What Are the Effects
of the ECB’s N
egative
Interest Rate Policy?
Policy Department for Economic, Scientific and Quality of Life Policies
Directorate-General for Internal Policies
Author: Grégory CLAEYS
PE 662.922 - June 2021
EN
IN-DEPTH ANALYSIS
Requested by the ECON committee
Monetary Dialogue Papers, June 2021
Abstract
Several central banks, including the European Central Bank since
2014, have added negative policy rates to their toolboxes after
exhausting conventional easing measures. It is essential to
understand the effects on the economy of prolonged negative
rates. This paper explores the potential effects (and side effects)
of negative rates in theory and examines the evidence to
determine what these effects have been in practice in the euro
area.
This paper was provided by the Policy Department for Economic,
Scientific and Quality of Life Policies at the request of the
committee on Economic and Monetary Affairs (ECON) ahead of
the Monetary Dialogue with the ECB President on 21 June 2021.
What Are the Effects
of the ECB’s N
egative
Interest Rate Policy?
Monetary Dialogue Papers
June 2021
This document was requested by the European Parliament's committee on Economic and Monetary
Affairs (ECON).
AUTHOR
Grégory CLAEYS
1
, Bruegel
ADMINISTRATOR RESPONSIBLE
Drazen RAKIC
EDITORIAL ASSISTANT
Janetta CUJKOVA
LINGUISTIC VERSIONS
Original: EN
ABOUT THE EDITOR
Policy departments provide in-house and external expertise to support European Parliament
committees and other parliamentary bodies in shaping legislation and exercising democratic scrutiny
over EU internal policies.
To contact the Policy Department or to subscribe for email alert updates, please write to:
Policy Department for Economic, Scientific and Quality of Life Policies
European Parliament
L-2929 - Luxembourg
Email: Poldep-Economy-[email protected]
Manuscript completed: June 2021
Date of publication: June 2021
© European Union, 2021
This document was prepared as part of a series on “Low for Longer: Effects of Prolonged Negative
Interest Rates”, available on the internet at:
https://www.europarl.europa.eu/committees/en/econ/econ-policies/monetary-dialogue
Follow the Monetary Expert Panel on Twitter: @EP_Monetary
DISCLAIMER AND COPYRIGHT
The opinions expressed in this document are the sole responsibility of the authors and do not
necessarily represent the official position of the European Parliament.
Reproduction and translation for non-commercial purposes are authorised, provided the source is
acknowledged and the European Parliament is given prior notice and sent a copy.
For citation purposes, the publication should be referenced as: Claeys, G, What Are the Effects of the
ECB’s Negative Interest Rate Policy? Publication for the committee on Economic and Monetary Affairs,
Policy Department for Economic, Scientific and Quality of Life Policies, European Parliament,
Luxembourg, 2021.
1
The author is grateful to Marta Dominguez Jimenez, Lionel Guetta-Jeanrenaud and Monika Grzegorczy for excellent research assistance
and to Rebecca Christie, Maria Demertzis, Francesco Papadia, Nicolas Véron and Guntram Wolff for their useful comments.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
3 PE 662.922
CONTENTS
LIST OF BOXES 4
LIST OF FIGURES 4
LIST OF TABLES 4
LIST OF ABBREVIATIONS 5
EXECUTIVE SUMMARY 6
INTRODUCTION 8
EFFECTS OF CENTRAL BANK NEGATIVE RATES IN THEORY 10
2.1. Potential positive effects on lending, output and ultimately inflation 10
2.2. Possible negative side effects for the financial sector 11
2.3. Possible reduced response of economic activity to interest rates 12
EFFECTS OF CENTRAL BANK NEGATIVE RATES IN PRACTICE 13
3.1. Transmission to market (interest and exchange) rates 13
3.2. Effects of negative rates on banks 15
3.2.1. Direct cost of a negative deposit facility rate for euro-area banks 15
3.2.2. Impact on other components of the balance sheets of commercial banks 18
3.3. Macro and financial stability impact 22
CONCLUSIONS: WHAT DO WE KNOW AND WHAT SHOULD THE ECB DO? 24
REFERENCES 27
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 4
LIST OF BOXES
Box 1: Estimating the pass-through of monetary policy to bank rates 20
LIST OF FIGURES
Figure 1: Interest rates 9
Figure 2: Policy and market interest rate in the euro area 14
Figure 3: Euro nominal effective exchange rate 14
Figure 4: Excess liquidity held at the ECB and direct costs for the banks of the euro area 16
Figure 5: Direct cost of a negative deposit facility rate per country 17
Figure 6: Composite bank interest rates for households and corporations (%) 19
Figure 7: Bank lending to the private sector in the euro area (YoY, %) 22
Figure 8: Estimated impact of three ECB rate cuts (by 10 bps) on macro variables 23
LIST OF TABLES
Table 1: Relationship between monetary policy rate and banking rates 20
Table 2: Relationship between and banking rates in case of negative rates 21
What Are the Effects of the ECB’s Negative Interest Rate Policy?
5 PE 662.922
LIST OF ABBREVIATIONS
European Central Bank
Effective lower bound
European Union
Gross domestic product
Negative interest rate policy
Non-financial corporations
Pandemic emergency purchase programme
Quantitative easing
Targeted longer-term refinancing operations
Zero lower bound
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 6
EXECUTIVE SUMMARY
• Since the global financial crisis, several central banks around the world have added negative
policy rates to their toolboxes after exhausting conventional easing measures.
• The European Central Bank introduced its negative interest rate policy (NIRP) in June 2014
when it decided to cut for the first time its deposit facility rate below 0%, to -0.1%. Since then, the
ECB has cut its deposit rate four more times, each time by 10 basis points, to reach -0.5% in
September 2019.
• NIRP has now been in place for seven years in the euro area and markets currently expect
rates to stay negative for at least five years. It is therefore crucial to fully understand the effects
on the economy of prolonged negative rates.
• Central banks which have adopted NIRP are generally positive about its use in helping them
fulfil their objectives. However, NIRP remains controversial and has been accused of causing
significant side effects in particular for the financial sector. Indeed, the existence of various
frictions (e.g. physical cash and possible cognitive biases) means that going below 0% could lead
to additional channels and non-linear effects of monetary policy.
• In this paper, we look at the potential (positive and negative) effects of negative rates and
describe the potential mechanisms at work when this measure is used, before turning to the
evidence.
• The experience of the euro area since 2014 shows that the negative deposit facility rate is
fully transmitted to the benchmark overnight rate and then propagates along the whole
yield curve.
• There is some evidence (in particular from Denmark and Switzerland) that NIRP also impacts
the exchange rate through a change in cross-border flows.
• As far as bank rates are concerned, it appears that the effects of rate cuts in negative territory
are not different to “standard” rate cuts. Like them, they reduce banks’ interest margins because
rates on banks’ assets are more sensitive to policy rates than rates on banks’ liabilities, but this effect
does not appear to be amplified below 0% (at least at the current minimal level of negative rates).
• A negative deposit facility rate implies some cost for banks that hold excess reserves at the
ECB. This cost has been growing significantly, in particular since the introduction of the pandemic
emergency purchase programme (PEPP) in 2020. Moreover, this cost is concentrated in the
countries that host the main European financial centres (where investors that have sold assets to
the ECB park their euro deposits).
• A more difficult and more fundamental question about negative rates is whether output,
employment and inflation are still sensitive to these financial variables when rates are very
low or negative. The ECB’s own research is confident that the effect of NIRP on these variables in
the short- to medium-term is positive. Potential negative side effects do not appear to have
materialised in a significant way for the moment.
