ArticlePDF Available

Central Bank Finances and Independence - How Much Capital Should a CB Have?



The issue of central bank (CB) capital and of rules for the distribution of CB profits to owners would appear, at first blush, to bear strong similarities to such questions for private corporations. After all central banks and private firms are incorporated within a similar legal structure and utilize similar accounting principles. However this
Central Bank Finances and Independence – How Much
Capital Should a CB Have?
Alex Cukierman
May 21 2006
1. Introduction
The issue of central bank (CB) capital and of rules for the distribution of CB
profits to owners would appear, at first blush, to bear strong similarities to such questions
for private corporations. After all central banks and private firms are incorporated within
a similar legal structure and utilize similar accounting principles. However this
resemblance in formal procedures hides several important differences. Unlike private
corporations, CB's are set up to achieve aggregate policy objective(s) rather than to
maximize profits. Unlike a private corporation, a negative net worth (or capital) at the CB
does not imply that the bank will go bankrupt and cease to operate. Finally, the main
owner of the CB is the government rather than private individuals implying that any
distribution of profits increases the spending power of government and that CB losses
ultimately translate into revenue losses or additional expenditures for the central
Until the mid eighties most CB's were dominated by governments and functioned
to a large extent as divisions of treasuries or ministries of finance. As such they were
utilized for a variety of (often conflicting) purposes such as helping finance government
expenditures, trying to stimulate economic activity and exports, and maintaining price
stability and financial stability. As a consequence the precise magnitude and sign of CB
capital was a relatively mute issue. The last two decades saw the emergence of a new
consensus according to which; (i) the CB should focus mainly on assuring price stability
even at the cost of substantial neglect of other objectives; (ii) the bank should be free to
set monetary policy instruments independently of other branches of government; (iii) as
was the case in earlier times, the CB is still expected to utilize its policy instruments to
safeguard the stability of the financial system, particularly during times of foreign
exchange and other financial crises.
This paper accepts those three principles as desirable features of modern
monetary policymaking institutions and focuses on two issues the first of which is
positive and the second normative. The positive issue concerns the implications of
alternative levels of CB capital and of rules for the distribution of profits to governments
for the extent to which the CB is able to set its monetary policy instruments without
interference from the political establishement. Obviously, when central banks functioned
mainly as arms of the treasury, this question was irrelevant since CB independence (CBI)
in the modern sense was mostly non existent and central banks were not expected to have
The normative issue concerns the existence of a potential tradeoff between
democratic accountability and CB independence. This tradeoff comes into play when the
occurrence of sufficiently large economic or political shocks force the CB to engage in
policies that have substantial fiscal implications like the bailout of a large segment of the
financial system. Since it involves a redistribution of resources such an action allows a
non elected institution (the CB) to make fiscal policy decisions. This is obviously
questionable on the grounds of democratic accountability and raises important questions
of institutional design and, given this design, the precise choice of a point along the
tradeoff between democratic accountability and CBI.
The remainder of the paper is organized in seven sections, the first three of which
focus on positive issues. Section 2 briefly discusses the conditions under which the level
of CB capital is likely to affect its independence and argues that independence is affected
only when this capital crosses a certain, probably negative, threshold. Section 3 reviews
sets of circumstances that lead to negative levels of CB capital. The channels through
which low, or negative, levels of CB capital might affect the conduct of monetary policy
are discussed in section 4.
The focus of the remaining four sections is mostly normative. Section 5 examines
ways to improve the thorny tradeoff between democratic accountability and CBI. Section
6 considers factors that should affect the desirable level of CB capital. The extent to
which concerns about CB capital should be relaxed (and if so, how) during extended
depressions is discussed in section 7. The last section discusses principles for the
allocation of profits between government and the CB with particular emphasis on
methods for rebuilding negative or insufficient levels of capital.
2. When does central bank capital affect its independence?
Unlike a private corporation, a CB is not liquidated when its capital becomes
negative. In spite of this, most modern central banks hold positive amounts of capital as
insurance against political interference.
The reason is that, when its capital is negative, it
is more likely that the CB will depend on infusions of funds from the Treasury opening
the door for various pressures on the bank to ease policy in order to contribute directly or
indirectly to financing of the deficit and/or to a (temporary) higher level of economic
activity. In such cases the Treasury may, explicitly or implicitly, condition the
recapitalization of the CB on certain policy actions.
When endowed with sufficient legal independence and positive levels of capital, it
is quite likely that, most contemporary CB's will be able to resist such pressures.
However, if at the time those policies are required, the bank already has a substantial
amount of negative capital, it is likely that the political establishement will have the
ability, and often the incentive, to stop, delay or severely limit such policy actions. This
suggests that the relation between independence and the level of CB capital is likely to be
discontinuous in the sense that below a certain threshold of negative capital the CB will
be seriously limited by political authorities, even if it enjoys a high level of legal
independence. But above this threshold the ability of the bank to conduct policy
independently does not depend, to a first approximation, on the level of CB capital.
It follows that maintainance of a sufficiently high level of capital is basically a
(probably partial) insurance against states of nature in which the bank’s ability to resist
There are some cases in which CB capital is negative. But, as we shall see later, those usually occur
following substantial changes in monetary regime and/or related institutions, suggesting that those
negative levels might not have been planned exante.
