Working Paper Series: The Role of Central Bank Capital Revisited

WO R K I N G PA P E R S E R I E S
N O. 3 9 2 / S E P T E M B E R 2 0 0 4
THE ROLE OF
CENTRAL BANK
CAPITAL REVISITED
by Ulrich Bindseil,
Andres Manzanares
and Benedict Weller
WO R K I N G PA P E R S E R I E S
N O. 3 9 2 / S E P T E M B E R 2 0 0 4
THE ROLE OF
CENTRAL BANK
CAPITAL REVISITED 1
by Ulrich Bindseil ,
Andres Manzanares
and Benedict Weller 2
In 2004 all
publications
will carry This paper can be downloaded without charge from
a motif taken http://www.ecb.int or from the Social Science Research Network
from the
€100 banknote. electronic library at http://ssrn.com/abstract_id=586763.
1 The views expressed in this paper are those of the authors and do no necessarily reflect those of the ECB.We wish to thank
Ignazio Angeloni, Denis Blenck, Carsten Detken,Vitor Gaspar, Ian Ingram, Joachim Keller, Simone Manganelli, Klaus Masuch,
Francesco Papadia , Leo von Thadden and an anonymous referee for useful discussions and comments on a previous
version of the paper.
2 Correspondence address: European Central Bank, Kaiserstrasse 29, D-60311, Frankfurt am Main, Germany;
e-mail: Ulrich.Bindseil@ecb.int, Andres.Manzanares@ecb.int, Benedict.Weller@ecb.int
© European Central Bank, 2004
Address
Kaiserstrasse 29
60311 Frankfurt am Main, Germany
Postal address
Postfach 16 03 19
60066 Frankfurt am Main, Germany
Telephone
+49 69 1344 0
Internet
http://www.ecb.int
Fax
+49 69 1344 6000
Telex
411 144 ecb d
All rights reserved.
Reproduction for educational and non-

commercial purposes is permitted provided
that the source is acknowledged.
The views expressed in this paper do not

necessarily reflect those of the European
Central Bank.
The statement of purpose for the ECB

Working Paper Series is available from the
ECB website, http://www.ecb.int.
ISSN 1561-0810 (print)

ISSN 1725-2806 (online)
CONTENTS
Abstract 4
Non-technical summary 5
1 Introduction 7
2 The literature on central bank capital 8
3 A simple model based on a central bank
with no liquidity constraints 13
4 Monte Carlo simulations and the long run
steady state of the central bank balance
sheet and inflation in the base model 17
5 Why central bank capital matters –
liquidity constraints, central bank
credibility and independence 23
6 Some considerations regarding profit
sharing rules 30
7 Conclusion 32
Annex 1: Proof of the proposition of long term
return to profitability and positive capital 33
Annex 2: Proof of the proposition that if a
central bank has larger starting capital it will
always have a longer balance sheet 34
References 35
European Central Bank working paper series 36
ECB
Working Paper Series No. 392
September 2004 3
Abstract
This paper explores the role of central bank capital in ensuring that central banks focus on price
stability in monetary policy decisions. The paper goes beyond the existing literature on this topic
by developing a simple, but comprehensive, model of the relationship between a central bank’s
balance sheet structure and its inflation performance. The first part of the paper looks at
solvency, i.e. under which conditions the “economic” capital (i.e. the discounted long term P&L)
of a central bank always remains positive, despite adverse shocks, assuming a stability oriented
monetary policy. The second part shows that in practice, capital is important for central banks
beyond the issue of positive economic capital, when taking realistic assumptions regarding
central bank independence. Capital thus remains a key tool to ensure that central banks are
unconstrained in their focus on price stability in monetary policy decisions.
Keywords: Central Bank Capital, Central Bank Independence

JEL classification: E42, E58
ECB
4 Working Paper Series No. 392

September 2004
Non-technical summary
Although it is generally acknowledged that central banks need a certain level of capital in
order to achieve their monetary policy objectives, i.e. a low level of inflation, the explanations
of the causation and exact nature of the relationship have often remained vague and the
underlying assumptions have not been spelled out.
This paper revisits the role of central bank capital using a specific model of the evolution of
the central banks’ profitability and balance sheet structure in the context of a basic
macroeconomic setting. This deliberately simple model of central bank capital intends to
illustrate under what conditions a low (or even negative) level of capital would not have
harmful effects on the ability of the central bank to achieve its monetary policy target. The
model describes the evolution of the central bank’s balance sheet within a macroeconomic
context, under certain assumptions about both the mechanics of the economy and the
institutional constraints faced by the central bank. Under these assumptions, it is shown that a
temporary shock creating negative capital and a loss-making situation is always reversed in
the long run with the central bank returning to profitability and a positive level of capital.
There are two exceptions: when the economy falls into a deflationary trap, from which it is
not possible to escape; and, second, when the growth rate of the demand for banknotes falls
short of nominal interest rates. However, a central bank with a loss-making balance sheet
structure would in this context still able to conduct its monetary policy in a responsible way,
even with a negative long-term profitability outlook. Hence, some other factors not included
in such a “mechanical” model must be considered in order to explain the empirical evidence
regarding the negative correlation between inflation performance and financial strength of
central banks.
In order to do so, we address the relationship between central bank capital and its credibility
to perform its monetary policy functions. It is likely that there is another set of factors, related
to the institutional environment in which the central bank exists, that is causing a relationship
between the weakness in the central bank’s financial position and its inability to control
inflation. In practice, a central bank can never achieve an absolute, guaranteed institutional
independence. In particular, no government can commit future governments (whether they
obtain power by election, war, or revolution) not to change the central bank law or abolish its
exclusive right to issue legal tender. Other, admittedly unlikely scenarios limiting the infinite
solvency of the central bank, like a de facto dollarisation, have never a zero likelihood in the
eyes of the public. It is hence important to distinguish between a situation where an assumed
absence of central bank’s concerns about balance sheet strength is shared by the public and
the State, and the more realistic situation where prolonged central bank losses would lead to a
complete loss of credibility. Even if a central bank is not subject to liquidity constraints,
ECB
September 2004 5
credibility is indeed positively correlated with the level of capital when some extensions to
the model are included, allowing some degree of concern about the profit and loss account to
affect the central bank’s interest rate setting behaviour.
Finally, the paper briefly discusses profit sharing rules, which are obviously closely linked to
the capital issue. The main conclusion is that a fully automated and fully credible rule of re-
capitalisation by the government of the central bank in case of losses can be regarded as a
substitute for positive capital. Since such rules are however difficult to implement in practice,
positive capitals seems to remain a key tool to ensure that independent central bankers always
concentrate on price stability in their monetary policy decisions.
ECB

September 2004
1. Introduction
In the last two decades the issue of commercial banks’ capital adequacy has become one of the most
analysed topics in banking and finance. Banks, regulators and academics have devoted substantial
resources to developing capital adequacy regulations that could be generally applied to banks
everywhere in the world. Central banks, too, normally have a capital position in their balance sheet.
However, there are no rules requiring central banks to hold a certain level of capital: indeed, cross-
country surveys reveal that they have been taking rather different approaches. The rationale for credit
institutions to hold capital obviously cannot be translated one-to-one to central banks. For instance, the
central bank normally cannot become illiquid as long as the currency it issues is legal tender.
Nevertheless, recent studies have concluded that positive central bank capital is required if the central
bank is to perform its tasks successfully, in particular, to achieve price stability. However, this
conclusion sometimes seems to be drawn from anecdotal evidence without providing an exact analysis
of the nature of the relationship and the underlying assumptions. This paper, therefore, revisits the role
of central bank capital on the basis of a specific model of the evolution of the central banks’
profitability and balance sheet structure in the context of a basic macroeconomic setting. Using this
approach, it is possible to investigate the interrelationship between capital and other balance sheet
items and macroeconomic variables with more precision than previous studies. The paper identifies
the long-run steady states for different levels of initial central bank capital, using inter alia Monte
Carlo simulations.
The paper is structured as follows. Section 2 reviews the existing literature on the topic. While this
literature stresses the empirical finding of an inverse relationship between the central bank’s financial
strength and the country’s inflation rate, it lacks a more formal derivation of causality between the two
variables. Section 3 thus presents a simple model of central bank capital in order to illustrate under
what conditions deficient capital would not have harmful effects on the central bank’s policy target.
The model explicitly describes the development of the central bank’s balance sheet within a
macroeconomic context, under the assumption that the central bank is unconstrained in liquidity terms.
A central bank is defined here as being liquidity unconstrained if it has the strictly non-revocable
privilege to issue legal tender. The model further assumes that the central bank sets interest rates
according to a Taylor rule and that there is a Wicksellian relationship between interest rates and
inflation. Section 4 provides Monte Carlo simulations and derives a proposition describing the
conditions under which central banks in the long run always manage to return to profitability and
positive capital. They do not in particular in two cases: when the economy falls into a deflationary
trap, from which is it not possible to escape in the basic Wicksellian model; and, second, when the
growth in the demand for banknotes falls short of nominal interest rates. In these cases, the central
bank never recovers profitability. In Section 5, the relationship between central bank capital or balance
sheet strength and its credibility to perform its monetary policy functions is addressed. This section
stresses the importance of distinguishing between a situation, where an assumed absence of central
ECB
September 2004 7
bank’s concerns about balance sheet strength is shared by the public and the State, and the more
realistic situation where prolonged central bank losses would lead to a credibility crisis. Therefore,
even if a central bank is not subject to liquidity constraints, credibility should be positively correlated
with the level of capital.
This gives rise to the observed relationship between the level of capital and inflation performance. In
essence, the reasons why a central bank might seek to increase profitability at the expense of higher
inflation are closely linked to the broader issue of central bank independence. Again, Monte Carlo
simulations are used to examine the quantitative effects under a range of different assumptions.
Section 6 discusses profit sharing rules between the central bank and the government, which is also
obviously linked to capital. Section 7 concludes.
2. The literature on central bank capital
The issue of central bank capital and financial independence has in the past not attracted the level of
public attention paid, for example, to the operational and statutory (or “institutional”) independence of
central banks. Nevertheless, over the last few years, there has been increasing discussion about
whether central banks need capital and, if so, how much. There are a number of possible reasons. First,
the trend over the last decade of greater independence for central banks to implement monetary policy
has raised the question of central bank financial independence. Second, central banks have been more
aware of the risks they face through the use of more sophisticated risk management techniques. Third,
there is a pressure, particularly resulting from the International Monetary Fund, for higher levels of
transparency in central banks’ accounting practices.2 Finally, interest in central bank financial
independence has also been spurred by public debates in a numbers of countries, such as Finland,
Japan and Switzerland. The following section reviews some of the main themes that have been raised
in the literature.
Stella [1997], [2002], an IMF staff member, was one of the first to analyse the fact that several central
banks had incurred such large losses that they had to be recapitalised by the government, or were in
the process of being recapitalised. Their balance sheets had become so weak – i.e., capital was
substantially negative – that further losses were inevitable and this consequently interfered with the
attainment of the central banks’ policy objectives, primarily maintaining a low inflation rate. For
instance in Uruguay in the late 1980s, the central bank’s losses were equal to 3% of GDP; in Paraguay
the central bank’s losses were 4% of GDP in 1995; in Nicaragua losses were a staggering 13.8% of
GDP in 1989. By end of 2000, the Central Bank of Costa Rica had negative capital equal to 6% of
2
See “Code of Good Practices on Transparency in Monetary and Financial Policies” on the IMF website at
http://www.imf.org/external/np/mae/mft/index.htm
ECB

