Monetary Policy Pass-Through to
Interest Rates: Stylized Facts
from 30 European Countries
Robert Beyer, Ruo Chen, Claire Li, Florian Misch, Ezgi O. Ozturk, and
Lev Ratnovski
WP/24/9
IMF Working Papers describe research in
progress by the author(s) and are published to
elicit comments and to encourage debate.
The views expressed in IMF Working Papers are
those of the author(s) and do not necessarily
represent the views of the IMF, its Executive Board,
or IMF management.
2024
JAN
* Corresponding author.
** The authors would like to thank Oya Celasun, Luis Brandao Marques, Vincenzo Guzzo, Sebastian Weber, Pelin Berkmen, Mahir
Binici, Ozge Emeksiz, Ezequiel Cabezon, Salvatore Dell'Erba, Fazurin Jamaludin, Rui Mano, Vina Nguyen, Yu Shi and ECB
staff for useful comments. Kayla Qin, Agnesa Zalezakova and Wei Zhao provided useful assistance.
© 2024 International Monetary Fund WP/24/9
IMF Working Paper
European Department
Monetary Policy Pass-Through to Interest Rates: Stylized Facts from 30 European Countries
Prepared by Robert Beyer, Ruo Chen, Claire Li, Florian Misch*, Ezgi O. Ozturk, and Lev Ratnovski**
Authorized for distribution by Helge Berger
January 2024
IMF Working Papers describe research in progress by the author(s) and are published to elicit
comments and to encourage debate. The views expressed in IMF Working Papers are those of the
author(s) and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.
ABSTRACT: The extent to which changes in monetary policy rates lead to changes in loan and deposit rates
for households and firms, referred to as ‘pass-through’, is an important ingredient of monetary policy
transmission to output and prices. Using data on seven different bank interest rates in 30 European countries,
different approaches, and the full sample as well as a subsample of euro area countries, we show that a) the
pass-through in the post-pandemic hiking cycle has been heterogenous across countries and types of interest
rates; b) the pass-through has generally been weaker and slower, except for rates of non-financial corporation
loans and time deposits in euro area countries; c) differences in pass-through over time and across countries
for most deposit rates are correlated with financial sector concentration, liquidity, and loan opportunities, and d)
the effects of pass-through to outstanding mortgage rates on monetary transmission on prices and output are
heterogenous across countries.
RECOMMENDED CITATION: Beyer, R.,Chen, R., Li, C., Misch, F., Ozturk, E., and Ratnovski, L. (2024).
Monetary Policy Pass-Through to Interest Rates: Stylized Facts from 30 European Countries. IMF Working
Paper WP/24/9
JEL Classification Numbers:
E43; E52
Keywords:
Monetary Policy Transmission; Monetary Policy Pass-Through
Author’s E-Mail Address:
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
3
WORKING PAPERS
Monetary Policy Pass-Through to
Interest Rates: Stylized Facts from
30 European Countries
Prepared by Robert Beyer, Ruo Chen, Claire Li, Florian Misch
1
, Ezgi O.
Ozturk, and Lev Ratnovski
2
1
Corresponding author.
2
The authors would like to thank Oya Celasun, Luis Brandao Marques, Vincenzo Guzzo, Sebastian Weber, Pelin Berkmen, Mahir
Binici, Ozge Emeksiz, Ezequiel Cabezon, Salvatore Dell'Erba, Fazurin Jamaludin, Rui Mano, Vina Nguyen, Yu Shi and ECB
staff for useful comments. Kayla Qin, Agnesa Zalezakova and Wei Zhao provided useful assistance.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
4
Contents
1. Introduction ..................................................................................................................................................... 5
2. The Role of Policy Pass-Through for Monetary Policy Transmission ....................................................... 6
3. Data .................................................................................................................................................................. 8
4. Stylized Facts .................................................................................................................................................. 8
4.1 In the post-pandemic hiking cycle, pass-through has been heterogeneous across sectors, loan and
deposit types, and countries ........................................................................................................................... 8
4.2 The pass-through has been weaker this time, except to NFC loan rates ............................................... 10
4.3 The pass-through has been slower this time, except to NFC loan rates ................................................ 13
4.4 Lower banking competition, ample household and NFC deposits, and high banking liquidity are
sometimes associated with lower deposit rate pass-through ....................................................................... 17
4.5 Pass-through to rates of existing mortgages weakened over time as the share of fixed rate mortgages
increased, with heterogeneous effects across countries .............................................................................. 21
5. Conclusions ................................................................................................................................................... 22
References ......................................................................................................................................................... 23
Annex I. Data Description ................................................................................................................................. 26
Annex II. Additional Charts and Results ......................................................................................................... 30
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
5
1. Introduction
The post-pandemic surge in inflation has triggered the most aggressive monetary policy tightening in decades.