Such concerns may appear unrealistic for, well capitalized, major central banks like the Fed and the ECB.
But, as documented by Stella (2005), they were highly relevant for several Latin American countries with
negative CB capital and budgetary deficits.
the pressures of political authorities is weakened. Central bankers are aware of the
possibility that negative levels of capital might jeopardize their ability to choose policies
independently. Pronouncements by CB governors suggest that avoidance of such
situations occasionally affects policy decisions even when CB capital is clearly in the
positive range. For example a speech by the governor of the Bank of Japan (Fukui
(2003)) suggests that recent monetary policy in Japan was motivated, inter alias, by a
desire to shield the bank’s capital position from going into negative territory for reasons
that are apparent from the following quote
“Consider a case where, for whatever reason, a central bank's capital becomes depleted,
and the bank requires financial support from the government. The central bank might
either run into difficulties in conducting its policy and other business operations or might
cause the view to spread that it will, and eventually it will become difficult to maintain
public confidence in the currency.”
Interestingly, governor Fukui made this statement at a time in which conventional
wisdom was that the monetary policy of the Bank of Japan was not sufficiently
expansionary, partly because it was excessively concerned with avoidance of losses and
the strength of its balance sheet.
3. Circumstances conducive to losses and negative capital at the CB
Sustained losses leading to negative net worth at the CB often arise following
structural changes in the financial sector and/or, as a consequence of changes in monetary
regime. Both kinds of changes often occur following financial or exchange rate crises.
This section describes alternative conjunctions of economic circumstances and policies
that commonly result in negative capital at the CB.
1. During periods of credibility buildup the CB conducts contractionary monetary policy
in order to convince the public that it is serious about attaining and maintaining a low rate
of inflation. Such stabilization of inflation can be supported by an explicit inflation target
(as was the case in the UK since 1992) or without it (as was the case in the US in the
Krugman (1998) and Cargill (2005) have argued that the Bank of Japan commited policy errors similar to
those made by the Fed during the great depression. In their view monetary policy in both cases was overly
restrictive because the bank was excessively concerned with its net worth position. A fuller discussion of
this point of view within the context of negative CB capital appears in section 7.
early eighties). In either case the CB must maintain high interest rates during the period
of stabilization. Often, this creates situations in which the CB pays high interest rates on
its liabilities in order to attract deposits from banks and mop up liquidity.
If most of the CB assets are in domestic currency this does not necessarily lead to
losses since the interest rate on CB assets is also high. This is the case of the Federal
Reserve. However, if a substantial fraction of assets of the CB is in foreign currency
denominated assets that carry lower interest and the exchange rate is pegged, the average
return on the bank’s assets is lower than what the bank pays on its liabilities – which
leads to losses. When the period of stabilization streches over several years the negative
differential between the average return on asset and the interest paid on liabilities can
move a comfortably positive CB capital into the red. Such cumulative losses, due to a
stabilization effort have been experienced during the nineties in Israel and in several
former socialist economies.
2. The CB net worth position is also weakened when it utilizes its resources in order to
maintain a fixed peg or a narrow band in the face of market pressures for adjustment but
eventually abandons this policy and lets the exchange rate float. Such conjunction of
policies tends to weaken the net worth position of the bank independently of whether the
pressures are for devaluation or revaluation of the domestic currency.
In the first case, the bank sells foreign currency in order to prevent a devaluation. As
a consequence the bank’s net assets (assets minus liabilities) in foreign currency go
down. When the devaluation occurs the bank either incurs losses if its net foreign
currency denominated assets are negative, or enjoys smaller devaluation profits if net
foreign assets are still positive. In either case the net worth position is weakened in
comparison to a situation in which the bank did not try to stop the depreciation. An
example is the CB of Mexico following the exchange rate attack of 1994.
When the bank sterilizes capital inflows in order to prevent an appreciation of the
currency but ultimately abandons this policy, the net worth position is weakened again.
During the period of sterilization the bank’s net asset position in domestically
denominated assets goes down. Hence, when the appreciation occurs, the bank’s net
worth is lower than what it would have been if it had not engaged in the sterilization of
capital inflows.
3. Following banking and related financial crises the CB often assumes some of the
non performing assets and obligations of financial institutions in order to prevent panics
and maintain the stability of the financial system. This is often reinforced by political
pressures. Such bailouts obviously increase the risk of loss and weaken the net worth
position of the CB. Non performing loans due to past soft budget constraint were quite
common during the nineties in the former socialist economies and other developing
economies. Fiscal bailouts of non performing loans can be done explicitly through the
budget or by absorbing them into the balance sheet of the CB. When bailouts are
substantial the second method is likely to erode, as a byproduct, the instrument
independence of the CB.
4. The impact of negative CB capital on monetary policy
When, due to some combination of the reasons discussed in the previous section,
CB capital becomes sufficiently negative it is likely that the ability of the bank to conduct
policies that lead to additional losses will be compromised. If, at the time this happens,
the state of the economy calls for expansionary monetary policy this is not a binding
constraint since expansionary policy normally raises CB income through seignorage and
increases in the bank’s interest earning net assets.