September 2004
GDP.3 However, he also found that there were some rather puzzling outliers: for example, the Central
Bank of Chile made losses over years but there was no effect on its very good inflation performance.
Stella argues that capital – when defined in the traditional accounting sense that is used for
commercial banks – is meaningless when applied to central banks. Instead, the central bank’s net
worth – defined as the price a fully informed risk-neutral investor would pay to purchase the bank
under normal market conditions – is a much more useful indicator of a central bank’s potential
profitability and financial independence. First, net worth takes into account the central bank’s
“franchise value” – the value of its special legal status of being able to print money and impose reserve
requirements on commercial banks. At the same time, it also takes into account the central bank’s off-
balance sheet rights and obligations – such as the obligation to bail out the banking sector in a crisis or
defend a fixed exchange rate peg – which would tend to reduce net worth.4 Therefore, a central bank
could have a balance sheet structure such that even though it had “large” capital, it would still make
considerable losses; conversely, it could have zero capital yet make very large profits. Thus, Stella
preferred to use a more vague concept of a “central bank’s financial strength”. According to this
definition, a financially strong central bank is one that possesses sufficient resources to attain its
fundamental policy objectives. Stella [1997] concludes:
“…Eventually, the [central bank’s] balance sheet may deteriorate to a point where they either
must abandon control over inflation, repress the financial system, become reliant on constant
infusions by the Treasury, or – the last alternative – be recapitalised.”
However, the link between the central bank’s financial strength and the need to abandon price stability
or repress the financial system appears to be drawn from the possibility that the central bank becomes
illiquid. It is argued that a financially weak central bank (which is likely to have negative capital and
repetitive losses) will consequently have to sell liquid interest-earning assets in order to cover its
operating costs and other financial obligations. As it sells these assets, ceteris paribus, it also reduces
its potential to makes profits in the future and “eventually the central bank will exhaust its supply of
valuable liquid assets”. At this point, it is argued that the central bank would have three options:
issuing its own debt certificates; repression of the financial system (e.g. through high non-remunerated
reserve requirements); or the abandonment of price stability by injecting excess liquidity into the
banking system. He rules out the first option of issuing debt certificates as it runs the risk of the central
bank accumulating an unsustainable debt burden:
“…The sustainability of central bank debt issuance is a function of the same factors that
determine the sustainability of government debt in general. These include expectations of the
3
See also Leone [1993], Dalton and Dziobek [1999]
4
Blejer and Schumacher [1999] provide a detailed analysis and taxonomy of the different types of central banks’ “contingent
liabilities” – defined as financial commitments triggered by the occurrence of an event whose realisation is uncertain, e.g.
guaranteeing the stability of the banking sector or holdings of derivatives. These liabilities tend to be off-balance sheet and
reduce the transparency of central bank accounts. They argue that this may result in serious problems regarding the proper
assessment of the financial position of the monetary authority.
ECB
September 2004 9
future income and expenditure stream of the central bank, the growth rate of demand for the
securities being purchased from the central bank, the reputation of the issuer of the security,
macroeconomic developments, the government’s commitment to guarantee obligations of the
central bank, budgetary development, etc.”
However, one could argue that debt sustainability issues are less of a concern to a central bank than to
a government. Unlike a government, a central bank can always obtain the funds it needs to cover its
expenses simply by crediting a credit institution’s current account at the central bank. If this were to
create excess liquidity in the banking system (i.e., with aggregate current accounts higher than reserve
requirements), overnight interest rates would consequently fall to the deposit facility rate or to zero if a
deposit facility were not available. Of course, if interest rates were to fall to zero, it could result in a
too “loose” monetary policy and lead to an increase in inflation. In order to avoid this, the central bank
could issue debt certificates to absorb the excess liquidity, offering an interest rate at a level which
would ensure price stability. Banks would have no alternative except to purchase the debt certificates,
as the alternative would be to earn zero interest on the excess reserves. One might argue naively that
this process could continue indefinitely, with the losses resulting from the interest paid on the debt
certificates being covered by further crediting banks’ current accounts and consequently issuing more
debt certificates to mop up the excess liquidity. There would be no technical restriction on the amount
of debt certificates which could be issued.
In the literature there is understandably no precise definition of what the optimal level of capital
should be. Stella identifies four different ways that central banks have used in practice to determine
their own level of capital: (1) An absolute nominal value of capital; (2) A target ratio of capital to
another central bank balance sheet item; (3) A target ratio of capital to a macroeconomic variable; (4)
According to the perceived risks to the “solvency” of the bank. According to the last approach, the
optimal level of capital would depend on a large number of qualitative and quantitative factors, such
as the macroeconomic economic environment, the central bank’s vulnerability to large financial
shocks, the bank’s historical legacy, the status of institutional relations with the government, the
bank’s policy obligations (e.g., defence of a currency peg), and its volatility of profits.5
Martínez-Resano [2004] also attempts to define the factors which determine the optimal level of
capitalisation. He surveys the full range of risks that a central bank’s balance sheet is subject to. Then,
using a simple VAR model, he analyses the interplay between capitalisation, accounting rules and
dividend distribution, so as to determine a simple benchmark for central bank financial strength. He
concludes that, in the long run, central banks’ financial independence should be secure as long as
demand for banknotes is maintained. However, in the short- and medium term, he finds that financial
vulnerability could impact on a central bank’s effective independence. In order to avoid this
possibility, he concludes that adequate capitalisation is key. In the model, the size of the optimal
5
A fuller discussion of the process of determining the level of required capital and specific practices is provided by Dalton
[1999].
ECB

September 2004
capital buffer depends on risk exposure, volatility, costs of underperformance (i.e. the impact of lack
of independence on inflation) and how the central bank’s profits are distributed. If a central bank were
guaranteed to retain all its profits, the central bank could operate without capital at all. Otherwise, a
certain level of capital is required or, as a second-best option, adequate institutional arrangements. He
finds that one of the preferred methods for profit distribution is “smoothing”, as it is consistent with
the assumed natural cycles of central banks’ earning generation and is also neutral in terms of income
for the Government. Stella [2002] and Sullivan [2002] also stress the importance of a clear set of rules
to govern the distribution of both profit and losses in this historically highly opaque area.6 Sullivan
[2002] argues that the central bank law should specify clearly how its profit is calculated and how net
losses are allocated, as well as how the bank will be recapitalised in the event of a crisis.
Ernhagen et al [2002] from the Sveriges Riksbank also examine whether having a buffer of capital is
of importance for a central bank’s independence. The authors argue that the central bank can always
pay its debts and operating costs with banknotes, but if the total demand for banknotes among the
public has not increased then these newly printed banknotes will quickly be exchanged at the central
bank. As argued previously by Stella [1997], the central bank will then be forced either to finance
itself by issuing “accelerating” interest bearing debt which would further reduce future expected
profitability or by lowering interest rates to levels which would trigger inflation and consequently
increase demand for banknotes. Reducing interest rates to increase demand for banknotes may not be
compatible with the central bank’s monetary policy objectives. However, as already mentioned, the
central bank does not need to “finance itself” through debt certificates, but only may want to absorb
excess liquidity through debt certificates for monetary policy purposes.
Ernhagen et al [2002] provide some crude calculations regarding the minimum level of capital for the
Riksbank, although they admit that it is impossible to provide a precise answer and that it depends to a
large extent on the safety margins that the government wishes to provide. First, they calculate the
amount of capital – assuming an interest rate of 5% – that would be needed to cover operating costs,
i.e. wages, rent, fixed capital etc. under the assumption that they cannot rely on seignorage income.
They assume no seignorage income because of the increasing use of e-money and other cashless
payments, which may cause an abrupt reduction in demand for banknotes.7 Second, they add the risks
of losses on the central bank’s foreign exchange portfolio based on value-at-risk calculations, although
it is acknowledged that the risk of losses depends significantly on whether Sweden has a floating or
fixed exchange rate. Currently Sweden has a floating exchange rate, but if it were to have a fixed
exchange rate, e.g. if it were to join ERM II, the risks of losses resulting from defending the exchange
rate could be substantial. Finally, they add the potential losses, which could result from providing
6
According to Vaez-Zadeh [1991], p.70, who reviewed central banking laws in some 60 countries, almost one third of
central banks do not have any specific provisions regarding the treatment of losses. Pringle and Courtis [1999] find that most
of the 27 central banks they surveyed provide rules for the distribution of profits (although there is no explicit mention of
what the rules are for losses).
7
Note that the Bank of England does not receive seignorage income from banknote issue at all. The assets that back the UK
banknotes are held by a separate department of the bank (the Issue Department) in an attempt to demonstrate that assets of
appropriate quality back the currency. The profits, once the expenses of banknote production and distribution have been
taken care of, go directly to the Treasury. The bank’s main source of income is from unremunerated cash ratio deposits which
commercial banks are required to hold at the Bank of England.
ECB
September 2004 11
emergency liquidity assistance if there were a banking crisis: this may be much more serious in terms
of cost than a loss on the foreign exchange portfolio.8 They use the losses from the previous banking
crisis in the late 1980s as a benchmark although they say there is much uncertainty: the true cost
would depend on the quality of the collateral provided, the amount of lending without adequate
collateral, and the extent to which the losses would be shared with the government. Having calculated
the minimum level of capital required, the authors also assess the costs of having too high a level of
capital. They conclude that as long as the return on the “excess” capital is as high as the interest rate
paid by the government on its debt, and that the profits of the central bank are transferred to the
government, there are no adverse consequences resulting from overcapitalisation of the central bank.
According to Pringle [2003], the Bank of Japan (BOJ) has also thought rather extensively about
central bank capital. While conducting monetary policy in recent years, the BOJ has taken on
considerable risks, purchasing vast amounts of long-term Government bonds at very low yields.9
When the economy recovers, bond yields will inevitably rise, sending bond prices – and thus the value
of the BOJ’s assets on a “mark-to-market” basis – plummeting. Pringle argues that the BOJ’s capital –
until recently only 3% of its total assets – could be wiped out if the BOJ had to report huge paper
losses, thus undermining the credibility of the BOJ.10 The issue has become more urgent as the BOJ’s
return on assets declined sharply last year, as a result of lower interest rates in the US and euro area,
and the BOJ made an unprecedented request to the Finance Ministry to be allowed to keep 15% of its
profit for 2002 to build up its capital (as a rule, the central bank only retains 5%). Although Pringle
questions the usefulness of capital when the BOJ balance sheet is exposed to massive “macro” risks
and that in the longer term there is unlikely to be a problem as the BOJ is a monopoly supplier of base
money, he concludes that “to support its credibility, an independent central bank needs adequate
capital as well” without however providing detailed arguments.11
Overall, one may conclude that the literature seems to have one main shortcoming: although stressing
that capital plays a substantially different role in a central bank than in a commercial bank, it seems to
rely on the assumption that a central bank would not be able to issue debt certificates indefinitely.
However, as argued above, this kind of “illiquidity” argument does not appear to be technically
obvious in the case of a central bank. Furthermore, no model has so far been developed to analyse the
exact interaction between the different variables. In the following sections, we will aim at giving more
8
Gros and Schobert [1999], in contrast, argue that the lender of last resort function is not a reason to “over-capitalise” central
banks. They argue that the only danger that a central bank might face as a lender of last resort is that it would be obliged to
create additional money, but this can always be done without limits independently of its capital. Then, even if the central
bank lends big amounts to banks in trouble during a financial crisis and has difficulties collecting the loans when the crisis is
over, these losses should not be covered through additional money creation, but they are anyway the responsibility of
national finance ministries, which would have to make up any losses national central banks incur as lenders of the last resort.
9 In a speech delivered on 25 October 2003, Kazuo Ueda, a Policy Board member of the Bank of Japan, remarked, “The BOJ
has, over the past few years, taken various risks in its operations seen unthinkable under normal circumstances and there are
risks that this will bring about consequences…If yields rose sharply and the value of JGBs holdings fell by 10% or so, the
BOJ will essentially be in negative net worth.”
10
See also JP Morgan, Japan Markets Outlook and Strategy, 24 January 2002, which raised questions as to the likely
duration of the Bank of Japan’s willingness to use its “rinban” operations to support the long end of the government bond
yield curve because of the potential losses that could be incurred.
11
Svensson [2004] also takes up the issue of the need for central bank capital in the context of escaping a deflationary trap.
He assumes, referring to Stella, that “a negative capital would require a capital injection and put the bank at the government’s
mercy. In order to avoid this, the bank never voluntarily allows its capital to fall below a certain minimum level.”
ECB