• However, a long period of negative rates could also entail some medium- to long-term risk,
in particular in terms of financial stability. Financial institutions seem to have increased their risk-
taking with the advent of negative rates. Whether this risk-taking is excessive remains to be seen
and will need to be monitored carefully.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
7 PE 662.922
• The direct cost incurred by banks that hold excess reserves appears to be the most tangible
side effect at this stage. The easiest solution to mitigate this would be to adjust the two-tier
system put in place in 2019 and to increase the quantity of excess reserves exempted from the
negative rate.
• If the ECB believes it is approaching the reversal rate but needs to provide more monetary
easing, it should refrain from cutting its deposit facility rate again, and instead cut further its
targeted longer-term refinancing operations rate.
• Negative rates for a too-long period could lead to financial instability. The best way to deal
with potential financial imbalances is to use macroprudential tools. However, the euro area’s
current macroprudential framework might not be capable of playing its role. It is therefore critical
to build a better set-up for the use of macroprudential tools, so that they can be used forcefully and
in a timely way when needed.
• Finally, given the current economic situation, the ECB should be extremely cautious and not
rush before exiting negative rates. Even if negative rates proved to be ineffective, it would be
extremely dangerous to exit negative rates too soon as it could harm the post-COVID-19 recovery
and destabilise European sovereign debt markets.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 8
INTRODUCTION
Since the global financial crisis, and after exhausting conventional easing measures, several central
banks around the world have added negative policy rates to their toolboxes, in addition to other
unconventional measures such as asset purchases and forward guidance. The central bank of Sweden,
in July 2009, was the first to move one of its policy rates into negative territory. It was followed by the
central banks of Denmark (in July 2012), Switzerland (in January 2015), Japan (in February 2016) and by
the European Central Bank (ECB).
In the euro area, the ECB introduced its negative interest rate policy (NIRP) in June 2014 when the ECB
Governing Council decided to cut for the first time the ECB’s deposit facility rate (DFR) – the main policy
rate to influence market rates since the global financial crisis – below 0% to -0.1%. Since then, the ECB
has cut its deposit rate four more times, each time by 10 basis points (bps), to reach -0.5% in September
2019.
How did the ECB end up resorting to negative rates? While the ECB’s main refinancing operations rate
averaged around 3% between the creation of the ECB in 1999 and the failure of Lehman Brothers in
September 2008, it has averaged less than 0.5% in the 13 years since then. One obvious reason for this
is that the ECB has in that time faced the two most important economic crises in almost a century and
had to provide highly accommodative monetary policy to fulfil its price stability mandate. Another
more fundamental reason is that interest rates have declined in the last four decades in advanced
economies and their central banks have had to adjust.
Indeed, according to the current macroeconomic consensus, the steady-state level of central bank
policy rates should be guided by the so-called neutral rate, i.e. the real rate compatible with inflation
around target and output at potential, which is driven by fundamental factors including productivity,
demographic growth and the saving behaviour of households. This rate is not directly observable, and
its measurement is highly uncertain, but most estimates point towards a significant decline in the
neutral rate, in particular since the global financial crisis (see, for example, the estimates of Holston et
al., 2016, in Figure 1 panel A). If a central bank’s reaction function can be characterised through a simple
Taylor rule
2
, with a neutral rate close to 0% and an inflation target of 2%, this means that the ECB’s
nominal steady-state policy rate would be around 2%, leaving the ECB without enough conventional
ammunition to face shocks, given that historically, central banks have cut rates by around 300 bps
during recessions
3
.
As a result, if the neutral rate were to remain at a low level, unconventional tools, and possibly NIRP,
would have to be used often in order to make monetary policy accommodative enough when there is
slack in the economy and inflation is below target. Actually, markets currently believe that the
overnight interest rate in the euro area will stay negative at least until 2026 (Figure 1, panel B).
2
The basic Taylor rule, following Taylor’s original specifications and coefficients (Taylor, 1993), looks like this: r=inflation+r*+0.5(inflation-
target) +0.5(output gap)
3
The US Fed cut its policy rates on average by 330 basis points during recessions from 1920 to 2018, the Bank of England by 290 bps from
1955 to 2018, and the central bank for Germany (the Bundesbank followed by the ECB) by 260 bps from 1960 to 2018 (Bruegel calculations
based on OECD, Fed, BoE and Bundesbank).
What Are the Effects of the ECB’s Negative Interest Rate Policy?
9 PE 662.922
Figure 1: Interest rates
Panel A: Neutral real rate estimates (%) Panel B: expectations for euro area overnight rate (%)
Note: Interest rate expectations as of May 2021, derived from EONIA zero-coupon swaps of different terms (1 year, 2 years, up
to 10 years), which provide information on market expectations of the compounded overnight EONIA over the contract term.
Expectations for 2022 interest rate, for instance, are derived through expected compounded EONIA over the next year (2021),
given by the 1-year swap, and expected compounded EONIA over the next two years (2021 and 2022), given by the 2-year
swap.
Source: Panel A: Holston et al. (2016), updated in 2021. Panel B: Bruegel based on Bloomberg.
If NIRP becomes an often-used monetary policy tool, it is crucial to fully understand the effects on the
economy of prolonged negative rates. Central banks which have been part of this experiment are
generally positive (see e.g., Schnabel, 2020) about the use of negative rates in helping them fulfil their
objectives (whether it is to bring inflation towards target, as in the euro area or in Sweden, or to stabilise
the exchange rate as in Denmark or Switzerland). However, NIRP remains controversial and has been
accused of causing significant side effects in particular for the banking sector. As a result, two major
central banks, the US Fed and the Bank of England (BoE), have refrained from using that tool despite
facing the same crises as the countries that have resorted to it.
This is because the effects of NIRP could be different to those of traditional rate cuts in positive territory,
and the net effect could more be ambiguous because of the presence of various frictions in the
economy: the existence of cash yielding a 0% nominal interest rate, cognitive biases of investors and
households, and financial and legal constraints. In this paper, we first look at the potential effects of
negative rates and describe the potential mechanisms at work when this measure is used (section 2),
before turning to the evidence (section 3). We conclude with a discussion on what the ECB can do to
reduce or avoid potential side effects (section 4).
0
1
2
3
4
5
6
1961
1965
1970
1974
1979
1983
1988
1993
1997
2002
2006
2011
2016
US Euro Area
-1
0
1
2
3
4
5
6
2001
2003
2006
2008
2011
2013
2016
2019
2021
2024
2026
2029
EONIA
(realised)
EONIA
(expectations)
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 10
EFFECTS OF CENTRAL BANK NEGATIVE RATES IN THEORY
2.1. Potential positive effects on lending, output and ultimately inflation
First, NIRP is the logical extension of the main monetary policy tool used in the two decades that
preceded the financial crisis: rate cuts. Compared to quantitative easing (QE), the transmission channels
for which are relatively complex, or to forward guidance, which relies on the role played by
expectations (and which could be fraught with time-consistency issues), NIRP is the most conventional
of the unconventional tools introduced since the financial crisis, and a much more mechanical tool.
Indeed, the main transmission channel of NIRP to the economy should be very similar to that of
traditional rate cuts. When needed, central bank rate cuts provide monetary accommodation through
the easing of financing conditions, which tends to boost credit demand for investment and
consumption, provide some fiscal space to governments, and thus increase aggregate demand and
inflation. Negative policy rates, as long as they are transmitted to bank and market rates, are supposed
to function in the same way.
However, the existence of various frictions (e.g. physical cash yielding a 0% nominal interest rate and
possible cognitive biases) means that going below 0% could lead to additional channels and non-
linear effects of monetary policy at the “zero lower bound” (ZLB), or more precisely at the “effective
lower bound” (ELB), which could be slightly below 0% because of the cost of cash storage. There could
therefore be several additional channels of central bank policy rate cuts in negative territory.