However, if in order to achieve an inflation target or for other reasons,
contractionary policy is called upon, the likelihood that effective pressures will be
exerted on the bank to desist from such policy becomes a serious possibility. One can
easily imagine scenarios in which an increase in the policy rate by the bank will trigger
calls in the legislative and/or executive branches of government to stop the bank from
conducting policies that ultimately raise the net outlays of the Treasury. Other things the
same, and given current levels of CBI, such pressures are likely to be substantially less
effective when central bank capital is comfortably within the positive range.
The Czech National Bank went through such a process during the nineties and the early 2000 (Frait
Levels of CB capital amounting to about 3 percent of GDP are not uncommon. In some extreme cases like
Hong Kong, Switzerland and Malaysia this proportion is at least 10 pecent of GDP.
The likelihood that, in the presence of negative central bank capital, the bank’s
ability to conduct restrictive policies will be seriously limited is likely to be more in
evidence when the bank does not possess a portofolio of marketable government bonds.
In such cases contractionary monetary policy is usually conducted through reverse
repurchase agreements operations or similar arrangements. Under such schemes private
banks are induced to place deposits with the CB through a sufficiently high interest rate
making it more evident to lay legislators and others that the bank is engaging in loss
creating policies.
Consequently, effective political coalitions against contractionary
monetary policy are more likely to form when the bank offers high interest on deposits at
the bank, than when they take the form of open market sales.
The risk that negative capital at the CB will limit the bank’s ability to conduct
contractionary monetary policy is also greater in countries with narrow domestic capital
markets. In such countries the supply of funds to government is relatively inelastic. As a
consequence, when government needs to run deficits it is more dependent on seignorage
than governments of countries with wide capital markets like the US and the UK. In the
presence of negative CB capital, such governments are likely to oppose contractionary
monetary policy more vigorously than the political establishements of countries with
wide capital markets.
Discussion in Stella (2005) suggests that, more often than not, CB losses leading
to substantial negative net worth at the CB interfered with the ability to conduct
restrictive policies and ultimately with the CB ability to maintain price stability.
Examples discussed by Stella include, among others, Peru, Bolivia, Uruguay, Paraguay,
Costa Rica and Venezuela during the nineties.
However this does not necessarily imply that negative levels of CB capital always
prevent the achievement of price stability. In spite of increasingly negative levels of CB
capital, Chile managed to successfully reduce inflation over the last decade by means of
inflation targeting.
This remarkable performance was made possible through a sustained
policy of budgetary surpluses on the part of government. In the presence of such
Open market sales of marketable securities have, in principle, a similar effect since they also reduce the
net interest bearing assets of the CB. However the link between CB losses and such sales is less visible, at
least for non specialists.
Stella notes that the CB of Chile consistently had losses of about one percent of GDP per year during the
last decade.
surpluses the risks to price stability due to negative CB capital are obviously negligible.
The reason is that a main governmental incentive to apply pressures against restrictive
monetary policy by the CB is non existent in the presence of sustained surpluses. The
Chilean case suggests that negative levels of CB capital may have very different impacts
on price stability depending on the long run stance of fiscal policy.
5. How to improve the tradeoff between independence and democratic
Macroeconomic theory usually treats fiscal and monetary policies as distinct from
each other. Although this is a useful first order approximation there are cases in which
monetary policy actions have fiscal implications. Conversely, decisions about the type of
exchange rate regime, which usually are under the authority of government and the
treasury, affect the degree of instrument independence of the CB. In what follows an
example of a situation in which monetary policy decisions have fiscal implications is
provided and an elaboration of the link between the exchange rate regime and the degree
of instrument independence of the CB are provided.
Since the Bank of Israel does not have a stock of marketable government
securities there is an agreement between the bank and the treasury according to which the
bank has the authority to issue and retire earmarked short term government obligations
solely for the purpose of conducting monetary policy.
Although the proceeds from such
issues cannot be used for the financing of budgetary deficits, restrictive monetary policy
decisions by the bank lead, given this arrangement, to increases in the size of government
In central banks that use open market sales or reverse repos to implement
restrictive monetary policy the fiscal implications of restrictive policies, although less
visible at first blush, are nonetheless present. They take the form of a smaller amount of
seignorage paid to government at the end of the year (when seignorage net of expenses is
positive) or a larger negative capital that ultimately has to be funded by the fiscal
authority (when seignorage net of expenses and CB capital are negative).
Further details appear in Cukierman (Forthcoming).
Consider next the impact of the exchange rate regime on the instrument
independence of the CB. When the exchange rate is pegged (through a currency board or
by some other means ) the bank is forced to subjugate its interest rate policy to the fixed
exchange rate objective. When the exchange rate is more flexible, as is the case under
wide bands or fully flexible regimes, the bank is free to move the interest rate in order to
achieve other objectives like stabilization of inflation around a target rate and
stabilization of the output gap. As we saw in section 3 a fixed peg that is ultimately
abandoned, normally because of a governmental decision, reduces the capital position of
the CB. Thus, decisions made by political authorities may, depending on the type of
exchange rate regime, induce negative levels of CB capital.
The previous discussion suggests that monetary and fiscal policy decisions are
more interwoven than textbook models would lead one to believe. In particular, both
fiscal and monetary policy decisions affect the profits of the CB and its capital position.
The existence of such interactions makes the choice of institutions designed to achieve a
reasonable tradeoff between democratic accountability and CBI particularly tricky.