This has, in turn, generated renewed interest in the aggregate effects of monetary policy. This paper analyzes
the pass-through of policy rates in European economies to lending and deposit rates of credit institutions,
referred to as bank interest rates, thereby providing insights into an important (but not the only) aspect of
monetary policy transmission.
The impact of monetary policy-induced changes in loan and deposit interest rates for households and non-
financial corporations (NFCs) is one of many channels through which monetary policy affects financial
conditions, and thereby output and prices. Bank interest rates can affect agents’ investment, consumption and
saving decisions, debt service costs and savings income, as well as asset values, all of which affect real
economic outcomes. The extent of pass-through of monetary policy rate changes to bank interest rates
therefore affects monetary policy transmission and effectiveness. In turn, pass-through to bank interest rates
can be influenced by financial sector concentration, liquidity, and the availability of deposits relative to profitable
lending opportunities. These characteristics can affect incentives for banks to compete for deposits and
potentially undermine pass-through in hiking cycles. Methodologically, analyzing pass-through has advantages:
the effects of monetary policy on bank loan and deposit rates are arguably more direct, often with shorter lags
and fewer confounding factors compared to monetary policy transmission on output and prices.
Focusing on monetary policy pass-through in European countries during the post-pandemic hiking cycle, this
paper makes two key contributions to the previous literature. First, we compile data on seven different interest
rates for 30 European countries both within euro area and non-euro area countries with independent monetary
policy (defined as having no official peg and not being fully euroized). This dataset contains interest rates for
key types of new loans and deposits of both households and NFCs. Second, we present a range of early
evidence on stylized facts on monetary pass-through during the post-pandemic hiking cycle using a mix of
descriptive statistics and regression analyses, both for the full sample of countries and a subset of euro area
countries (EA-12 countries). In particular, we document that: a) the pass-through in the post-pandemic hiking
cycle has been heterogenous across countries and types of interest rates; b) the pass-through has been
weaker and slower relative to past hiking cycles, except for rates of new loans to NFCs and NFC time deposits
(and controlling for the peculiarities of the current cycle including possible anticipation effects of low initial rates
does not alter these finding qualitatively); c) the weakening of pass-through to deposit rates is in part
associated with higher financial sector concentration, more ample deposits relative to loan opportunities, and
liquidity; and d) using data on outstanding mortgages as an example, the effects of pass-through on aggregate
monetary policy transmission are heterogenous across countries. Taking account of the peculiarities of the
post-pandemic hiking cycle including possible anticipation effects and very low initial rates does not change
these stylized facts.
Our paper contributes to the empirical literature studying monetary policy pass-through to bank interest rates,
including in Europe. One strand of this literature focuses on the euro area and mostly covers the 2000s. For
example, Sørensen and Werner (2006), Bernhofer and van Treeck (2013), Sander and Kleimeier (2006)
document how pass-through differs across countries, interest rate types, and in the short vs. long run (see for
example de Bondt, 2005, for a survey), and Saborowski and Weber (2013) examine the determinants of pass-
through to the aggregate bank lending rate in a large panel of countries. Other papers focus on European
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
6
countries outside of the euro zone.
3
A meta-data study on interest rate pass-through, based on more than a
thousand estimates from 50 studies, finds that the average pass-through from monetary policy rates to bank
lending rates is around 0.8 (Gregor and Melecký, 2021).
Our paper is also related to another strand of the literature which exploits heterogeneity in the pass-through
across banks within countries. For instance, Wang et al. (2022) use data on U.S. banks and show that bank
market power affects pass-through, and that the effects depend on the level of the policy rate. Kho (2023) uses
country- and bank-level data from 13 euro area countries from before the post-pandemic hiking cycle and
shows that the degree of banking sector concentration has asymmetric effects on pass-through to overnight
(O/N) deposit rates during tightening and loosening cycles. López-Quiles (2021) estimates pass-through to
money market interest rates using transaction-level data of the largest banks in the euro area.