First, going negative could have a stronger impact on the whole yield curve. The reason is that
breaching the ZLB for the first time and/or announcing that negative rates will be part of the central
bank toolbox in the future, could lead investors who might not have thought negative rates were
possible to revise their expectations about what minimum rates can be in the future (by removing
any non-negativity restriction), and thus reduce rates along the whole yield curve.
Second, the traditional portfolio rebalancing effect that already exists with standard rate cuts
(pushing investors towards riskier assets in a search for yield) could be boosted by the aversion of some
investors to negative rates. This could be for contractual reasons, for instance if redemption at par is
guaranteed on some savings products. It could also be for behavioural reasons if economic agents
(households or corporations) are subject to some form of loss aversion. This could, for instance, lead
cash-rich corporations to increase fixed investment to avoid negative rates. At the bank level, portfolio
rebalancing could also be boosted by negative rates because of a “hot-potato” effect pushing banks to
purchase various assets in order to shift negative-yielding reserves to other banks (even if the
aggregate level of reserves cannot be controlled by banks and is now mainly determined by the pace
of ECB asset purchases).
Third, the exchange rate channel, which matters a lot in small open economies like Denmark, Sweden
and Switzerland, could also be stronger when rates are negative. Conventional rate cuts influence
exchange rates through the interest rate parity: a negative expected interest rate differential with
partners leads to a depreciation of the currency, which in turn tends to boost exports and increase the
price of imports, therefore impacting positively output and inflation. This effect could be increased if
cross-border capital flows are more sensitive to negative rates (again because of investors’ aversion to
them and greater portfolio rebalancing effects). In fact, this is how Denmark justified the introduction
of NIRP, as a tool to defend its peg with the euro, at a time when inward capital flows (because Denmark
was perceived at a safe haven during the euro crisis) were leading to an appreciation of its currency
that could have hampered its exports and reduced inflation below a desired level.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
11 PE 662.922
Finally, if all these positive effects were to materialise and result (after also taking into account potential
negative effects discussed below) in an improved macroeconomic outlook, general equilibrium
effects may also have knock-on benefits for banks’ profitability. Higher economic activity could lead to
an increase in the demand for credit and in banks’ non-interest income, but it would also improve their
borrowers’ creditworthiness, thereby improving the credit quality of their assets and reducing non-
performing loans and loan-loss provisioning. Moreover, capital gains derived from the increase in the
value of the securities held by banks would increase their net worth and increase their distance to
default (as discussed by Chodorow-Reich, 2014). This improved health of the banking sector could in
turn enhance the ability of banks to finance the economy.
2.2. Possible negative side effects for the financial sector
Most criticism of negative rates focuses on the potential negative side effects for the financial sector,
which could at some point hinder the crucial role the sector plays in financing the economy.
First, it is true that a negative deposit facility rate represents a direct cost for banks that hold reserves
above the minimum reserves required by the central bank. The problem for banks is that with the
adoption of various asset purchase programmes, on aggregate, the banking sector is flooded with
excess liquidity and cannot avoid that cost.
However, although banks’ reserves at the ECB have increased very significantly in recent years, they
still represent a relatively limited share of the total assets held by banks in the euro area. What really
matters is thus not only the direct cost of these reserves held at the ECB, but how negative rates
influence other components of banks’ balance sheets that could reduce their overall profitability.
Indeed, negative rates could also lead to a potential decline in net interest income of banks if there
is a non-linear threshold effect at the ZLB (or near it, at the ELB). This would happen if the returns on
their assets are reduced (even more so because negative rates have led to a flattening of the curve, as
discussed above), while the policy rate cut does not fully translate into a fall in the interest rates paid
on their liabilities, because of a downward rigidity of deposit rates below 0%.
This could happen if banks are reluctant to pass on negative rates to deposits because they fear
households or corporations will start to hoard cash or move their accounts to another bank to avoid
negative rates. A reduction of net interest margins of banks could then lead either to reduced lending
volumes or to more expensive lending, as banks try to pass the cost of negative rates to their customers
to restore their intermediation margins. Both possibilities would reduce the transmission of the
monetary policy easing to the real economy. However, as discussed before, this could in part be
compensated for by the positive macro impact on lower rates that could lead to a reduction in non-
performing loans and to an increase in asset values held by banks, which would improve the banks’
financial health. Which effect dominates is not clear in theory and is ultimately an empirical question.
Negative rates could also impact non-bank financial institutions such as money market funds. Their
business model (i.e. issuing short-term, very liquid, almost cash-like, liabilities to invest in liquid safe
assets) could be particularly compromised by the combination of negative rates and a flattening of the
yield curve, given their already thin interest margins.
Finally, another important potential side effect is simply the corollary of the increase in the portfolio
rebalancing effect: an increase in risk-taking by banks and by other financial institutions such as
insurance companies, could if it becomes excessive lead to the emergence of bubbles and financial
instability episodes. This is a crucial element to take into account as this could make the use of
negative rate counterproductive in the long run, even if positive effects dominate in the short run.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 12
2.3. Possible reduced response of economic activity to interest rates
Conventional monetary policy, as it has been applied by all major central banks in the last decades,
relies on the conjecture that reducing nominal rates stimulates economic activity (see e.g., Woodford,
2003). This notion is also what supports the use of negative rates, if the fall in the policy rate passed to
market/bank rates then translates to higher demand for credit for consumption and investment, and
thus to increased output and inflation (as discussed in section 2.1). This is the idea behind the neutral
rate discussed above: as long as the central bank can reduce interest rates sufficiently (either through
standard rate cuts, or through negative rates, QE or forward guidance when the ZLB is reached) it
should be able to bring inflation back towards target and ensure that the economy is at full
employment. The question in that case is more how to go sufficiently low than whether it can be done.
However, after years of too-low inflation, the idea that such a neutral rate can be reached is now being
questioned. Maybe full employment and targeted inflation cannot be attained through the lowering
of rates only, without the support of other policies (fiscal and structural). This could be the case either
because the “reversal rate”
4
, at which negative effects start dominating, would be above the
presumptive neutral rate, or more simply because the sensitivity of output to interest rates would not
be as large as usually considered.
There could be multiple reasons behind this (Stansbury and Summers, 2020). The sensitivity of output
and inflation to interest rate cuts could be lower than it used to be because of sectoral changes in the
economy (towards less interest-sensitive sectors), or because fixed-capital investment is less sensitive
than before (e.g., because intangible capital depreciates more quickly and needs to be replaced often
whatever the interest rate is)
5
. It could also be diminished when rates are low or negative (e.g., because
some households target a minimum level of savings which is negatively affected by low rates
6
). The
responsiveness of output to rates could also be reduced during recessions (if firms are too pessimistic
about future demand). Or it could also decrease after a period of prolonged low rates (for instance
because demand for durable goods or housing is satiated or because agents are already overly
indebted). Finally, the response of the real economy to rates could be asymmetric: economic agents
might react differently to rate increases than to rate cuts (e.g., because credit constraints become
binding when central banks tighten their policies). For all these possible reasons, negative rates (and
monetary policy in general) could be an insufficient tool to achieve full employment and price stability.
4
An expression coined (and a concept formalised) by Brunnermeier and Koby (2018).
5
Or, as discussed by Geerolf (2019), it could be that fixed capital investment has never really been sensitive to interest rates and that firms
invest only when they need to produce more to meet increasing expected demand, as posited in the accelerator model of investment.
6
Guerrón-Quintana and Kuester (2019), exploring the implications of pension systems for the design of monetary policy, show that if the
public pension system is not generous enough, low for long interest rates can reduce aggregate activity.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
13 PE 662.922
EFFECTS OF CENTRAL BANK NEGATIVE RATES IN PRACTICE
In practice, these various effects, positive and negative, coexist. Which one dominates depends on
multiple parameters, including in particular the magnitude of negative rates. The ‘reversal rate’
represents the interest rate at which negative effects start dominating and at which further rate cuts
become counterproductive and contractionary
7
.