September 2004
precision to analysis through a simple modelling approach encompassing both the central bank
balance sheet and a stylised macro-economy.
3. A simple model based on a central bank with no liquidity constraints
The mechanisms by which central bank capital can impact on a central bank’s ability to achieve price
stability can be best illustrated by first developing a simplistic model in which there is a kind of
dichotomy between the level of capital and inflation performance. To create this dichotomy requires
the strong assumption that the central bank is “liquidity unconstrained”, which could be seen to reflect
total certainty that the central bank will maintain the right to issue legal tender over an infinite
horizon. Making such an assumption implies that the level of capital and the profit and loss situation
should not become a problem as they could for a private company, which eventually would face
illiquidity.
It needs to be highlighted that the model obtains the dichotomy result quasi by assumption, and that to
approach reality, some of the assumptions will need refinement, which will lead to a breakdown of the
dichotomy result. Still, the model is an appropriate starting point to derive the actual reasons for the
relevance of central bank capital in the most transparent way.
Most importantly, the model does not yet encompass the interaction between a central bank’s financial
strength and the public’s expectations, which could lead to a breakdown of the model’s hypothesised
Wicksellian relationship between inflation and interest rates. Since ultimately the assumption of the
central bank as a potentially unlimited source of money depends on the public’s trust in this money’s
worth, it can indeed prove unrealistic that a loss making central bank is believed to ignore balance
sheet concerns and rely on a very long-term profitability outlook. In fact, even within the framework
of the model presented here, there is still a non-zero probability of the economy falling in a
deflationary trap and the central bank sustaining permanent losses. In addition, it could happen,
especially after a series of capital losses, that even unjustified private sector expectations of a risk that
the central bank might lose the right to issue legal tender at some distant point in time could already
today undermine the reputation of the central bank. Hence, a feasible explanation for the importance
of central bank capital could well rely on a self-fulfilling expectations kind of argument. The public
does not fully give credit to the theoretical possibility – from an accounting point of view – of a
central bank conducting a successful monetary policy while incurring repeated capital losses.
Having made this caveat, we now turn to the specification of the model, by considering the following
simplistic central bank balance sheet.
ECB
September 2004 13
Stylised balance sheet of a central bank
Assets Liabilities
Monetary policy operations (“M”) Banknotes (“B”)
Other Financial Assets (“F”) Capital (“C”)
Banknotes are assumed to always appear on the liability side, while the three other items can be a
priori on any side of the balance sheet. For the purpose of the model, a positive sign is given to
monetary policy and other financial assets when they appear on the asset side and a positive sign to
capital when it appears on the liability side. The following assumptions are taken on each of these
items:
• Monetary policy operations can be interpreted as the residual of the balance sheet. It is
remunerated at iM %, the operational target interest rate of the central bank. Assume that the
central bank, when setting iM , follows a kind of simplified Taylor rule of the type
iM ,t = 4 + 1.5(π t −1 − 2) . According to this rule, the real rate of interest is 2% and the inflation
target is also 2%.12 An additional condition has also been introduced in the Taylor rule, namely
that in case it would imply pushing expected inflation in the following year into negative values,
the rule is modified so as to imply an expected inflation of 0%. It will later be modelled that for
profitability/capital reasons, i.e., reasons not relating directly to its core task, the central bank may
also deviate from this interest rate setting rule.
• Other financial assets contain foreign exchange reserves including gold but possibly also domestic
financial assets clearly not relating to monetary policy. Assume it is remunerated at iF %. The rate
iF % may be higher or lower than iM %, which depends inter alia on the yield curve, international
imbalances in economic conditions, the share (if any) of gold in F, etc. Also, F can be assumed to
produce revaluation gains/losses each year. One may assume that iF , t = iM ,t + ρ + ω t with
normally, but not necessarily, ρ > 0 , implying that the rate of return on F would tend to be higher
than the interest rate applied to the monetary policy instruments, and ωt is a random variable with
zero mean reflecting the associated risks. F can in principle be determined by the central bank, but
it may also be partially imposed on the central bank through its secondary functions or ad hoc
requests of the Government. Indeed, F may include, especially in developing countries, claims
resulting from bank bailouts or from direct lending to the Government, or also from special
industry borrowing programs imposed on the central bank by the Government. Typically, such
assets are remunerated at below market interest rates, such that one would obtain ρ < 0 .
• Banknotes are assumed to depend on inflation and normally follow some increasing trend over
time, growing faster when inflation is high13. Assume that
12
See e.g. Woodford [2003] for a discussion of the properties of such policy rules.
13
This assumption would not hold in case money velocity accelerates when inflation increases (Cagan-Bresciani Terroni
thesis). In such a case, the profitability outlook of the central bank becomes less optimistic, as in the general case when
ECB

September 2004
Bt = Bt −1 + Bt −1 (2 + π t ) / 100 + Bt −1ε t , whereby π t is the inflation rate, “2” is the assumed real
interest or growth rate and ε t is a noise term. It is assumed that the real interest rate is exogenous.
Despite the development of new retail payment technologies over many years, and speculation
that banknotes could vanish in the long run, banknotes have continued to increase in most
countries at approximately the rate of growth of nominal GDP. Our stylised balance sheet does not
contain reserves (deposits) of banks with the central bank, but it can be assumed alternatively that
reserves are implicitly contained in Banknotes (which may thus be interpreted as the monetary
base). The irrelevance of the particular distribution of demand between banknotes in circulation
and reserves with the central bank would thus add robustness to this assumption on the dynamics
of the monetary base. 14
• Capital depends on the previous year’s capital, the previous year’s profit (or loss), and the profit
sharing rule between the central bank and the Government. In the basic model setting, it is
assumed that the profit sharing rule is as follows: if profit is positive, i.e. Pt −1 ≥ 0 then
Ct = Ct −1 + αPt −1 (with 0 < α < 1) , else Ct = Ct −1 + Pt −1 , and α is set to 0.5. Profits depend on
the returns on the different balance sheet positions and on operating costs. With regard to
operating costs, q, it is assumed that they grow over time at the inflation rate. Profit and thus
Capital is likely to contain a further random element, which reflects that extraordinary costs may
arise to the central bank when the Government manages to assign additional duties to the bank. In
the less industrialised countries, these costs could typically be the support of insolvent banks, or
the forced granting of credit to the Government. As mentioned above, such factors can also be
modelled as affecting the remuneration rate of financial assets.
To complete the model, one needs an equation that explains the evolution across time of the inflation
rate. A Wicksellian relationship between inflation tomorrow and inflation today is assumed, i.e.
π t +1 = π t + β (2 + π t − iM , t ) + µ t , i.e. inflation normally accelerates if interest rates on monetary
policy operations are below the sum of the real rate on capital (2%) and the current inflation rate.15
The noise term µ t means that inflation is never fully controlled. The equation also implies that there
is a risk of ending in a deflationary trap: when π t < −2 , then, due to the zero constraint to interest
rates, prices should start falling further and further, even if interest rates are zero. If µ t ~ N (0, σ µ2 ),
then this can always happen theoretically, but of course the likelihood decreases rapidly when the sum
of the present inflation and of the real rate is high.
Adding a time index t for the year, the time series are thus determined as follows over time16:
demand for banknotes decreases for technological or other reasons. Note as well that in a macroeconomic context of
hyperinflation, the opportunity cost of holding banknotes could reverse this positive relationship between inflation and
banknote demand. In extreme situations, hyperinflation could cause the surge of privately issued money substitutes or the
return to money-less trade in goods.
14
A switch from banknotes holdings to reserve holdings would imply that seignorage revenues would in the first case stem
from a general tax to the holders of banknotes, while in the second case they would be comparable to a tax on the banking
sector.
15
See Woodford [2003] for a discussion of such Wicksellian inflation functions.
16
The order of the equations, although irrelevant from a conceptual point of view, reflects how the eight variables can be
updated sequentially and thus how simulations can be obtained.
ECB
September 2004 15
Eq. 1 π t = π t −1 + β (2 + π t −1 − iM ,t −1 ) + µ t
Eq. 2 qt = (1 + π t / 100)qt −1
Eq. 3 Ft = F
Eq. 4 if Pt −1 ≥ 0 then Ct = Ct −1 + αPt −1 (with 0 < α < 1) , else Ct = Ct −1 + Pt −1
Eq. 5 Bt = Bt −1 + Bt −1 (2 + π t ) / 100 + ε t
π t −1
Eq. 6 if max(4 + 1.5(π t −1 − 2),0) < + 2 + π t −1 , iM ,t = max(4 + 1.5(π t −1 − 2),0)
β
π t −1
else iM = + 2 + π t −1
β
Eq. 7 iF , t = iM ,t + ρ + ω t
Eq. 8 M t = Bt + Ct − Ft
Eq. 9 Pt = iM , t M t + iF , t Ft − qt
This simple modelling framework captures all basic factors relevant for the profit situation of a central
bank and the related need for central bank capital. It can also be used to analyse the interaction
between the central bank balance sheet, interest rates and inflation. It should be noted that, from
equation 1 and i M ,t = 4 + 1.5(π t −1 − 2) , a second order differences equation can be derived of the
form π t +1 − (1 + β )π t −1 + 1.5π t − 2 = β + µ t . Disregarding the stochastic component, µ, this equation
has a non-divergent solution whenever − 2 3 < β < 2 / 3 . The constant solution π t = 2, ∀t , is a
priori a solution in the deterministic setting. However, it has probability 0 when considering again the
shocks µt.
Simulations can be performed to calculate the likelihood of profitability problems arising under
various circumstances. The model can be calibrated for any central bank and for any macroeconomic
environment. In this paper, we focus on a few important illustrative scenarios.
Before proceeding with the simulations, the impact of capital on the central bank’s profitability and
hence financial independence should be described briefly. First, note that as long as we do not foresee
the case of bankruptcy of the central bank, then by definition, negative capital is not a problem per se.
Indeed, as long as the central bank can issue legal tender, it is not clear what could cause bankruptcy.
By substitution using the balance sheet identity, one obtains the profit function:
Pt = iM ,t ( Bt + Ct ) + (iF − iM ).Ft − qt
Therefore, a higher capital means higher profits since it increases the size of the (cost-free) liability
side. For given values of the other parameters, one may therefore calculate a critical value of central
bank capital, which is needed to make the central bank profitable at a specific moment in time:
ECB

