relies on the quantity theory of money, does not apply
to a system of parallel currencies. On the economics of parallel currencies,
see Hayek (1978) and Vaubel (1978). (Earlier, Hayek [1937: 91-2] advocated a
type of currency board system.)
26. See
Bordo and Jonung (1987: Appendix) and the IMF's International Financial
Statistics.
27. For
more details on the operations of past currency boards mentioned in this
chapter, see Schuler (1992b).
28. For
some mathematics of seigniorage, see Osband and Villanueva (1992: Appendix A).
For an empirical study of seigniorage, see Fischer (1982).
29. Inflation
rates within the range should be tolerated because changing reserve currencies
is costly for the economy. The "menu costs" are probably small, but
the cost of discovering an appropriate new structure of prices to reflect the
new reserve currency is probably large. Readers who think our proposed range
for tolerating inflation is too large should compare it with actual inflation
in many developing countries.
30. Because
we think that appropriate rules for changing the reserve currency can be
devised, we do not share the worry of Schwartz (1992b: 20, 23) that a reserve
currency, once chosen, may cause disruptions if the reserve country becomes no
longer an important trading partner of the currency board country. Anyway, for
most developing countries those potential disruptions are small compared to the
disruptions caused by the unsound condition of their currencies at present.
31. The
best older criticisms are Analyst (1953), Basu (1971: 54-66, 240-4), Hazlewood
(1954), and Nevin (1961: 1-44, 67-71). See also some of the essays in Drake
(1966). The best recent criticism is Williamson (1995); see also Schwartz
(1992b: 18-21). For refutations, see Greaves (1953), King (1957: 61-99), and
especially Ow (1985: 54-86).
32. See
Ghosh and others (1998), Hanke (1999), Schuler (1992, 1996). Since 1998, Kurt
Schuler has posted on his Web site an offer to pay up to US$1,000 to anybody
who can find an important measure of monetary performance by which central
banking systems have done better than currency board and currency board-like
systems over the long run. So far, nobody has responded.
33. A
good discussion of the role of a central bank as a lender of last resort to
commercial banks is Goodhart (1988: 96-102).
34. On
the performance of floating exchange rates since the end of the Bretton Woods
system of pegged exchange rates in 1973, see MacDonald (1988).
35. Milton
Friedman (1997), the leading advocate of floating exchange rates for the major
developed countries, has for many years said that for developing countries
fixed exchange rates, as provided by currency boards or dollarization, are more
appropriate.
36. No
generally agreed criteria exist for determining optimum currency areas (Kawai
1992). One criterion that has been suggested is that labor should be mobile
within an optimum currency area. The experience of currency board systems has been
that labor mobility with the reserve country is unimportant. For example, labor
has not been mobile between Hong Kong and
37. The
converse proposition is that in a contracting economy with a currency board,
the money supply operates in an inflationary manner, which some critics of the
currency board system may think is a desirable countercyclical response.
38. The
discussion assumes that the deposits of the public would remain unchanged, and
that the public would not try to convert a large proportion of its deposits
into currency at the same time as it was converting currency board notes and
coins into reserve currency.
If the government had obtained the initial foreign reserves
by borrowing from the IMF, it would have to repay the IMF with funds other than
the net seigniorage of the currency board, because net seigniorage would now be
zero.