However, this will also depend on structural features of the economy: the prominence of banks in the
financial sector (vs markets), the financial structure of banks, the aversion of households and the
behaviour of non-financial corporations with respect to negative rates, the costs of holding cash
(storage, transportation, insurance, but also the convenience of electronic payments provided by bank
deposits), etc. This means that the net effect of negative rates is probably time- and country-specific. In
this section, we look at the evidence to see what the effects of negative rates have been in practice in
the euro area since their introduction in 2014.
3.1. Transmission to market (interest and exchange) rates
How have negative policy rates been transmitted to the main benchmark market rates? Given the
abundance of liquidity in the euro area’s banking sector due to the introduction of a very large amount
of reserves by the ECB through its various refinancing operations (LTROs and TLTROs) and asset
purchase programmes (in particular since 2015), the deposit facility rate has de facto become the main
ECB policy rate influencing market rates. As far as the short-term rate is concerned, we observe a full
pass-through from the ECB’s deposit facility rate to the operational target rate of the ECB, the
EONIA (replaced recently by the €STR). As Figure 2 Panel A shows clearly, breaching the ZLB did not
affect the transmission from the policy rate to the overnight rate (on the contrary, it even appears that
the volatility of the EONIA has been reduced).
Concerning the rest of the yield curve, a fall in long-term yields of euro area countries is visible to the
naked eye in Figure 2 Panel B after the introduction of negative rates, suggesting that this has led
investors to revise their expectations about the possible future path of policy rates. This observation is
confirmed formally by the literature. Christensen (2019), who studied the reaction of markets rate in all
countries that have adopted negative rates, found that the entire cross section of government bond
yields exhibits an immediate, significant, and persistent response to the introduction of negative
rates, with a maximum impact on yields with a maturity of 5 years. Rostagno et al. (2021) confirmed
that the negative ECB rate propagated across the whole sovereign yield curve. According to them, the
response of long-term rates to the NIRP even surpasses the response to conventional rate cuts by a
wide margin.
7
The concept of “reversal rate” is related but different from the concept of ELB, as the latter represents the interest rate at which agents
are indifferent between holding deposits and cash and which should be slightly negative to account for the costs/risks of holding cash.
The level of the reversal rate could thus be above or below the ELB.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 14
Figure 2: Policy and market interest rate in the euro area
Panel A: ECB rates and EONIA (%) Panel B: Government bond yields: 10-year (%)
Note: black vertical line indicates the introduction of NIRP in the euro area in June 2014.
Source: Bruegel based on ECB and FRED (https://fred.stlouisfed.org
).
On the effects of negative rates on the exchange rate, although their introduction was followed by a
strong depreciation of the euro clearly visible in Figure 3, the evidence from the literature is mixed.
While Hameed and Rose (2017) did not find a significant impact of NIRP on the evolution of the
exchange rate, it appears that in Denmark and Switzerland the appreciation of their respective
currencies was halted by the introduction of negative rates, which led to an adjustment in cross-border
bank flows (see Khayat, 2018, on Denmark; Basten and Mariathasan, 2018, on Switzerland). More
generally, Ferrari et al. (2017) also showed that the sensitivity of exchange rates to changes in policy
rates is stronger when rates are lower.
Figure 3: Euro nominal effective exchange rate
Notes: Index 2010 = 100. Nominal effective exchange rates are calculated as geometric weighted averages of bilateral
exchange rates. Black vertical line indicates the introduction of NIRP in the euro area in June 2014.
Source: Bruegel based on BIS and retrieved from FRED (https://fred.stlouisfed.org
).
-1
0
1
2
3
4
5
6
7
1999
2001
2003
2005
2007
2009
2011
2013
2015
2017
2019
2021
Deposit Facility Rate
Marginal Lending
Rate
Main Refinancing
Rate
EONIA
-2
-1
0
1
2
3
4
5
6
7
8
2013
2013
2014
2015
2015
2016
2017
2017
2018
2019
2019
2020
2021
Spain Germany
France Portugal
Euro area Italy
90
95
100
105
110
115
2004
2004
2005
2006
2007
2008
2009
2009
2010
2011
2012
2013
2014
2014
2015
2016
2017
2018
2019
2019
2020
What Are the Effects of the ECB’s Negative Interest Rate Policy?
15 PE 662.922
3.2. Effects of negative rates on banks
3.2.1. Direct cost of a negative deposit facility rate for euro-area banks
First, as its name suggests, a negative deposit facility rate implies a direct cost charged on excess
liquidity deposited by commercial banks at the ECB. In the first decade of the euro, banks did not hold
excess liquidity at the ECB and held only enough reserves to fulfil the reserves officially required by the
central bank (Figure 4 panel A).
However, with the financial crisis and the COVID-19 crisis, the ECB injected a very large quantity of
reserves into the banking sector. Total reserves of banks at the ECB went up from around EUR 10 billion
at the beginning of 1999 (compared to total assets of euro area banks of EUR 14 trillion, i.e. 0.7%) to
above EUR 4 trillion in April 2021 (compared to total assets of EUR 36 trillion i.e. 11.3%).
At first, the ECB injected reserves mainly through refinancing operations (MROs and then LTROs). This
meant that the banks could control the quantity of reserves they wanted to hold because they were
given the choice to participate in these operations. Thus, that before the financial crisis, no bank sought
to hold reserves above what was required (see Figure 4 panel A). During the financial crisis (2008-12),
some banks decided to hold extra reserves to insure themselves against possible liquidity shortages,
but gradually reduced these holdings after the euro crisis ended. Nevertheless, with the launch of the
public sector purchase programme (PSPP) in 2015, banks lost control of the aggregate level of reserves.
Since then, this level is fully determined by the ECB because central bank reserves created for the
purchases inevitably end up as deposits in a bank. With the addition of the pandemic emergency
purchase programme (PEPP) in 2020, excess liquidity is now quickly approaching EUR 4 trillion (Figure
4, panel A).
As a result, the direct cost of holding excess liquidity for banks has also risen significantly. After
levelling off at about EUR 7.5 billion annually between 2017 and 2019, the ECB decided in September
2019 to put in place a “two-tier system” by exempting part of the excess reserves from the negative
deposit rate (equal to six times required reserves). This led to a significant fall in the direct cost incurred
by banks, from around EUR 8.5 billion to EUR 4.6 billion. However, with the launch of the PEPP in March
2020, the cost for banks (even with tiering) has now increased to around EUR 15 billion per year (Figure
4 panel B).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 16
Figure 4: Excess liquidity held at the ECB and direct costs for the banks of the euro area
Panel A: Excess liquidity (in EUR bn) Panel B: Direct cost of negative DFR for banking sector (in EUR bn)
Notes: Excess liquidity is defined as deposits at the deposit facility net of the recourse to the marginal lending facility plus
current account holdings in excess of those contributing to the minimum reserve requirements. Direct cost is measured as
the excess liquidity (as defined in Panel A) x (- deposit rate). Direct cost net of tiering is measured as direct cost - tiered reserves
x (- deposit rate), with tiered reserves = excess liquidity exempted from negative rates, i.e., 6 times required reserves of the
whole banking sector since June 2019 and 0 before tiering was introduced.
Source: Bruegel based on ECB.
Moreover, this cost is unevenly distributed across the euro area (Figure 5 Panel A) and is mainly
concentrated in Germany, France, the Netherlands, Luxembourg and Finland (up to 95% of the cost
was incurred in these countries in July 2019, but it is down to 72%). As discussed in Darvas and Pichler
(2018), the main reason why excess liquidity is concentrated in these five countries is that for a large
share of asset purchases made by the ECB counterparties have headquarters outside of the euro area
and their euro liquidity is parked in bank accounts in a few euro area financial centres.