September 2004
(iF − iM ) 1
Pt > 0 Þ Ct > Ft − qt − Bt .
iM iM
Unsurprisingly, the higher the monetary policy interest rates, the lower the critical level of capital
required to avoid losses, since the central bank does not pay interest on banknotes (or excess reserves,
i.e. reserve holdings in excess of the required reserves). A priori this level of capital can be positive or
negative, i.e. positive capital is neither sufficient nor necessary for a central bank to make losses. It
would also be possible for a central bank with positive capital to suffer losses over a long period,
which could eventually result in negative capital. Likewise, a central bank with negative capital could
have permanent profits, which would eventually lead to positive capital. Moreover, when considering
the longer-term profitability outlook of a central bank in this deterministic set-up, it will turn out that
initial conditions for capital and other balance sheet factors are irrelevant and the only crucial aspect is
given by the growth rate of banknotes as compared to the growth rate of operating costs. The intuition
for this result (stated in proposition 1 below) is that, when considering only the long term, in the end
the growth rate of banknotes needs to dominate the growth rate of costs, independently of other initial
conditions.
4. Monte Carlo simulations and the long run steady state of the central bank balance sheet and
inflation in the base model
Obviously, the starting value of the array ( M 0 , F0 , B0 , C0 , π 0 , i0 ) as well as the level of the parameters
(α , β , ρ , σ ε2 ,σ ω2 , σ µ2 ) will be crucial for determining the likelihood that a central bank will be at a
certain moment in time in the domain of positive capital and profitability. Consider the following two
contrasting examples: (1) A profitable central bank with positive capital; (2) A non-profitable central
bank with negative initial capital. In both cases it will be assumed, for the sake of simplicity, that other
financial assets (F) are zero.
Example 1. A profit making central bank with positive capital
The classical central bank of an industrialised country with a floating exchange rate regime (e.g., US,
euro area, UK) has positive capital and monetary policy operations on the asset side. To illustrate the
case, assume that M 0 = 120 , B0 = 100 , C 0 = 20 . Assume furthermore that ρ = 2% , q = 1 ,
π 0 = 2% , i0 = 4% and (σ ε = 1,σ ω = 0.5,σ µ = 0.5) and that α=0.5, β=0.2.
Thanks to its Taylor rule, the monetary policy performance of the central bank is very good: inflation
remains close to the target rate of 2% and the nominal interest rate will remain at 4% on average.
Banknotes will show steady growth in line with the nominal interest rate of 4%, while monetary policy
operations will grow somewhat slower due to the de-leveraging effect of capital. The following two
charts (figures 1 and 2) show the median and 95% confidence intervals of interest rates, inflation, the
size of outstanding monetary policy operations, and profits for 1000 Monte Carlo simulations over 100
periods.
ECB
September 2004 17
Figure 1: Evolution of inflation and interest rates: a profitable bank with positive initial capital
inflation
yield of mon. pol. operations
11
10
9
annual percentage points
8
7
6
5
4
3
2
1
0
-1
-2
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96
simulation period
Note: The paths shown are the median (thick lines) and 0.05, 0.95 quantiles (narrow lines) for the cross-section of the simulated data in
each period, with 1000 replications. Initial values were: C0 = 20, B0 = 100, F0 = 0, q0 = 1
Figure 2: Evolution of monetary policy operations and profits: a profitable bank with positive
initial capital
monetary policy operations (left-hand scale)
Series2
Series5
profit (right-hand scale)
12000 1400
1200
10000
1000
accounting units
8000
800
6000 600
400
4000
200
2000
0
0 -200
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96
simulation period
Note: The paths shown are the median (thick lines) and 0.05, 0.95 quantiles (narrow lines) for the cross-section of the simulated data in
each period, with 1000 replications. Initial values were: C0 = 20, B0 = 100, F0 = 0, q0 = 1
It is noteworthy that, even in the case of an initially profitable central bank with positive capital, there
is a probability of more than 2.5% that the central bank will fall into a deflationary trap from which it
is impossible to escape. Their income consequently falls to zero and each year it would make a loss
equal to the level of operating costs. The likelihood of ending in a deflationary trap, however, could be
reduced by assuming a lower variance of shocks, by modifying further the Taylor rule to make the
central bank even more prudent in this respect, or by letting the central bank target a higher inflation
rate.
ECB

September 2004
Example 2. A non-profitable central bank with negative initial capital
Assume that the central bank above was at its starting position forced to bail out banks in a crisis and
made a loss of 100.
Its initial balance sheet is assumed to be as
follows
M 0 = −80, B0 = 20, C0 = −100 . Assume again that q = 1,π 0 = 2%, i0 = 4% and
(σ ε2 = 1,σ ω2 = 0.5,σ µ2 = 0.5) . Now, this central bank will make an expected loss of around 4.2 and it
will continue making losses in the foreseeable future. However, after a long period of time, it is still
able to return to profitability in most of the cases, unless adverse random shocks push it into a
deflationary trap17. Before looking again at Monte Carlo simulations, consider first the deterministic
case in which economic shocks would be absent (σ ε2 = 0, σ ω2 = 0, σ µ2 = 0) . It is easy to calculate in a
spreadsheet that losses in year 2 will be 4.3, in year 10 will be 5.8, in year 50 will be 22.2. From that,
one would expect a further monotonous increase of losses, but one obtains that the maximum loss is
reached in year 137 with 162.4.. After that, losses decline and profits become positive from year 163
onwards. The reason for this, as pointed out above, lies in the overriding importance of the growth
rates of banknotes and operating costs, rather than the initial state of the balance sheet. Since eq. 2
assumes that operating costs only grow at the inflation rate and eq. 5 sets the growth of banknotes at
the nominal interest rate, the condition of long-term profitability is satisfied in the deterministic
context. This property of a long-term return to profitability and positive capital is obtained, in the case
of absence of shocks, for the model assumptions made above for any initial values of the capital and
operating costs. The following proposition summarises the result.
Proposition (long run return to profitability): A central bank in an economy as described by the
system of equations (1-9), absence of shocks, (σ ε2 = 0, σ ω2 = 0, σ µ2 = 0) , and initial values of
inflation and key rates at their equilibrium levels of π 0 = 2, i0 = 4 will always return to
profitability and positive capital at a certain moment in time, regardless of starting values, i.e.
∀C0 ∈ ℜ, ∀B0 ∈ ℜ + , ∀q0 ∈ ℜ : (∃τ ∈ ℜ+ : ∀t > τ : Pt > 0), (∃τ '∈ ℜ+ : ∀t > τ ': Ct > 0) .
The proof is provided in the annex. It is thus shown that under plausible assumptions, central banks
return in the long run to profitability regardless of the starting level of negative capital and regardless
of the staring level of operating costs. How long it takes for the central bank to return to profitability
depends on operating costs and initial capital. The following charts show, starting from the
deterministic specification above, the years of return to profitability for different values of initial
capital (figure 3) and different values of operating costs (figure 4).
17
It will return to profitability given the hypothesis made here that banknotes grow by a rate corresponding to the real growth
rate plus inflation, which is under normal circumstances approximately equal to the key central bank rate.
ECB
September 2004 19
Figure 3: Period of first positive profit depending on initial capital
500
first period in which a positive profit
450 Initial banknotes: 20 Initial banknotes: 10

operating costs: 1 operating costs: 1
400
350
Initial banknotes: 30
is achieved
300 operating costs: 1

250 Initial banknotes: 40
150
100
Initial banknotes: 50
0
-300 -280 -260 -240 -220 -200 -180 -160 -140 -120 -100 -80 -60 -40 -20 0
initial capital
Note: The scatterplot shows the first year when profits are achieved in a deterministic context with fixed initial banknotes
and varying initial negative capital.
Figure 4: Period of first positive profit depending on initial capital
500
first period in which a positive profit
Initial capital: -200