The main issue can be stated as follows. Instrument independence on the part of
the CB allows it (at least in some cases) to make decisions that have fiscal implications.
This violates the principle of democratic accountability since CB officials are non elected
technocrats. On the other hand, decisions made by political authorities with respect to
matters such as the type of exchange rate regime, or the financing of bailouts affect the
instrument independence of the CB through its balance sheet position. This violates the
principle of CBI. Obviously there is a tradeoff between those two principles.
This raises important institutional design questions about ways to make this
tradeoff less acute. Those questions have not received much treatment in the existing
economic literature and I do not have definitive institutional blueprints for their
resolution. Instead I offer, in what follows, several practically oriented principles that, in
my view, are likely to ameliorate the tradeoff between CBI and democratic
1. In non extreme circumstances let the CB alone decide about monetary policy, even if
some of those decisions have fiscal implications.
2. When extreme circumstances like financial crises necessitate the enactment of
monetary policies that have substantial fiscal implications a joint decision of the CB and
of the political authorities is advisable.
3. Delimitation of threshold levels of monetary policy with fiscal implications above
which government and the CB would be urged to reach a joint decision should be
formulated in advance. These thresholds should be set at levels that would make the
probability of fiscal interference with monetary policy matters small. Obviously the
precise circumstances and levels of thresholds above which government would
temporarily become involved in monetary decisions generally involve both positive and
normative considerations.
4. If, due to a financial crisis or other reasons, government decides to rescue financial
institutions the implementation of such operations should not affect the net capital
position of the CB. This implies that such operations should appear as explicit items on
the government’s budget. Such an arrangement is desirable not only because it protects
the instrument independence of the bank, but also because of transparency and
accountability considerations of politicians to the public that elected them. Fry, Goodhart
and Almeida (1996) note that governments in many developing countries tend to burden
the CB with such operations precisely because they try to avoid transparency and
accountability. In their language, governments “are quite content to hide the fact that they
are squeezing the goose that lays the golden eggs”.
5. More generally, the CB net worth better be shielded from the impact of decisions that
are made by other authorities. For example if government decides on an exchange rate
peg forcing the CB to defend it, and then decides to abandon the peg, creating CB losses,
those losses should be recapitalized.
6. How much capital should a CB have?
The previous section proposed a number of principles designed to obtain a
favorable tradeoff between CBI and democratic accountability but did not address the
more specific question regarding the desirable magnitude of central bank capital given
the existing institutions of each particular country. Since countries differ in terms of their
political systems, economic shocks, exchange rate regimes, financial institutions and
other dimensions one answer does not fit all. This section brings up some qualitative
considerations that are relevant for the choice of CB capital for countries with given
economic structure and institutions.
1. An important determinant of the desirable level of CB capital is the magnitudes of
shocks that monetary policy is expected to react to . Other things the same, larger
variances of such shocks call for larger levels of capital to achieve a given level of
insurance against loosing the ability to conduct monetary policy independently.
2. The width of areas of responsibility of the CB is a second factor. The more numerous
those areas, the larger the recommended level of capital. For example, CB's that do not
manage foreign exchange reserves and/or are not involved in the supervision and
regulation of the financial system can have lower levels of capital. Since, in developing
countries it is likely that the CB will have wider responsibilities, a corrolary of this point
is that, other things the same, the desirable level of CB capital should be higher in
developing countries. Further, since the CB is also likely to have wider responsibilities in
countries with relatively narrow capital markets, an additional corrolary is that, other
things the same, CB capital should be higher in countries with narrow capital markets.
3. The higher the tendency of government to create deficits the more important it is to
protect the independence of the bank, and therefore, the higher should its level of capital
be. Similarly, the more politically unstable is government the higher the importance of
CBI and the higher the desirable level of CB capital.
4. The nature of other institutional arrangements governing the relation between
government and the CB should also affect the desirable level of capital. CB capital is
only one component of a package of institutional arrangements governing the relation
between government and the CB. Of particular importance is the nature of the exchange
rate regime. If the institutional arrangements between the bank and the treasury are such
that the bank is expected to be the sole defendent of a fixed peg, it should have a higher
level of capital than if the responsibility for defending the exchange rate rests wholly or
partially with the treasury.
5. By affecting the probability of sizable losses the structure of the bank’s balance sheet
by currency of denomination should also affect the desirable level of CB capital. Since
they issue currency and hold the reserves of the banking system, the bulk of CB liabilities
is generally in domestic currency. However, there are substantial differences between
central banks in the fraction of their assets that is denominated in foreign exchange. At
one extreme of the spectrum is the US in which the bulk of CB assets is in domestic
currency. At the other extreme are small open economies with fixed pegs, like Hong
Kong, in which the bulk of CB assets is denominated in foreign exchange. Sims (2004),
who refers to those two extreme types as F (for Fed) and E central banks respectively,
notes that the first type of bank is perfectly hedged in terms of currency risk while the
second type assumes substantial currency risks.
The larger the currency mismatch between the currency composition of assets and
liabilities, the larger the level of CB capital needed to cushion against CB losses due to
changes in the exchange rate. Consequently central banks with larger fractions of foreign
exchange denominated assets should have higher capital. A comparison of the levels of
CB capital in the US and Hong Kong as representatives of extreme types of central banks
is consistent with the view that CB actually follow this principle. The Fed’s capital is less
than one quarter of a percent of GDP while the capital of Hong Kong’s monetary
authority is over thirty percent of GDP.