In some countries these costs represent a non-negligible share of the profits generated by European
banks (Figure 5 Panel B). However, again, the share of profits that these costs represent varies greatly
across the euro area. In that regard, the comparison between France and Germany is enlightening. Even
though the banks from these two countries both represent a high share of excess reserves and thus
bear a high cost, the situation is very different when these costs are compared to the profits generated
by their banking sectors. While the cost associated with the negative deposit facility rate of the ECB
represented 28.2% of the profits of German banks, this only represented only 4.8% of the profits of
French banks in 2019. This suggests that a negative deposit facility rate is not the main driver of the
low profitability of banks in some countries, but that there are some structural (efficiency) issues
affecting the banks’ profitability in these countries.
0
500
1.000
1.500
2.000
2.500
3.000
3.500
4.000
2021-Apr
2019-Apr
2017-May
2015-Apr
2013-Nov
2012-Jul
2011-Mar
2009-Nov
2008-Jul
2007-Mar
2005-Nov
2004-Jul
2003-Feb
2001-Oct
2000-Jun
1999-Feb
Deposit facility
Excess reserves on current
account facility
Excess liquidity
0
5
10
15
20
2021-Apr
2020-Nov
2020-May
2019-Oct
2019-Apr
2018-Oct
2018-May
2017-Oct
2017-May
2016-Oct
2016-Apr
2015-Oct
2015-Apr
2014-Nov
2014-Jul
2014-Mar
direct cost of negative
deposit rate on excess
liquidity
direct cost net of tiering
What Are the Effects of the ECB’s Negative Interest Rate Policy?
17 PE 662.922
Figure 5: Direct cost of a negative deposit facility rate per country
Panel A: Annualised direct cost per country (in EUR millions)
Panel B: Direct cost per year and per country compared to profits (in EUR millions and % of profits)
Notes: Direct costs are calculated in the same way as in Figure 4. Profits are derived indirectly through the return on equity of
banks.
Source: Bruegel based on ECB.
The only way for banks to circumvent the cost related to a negative DFR is to convert ECB reserves into
cash. Actually, cash held by monetary and financial institutions in the euro area has increased since
the ECB started imposing a negative rate on excess reserves. Nonetheless, amounts are still marginal
compared to excess reserves, and the movement is still mainly circumscribed to German banks (ECB,
0
2000
4000
6000
8000
10000
12000
14000
16000
Austria Belgium Cyprus Germany Estonia
Spain Finland France Greece Ireland
Italy Lithuania Luxembourg Latvia Malta
Netherlands Portugal Slovenia Slovakia
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 18
2018). This probably means that the policy rate is still above or around the ELB
8
. This remains a fairly
limited phenomenon at the moment, but if this shift towards cash becomes more widespread, it could
reduce the transmission from policy rates to the overnight market rate.
3.2.2. Impact on other components of the balance sheets of commercial banks
However, even if the share of excess reserves on the banks’ balance sheet has increased considerably
in the last 10 years, it is still relatively limited (at around 11%). The costs associated with excess reserves
could thus be manageable if the maturity transformation business model of banks is not affected by
negative rates. Indeed, a compression of the spread between credit and deposit rates and a
general flattening of the curve, which reduce the returns from maturity transformation, would be
much more damaging for the banks’ overall profitability than the mere negative DFR imposed on
excess liquidity.
How has the negative policy rate been transmitted to the euro-area banks’ rates? In order to answer
this question and determine if the net interest margins of banks have been compressed, we look
separately at the rates on loans and on deposits for both households and corporations. Figure 6
displays these four different bank rates. It is difficult to see precisely with the naked eye how policy
rates influence these bank rates and if the effects change below the ZLB/ELB, but our simple
econometric analysis (see Box 1 for details) confirms that in general the pass-through is incomplete for
all interest rates. In other words, a 1-point decrease in the policy rate (proxied by the EONIA) does not
fully translate into a 1-point decrease in bank rates. But we find that the pass-through is significantly
higher for rates on banks’ assets (loans) than on their liabilities (deposits). It is also higher for
corporations than for households (see detailed results in Table 2).
Overall, our results imply that rate cuts indeed compress banks’ interest margins. However, we also
check if the relationships change when the EONIA becomes negative (and when the EONIA falls below
other more negative thresholds down to -0.5%) and we fail to find any evidence of a non-linearity of
this effect below 0%. In other words, monetary policy easing appears to have a negative impact on
banks’ net interest margins (at least directly, because indirectly an improvement in the economy due
to the easing could counterbalance this effect in the medium to long run
9
), but rate cuts in negative
territory do not seem to amplify this negative impact on banks’ margins in a significant way.
8
This would be compatible with the estimation of the ELB at -0.7% by Rostagno et al. (2016) based on the cost of holding cash at 0.4% plus
some extra inconvenience of transacting in cash.
9
One important channel leading to the improvement at the macro level is that the banks’ loss of interest margins represents a gain for
households and corporations which can borrow more cheaply, while the rate on their deposits do not fall as much.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
19 PE 662.922
Figure 6: Composite bank interest rates for households and corporations (%)
Notes: Rates are weighted averages of rates applied by maturity/type of loan. For instance, "Household Deposit Rate" is an
average of the annualised agreed rates with maturities below and above 2 years weighted by outstanding amounts remaining
at the end of the periods. See details in box 1. Black vertical line indicates the introduction of NIRP in the euro area in June
2014.
Source: Bruegel based on ECB.
The fact that rates on households’ deposits are still above 0% is generally interpreted as a non-linearity:
banks are not passing negative rates to households’ deposits because they are afraid of cash
withdrawal (because cash cannot be submitted to negative rates). But it is also possible that this rate is
still above 0% because of its usual stickiness. Rate cuts, which are generally passed more quickly to
corporate deposits, have actually led some banks to impose negative rates on their deposits despite a
similar risk of cash hoarding for non-financial corporations (NFC) than for households: the average euro
area rate on NFC’s overnight deposits is still positive, but the averages in Germany and in particularly
the Netherlands are negative (Schnabel, 2020).
Concerning cash hoarding, as seen earlier, the conversion of reserves into cash by banks is still
marginal. This is also true for deposits of households and corporations. Cash hoarding is not happening
yet, probably because the negative rate applied to deposits is still lower than the cost of storing cash
and because of the convenience of bank deposits and electronic payments (especially with the recent
increase in online shopping).
Nevertheless, our results also mean that the effect of rate cuts (whether below or above the ZLB) on
banks will depend on their financial structures, and in particular on the composition of their assets
and liabilities (which differs across banks and across countries): if a bank is heavily reliant on
households’ deposits and if lending rates to households and corporations decrease more steeply than
the rates applied to deposits, then its net interest income will be particularly reduced.
The literature generally shows that, on average in the euro area, banks’ overall profitability has not been
particularly affected by the introduction of negative rates (Jobst and Hin, 2016; Stráský and Hwang,
2019) in part because increased asset values and stronger economic activity have offset the negative
effects. But the literature also suggests that banks with higher shares of households’ deposits have
-2
0
2
4
6
8
10
2003
2003
2004
2005
2005
2006
2007
2007
2008
2009
2009
2010
2011
2011
2012
2013
2013
2014
2015
2015
2016
2017
2017
2018
2019
2019
2020
2021
EONIA
Corporate Loans Interest Rate
Household Loans Interest Rate
Household Deposits Interest Rate
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 20
either seen their profitability more affected by the fall in rates (Heider et al., 2019) or that they have not
passed the rate cuts to their loans to try to compensate (Amzallag et al., 2019).