450 initial banknotes: 20
400 Initial capital: -250
initial banknotes: 20
350
is achieved
300
250
200
initial banknotes: 20
150
100 Initial capital: -50 initial banknotes: 20
50 initial banknotes: 20
0
0 0.25 0.5 0.75 1 1.25 1.5 1.75 2 2.25 2.5 2.75 3 3.25 3.5 3.75 4 4.25 4.5 4.75 5
initial operating costs
Note: The scatterplot shows the first year when profits are achieved in a deterministic context with fixed initial banknotes
and varying initial operating costs.
Monte Carlo simulations tend to confirm these properties of the deterministic case. For instance in the
initial specification with initial C0 = −80, B0 = 20, q0 = 1 , the median central bank reaches
profitability again after 144 years. Figure 5 displays the evolution of profits and the sum of monetary
policy operations over 200 periods.
ECB

September 2004
Figure 5: Evolution of monetary policy operations and profit: a non-profitable bank
with negative initial capital and no preference for returning to profitability
Series2
Series1
2000 350
1500 250
1000 150
accounting units
500 50
0 -50
-500 -150
-1000 -250
-1500 -350
1
9
17
25
33
41
49
57
65
73
81
89
97
105
113
121
129
137
145
153
161
169
177
185
193
simulation period
Note: The paths shown are the median (thick lines) and 0.05, 0.95 quantiles (narrow lines) for the cross-section of the
simulated data in each period, with 1000 replications. Initial values were: C0 = -80, B0 = 20, F0 = 0, q0 = 1
It should be noted, however, that the assumed growth rates of the individual items are key to this
result. A return to profitability would no longer be guaranteed if the growth rate of banknotes were
assumed to be lower or if the growth rate of operating costs were higher. Furthermore, the point of
return to profitability can occur a long time in the future – beyond the work- and life-time expectations
of central bankers – and thus, in practice, may be irrelevant for their decision making.
As in the first example, the monetary policy performance of the central bank is obviously untouched
by its poor profitability and capital outlook, as the model practically by assumption precludes any
interrelation in this direction. Proponents of a role of the monetary base in monetary policy
implementation may argue that the Wicksellian inflation function on which the present model is based
is invalid, or that, at a minimum, it is necessary to check whether its implications do not lead to
contradictory results. For instance, they might argue that a sustained growth of the monetary base at
rates different from the sum of the real growth rate and the rate of inflation would be counterintuitive.
However, banknotes in circulation, which are the only element of the monetary base in our model, are
indeed assumed to increase at the rate of inflation plus the real growth rate of the economy, such that
no contradiction appears. Moreover, monetarists might argue that liquidity-absorbing open market
operations are a kind of substitute to the monetary base, and therefore increase it. However, it can be
counter-argued that the central bank would have the option to conduct operations with a very long
maturity, such as issuing debt certificates with a maturity of five or ten years. These should definitely
not be considered as substitutes for the monetary base.
ECB
September 2004 21
What is interesting to note in this context is that the length of the balance sheet of a loss-making
central bank will always in the long-run be shorter than that of a profitable central bank. This follows
directly from the reversal towards profitability at some moment in time, which is postulated in the
proposition. This reversal implies that there is a point in time after the reversal where capital will be
zero, and thus the length of the balance sheet will be exactly equal to banknotes, which is the
minimum length of the balance sheet at that moment in time. At that moment, the central bank which
started as a profitable one has already built up a huge capital position on the liability side, and the
central bank which originally made losses will never be able to catch up.
Proposition 2 (length of balance sheet). Consider two central banks which are identical but for
their initial capital C0a , C0b , with C0a < C0b . If C0a > 0 , then, ∀t ∈ ℜ+ : Lat < Lbt . If C0a < 0 , then,
∃τ ∈ ℜ+ : Lat < Lbt , ∀t > τ .
The proof is given in the annex. One may note again that the point in time in which the reversal occurs
may be far in the future. Figure 6 displays for the deterministic case with B0 = 20 the development of
the balance sheet length for alternative levels of initial capital, namely C0 = −100,−50,0,50 . It
appears that first, the length of the balance sheet grows fastest for C0 = −100 , but that this is reversed
over time.
Figure 6: Evolution of length of balance sheet for alternative initial capital (logarithmic scale)
100000 Evolution of length of balance sheet for alternative C(0) in logarithmic scale
10000
C(0)=-100
1000
C(0)=-50
C(0)=0
100
C(0)=50
10
1 11 21 31 41 51 61 71 81 91 101 111 121 131 141 151 161 171 181 191
ECB

September 2004
Finally, it may be interesting to consider the scenario if banknotes in circulation were to disappear due
to technological innovations, such as e-money.
Assume that the starting values of M, B and C are as follows: M 0 = −100 , B0 = 0 , C 0 = −100 ,
q0 = 1 . How would the central bank balance sheet and profit evolve in this even simpler case?
Obviously, this central bank could never get out of its loss-making situation, with B remaining at zero,
and M and the length of the balance sheet both growing at the same rate. If the growth rate of
operating costs again equals the rate of inflation, and the nominal interest rate is 4%, then in the long
run the weight of operating costs in the growth rate of the positions vanishes, and all quantities grow
by 4%, which is simply the nominal growth rate of the economy. So in this case, the “indebtedness” of
the central bank is growing at a steady rate.18 Of course, the fact that the rate of central bank
indebtedness with respect to the overall economy would remain constant still does not alleviate the
increasing absolute income gap of an eternally loss making central bank.
However, the point so far developed is that, in such a model, central bank capital still does not seem to
matter for monetary policy implementation, in essence because negative levels of capital do not
represent any threat to the central bank being able to pay for whatever costs it has.19 Although losses
may easily accumulate over a long period of time and lead to a huge negative capital, no reason
emerges why this could affect the central bank’s ability to control interest rates. Under this setting, it
was shown that in the long run, subject to a central assumption on the growth rate of banknotes, losses
and negative capital are always reversed into profits and positive capital unless unexpected shocks
lead to a deflation trap. Operating costs can be covered indefinitely provided that their growth rate
does not exceed that of banknotes. One could therefore conclude that the model implies a perfect
dichotomy between the central bank balance sheet structure and its ability to fulfil its monetary policy
tasks. This is however not what experience has suggested so far. The next section tries to find out why
the model’s outcome contrasts with the empirical evidence.
5. Why central bank capital matters – liquidity constraints, central bank credibility and
independence
Having shown that in the model above, a perfect dichotomy exists between the central bank’s balance
sheet and its monetary performance, how does one explain the observation, made for instance by
Stella (2003), that many financially weak central banks are associated with high inflation rates? It is
likely that there is another set of factors, related to the institutional environment in which the central
bank exists, that is causing a relationship between the weakness in the central bank’s financial position
18
Ernhagen [2002] argues that indebtedness keeps “accelerating” in such a case. In the context of the base model presented,
however, it would grow at a comparable rate as the whole economy.
19
See section 2 for a more detailed discussion on why a central bank is not directly constrained in the amount of
credit it can sustain, unlike any other economic agent.
ECB
September 2004 23
and its inability to control inflation. In many cases, central banks have a weak financial position for
one or more of the following reasons: (1) they have been forced to bail out the banking system after a
crisis; (2) the economy is heavily dollarised, reducing seignorage revenues; (3) they are large lenders
to the Government at low or negligible interest rates in order to finance the budget deficit. Yet, these
elements are the direct result of a general lack of either institutional independence of the central bank
from the Government or public trust in monetary policy. In fact, one can assume that the degree of
public trust in an effective inflation control is to a great extent dependent on the perceived institutional
independence of the central bank, whereby the public will generally judge from the historical track
record. Lack of public trust and a low level of institutional independence seem to be key factors
explaining poor inflation performance. Thus, in order to show the importance of central bank capital, it
is necessary to establish a causal link between a central bank’s financial strength and its perceived
level of institutional independence. This section does not attempt to model fully the relationship
between a central bank and the Government, but instead highlights some of the factors that could be
important in the present context. These are incorporated into the model developed in section 3 to
analyse the quantitative effects.
We start by examining three possible scenarios of the relationship between the central bank and the
government: first, total lack of public trust in the central bank (possibly full institutional dependence);
second, unlimited public trust (what could be called perfect independence of the State and the central
bank); and, third, some point between these two extremes, which is probably the most realistic. First,
in case of dependence, i.e. if there were no separation between the central bank and the government,
the capital of the central bank is obviously irrelevant since one then has to consider only the aggregate
capital of the State (including the central bank and the government). This may best be illustrated by
comparing the balance sheet of the State and its sub-units in case of a capital-rich central bank and a
capital-poor central bank.
The case of a central bank with negative capital

Central Bank Government Consolidated state
CB Capital 500 Banknotes 100 Gvt. Assets 1000 Gvt debt 500 Gvt. Assets 1000 Banknotes 100
MPOs 400 CB Capital 500 Gvt.. Debt 500
MPOs 400
The well-capitalised central bank

Central Bank Government Consolidated state
MPOs 600 Banknotes 100 Gvt Assets 1000 Gvt. Debt 1500 Gvt. Assets 1000 Banknotes 100
CB Capital 500 CB Capital 500 MPOs 600 Gvt. Debt 1500
Assuming that the central bank and the government are in practice one, such that the accounting
separation does not reflect substance, we still obtain the result that the State’s balance sheet is longer if
the central bank is rich, i.e. well equipped with capital. It could be argued that if there is no separation
of substance anyway (which has never appeared to be an appropriate institutional setting), capitalising
ECB