Importantly, our paper overcomes a critical limitation of the existing literature by covering the post-pandemic
hiking cycle and uses data on individual countries or covering only a subset of euro area countries. Hence, our
paper is related to recent studies analyzing the pass-through during the post-pandemic hiking cycle, which all
report findings that are consistent with ours. For example, for the euro area, Byrne and Foster (2023) find that
pass-through is generally weaker now relative to previous cycles, Messer and Niepmann (2023) find that pass-
through to deposit rates has been more sluggish than in the past, Lane (2023a and 2023d) reports that the
pass-through to NFC loans has been strong and that loan rates started to rise before the ECB hiked its policy
rates in July 2022, and Lane (2023c) finds weaker pass-through to O/N deposit rates than to time deposit rates
in the post-pandemic hiking cycle. For other European economies, Cabezon and Kovachevska (2024) finds
that pass-through in North Macedonia is weaker compared to regional peers and the euro area, and Binici
(2024) attributes relatively weak pass-through in Malta in the post-pandemic tightening cycle to abundant
liquidity in the banking system.
The paper is organized as follows: Section 2 discusses conceptually the role of policy pass-through in overall
monetary policy transmission. Section 3 describes the data. Section 4 presents stylized facts on monetary
policy pass-through. Section 5 concludes.
2. The Role of Policy Pass-Through for Monetary
Policy Transmission
Monetary policy usually uses the short-term money market interest rate as its operational target. The money
market rate is transmitted via financial markets to deposit and lending rates of credit institutions (‘bank interest
rates’), and through them to the broader economy. The ability of the central bank to achieve its ultimate
objectives—price, financial, and overall macroeconomic stability—hinges in part on the transmission of the
policy rate to other interest rates. In turn, bank interest rates affect the macroeconomy through several related
channels:
1. Interest rate channel: Changes in interest rates alter firm investment behavior and household
consumption and saving choices. Higher real interest rates increase the hurdle (break-even) rates for
investment, thereby reducing it, and they make saving more attractive, which in turn reduces
3
For example, Gregor and Melecký (2018) focus on pass-through to lending rates in Czech Republic, Egert et al. (2007) study the
interest rate pass-through in five central and eastern European countries, Becker et al. (2012) analyze the pass-through to
mortgage rates in the UK, and Hansen and Welz (2011) analyze the pass-through to lending and deposit rates in Sweden.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
7
consumption. This is the standard neo-classical channel of monetary policy (Goodfriend and King,
1997).
2. Cash flow channel: Changes in interest rates affect agents’ income and costs relating to existing
assets and liabilities that carry variable interest rates. In other words, tighter monetary policy can
directly increase interest income and debt service costs, but the cash flow effects depend on whether
households and firms are net borrowers or savers and their exposure to variable rate debt and
deposits. With different propensities to consume or invest across households and firms, this channel
can affect aggregate macroeconomic outcomes (see, e.g., Auclert, 2019).
4
Low pass-through to
deposit rates may matter in this context given that many households in Europe are net savers.
3. Balance sheet channel: Higher interest rates reduce the value of assets used as collateral, thereby
tightening agents’ financial constraints. For firms, this amplifies the neoclassical impact of interest
rates on investment, while for households this can alter the consumption of durable goods (Kiyotaki
and Moore, 1997). Additionally, wealth effects (lower asset valuations due to higher interest rates)
reduce the households’ propensity to consume (Lettau and Ludvigson, 2004).
4. Banking channel: Higher interest rates tend to be associated with higher bank net interest rate
margins, thereby increasing bank profits (Borio et al., 2017). Interest rates may also affect bank
funding conditions, as higher deposit rates may increase deposit supply. Higher bank profits and
ample deposits may increase bank lending capacity (Kashyap and Stein, 2000; Drechsler et al.,
2017).