Box 1: Estimating the pass-through of monetary policy to bank rates
In order to assess the impact of changes of central bank policy rates to bank rates, we first construct
composite interest rates for households and corporations on loans and deposits. Using ECB data, we
weight various interest rates with different maturities and destination of use (e.g., loans for house
purchase with a maturity of more than 10 years) by the importance of its amount of relative to its
category (other loans made to households).
This provides us with four composite time series of interest rates for the euro area. Next, we explore
the relationship between the monetary policy rate (proxied by the EONIA) and these different bank
rates. First, we estimate the long-run relationship between the two variables with the following
regression:
= +
+
where
is a given bank rate (e.g., loans for households),
is the monetary policy rate (proxied here
by the EONIA). The table below summarizes the results for all four types of interest rates.
Table 1: Relationship between monetary policy rate and banking rates
Household
Deposits
Corporate
Deposits
Household
Loans
Corporate Loans
0.46***
0.73***
0.92***
0.92***
*** statistically significant at the 1% threshold
Source: Bruegel.
The results indicate that the pass-through is incomplete for all interest rates. In other words, a 1-
point decrease in the policy rate is not translated into a 1-point decrease of banking rates. In addition,
these results indicate that the pass-through is significantly higher for rates applied on banks’ assets
(loans) than on their liabilities (deposits). Similarly, the pass-through is higher for corporations than
for households.
Next, we look at the short-term effect of changes in the policy rate on bank rates in order to assess
whether this relationship has changed since rates are negative. We estimate the following
regression:

=
+



+


+


+

+
(

× 

) +
where 
=

for =
{
,
}
and (


) is a vector with the error terms of equation 1
(in other words the difference between the actual value of the given bank rate in t-1 and the
predicted bank rate estimated above).

is a dummy variable equal to 1 when the policy rate is
below 0, making
(

× 

)
the interaction variable between changes in the policy rate when
the rate is negative. In other words,
captures the effect of a change in t-1 of the monetary policy
What Are the Effects of the ECB’s Negative Interest Rate Policy?
21 PE 662.922
rate on changes in the bank rate and
the additional effect when the rate is negative. If
is not
null and statistically significant, this would mean that the pass-through changes when rates are
negative.
Table 2: Relationship between and banking rates in case of negative rates
Household
Deposits
Corporate
Deposits
Household
Loans
Corporate Loans
0.46***
0.73***
0.92***
0.92***
Short-term effect
(
)
0.07*** 0.18*** 0.34*** 0.52***
Additional effect
when rates are
negative (
)
-0.01 -0.27 -0.6 -0.65
***, ***, *: statistically significant at the 1%, 5% and 10% threshold respectively.
Source: Bruegel.
While the coefficients of the short-term changes in the policy rate confirm that the pass-through is
stronger for banks’ assets (loans) than for their liabilities (deposits) as well as for corporate rates, the
effect of the interaction variable is not statistically significant. This suggests that there is no change
in the relationship between the policy rate and bank rates when the policy rate is negative. In other
words, nothing indicates the presence of non-linearity of the effect below the threshold of zero. We
test other thresholds by steps of 10 bps below 0% (-0.1%, -0.2%, etc.) and find similar results.
What has been the effect on banks’ lending volumes? Whether for households or NFCs, lending has
grown steadily from the introduction of negative rates to the COVID-19 crisis (Figure 7)
10
. However, it
is extremely difficult to disentangle the effect of NIRP from other potential drivers. Part of this strong
growth can surely be explained by the recovery that took place in the euro area from 2014 to 2020
(which increased demand for loans), but at least at the aggregate level, we observe neither reduced
lending nor an increase in rates to compensate for lost profits (Figure 6). This suggests, at first sight,
that in terms of bank lending volume, the economy has reacted to negative rates similarly than to rate
cuts in positive territory. The results from the literature on that question are mixed, but some studies
confirm that banks have indeed increased lending volumes after the introduction of negative rates, in
particular banks that rely less on deposits (e.g. Heider et al., 2019).
10
The data after March 2020 is clearly influenced by economic policies (e.g. government guarantees on bank loans) and lockdowns put in
place to respond to the COVID-19 crisis which have respectively led to high borrowing from firms and low borrowing from households
for consumption.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 22
Figure 7: Bank lending to the private sector in the euro area (YoY, %)
Note: black vertical line indicates the introduction of NIRP in the euro area in June 2014.
Source: Bruegel based on ECB.
To conclude on the effects of negative rates on banks, there are nevertheless two important caveats.
First, the evidence discussed and the literature (in the euro area but also in other jurisdictions that have
adopted NIRP) has been gathered with negative rates relatively close to 0%. This means that a strong
non-linearity could still arise at lower rates. Second, the impact on banks could also vary over time.
Effects of prolonged negative rate could be different: at the beginning, the effect is positive as the one-
off mark-to-market revaluation dominates, but, if negative rates are prolonged, maturing assets are
gradually replaced by low-yielding loans, which could lead to a persistent decline in interest margins.
In addition, the reduction of net interest income compensated for at the beginning by higher volumes
and increased fees might not last (or banks will have to adjust their business model in order to rely
more on fees than on interest margins).
3.3. Macro and financial stability impact
This is undoubtedly the most difficult and the most crucial question to answer. There is now a fast-
expanding literature on the effects of negative rates, but it has not yet led to a clear consensus on the
macro effect of negative rates or on its long-term effects on financial stability.
It is still very difficult to measure precisely the macro effects of NIRP for various reasons. First, various
other unconventional measures were implemented at the same time and are difficult to disentangle
from negative rates. Second, NIRP was only adopted in a few countries. And, finally, there could also be
a possible selection bias, as central banks that that have concluded less favourable cost-benefit
analyses might have decided not to use NIRP (this could be the case with the Fed and the BoE).
On the macro effects, ECB papers are generally (and not surprisingly) positive about the effects of
NIRP. They generally find a positive (although not very high) impact of negative rates on output,
employment and inflation compared to a counterfactual without negative rates (see Figure 8 borrowed
from Rostagno et al., 2021), and that the reversal rate is not yet a valid concern in the euro area at this
stage.
-6
-4
-2
0
2
4
6
8
10
12
14
16
Loans to non-financial corporations
Loans to households (consumption)
Loans to households (house purchase)
What Are the Effects of the ECB’s Negative Interest Rate Policy?
23 PE 662.922
Figure 8: Estimated impact of three ECB rate cuts (by 10 bps) on macro variables
Notes: Their analysis is based on a Bayesian vector autoregression (BVAR) model with 17 variables and a dense, controlled
event-study approach to identification. The shaded area reflects the dispersion of the median responses obtained over the
simulation samples. See Rostagno et al. (2021) for details on the methodology used.
Source: Figure 12.1 from Rostagno et al. (2021).
However, other papers are more doubtful about the positive macro effects of negative rates, especially
in the long run. Stansbury and Summers (2020) in particular documented extensively all the possible
reasons why the sensitivity of output, employment and inflation to interest rates might have
declined in recent decades, and why this could run counter to the argument for negative policy rates.
The sum of evidence collected by the authors is definitely thought-provoking, but it does not provide,
at this stage, a clear picture and a quantification of the overall macro effect. Moreover, the evidence
collected is mainly about the US and could be very different in Europe given the massive differences in
the functioning of the two economies, e.g. in the financial sector, in the housing market, in the social
safety net and pension systems, and in households’ behaviours.
Nevertheless, one particular reason why negative rate could be counterproductive is if low and
negative rate were to have long-term consequences for financial stability and lead to the formation of
bubbles and ultimately to financial stress episodes. There is strong evidence that financial institutions
in the euro area have increased their risk taking and have engaged in search-for-yield as a result of
negative rates (Bubeck et al., 2020; Heider et al., 2019). However, as discussed earlier, increasing risk-
taking and pushing financial institutions to rebalance their portfolios is essentially a feature not a bug
of accommodative monetary policy, but it is very difficult to say if/when this additional risk-taking
becomes excessive.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 24
CONCLUSIONS: WHAT DO WE KNOW AND WHAT SHOULD THE
ECB DO?