September 2004
the central bank introduces redundancies in the State’s financial position and does not increase the
public’s confidence in the worthiness of money issued by the central bank. It would thus be preferable
to not equip the central bank with capital in that case. Of course, the two cases illustrated above also
make a difference for financial markets: the second case would lead to a higher amount of
Government paper in circulation and, ceteris paribus, to a longer balance sheet of the banking system.
The minimum length of the State’s balance sheet corresponds to the government’s assets, and will be
obtained whenever central bank capital is equal or lower than the negative of banknotes. When central
capital is above this level, the length of the State balance sheet grows proportionally.
At the other end of the spectrum, if unconstrained liquidity and absolute independence of the
central bank from the Government were generally acknowledged, capital would also be irrelevant.
This is the case modelled in section 3. Absolute independence implies that the central bank will never
need to approach the government to get additional funds in order to regain the public’s trust in the
value of the currency. In particular, this condition requires that the Government can never change the
central bank’s law, i.e. never change the rights and functions of the central bank, and, as defined in
section 3, in particular never can withdraw the central bank’s licence to issue legal tender. Otherwise,
the central bank could not risk sustaining repeated losses. This also includes the assumption that
external events do not exist which can do so, such as wars or revolutions.20 As shown in section 3, the
central bank can in that case always continue performing its tasks regardless of the level of capital and
normally will even return to profitability in the long run. Regardless of its profitability and capital
position, it can also fund whatever operating expenses it deems to be necessary. It should be noted that
there are exceptions to this assignment of the right to issue legal tender to the central bank, such that
one should not regard the threat of losing the licence to issue legal tender as completely irrelevant.
Where central bank capital matters is probably in between these two extremes. This, indeed, is
probably where to place most of the world’s central banks. In practice, total independence of the
central bank now and in the future, as assumed in our basic model, is impossible, since no government
can commit future governments (whether they obtain power by election, war, or revolution) not to
change the central bank law, even if that is written in a constitution. Complete lack of independence,
on the other hand, is undesirable because of the conflict between the Government’s short-run
objectives at any point in time, and its objectives in the long run (i.e., the problem of time
inconsistency developed in the Barro-Gordon model). Thus, partial independence of the central bank
is the only possible and desirable solution, leaving us with the question why exactly capital matters in
this case for monetary policy.
Understanding precisely the way central bank capital becomes relevant in this case would require
modelling the relationship between the Government and the central bank. We will not aim here at
modelling this relationship thoroughly; nevertheless, some of the key aspects are discussed and their
impact on the model developed in the previous section is examined. Consider first what exactly
20
Technical innovations that lead to lower growth rates of banknotes or even to a disappearance of banknotes are not the
issue here, since the central bank then can still issue deposits. That is in any case the only thing it is doing actively, already
today.
ECB
September 2004 25
happens in case the privilege to issue legal tender is withdrawn as could happen de facto as a result of
dollarisation. Although the right to issue legal tender is not traditionally seen as a relevant for
institutional independence, it would be an important factor when assessing a central bank’s financial
independence (see Martínez-Resano [2004]). If the central bank lost the right to issue currency, it
would still need to pay its expenses (salary, etc.) in a new legal tender that it does not issue. Also,
banknotes and outstanding credits would need to be redeemed in the new currency at a certain fixed
exchange rate. Consider here the two cases of central banks with positive and with negative capital
with a very simple balance sheet consisting only in Capital, Banknotes, and monetary policy
operations.
Two central banks, before their right to issue legal tender is withdrawn
Positive Capital Central Bank Negative Capital Central Bank

Monetary Banknotes Capital Banknotes
policy Capital (negative) Monetary policy
operations operations
After the withdrawal of the right to issue legal tender, both central banks become normal financial
institutions. After liquidating their banknotes and monetary policy operations, their balance sheets take
the following shape:
Two central banks, before their right to issue legal tender is withdrawn
Positive Capital (former) Central Bank Negative Capital (former) Central Bank
Financial assets Capital Capital (negative) Financial debt
Obviously, the second institution is bankrupt, and the holders of its banknotes and of liquidity
absorbing monetary policy operations are not likely to recover their claims (as the no Ponzi game
condition should hold in this case21). Also, the institution will immediately have to stop paying salaries
and pensions, etc. In case of a positive probability of withdrawal of the right to issue legal tender,
central bank capital and profitability will thus matter. In the case of negative capital, the following
issues will arise:
• The staff and decision making bodies of the central bank have incentives to get out of the negative
capital situation by lowering interest rates below the neutral level, which in turn triggers inflation,
and eventually an increase of the monetary base up till positive capital is restored.
21
It has also been argued that even for a central bank with certainty about the right to issue legal tender, the no-Ponzi game
condition should hold. While in practice, this is probably true, we argue that in theory, if taking indeed the strict assumption
of an eternal guarantee of the right to issue legal tender, Ponzi game considerations are not relevant for the central bank as it
will with certitude, by definition never encounter liquidity problems.
ECB

September 2004
• Negative capital weakens the bargaining position of the central bank towards the Government,
since the Government’s admittedly somewhat remote threat to withdraw from the central bank the
right to issue legal tender becomes more worrisome for the central bank. Thus, with negative
capital, the central bank will tend to be more pliable towards the Government, even if it does not
want to ask the Government to re-capitalise it.
• The markets will have reasons to anticipate less stability-oriented behaviour of the central bank,
which drives up inflationary expectations.
• The holders of the monetary base may also feel uncomfortable from the mere risk of losing their
money in case that the central bank loses its right to issue legal tender – even if this central bank
would never be tempted to trigger inflation.
One may thus conclude from this approach that the higher the likelihood of a central bank to lose its
right to issue legal tender, the more important central bank capital becomes. As the likelihood of such
an event will however never be zero, central bank capital will always matter.22 Once this conclusion is
drawn, one can start deriving, through simulations, which level of central bank capital is adequate to
ensure a monetary policy aiming exclusively at maintaining price stability.
Assuming that a financially weak central bank wishes to restore its capital to a positive level, how
could this be done? It could get out of a loss-making situation, for instance, by buying back all its
outstanding liquidity-absorbing open market operations (e.g. debt certificates). This would lead to
corresponding excess reserves and zero interest rates in the money market. Since excess reserves do
not need to be remunerated, this frees the central bank immediately from this source of losses. If the
zero interest rates implied by the excess reserves are too low from a macroeconomic point of view
(which they normally are), they will trigger inflation. If this inflation is not stopped by the central bank
through an increase in interest rates, and even accelerates, then the central bank will eventually return
to profits since banknotes will increase such as to allow the central bank to again purchase interest rate
bearing assets (M>0) after some time. For instance, when inflation rate has reached 1000%, the central
bank will soon need to do liquidity-providing operations to provide enough liquidity for banknotes,
and it can then again set interest rates at some level which at least avoids that inflation rates accelerate
further (which would mean, in the model, setting interest rates at 1002%).
In order to gain the public’s trust in the currency, the central bank will thus normally care about
profitability and positive capital. Therefore, one may, in the case of negative capital, substitute the
~
interest rate generated by the Taylor rule iM ,t by an interest rate iM ,t determined as follows (with θ<0
a constant):
~
iM ,t = min(4 + θ , iM ,t )
22
One may add to that logic softer arguments explaining the preference of central bankers for positive capital, and thus their
possible readiness to deviate from a price stability-oriented interest rate policy in the event that capital and profits are
negative. For instance, one may conjecture that higher salary levels of central bankers and job-related privileges may be
easier to justify in an organisation that is profitable, than in one that makes losses and survives only thanks to the right to
issue legal tender that is assigned to it by the Government. There may also be higher levels of non-monetary income, e.g., in
form of prestige, if one works for a profitable organisation, rather than a permanently loss-making state bureaucracy.
ECB
September 2004 27
The functional form given to the capital term in this equation is, of course, ad hoc. It implies that if
capital is negative, the central bank no longer reacts to an increase of inflation (reflected in the
suppression of the inflation term) and even reduces rates further, by an amount corresponding to θ.
Simulating again the major time series in the case of such a modified policy function yields the
following charts (figures 7 and 8), obtained with θ = -1:
Figure 7: Evolution of inflation and interet rates: a non-profitable bank with

negative initial capital and preference for returning to profitability
inflation
yield of mon. pol. operations
S i 4
80
70
annual percentage points
60
50
40
30
20
10
-10
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96
simulation period
Figure 8: Evolution of monetary policy operations and profit: a non-profitable bank

with negative initial capital and preference for returning to profitability
Series2
Series1
12000 350
9000 250
6000
150
accounting units
3000
50
0
-50
-3000
-150
-6000
-9000 -250
-12000 -350
1 6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96
simulation period
ECB

September 2004
Obviously, the central bank ends up much faster in the domain of positive capital and profitability – at
the expense of temporarily higher inflation rates.
To see what the initial capital implies for the likelihood of the central bank falling temporarily into the
domain of negative capital (triggering inflation), we simulated the evolution of the central bank
balance sheet and of interest rates and inflation for initial capital levels between -100 and + 100. The
following chart reveals that the initial level of capital is relevant for the expected (more precisely: the
median) level of inflation, whereby the point where the inflation curve reaches the 2% target level
depends on the parameters of the model. In the case of the standard specification of parameters used so
far, such a relevance is only obtained for negative values of capital. This is due to the fact that our
central bank over time quickly accumulates positive capital and that there are no direct shocks to
profits. If, however, there is a certain likelihood of a large negative shock to profit (due e.g. to a
foreign exchange revaluation or “contingent liabilities” as formulated by Blejer and Schumacher
[1999]), then the positive relationship between capital and inflation performance extends into positive
capital whenever the central bank has a preference for profitability at the expense of higher inflation
(modelled by θ<0). One may then calculate the “value at risk” of the central bank (not only taking into
account its risky assets, but also the functions it may have to fulfil), and determine a capital that with,
say, a 95% probability ensures that within one year capital will not be exhausted. This is the approach
basically taken by Ernhagen et al [2002]. Here, we simply simulate the threat of losses by assuming
that, with a likelihood of 10%, the bank makes an annual loss equal to 33% of its banknotes in
circulation. Then, as figure 9 reveals, inflation performance becomes satisfactory ex ante only with an
initial capital level of around 40.
Figure 9: Median of average inflation over 100 periods

median of average inflation over 100 periods (no loss making episodes)
median of average inflation over 100 periods (10% likelihood of loss making episode)
16
14
12
10
inflation
0
-100 -80 -60 -40 -20 0 20 40 60 80 100
initial capital
Note: For each initial capital, a simulation with 200 replications was run and the median 100-period-average inflation was
computed. In the first scenario, the profit rule is given strictly by equation 9 in the system, while the second scenario
incorporates a 10% likelihood (independent draw each period) of making a loss equivalent to one third of banknotes and a
preference of the central bank for returning fast to profitability (θ=-1). In both simulations, B0=100, F0=0.
ECB
September 2004 29
Selecting a level of capital that is deemed sufficient to keep the probability of inflation being triggered
by lack of capital sufficiently low therefore means choosing some non-negative level of capital such
that the likelihood that the central bank falls back into negative capital in the future remains negligible.
Indeed, central banks tend to have non-negative capital. The implication of temporary huge losses and
implied negative capital tend to be mitigated by rules that the government may re-capitalise the central
bank in such cases. A re-capitalisation rule is a full substitute for capital in so far as it is
unconditionally automatic. The more discretion that is given to the government in its decision to re-
capitalise, the less such a rule will work as a substitute for capital to achieve a good inflation
performance.
Besides the described relationship between central bank capital and the credibility of monetary policy
objectives, it should be noted that the level of capital has also two more direct operational
consequences, which might also be integrated into a model:
• If the profitability of a central bank is at risk, there may not only be incentives to neglect monetary
policy objectives, but also to choose monetary policy assets or other financial assets with higher
expected yield. This may appear undesirable because it implies allocation of resources of the
central bank away from its core tasks, or because one may generally feel that public institutions
should only engage in the most basic investment activities and take no unnecessary risk, even if
this is at the detriment of expected profits.
• One may argue that defining a central bank with no constraint in terms of capital means
introducing another state institution without balance sheet restriction, which may thus have leeway
to waste scarce resources. On the other hand, if it is clear that whenever capital falls below some
threshold, the central bank needs to be bailed out by the government in a somewhat painful
procedure, then capital restrictions may provide incentives to the central bank to act efficiently (an
effect which is however limited by the fact that under normal conditions, central banks generate
profits even if they are not overly efficient).
As a final remark, it can be conjectured that the origin of the public’s perception about the necessity of
a strong financial position of the central bank may be related to the era of the gold standard when a
central bank could indeed become insolvent and negative capital thus indicated a serious problem and
should probably have triggered a run on its gold reserves. Financial markets may also perceive a
reduction of central bank capital as increasing the probability that the relationship (of independence)
between the State and central bank will be reviewed and possibly changed.
6. Some considerations regarding profit sharing rules