6. The level of CB capital may also affect the credibility of the CB for being committed
to the maintainance of a low rate of inflation. If individuals understand that a sufficiently
negative level of CB capital impairs the ability of the bank to conduct policy
independently, a sufficiently negative level of CB capital may induce an increase in
inflationary expectations and reduce credibility. This may occur even if the bank enjoys
high legal independence. However such a scenario need not necessarily take place if
other factors suggest to the public that government policy at the time is likely to be
conservative. For example if, as was the case in Chile, government consistently maintains
budgetary surpluses and is perceived to be generally tough on inflation, negative capital
at the CB need not affect credibility. The upshot is that, in assessing the impact of
negative CB capital on credibility one should consider it in conjunction with the stance of
fiscal policy, related institutional arrangements and the state of the economy.
The ECB, with about one third of assets in foreign exchange, occupies an intermediate
position between those polar cases.
7. Should one be concerned about negative capital during extended
The discussion of the previous section was geared towards the avoidance of
situations in which, due to the accumulation of substantial losses, it is politically difficult
for the bank to engage in further loss creating policies, when such policies are called
upon. Underlying the focus on avoidance of such scenarios is the belief that more often
than not, when expansionary policies are called upon, they are not met by the same level
of resistance from the political establishement as is the case for contractionary policies.
However during deep depressions conventional economic wisdom is that the CB
should flood the economy with liquidity even at the cost of absorbing a substantial
amount of bad debts. Friedman and Schwartz (1963) convincingly argue that the severity
of the great depression in the US would have been smaller if the Fed had been less
concerned with safeguarding its net worth. Krugman (1998), Cargill (2005) and others
have argued that in the face of the severe recent depression in Japan, the Bank of Japan
made similar policy errors partly because of an excessive concern with CB losses that
could lead to negative levels of capital. In particular, Bernanke (2003) argues that the
Bank of Japan held down the level of open market purchases to avoid potential capital
losses due to rising interest rates after the depression is over.
Does this imply that, in the face of a deep and extended depression, the CB should
forget about the impact of negative capital on CBI? The answer is not a simple yes or no.
On one hand, massive open market purchases in the face of a deep depression should not
be inhibited. On the other hand, in its role as the authority in charge of long run price
stability the Bank of Japan concerns about the impact of its capital on long run CBI is not
This is clearly one of the situations that calls for consultations and cooperation
between the CB and the Treasury since the large capital losses to be incurred by the bank,
if massive open market purchases are successful, ultimately have implications for the
balance sheet of consolidated government as well as for CBI. Bernanke (2003) makes a
creative suggestion that addresses both issues. He proposes a quid pro quo, in which the
Ministry of Finance acts to immunize the Bank of Japan’s balance sheet from interest rate
risk, and in return the bank increases its purchases of government bonds. This opens the
door for larger open market purchases while addressing the legitimate concerns of the
Bank of Japan about its balance sheet position.
8. Allocation of profits between the CB and government
Optimal taxation considerations of the Ramsey type imply that seignorage should
be used along with other taxes up to the point at which the marginal distortions of each
type of tax are equalized. This point of view implies that government and/or the treasury
should be allowed to determine monetary policy along with other types of taxes.
However, as is well known, due to dynamic inconsistency and short horizons
governments are subject to an inefficient inflation bias (Kydland and Prescott (1977),
Barro and Gordon (1983)).
In addition, leaving the discretion over seignorage in the hands of government
injects electoral considerations and political instability into monetary policy. This implies
that governments are not very likely to manage seignorage in line with the Ramsey
principle. As a matter of fact, since the inflation tax can be imposed without any
legislation and is less visible than other taxes, it is likely to be overused by political
authorities. Current conventional wisdom recommends, therefore, rules rather than
discretion in the allocation of seignorage between the CB and government.
Due to secular growth and the associated increase in the demand for money,
central banks accumulate positive amounts of seignorage which normally are
substantially higher than the operating expenses of the bank. Those funds ultimately
belong to society and could be used to finance part of government expenditures but, due
to reasons discussed above, should be governed by transparent rules rather than by
governmental discretion.
Today most countries recognize the wisdom of this approach
In related work Jeanne and Svensson (2005) propose to utilize the balance sheet concerns of independent
central bankers to produce a credible commitment to high inflation in order to make the real rate of interest
negative in spite of the fact that, during depressions, the zero bound on the nominal rate of interest is
The first four sections in Chapter 4 of Cukierman (1992) summarize the argument in favor of rules
rather than discretion in the management of seignorage. Due to related concerns most central bank laws
also prohibit the CB from lending to government (Cukierman (1992), chapter 19).
This is the case in the US and the ECB. Due to the fact that it is the central banker of many countries
whose governments are potential competitors over its seignorage revenues, the rules for distribution of the
ECB seignorage do not leave much room for discretion and are highly transparent.
and have fixed rules for the allocation of profits between the CB and government.
However those rules occasionally leave some room for expost negotiations between
government and the CB, and are not always very transparent. In some cases they open the
door for evasion of deficit limits.