Despite the burgeoning literature, the debate on the effects of NIRP is not yet fully settled.
Nevertheless, we have already learned a few things about how negative rates work. First, the
experience of the euro area since 2014 shows that the negative deposit facility rate is fully transmitted
to the benchmark overnight rate and then propagates along the whole yield curve. Second, there is
also some evidence (in particular from Denmark and Switzerland) that NIRP also impacts the exchange
rate through a change in cross-border flows. Third, as far as bank rates are concerned, despite some
concerns about a stronger negative impact of NIRP on net interest margins and banks’ profitability, it
appears that the effects of rate cuts in negative territory are not different to “standard” rate cuts. Like
them, they reduce banks’ interest margins because rates on banks’ assets are more sensitive to policy
rates than rates on banks’ liabilities, but this effect does not appear to be amplified below 0% (at least
at the current relatively low level of negative rates). Finally, a negative deposit facility rate implies some
cost for banks that hold excess reserves at the ECB. This cost has been growing significantly, in
particular since the introduction of the PEPP in 2020. Moreover, this cost is concentrated in the
countries that host the main European financial centres (where investors which have sold assets to the
ECB park their euro deposits). In some of these countries, Germany in particular, where the banking
sector is not as profitable as in other jurisdictions, this direct cost can represent a significant share of
banks’ profits (28% in 2019 in Germany). However, other examples, such as France, show that if the
banking sector is sufficiently profitable, the cost as a share of profits remain small (4.8% in 2019).
A more difficult and more fundamental question about negative rates, and for monetary policy in
general at this stage, is whether output, employment and inflation are still as sensitive to these
financial variables as they used to be when rates are very low or even negative. The ECB’s own
research is confident that the effect of NIRP on these variables in the short- to medium-term is positive
at this stage. Potential negative side effects do not appear to have materialised in a significant way for
the moment, which might mean that the ECB rate is still above the reversal rate. However, a long period
of negative rates could also entail some medium- to long-term risk, in particular in terms of financial
stability. Financial institutions seem to have increased their risk-taking with the advent of negative
rates. Whether this risk-taking is excessive remains to be seen and will need to be monitored carefully.
What should the ECB do at this stage? The ECB, like other central banks which have adopted NIRP, will
need to continue to monitor carefully the potential side effects, see if its policy rate approaches the
reversal rate, or even if this reversal rate is evolving (and potentially increasing) as negative rates are in
place for a prolonged period, in order to update its cost-benefit analysis.
The direct cost incurred by banks that hold excess reserves appears to be one of the most tangible side
effects at this stage. So, the first thing the ECB could do is to mitigate this visible side effect (which is
problematic in a few countries and which makes this policy very unpopular despite its potential
benefits). The easiest solution would be to adjust the two-tier system put in place in 2019 and to
increase the quantity of excess reserves that are exempted from the negative deposit facility rate
(currently equal to six times required reserves). Potentially, the ECB faces a trade-off between
mitigating the direct cost for banks and reducing the influence of the deposit facility rate on short-term
market rates if too much reserves are exempted. However, for the moment, despite the current tiering,
the overnight rate continues to trade near the DFR. The ECB could thus experiment and try to increase
the level of exempted reserves in a gradual way (for instance by raising the tiering multiplier from 6
to 8) to see if this trade-off materialises and if the influence of the DFR on the overnight rate diminishes.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
25 PE 662.922
But apart from this, what could the ECB do to avoid or overcome the reversal rate problem (if it were
to materialise)? The ECB has actually introduced an innovative tool in recent years: the TLTROs. This
new type of refinancing operation has attracted less attention than other unconventional tools, such
as QE or its negative DRF, but we believe it could actually help overcome the reversal rate issue. With
the launch of the TLTRO III in March 2019
11
, the ECB decided for the first time that the rate could be
negative and as low as the deposit facility rate (if some conditions were fulfilled in terms of lending
volume). In practice, this means that the banks resorting to TLTROs are paid to borrow from the ECB,
which partly compensates for the negative DFR on excess reserves.
At the start of the COVID-19 crisis, on 12 March 2020, the ECB went further and for the first time reduced
its TLTRO rate by 25 bps below the DFR, for banks fulfilling their lending benchmarks to the real
economy. On 30 April 2020, the ECB decided to further ease the conditions on its TLTROs by cutting
the applicable rate by a further 25 bps to as low as -1% (i.e. 50 bps below the DFR).
As discussed in Claeys (2020), making the level of the TLTRO rate independent from the DFR provides
a new way for the ECB to ease financing conditions. This allows the central bank to adopt a more
accommodative stance without having to cut its DFR further, thus avoiding its potential side effects.
Given that the rate is lower than what banks pay on their excess liquidity, this provides banks with a
strong incentive to borrow long-term from the ECB and to grant more loans. This, in turn, should
mechanically increase their reserve requirements, since these are calculated as a ratio of a bank’s
liabilities – mainly its customers’ deposits. Considering the new tiering system on reserve
remuneration, their exempted reserves would also be increased, even more than proportionally. This
ultimately should create a virtuous cycle for bank profitability and incentivise banks to lend to the
economy, despite negative policy rates (or more precisely thanks to a negative spread between the
DFR and the TLTRO rate). The main caveat is that the ECB will actually lose money on these operations.
However, this should not be a major source of concern, given that, as discussed in Chiacchio et al.
(2018), while it is preferable for the ECB to make profits rather than to record losses, it is not a profit-
maximising institution and its overriding mandate is price stability. As such, recording losses in the
short-to-medium term
12
when seeking to fulfil its macroeconomic function, should not stop the ECB
from using such a policy if it is effective.
If the ECB believes it is approaching the reversal rate but it needs to provide more monetary easing, it
should therefore refrain from cutting its DFR again, and instead reduce further its TLTRO rate below
its DFR.
Next, how to deal with potential financial stability issues? As discussed before, negative rates for a
too-long period, which could be necessary to fulfil the ECB’s price stability mandate, could nonetheless
lead to financial instability in the medium- to long-run. In our view, the best way to deal with financial
imbalances is to use macroprudential tools as the first line of defence. However, the euro area’s
macroprudential institutional framework might not be able of playing this crucial role, given its
decentralised nature and heterogenous functioning across European countries. It is therefore critical
to build a better set-up for the use of macroprudential tools in Europe to ensure that they can be used
forcefully and in a timely way when needed.
11
TLTROs I, which were the first refinancing operations conditional on new lending by the banks, were announced in June 2014. TLTROs II
were launched in March 2016.
12
Assuming that the negative spread of 50 basis points between the deposit rate and the TLTRO rate would in the end apply to a volume
of between €500 billion and €1 trillion in loans, this would lead to losses on these operations of between €2.5 and 5 billion, compared
with an average of €14 billion of distributable profits per year for the Eurosystem from 1999 to 2017 (Chiacchio et al, 2018).
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 26
Finally, at the current juncture, given the situation after one and a half years of the COVID-19 pandemic
and in 2020 the largest decline in GDP in the euro area since the Second World War, the ECB should be
extremely cautious and not rush before exiting unconventional monetary policies including
negative rates. Even if negative rates proved to be ineffective, or insufficient to bring the economy to
full employment, given the full pass-through between the deposit rate and market rates
13
, it would be
extremely dangerous to exit negative rates too soon as it could harm the post-COVID-19 recovery as
well as destabilise European sovereign debt markets. This would be highly damaging given the
increased role played by fiscal policy in terms of macroeconomic stabilisation.