For most starting points of central bank balance sheets and capital, the rule governing the sharing of
profits between the central bank and the government will be critical for determining the evolution of
ECB

September 2004
capital and profitability, and thus the likelihood that central bankers may be tempted to deviate from a
stability-oriented interest rate policy due to worries about their bank’s capital. So far, we had assumed
that the government receives simply one half of the profit of the central bank, as long as profits are
positive, and that otherwise, no transfer takes place. However, alternative rules can be built in and
simulated easily. A more general class of profit sharing rules would be, for G being the transfer to the
government, λ and ψ being positive coefficients between 0 and 1 and Pt + , Pt − being positive and
negative profits, such that always Pt + × Pt − = 0 .
Gt = λPt + + ψPt −
The profit-sharing rule assumed above is a special case of this rule with λ = 0.5;ψ = 0 . Many central
banks have profit-sharing rules that make reference to some balance sheet indicator, such as the level
of capital or a capital adequacy ratio. For instance, often the share of profits to be transferred to the
Government increases if the capital has reached a certain level, either in absolute terms, or in relation
to some other balance sheet items. There may even be a “bail-out” rule in case of negative capital.
ì if Ct ≥ f ( X t ) : Gt = λ1Pt +
ï
ï
íif 0 < Ct < f ( X t ) : Gt = λ2 Pt +
ï
ïif 0 > Ct : Gt = − ρCt + λ2 Pt + − ψPt −
î
with λ1 > λ2 and X t being the array of relevant balance sheet variables. If the constant ρ is set to 1,
then the central bank is fully bailed out whenever it reaches negative capital. One simple specific rule
is the one which defines a capital ratio, as for instance the capital divided by the sum of net financial
assets, including monetary policy assets, and assigns a share 1 > λ1 > 0 of the profit whenever this
“capital adequacy ratio” is ensured, but leaves all profit with the central bank when it is not. Assuming
that the capital adequacy ratio is 8%, one thus obtains the following rule:
ìif Ct ≥ 8% * ( M t + Ft ) : Gt = λ1Pt +
ï
í
ï else : Gt = 0
î
There are no limitations to the design of profit-sharing rules. Each profit-sharing rule will, for a given
initial structure of the balance sheet and an evolution of exogenous shocks, lead to a different further
evolution of the balance sheet and of capital and profits in particular. If one adds the assumption that
central bankers will be influenced by negative capital in their monetary policy decisions, i.e. that they
will tend to be too loose whenever capital is negative, then also the inflation and interest rate time
series may be influenced by the choice of the profit sharing parameters. In principle, the optimality of
the parameters of the profit-sharing rule can thus be investigated by simulating the model presented
above.
ECB
September 2004 31
7. Conclusion
In this paper, the role of central bank capital was revisited by setting up a simple model of the
relationship between the central bank balance sheet, interest rates and inflation. It builds on work of
previous papers on central bank capital, which did however not attempt to formally model these
relationships. The first part of the paper showed that under a series of strong assumptions a dichotomy
between the central bank’s balance sheet structure and its ability to maintain low inflation could be
constructed (what Martinez-Resano [2004], calls a Modigliani-Miller theorem for central banks). The
assumptions were analysed under which one can show that in the steady state, central banks always
return to profitability in the long run, regardless of starting levels of operating costs and capital
(whereby the “long run” was however shown to easily go beyond human life expectancy, and thus
having doubtful implications on actual decision making). In this model, the only real dangers of not
returning to profitability in the long run are the case of a deflationary trap (from which there is no
escape in our simple Wicksellian setting), or if banknote growth slows down too much (namely below
the growth rate of operating costs). In addition, it was shown that the length of the central bank
balance sheet in the long run is positively correlated with initial capital, contradicting the impression
given in the previous literature that loss accumulation and thus the growth of the “debt” of the central
bank would be of an exceptionally dynamic nature.
The second part of the paper discusses a model variant under more realistic assumptions. It then
becomes rational for central bankers to care about the level of capital, and to use the tools in their
hands, namely open market operations and interest rate policy, to influence the evolution of capital.
This idea is integrated in a simple way into the model to show how, indeed, the initial level of central
bank capital becomes relevant for the long run average inflation performance of a central bank. If
shocks to central bank profits are rather limited, a non-negative capital is sufficient. If however also
some non-zero probability is assigned to large losses, then the positive relationship between inflation
performance and capital is also valid for positive levels of capital. The required level of positive
capital ensuring good inflation performance will depend on the risks in the central bank balance sheet
and on “contingent liabilities” i.e. possible off-balance sheet obligations. Finally, the paper briefly
discusses profit sharing rules, which are obviously closely linked to the capital issue. Although no
simulations are performed, it is argued that the impact of profit-sharing rules on balance sheet items,
interest rates and inflation can be simulated in the proposed model in exactly the same way as the
impact of initial capital. A fully automated and fully credible rule of re-capitalisation by the
government of the central bank in case of losses can be regarded as a substitute for positive capital.
Since such rules are however difficult to implement in practice, positive capitals seems to remain a
key tool to ensure that independent central bankers always concentrate on price stability in their
monetary policy decisions.
ECB

September 2004
Annex 1: Proof of the proposition of long term return to profitability and
positive capital
Proof: We assume that the dynamics of the system follows equations 1-9 (section 3). Since
inflation and monetary policy rate are fixed at 2% and 4% (this follows from equations 1 and 6,
respectively) and financial assets are assumed to be constantly zero, we can simplify the system to 23:
ìC t = C t −1 + Pt −1
ï B = B + B (0.02 + π~) = (1 + ~ i ) Bt −1 ~
ï t t −1 t −1 ì i = i / 100
(1) í , where í~
ïM t = Bt + C t îπ = π / 100
ïî Pt = i M ,t M t − q 0 (1 + π~ ) t
The goal is to prove that from some t0 onwards, profit, Pt , becomes positive (and tends to infinite as
t tends to infinite). We will obtain an expression for M t , the first summand of Pt (see fourth equation
in system (1)).
Substituting Pt by its value, we can further reduce the system to equations:
ìM t = Bt + C t
ï ~ ~
(2) í(1 − (1 + i ) L)C t = i LBt − q (1 + π~ ) t −1
ï(1 − (1 + ~
i ) L) Bt = 0
î
where L denotes the backward operator, e.g. LBt = Bt −1 .
~
Multiplying the first equation in (2) by the operator (1 − (1 + i ) L) we obtain:
~ ~
(1 − (1 + i ) L) M t = i LBt − q (1 + π~ )t −1
Developing the autoregressive part on the left-hand side, we obtain
~
( )~ ~
( )
Þ M t = i LBt − q (1 + π~ )t −1 + (1 + i ) i LBt −1 − q (1 + π~ )t − 2 + ... + (1 + i )t −1 (i LB1 − q ) + (1 + i )t M 0
~ ~ ~
~
and using using Bt = (1 + i )t B0
( ) ( )
Þ M t = i (1 + i )t −1 B0 − q(1 + π~ ) t −1 + (1 + i ) i (1 + i ) t −2 B0 − q (1 + π~ )t −2 + ... + (1 + i ) t −1 (i B0 − q ) + (1 + i )t M 0
~ ~ ~ ~ ~ ~ ~ ~
Now we regroup the summands,
~ ~ ~ ~ ~
Þ M = i t (1 + i )t −1 B − [(1 + π~ )t −1 + (1 + i )(1 + π~ ) t −2 + ... + (1 + i )t −1 ]q + (1 + i )t M
t 0 0
and observe that the geometric sum of the second term can be simplified, obtaining
é ~ t ù
æ (1 + i ) ö
ê (1 + π~ ) t −1[çç ~ ÷÷ − 1] ú
~ ~
Þ M t = i t (1 + i )t −1 B0 − êê è (1 + π ) ø ú q + (1 + ~
i )t M 0
~ ú
1+ i
ê −1 ú
êë 1 + π~ úû
23
Since capital is negative, there is no need to make assumptions about the rate at which profit is distributed to
the Government, at least for describing the dynamics until profitability is reached.
ECB
September 2004 33
Returning to the fourth equation in system (1), we obtain the following expression for Pt :
é ~ t ù
æ (1 + i ) ö
ê (1 + π~ ) t −1 [çç ÷ − 1] ú
~ ~ t −1 ê è (1 + π~ ) ÷ø ú ~
Þ Pt = i t (1 + i ) B0 − ê ~ ú q + (1 + i ) t M 0 − (1 + π~ ) t q 0
1+ i
ê −1 ú
ê 1 + π~ ú
ë û
It is clear that the first term (positive part) dominates the other three as t increases (since its
exponential term is multiplied by t , while the other two summands only have an exponential term
with at most the same base). Hence Pt grows monotonically and Pt0 > 0 from some t0 onwards.
The intuition for this result is as follows: The only source contributing to profit in a positive way is
~
banknotes, which grow each year by a factor of (1 + i ) . This growth factor is to be compared to the
~ ~
one of capital, which is in any case lower than i (1 + i ) , and to the one of operating costs, namely
~
(1 + π~ ) < (1 + i ) .
However, it can be noted from the second last expression in the proof, if operating costs grew at any
~
rate higher than i = 0.04 , then profits would inevitably remain negative. For the limit case when
~
operating costs grow at exactly the same rate as banknotes, i = 0.04 , the sign of the profit would
only depend on the initial conditions. It can therefore be concluded that, apart from the general
underlying assumption in the model of steady growth in banknotes coupled with inflation, productivity
gains of the central bank over time (operating costs growing only at the inflation rate, and not at the
nominal interest rate) are key to the proposition.
Annex 2: Proof of the proposition that if a central bank has larger starting
capital it will always have a longer balance sheet
Proof: As shown before, assuming that inflation and monetary policy rate are fixed at 2% and 4% and
financial assets are constantly zero and all profit is retained, we can simplify the system to:
é ~ t ù
~ t −1 æ (1 + i ) ö
ê (1 + π ) [çç ~ ÷÷ − 1] ú
~ ~
Þ M t = i t (1 + i )t −1 B0 − êê è (1 + π ) ø ú q + (1 + ~
i )t M 0
~ ú
1+ i
ê −1 ú
êë 1 + π~ úû
Assume that there are two central banks “a” and “b”, and that C0a , C0b is their initial and capital,
respectively (etc). First, if 0 < C0a < C0b , then ∀t ∈ ℵ+ : M ti = Lit , i = a, b . It is clear that since
ECB