There is little doubt that clear rules for the allocation of CB profits between
government and the CB constitute a good practice benchmark in normal times. However
such rules should also contain contingencies for cases in which, due to exceptional
circumstances, CB capital becomes negative. In the absence of such contingencies the
rules should be sufficiently flexible to allow the bank to rebuild its capital to the desirable
level within a reasonable period of time.
This could be done by allowing the bank to retain a larger fraction of its profits
following substantial losses.
Implementation of such capital rebuilding procedures
appears to be particularly important when the negative capital at the bank largely
originates from policy decisions taken by government in areas that are not under the
direct authority of the CB. Examples include decisions to defend a fixed peg or to bail out
financial institutions through the CB. In the absence of such capital rebuilding
mechanisms government would have the ability to severely reduce the actual
independence of the bank through the negative capital “back door” even if the legal
independence of the bank was high.
The argument is sometimes made that the CB should not worry about the effect of
negative capital on its independence since it can always print money to replenish
However this argument disregards the fact that such policy may conflict with
the policy objectives facing the bank at the time. In particular, an attempt to rebuild
negative CB capital by relying on money creation may conflict with policy measures
needed to achieve an inflation target. It would appear, therefore, that other methods are
Although it is not possible to anticipate all possible contingencies, an exante
specification of the mechanisms through which CB capital is to be rebuild in case of need
Some CB have actually implemented crude versions of such a mechanism.
See Frait (2005) for example.
(subject to feasibility constraints), is better than expost negotiations between government
and the CB. Clearly specified and transparent exante principles regarding the allocation
of profits between the CB and government in case CB capital is seriously depleted,
safeguards CBI and enhances the credibility of the commitment to price stability by both
government and the CB.
Bernanke B. S. (2003), “Some Thoughts on Monetary Policy in Japan”, Speech before
the Japan Society of Monetary Economics, Tokyio, Japan, May 31, Web address:
Barro R. J. and Gordon R. (1983), "A Positive Theory of Monetary Policy in a Natural
Rate Model", Journal of Political Economy, 91, 589-610.
Cargill T. (1985), “Is the Bank of Japan Financial Structure an Obstacle to Policy?” IMF
Staff Papers, 52, 2, Novenber, 311-334.
Cukierman A. (1992), Central Bank Strategy, Credibility And Independence-Theory
and Evidence. The MIT Press, Cambridge, MA.
Cukierman A. (Forthcoming), "Legal, Actual and Desirable Independence: A Case Study
of the Bank of Israel " in N. Liviatan and H. Barkai (eds), The Bank of Israel: A Monetary:
History, Volume II, Oxford University Press.
Web address
Frait J. (2005), “Exchange Rate Appreciation and Negative Central Bank Capital: Is
There a Problem?” Paper presented at the Expert Forum on Central Bank Finances and
Impact on Independence, Bank of England, August/September 2005.
Friedman M. and A. Schwartz (1963), A Monetary History of the United States,
Princeton University Press, Princeton, NJ.
Fry M., C. A. E. Goodhart and A. Almeida (1996), Central Banking in Developing
Countries: Objectives, Activities and Independence, Routledge, London, UK.
Fukui T. (2003), “Challenges for Monetary Policy in Japan”, Speech at the Spring
: Web address. 1June , Meeting of the Japan Society of Monetary Economics
Jeanne O. and L. E. O. Svensson (2005), “Credible Commitment to Optimal Escape from
a Liquidity Trap: The Role of the Balance Sheet of an Independent Central Bank”,
November. Web address:
Krugman P. (1998), “Its Baaack! Japan’s Slump and the return of the Liquidity Trap”,
Brookings Papers on Economic Activity, 2,, 137-187.
Kydland F. E. and Prescott E. C. (1977), "Rules Rather Than Discretion : The
Inconsistency of Optimal Plans", Journal of Political Economy, 85, 473-9.
Sims C. (2003), “Fiscal Aspects of Central Bank Independence”, in: Sinn H. W., M.
Widgren and M. Kothenburger (eds.), European Monetary Integration, 2004, MIT
Press. Web address:
Stella P. (2005), “Central Bank Financial Strenght, Transparency and Policy Credibility”,
IMF Staff Papers, 52, 2, November, 335-365.
... When central bank loses its autonomy, public may also doubt on its policy flexibility. Cukierman (2011) suggested that most central banks hold positive amount of capital as insurance against political interference in their policy. For instance, the recapitalization from government can probably make central bank fall into pressure to ease monetary policy in order to contribute to finance deficit or to stimulate economic growth. ...
... Some other literatures also suggested that having 24 Many mistakenly interpret the Bank of Japan governor speech in 2003 as a case of excessive concern with avoidance of losses in the balance sheet as part of the reason for the monetary policy of the Bank of Japan that was not sufficiently expansionary. (Cukierman, 2011) 25 Non-linear relationship implies that the impact of financial status on policy outcome occurs in the extreme financial loss case. an outstanding communication with stakeholders about the rationale of its policy and having strong governance are essential for the central bank survival. ...
... This also gives implication on central bank financial strength and its credibility. Cukierman (2011) suggested that greater width of areas of responsibility, the larger recommended level of capital as a risk buffer. ...