13
This could be particularly relevant in particular if, as discussed above, the effects of interest rate changes are asymmetric and if the effects
of tightening are much stronger than the effects of rate cuts.
What Are the Effects of the ECB’s Negative Interest Rate Policy?
27 PE 662.922
REFERENCES
• Amzallag, A., Calza, A., Georgarakos, D. and Soucasaux Meneses e Sousa, J.M. (2019). “Monetary
Policy Transmission to Mortgages in a Negative Interest Rate Environment”, ECB Working Paper
2243. https://www.ecb.europa.eu/pub/pdf/scpwps/ecb.wp2243~a1f298ab9d.en.pdf
• Basten, C., and Mariathasan, M. (2018). “How Banks Respond to Negative Interest Rates: Evidence
from the Swiss Exemption Threshold”. CESifo Working Paper 6901, Munich.
• Bubeck, J., Maddaloni, A. and Peydró, J.-L. (2020). “Negative monetary policy rates and systemic
banks’ risk-taking: evidence from the euro area securities register”. ECB Working Paper Series No
2398 / April 2020
• Brunnermeier, M., and Koby, Y. (2018). “The Reversal Interest Rate”. NBER Working Paper No. 25406.
December 2018. https://www.nber.org/papers/w25406
• Chodorow-Reich, G. (2014). “Effects of unconventional monetary policy on financial institutions”,
Brookings Papers on Economic Activity (Spring), 155–204.
https://www.brookings.edu/wp-
content/uploads/2016/07/2014a_ChodorowReich.pdf
• Chiacchio, F., Claeys, G. and Papadia, F. (2018). “Should we care about central bank profits?”, Policy
Contribution 2018/13, Bruegel.
https://www.bruegel.org/2018/08/should-we-care-about-central-
bank-profits/
• Claeys, G. (2020). “The European Central Bank in the COVID-19 crisis: Whatever it takes, within its
mandate”, Study for the Committee on Economic and Monetary Affairs, Policy Department for
Economic, Scientific and Quality of Life Policies, European Parliament, Luxembourg, 2020.
https://www.europarl.europa.eu/RegData/etudes/IDAN/2020/648811/IPOL_IDA(2020)648811_EN
.pdf
• Christensen, J.H.E. (2019). “Yield Curve Responses to Introducing Negative Policy Rates”, FRBSF
Economic Letter 2019–27, Federal Reserve Bank of San Francisco.
• Darvas Z. and Pichler, D. (2018). “Excess Liquidity and Bank Lending Risks in the Euro Area”, Study
for the Committee on Economic and Monetary Affairs, Policy Department for Economic, Scientific
and Quality of Life Policies, European Parliament, Luxembourg, 2018.
https://www.europarl.europa.eu/RegData/etudes/STUD/2018/626069/IPOL_STU(2018)626069_E
N.pdf
• ECB. (2018). “The cash holdings of monetary financial institutions in the euro area”, European
Central Bank, Economic Bulletin Issue 6/2018.
https://www.ecb.europa.eu/pub/economic-
bulletin/articles/2018/html/ecb.ebart201806_03.en.html
• Ferrari, M., Kearns, J. and Schrimpf, A. (2017). “Monetary policy’s rising FX impact in the era of ultra-
low rates”, BIS Working Paper, No 626. https://www.bis.org/publ/work626.pdf
• Geerolf, F. (2019). “A Theory of Demand Side Secular Stagnation”. Working Paper.
https://fgeerolf.com/hansen.pdf
• Guerrón-Quintana, P., and Kuester, K. (2019). “The dark side of low(er) interest rates”. Working
Paper. https://www.frbsf.org/economic-research/files/4-Guerron-Quintana.pdf
• Hameed, A., and Rose, A. K. (2018). “Exchange Rate Behaviour With Negative Interest Rates: Some
Early Negative Observations”. Pacific Economic Review 23 (1): 27-42.
IPOL | Policy Department for Economic, Scientific and Quality of Life Policies
PE 662.922 28
• Heider, F., Saidi, F. and Schepens, G. (2019). “Life below Zero: Bank Lending under Negative Policy
Rates”, The Review of Financial Studies 32 (10): 3728–61.
• Holston, K., Laubach, T. and Williams, J. (2016). “Measuring the Natural Rate of Interest: International
Trends and Determinants”. Federal Reserve Bank of San Francisco, 2016-11, 43.
https://www.frbsf.org/economic-research/publications/working-papers/2016/11/
• Jobst, A., and Lin, H. (2016). “Negative Interest Rate Policy (NIRP): Implications for Monetary
Transmission and Bank Profitability in the Euro Area”. IMF Working Papers, 16(172), 48.
https://www.imf.org/external/pubs/ft/wp/2016/wp16172.pdf
• Khayat, G. A. (2018). “The Impact of Setting Negative Policy Rates on Banking Flows and Exchange
Rates”. Economic Modelling 68: 1–10.
• Rostagno, M., Bindseil, U. Kamps, A., Lemke, W., Sugo, T. and Vlassopoulos, T. (2016). “Breaking
through the Zero Line: The ECB’s Negative Interest Rate Policy”, Presentation at Brookings
Institution, June 6.
https://www.brookings.edu/wp-
content/uploads/2016/05/20160606_Brookings_final_background-Compatibility-Mode.pdf
• Rostagno, M., Altavilla, C., Carboni, G. , Lemke, W., Motto, R. and Saint Guilhem, A. (2021).
“Combining negative rates, forward guidance and asset purchases: identification and impacts of
the ECB’s unconventional policies”, ECB Working Paper Series No 2564 / June 2021.
https://www.ecb.europa.eu/pub/pdf/scpwps/ecb.wp2564~e02f3aad4c.en.pdf
• Schnabel, I. (2020). “Going negative: the ECB’s experience”, Speech at the Roundtable on Monetary
Policy, Low Interest Rates and Risk Taking at the 35th Congress of the European Economic
Association, Frankfurt am Main, 26 August 2020.
https://www.ecb.europa.eu/press/key/date/2020/html/ecb.sp200826~77ce66626c.en.html
• Stansbury, A. and Summers, L. (2020). “The End of the Golden Age of Central Banking? Secular
stagnation is about more than the zero lower bound”, unpublished manuscript
• Stráský, J., and Hwang, H. (2019). “Negative interest rates in the euro area: Does it hurt banks?”,
OECD Economics Department Working Papers, 1574, 35.
https://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=ECO/WKP(2019)44&
docLanguage=En#:~:text=This%20conflation%20is%20confirmed%20by,of%20the%20ECB%20a
sset%20purchase
• Taylor, J.B. (1993). “Discretion versus policy rules in practice”, in Carnegie-Rochester conference
series on public policy, Vol. 39: 195-214, North-Holland.
https://web.stanford.edu/~johntayl/Onlinepaperscombinedbyyear/1993/Discretion_versus_Polic
y_Rules_in_Practice.pdf
• Woodford, M. (2003). Interest and Prices, Princeton University Press, Princeton, NJ.
PE 662.922
IP/A/ECON/2021-24
PDF ISBN 978-92-846-8184-6 | doi:10.2861/041324 | QA-09-21-214-EN-N
Several central banks, including the European Central Bank since 2014, have added negative policy
rates to their toolboxes after exhausting conventional easing measures. It is essential to understand
the effects on the economy of prolonged negative rates. This paper explores the potential effects
(and side effects) of negative rates in theory and examines the evidence to determine what these
effects have been in practice in the euro area.
This paper was provided by the Policy Department for Economic, Scientific and Quality of Life
Policies at the request of the committee on Economic and Monetary Affairs (ECON) ahead of the
Monetary Dialogue with the ECB President on 21 June 2021.