September 2004
M 0a < M 0b then ∀t ∈ ℵ+ : Lat < Lbt . Second, if C0a < C0b and C0a < 0 , then
∀t ∈ ℵ+ : M ti = Lit , i = a, b .
References
Bank of Japan (2003) “Challenges for Monetary Policy in Japan”, speech given at the 60th anniversary
of Japan Society of Monetary Economics, June 1st.
Blejer, M. and Schumacher, L. (2000) “Central Banks Use of Derivatives and Other Contingent
Liabilities: Analytical Issues and Policy Implications”, IMF Working Paper, no. 66, International
Monetary Fund, Washington D.C.
Bresciani-Turroni, C. 1937, The Economics of Inflation, London, Allen & Unwin
Cagan, P. D. 1956, The monetary dynamics of hyperinflation, in Friedman, M. ed., Studies in the
Quantity Theory of Money, Chicago, University of Chicago Press
Dalton, J. (1999) “Central Bank Financial Soundness, Profit Measurement and Distributional Issues”,
Briefing Paper for MAE Central Bank Accounting Workshop in Abu Dhabi (Washington:
International Monetary Fund).
Dalton, J. and C. Dziobek (1999) “Central Bank Losses and Experiences in Selected Countries”, MAE
Technical Note TN/99/1 (Washington: International Monetary Fund).
Ernhagen, T., Vesterlund, M. and Viotti, S. (2002) “How Much Equity Does a Central Bank Need?”,
Sveriges Riksbank Economic Review, 2/2002.
Gros, D. and Schobert, F. (1999) “Excess Foreign Exchange Reserves and Overcapitalisation in the
Eurosystem”, CEPS Working Document, no. 128.
Hawkins, J. (2004) “Central Bank Balance Sheets and Fiscal Operations”, Bank for International
Settlements, BIS Papers, no. 20.
Jeanne, O. and Svensson, L.E.O. (2004), “Credible commitment to optimal escape from a liquidity
trap: the role of the balance sheet of an independent central bank”, manuscript,
http://www.princeton.edu/~svensson/papers/js.pdf
Leone, A. (1993) “Institutional Aspects of Central Bank Losses”, IMF Paper on Policy Analysis and
Assessment 93/14 (Washington: International Monetary Fund)Martínez-Resano, R. J. (2004)
“Central Bank Financial Independence”, Banco de España Occasional Paper, no. 0401.
Martínez-Resano J. R. (2004) “Central Bank Financial Independence”, Banco de España, Occasional
Paper no. 0401
Pringle, R. and Courtis, N. (1999) Objectives, Governance and Profits, Central Banking Publications.
Pringle, R. (2003) “Why Central Banks Need Capital”, Central Banking, XIV, no. 1.
Stella, P. (1997) “Do Central Banks Need Capital?”, IMF Working Paper, no 83, International
Monetary Fund, Washington D.C.
Stella, P. (2002) “Central Bank Financial Strength, Transparency, and Policy Credibility”, IMF
Working Paper, no. 137, International Monetary Fund, Washington D.C.
Stella, P. (2003) “Why Central Banks Need Financial Strength”, Central Banking, XIV, no. 2, p.27.
Sullivan, K. (2002) “Profits, Dividends and Capital – Considerations for Central Banks”, speech given
at the LEG Seminar for Central Bank Lawyers.
Taylor, J. (1993) “Discretion versus Rules in Practice”, Carnegie-Rochester Series on Public Policy,
39, pp. 195-214.
Vaez-Zadeh, R. (1991) “Implications and Remedies of Central Bank Losses”, in Downes, Patrick and
Reza Vaez-Zadeh, The Evolving Role of Central Banks, International Monetary Fund, Washington
D.C.
Woodford, M. (2003), Interest rates and prices, Princeton University Press.
ECB
September 2004 35
European Central Bank working paper series
For a complete list of Working Papers published by the ECB, please visit the ECB’s website
(http://www.ecb.int)
366 “The informational content of over-the-counter currency options” by P. Christoffersen and

S. Mazzotta, June 2004.
367 “Factor substitution and factor augmenting technical: progress in the US: a normalized supply-side
system approach” by R. Klump, P. McAdam and A. Willman, June 2004.
368 “Capital quality improvement and the sources of growth in the euro area” by P. Sakellaris
and F. W. Vijselaar, June 2004.
369 “Sovereign risk premia in the European government bond market” by K. Bernoth, J. von Hagen
and L. Schuknecht, June 2004.
370 “Inflation persistence during periods of structural change: an assessment using Greek data”
by G. Hondroyiannis and S. Lazaretou, June 2004.
371 “Inflation persistence: facts or artefacts?” by C. R. Marques, June 2004.
372 “The operational target of monetary policy and the rise and fall of reserve position doctrine”
by U. Bindseil, June 2004.
373 “Technology shocks and robust sign restrictions in a euro area SVAR” by G. Peersman and
R. Straub, July 2004.
374 “To aggregate or not to aggregate? Euro area inflation forecasting” by N. Benalal,
J. L. Diaz del Hoyo, B. Landau, M. Roma and F. Skudelny, July 2004.
375 “Guess what: it’s the settlements!” by T. V. Koeppl and C. Monnet, July 2004.
376 “Raising rival’s costs in the securities settlement industry” by C. Holthausen and
J. Tapking, July 2004.
377 “Optimal monetary policy under commitment with a zero bound on nominal interest rates”
by K. Adam and R. M. Billi, July 2004.
378 “Liquidity, information, and the overnight rate” by C. Ewerhart, N. Cassola, S. Ejerskov
and N. Valla, July 2004.
379 “Do financial market variables show (symmetric) indicator properties relative to exchange
rate returns?” by O. Castrén, July 2004.
380 “Optimal monetary policy under discretion with a zero bound on nominal interest rates”
by K. Adam and R. M. Billi, August 2004.
381 “Fiscal rules and sustainability of public finances in an endogenous growth model”
by B. Annicchiarico and N. Giammarioli, August 2004.
ECB

September 2004
38 2 “Longer-term effects of monetary growth on real and nominal variables, major industrial
countries, 1880-2001” by A. A. Haug and W. G. Dewald, August 2004.
383 “Explicit inflation objectives and macroeconomic outcomes” by A. T. Levin, F. M. Natalucci

and J. M. Piger, August 2004.
384 “Price rigidity. Evidence from the French CPI micro-data” by L. Baudry, H. Le Bihan,
P. Sevestre and S. Tarrieu, August 2004.
385 “Euro area sovereign yield dynamics: the role of order imbalance” by A. J. Menkveld,
Y. C. Cheung and F. de Jong, August 2004.
386 “Intergenerational altruism and neoclassical growth models” by P. Michel, E. Thibault

and J.-P. Vidal, August 2004.
387 “Horizontal and vertical integration in securities trading and settlement” by J. Tapking
and J. Yang, August 2004.
388 “Euro area inflation differentials” by I. Angeloni and M. Ehrmann, September 2004.
389 “Forecasting with a Bayesian DSGE model: an application to the euro area” by F. Smets
and R. Wouters, September 2004.
390 “Financial markets’ behavior around episodes of large changes in the fiscal stance” by
S. Ardagna, September 2004.
391 “Comparing shocks and frictions in US and euro area business cycles: a Bayesian DSGE
approach” by F. Smets and R. Wouters, September 2004.
392 “The role of central bank capital revisited” by U. Bindseil, A. Manzanares and B. Weller,
September 2004.
ECB
September 2004 37

Working Paper Series: The Role of Central Bank Capital Revisited

Uploaded by

Copyright:

Available Formats

Working Paper Series: The Role of Central Bank Capital Revisited

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Working Paper Series: The Role of Central Bank Capital Revisited

Uploaded by

Copyright:

Available Formats

WO R K I N G PA P E R S E R I E S

All rights reserved.

Reproduction for educational and non-

The views expressed in this paper do not

The statement of purpose for the ECB

ISSN 1561-0810 (print)

Keywords: Central Bank Capital, Central Bank Independence

4 Working Paper Series No. 392

6 Working Paper Series No. 392

2. The literature on central bank capital

8 Working Paper Series No. 392

10 Working Paper Series No. 392

12 Working Paper Series No. 392

3. A simple model based on a central bank with no liquidity constraints

Monetary policy operations (“M”) Banknotes (“B”)

Other Financial Assets (“F”) Capital (“C”)

14 Working Paper Series No. 392

16 Working Paper Series No. 392

Example 1. A profit making central bank with positive capital

18 Working Paper Series No. 392

450 Initial banknotes: 20 Initial banknotes: 10

300 operating costs: 1

Figure 4: Period of first positive profit depending on initial capital

Initial capital: -200

20 Working Paper Series No. 392

22 Working Paper Series No. 392

The case of a central bank with negative capital

MPOs 400 CB Capital 500 Gvt.. Debt 500

The well-capitalised central bank

CB Capital 500 CB Capital 500 MPOs 600 Gvt. Debt 1500

24 Working Paper Series No. 392

Positive Capital Central Bank Negative Capital Central Bank

26 Working Paper Series No. 392

Figure 7: Evolution of inflation and interet rates: a non-profitable bank with

Figure 8: Evolution of monetary policy operations and profit: a non-profitable bank

28 Working Paper Series No. 392

Figure 9: Median of average inflation over 100 periods

6. Some considerations regarding profit sharing rules

30 Working Paper Series No. 392

32 Working Paper Series No. 392

34 Working Paper Series No. 392

366 “The informational content of over-the-counter currency options” by P. Christoffersen and

371 “Inflation persistence: facts or artefacts?” by C. R. Marques, June 2004.

36 Working Paper Series No. 392

383 “Explicit inflation objectives and macroeconomic outcomes” by A. T. Levin, F. M. Natalucci

386 “Intergenerational altruism and neoclassical growth models” by P. Michel, E. Thibault

You might also like