... Unlike a private institution that depends on earnings and solvency to function, strictly operational concerns are less important for the Fed largely because of the unique status of a central bank as the monopoly holder of a currency franchise. The central bank can "print money" to satisfy financial obligations and even operate with negative equity (see Cukierman, 2011;Archer and Moser-Boehm, 2013;Hall andReis, 2013, andDel Negro andSims, 2015 for further details). Furthermore, regardless of its portfolio losses or income expenses, the Fed still has operational control of short-term interest rates because its ability to pay interest on bank reserves allows it to conduct monetary policy independently of the size of its balance sheet. ...
... However, as noted above, severe political fallout could be triggered by large declines in the Fed's income and remittances to the Treasury or by large capital losses (even mark-to-market unrealized losses, which the Fed does report as an adjunct to its official balance sheet). The repercussions of such political problems are difficult to gauge but could be quite severe (Cukierman, 2011). Goodfriend (2014) argues that the Fed's "operational credibility" depends on the public's confidence that the independent pursuit of monetary policy involves little fiscal cost and so does not invade the province of the fiscal authority. ...
... A negative balance sheet will therefore not have the central bank cease to exist but rather impacts the spending power of the national government. 189 In the Eurozone, the ownership of the NCBs is regulated in national law. Some NCBs such as the Dutch and German bank are state-owned, others have private or mixed ownership. ...
Full-text available
This article will discuss the legal framework of Central Bank Digital Currency (CBDC) in the Eurozone. The different designs proposed by the European Central Bank will be examined against the legal framework. The first question that will be examined is whether CBDC could be introduced as part of the issuance of banknotes and be classified as legal tender. It will then consider whether the CBDC can be used as a monetary policy instrument under the ECB's monetary mandate. To do so this paper will consider two different types of CBDC. The wholesale CBDC is only accessible to financial institutions and the general purpose CBDC which would be accessible to consumers. This paper concludes that CBDC can be introduced as legal tender.
... Two well-known thresholds are considered for the maximum loss a central bank may bear: the central bank equity level and the central bank insolvency threshold. Breaching the first threshold entails political and credibility risks (Jeanne & Svensson (2007), Cukierman (2011)), while going insolvent implies a loss of control over inflation (Reis, 2013). We then develop a model of exchange market pressure suitable for a peg and consistent with the new central banking framework that is able to forecast the amount of foreign exchange interventions a central bank should carry out to maintain a peg. ...
... Another strand of the literature focuses on the central bank balance sheet as a supplemental tool to the policy rate in affecting macroeconomic outcomes. In many advanced economy central banks, the main policy instrument is a target rate for a short-term money 6 Refer to Archer and Moser-Boehm [2013], Cukierman [2011], and Stella and Lonnberg [2008]. 8 market rate. ...
... Schobert (2008), for example, reports that of the 8% of annual financial statements surveyed (of 108 central banks, between 1984 and 2005) where losses were reported, the great majority had sterilisation costs or exchange rate losses as the biggest expenditure items. 27 Cukierman (2011) suggests that monetary regime changes and structural changes to the financial sector are both conducive to loss-making by the central bank, especially in countries with narrow financial markets. We will also suggest that part of the reason is grounded in the nature of financial systems in less advanced economies, and is thus structural (Section 1 in Part C). ...
Technical Report
Full-text available
Full-text available
This paper explores the level of independence of the National bank of the Republic of Macedonia by primarily focusing on the legal provisions that pertain to the key aspects for achieving and maintaining price stability. It provides a historical perspective of the evolution of the independence since the first years of transition. The assessment of the independence of the NBRM is based on the index of Cukierman, Webb, and Neyapti (1992), as one of the most commonly used indices, and the index of Jacome and Vazquez (2005), which incorporates some specific aspects relevant for transition economies. Both indices indicate that the legal independence of the NBRM has increased over the years and that the current legal framework provides a high level of independence. Yet, it should be emphasized that there is a room for further strengthening, in particular in the areas of policy formulation and the process of appointment of the non-executive members of the council of the NBRM. As the indices are based on the legal provisions, they can serve only as an indication of the actual independence of the central bank.
Looking at the evolution of the central bank’s balance sheet gives us a unique window on the forces that have shaped our economy and central bank reaction functions. This paper considers the evolution of the NBU’s balance sheet over the period from 2001-2016, focusing on explicit and implicit monetary policy priorities at different periods. We then make simulations on the NBU’s balance sheet for the next five years assuming current NBU priorities for monetary policy. We then draw conclusions on the likely financial position of the NBU in the future and recommendations to ensure the NBU’s financial strength – essential for its continued independence.
The Dutch bank florin was the dominant currency in Europe over much of the seventeenth and eighteenth centuries. The florin, a fiat money, was managed by an early central bank, the Bank of Amsterdam. We analyze the florin’s loss of “reserve currency” status over the period 1781-92, using a new reconstruction of the Bank’s balance sheet. The reconstruction shows that by 1784, accommodative policies rendered the Bank policy insolvent, meaning that its net worth would have been negative under continuation of its policy objectives. Policy insolvency coincided with the Bank’s loss of control over the value of its money. © 2016 International Journal of Central Banking. All Right Reserved.
In this paper we show that the degree of central bank independence influences the optimal choice of monetary policy strategy during potentially unsustainable asset price booms. We assume that central bankers have to choose between a policy that preemptively raises short-term real interest rates in the boom phase to prevent the build-up of a financial market crisis scenario and the cleaning-up strategy that ignores its impact on the likelihood of a future crisis. We find that the more independent central bankers are, the more likely it is that they refrain from implementing preemptive monetary tightening to maintain financial stability. These results stand in sharp contrast with the seemingly predominant view that central bank independence fosters financial stability. The intuition underlying these results is that a preemptive interest rate hike gives rise to, among other things, a lower inflation rate in the boom period. Whether this disinflation creates additional costs or benefits depends on the degree of central bank independence. It can benefit dependent central bankers who otherwise would suffer from a higher inflation bias; however, for independent central bankers, this disinflation leads to an undesirable undershooting of their inflation target.
Full-text available
Natural-rate models suggest that the systematic parts of monetary policy will not have important consequences for the business cycle. Nevertheless, we often observe high and variable rates of monetary growth, and a tendency for monetary authorities to pursue countercyclical policies. This behavior is shown to be consistent with a rational expectations equilibrium in a discretionary environment where the policymaker pursues a "reasonable" objective, but where precommitments on monetary growth are precluded. At each point in time, the policymaker optimizes subject to given inflationary expectations, which determine a Phillips Curve-type tradeoff between monetary growth/inflation and unemployment. Inflationary expectations are formed with the knowledge that policymakers will be in this situation. Accordingly, equilibrium excludes systematic deviations between actual and expected inflation, which means that the equilibrium unemployment rate ends up independent of "policy" in our model. However, the equilibrium rates of monetary growth/inflation depend on various parameters, including the slope of the Phillips Curve, the costs attached to unemployment versus inflation, and the level of the natural unemployment rate. The monetary authority determines an average inflation rate that is "excessive," and also tends to behave countercyclically. Outcomes are shown to improve if a costlessly operating rule is implemented in order to precomrnit future policy choices in the appropriate manner. The value of these precommitments -- that is, of long-term agreements between the government and the private sector -- underlies the argument for rules over discretion. Discretion is the sub-set of rules that provides no guarantees about the government's future behavior.
Full-text available
A central bank is financially strong if it possesses resources sufficient to attain its fundamental policy objective(s). Once endowed with those resources, relations between government and central bank should be designed so that significant changes in central bank financial strength do not occur unless necessitated by changes in policy objectives. The level of strength required depends on the array of policy objectives (for example, the exchange rate regime) as well as the constraints and risks presented by the operational environment. Attaining credibility is facilitated if the public can easily determine the financial strength of the bank, yet for a variety of reasons this is often difficult. Transparency requires institutional arrangements that ensure the central bank generates profit in most states of the world, is subject to strict ex post independent audit, and transfers regularly all profits, after provisions, to the treasury.
Central bank capital and accounting measures of capital adequacy potentially constrain central bank policy outcomes. Historical and institutional factors explain why central banks are organized as public corporations; however, capital structure design provides little predictive insight into policy outcomes. In fact, focusing on accounting measures of capital adequacy and similar performance indicators potentially interferes with monetary policy, especially in extraordinary economic circumstances such as deflation. The Bank of Japan, like the Federal Reserve in the 1930s, has overemphasized accounting measures of central bank performance at the cost of nonoptimal policy outcomes.
Alex Cukierman is well known for his work on central bank behavior. This book brings together a large body of Cukierman's research and integrates it with recent developments in the political economy of monetary policy. Filled with applications and carefully worked out technical detail, it provides a valuable comprehensive analysis of central bank decisions, of the various effects of policy on inflation, and of the feedback from inflationary expectations to policy choices. Cukierman uncovers and analyzes the reasons for positive inflation and rates of monetary expansion. He shows that the money supply, and therefore inflation, are not exogenous. They are influenced by interactions involving distributional considerations, private information, personal motives, and the political environment. This point of view makes it possible to identify the institutional, political, and other features of a country that may be conducive to inflationary environments. Cukierman presents new multidimensional evidence on both legal and actual central bank independence for a sample of up to 70 countries and uses it to investigate the interconnections between the distributions of inflation and of central bank independence. He takes up such issues as why some countries have more independent central banks than others and identifies reasons for the substantial cross country variation in seigniorage. He provides positive explanations for the tendency of central banks, like the US Federal Reserve, to smooth interest rates and to be secretive. Observing that it is likely that the European Economic Community will have a monetary union before the turn of the century, Cukierman applies the techniques of modern political economy to discuss the effect of this change on the commitment to price stability. The book includes simple and advanced materials as well as informal summaries of the major technical results. The introduction contains a modular guide for reading and teaching the material.
Sumario: Motives for monetary expansion under perfect information -- Asymmetric information and changing objetives under discretion -- Velocity shocks, politics, signaling, inflation persistence, and accommodation -- Central Bank independence and policy outcomes: theory and evidence
Even if there is an agreed-upon, fixed social objective function and policymakers know the timing and magnitude of the effects of their actions, discretionary policy, namely, the selection of that decision which is best, given the current situation and a correct evaluation of the end-of-period position, does not result in the social objective function being maximized. The reason for this apparent paradox is that economic planning is not a game against nature but, rather, a game against rational economic agents. We conclude that there is no way control theory can be made applicable to economic planning when expectations are rational.