5
Table 1: Effects of monetary policy pass-through to bank interest rates onto the real economy
Channels that directionally weaken the effects of monetary policy tightening shown in orange
Channels of MP
transmission
Mechanism of channel
during MP tightening
Effects of higher pass-through during MP tightening
Effects of higher pass-
through to loan rates
Effects of higher pass-
through to deposit rates
Interest rate channel
Higher hurdle rate for new
investment, and higher
savings remuneration
Less investment
More saving →
Less consumption
Cash flow channel
Higher interest income and
debt services cost for
existing exposures
Lower cash flow →
Less consumption and
investment
Higher cash flow →
more consumption and
investment
Balance sheet channel
Lower value of collateral
Tighter non-price credit
conditions → Less investment
Banking channel
Lower banks’ net worth
Tighter bank funding
conditions → Less lending
Higher cost of bank
funding
More supply of deposits →
More lending
On balance, a higher pass-through of monetary policy to interest rates tends to strengthen monetary policy
transmission through each of these channels (Table 1). However, a higher pass-through to deposit rates can
4
The non-financial sector in a given economy may predominantly consist of either losers or winners, so that the aggregate effects
do not net out even if the marginal propensities to invest and consume are identical. In that case, the aggregate consumption
and investment effects could be offset by changes in the volume of loans through changes in credit conditions.
5
See also Lane (2023a) for a discussion of this channel. Obviously, there are other monetary policy transmission channels which do
not directly depend on interest rate pass-through; see, for example, Cevik et al. (2023), Lane (2023a), and Mishkin (1996) for
more details on the interest rate channels discussed here and a comprehensive list of monetary transmission channels more
broadly.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
8
create ambiguous effects in some cases. For instance, higher pass-through in a hiking cycle reduces private
investment and consumption through a higher hurdle rate for new investment and savings remuneration, but
this effect is weakened by higher cash flows from the existing stock of savings.
3. Data
We compile a dataset on loan and deposit interest rates of credit institutions in 30 European economies using
data from central banks and statistical agencies, referred to as bank interest rates. Our data includes seven
different types of interest rates: interest rates on overnight deposits and time deposits, both for households and
NFCs, consumer loans, mortgages, and loans to NFCs. The interest rates generally refer to new deposits and
loans, but we also collect data on interest rates of outstanding mortgages.
6
Our data is at monthly frequency and constitutes an unbalanced panel, spanning 30 European countries that
are either part of the euro area or have independent monetary policy (in the sense that they do not have hard
pegs or are completely euroized). Two thirds of countries have data on all seven types of interest rates (Annex
Table A2). Some series reach back to 2003, but for some countries the time coverage is considerably shorter.
The average number of years per country is 17.8. We complement this data with information on the monetary
policy rate, prices (HICP and core HICP inflation) and economic activity (industrial production). To examine the
role of potential determinants of pass-through, we also collect longer time series on banking sector
concentration (Herfindahl–Hirschman index using bank assets) and the loan-to-deposit ratio as a proxy of
financial sector liquidity.
Annex Tables A2, A3 and A4 contain details about the country coverage of bank interest rates and the
complete list of variables with broad definitions and sources. There are some minor differences in the exact
definition of interest rates. For instance, some interest rates are collected using the universe of credit
institutions, whereas others refer to all or only a subset of banks. In addition, in some countries, the household
sector encompasses non-profit organization serving households, but in others it does not. While these
differences imply that the interest rate series are not perfectly comparable across countries, they are too small
to materially affect our results. Reassuringly, the result for the euro area, which should be perfectly comparable
across countries, are qualitatively similar to the results based on the overall sample.
4. Stylized Facts
4.1 In the post-pandemic hiking cycle, pass-through has been heterogeneous
across sectors, loan and deposit types, and countries
We document the pass-through to each of the seven bank interest rates (time and O/N deposit rates for
both households and NFCs, and rates for mortgages, consumer loans and NFC loans) in the post-
pandemic monetary policy tightening cycle. We compute bank interest rate betas, defined as the ratio of
the cumulative increase in bank interest rates to the cumulative increase in the policy rate. For the
purpose of this exercise, we only use data from the post-pandemic hiking cycle and truncate the data for
6
In some time series, there is the distinction between pure new business and new business combined with renegotiations. We
generally use pure banks rates for new business, but for NFC loan rates in Czech Republic, Cyprus, Estonia, Finland and
Romania countries, we use information from both series to compile data series with fewer gaps.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
9
each country in the month of the first rate hike until three months after the last hike (or the first rate cut or
the end of our sample, whatever is shorter); see Annex Table A1 for the start and end dates of the post-
pandemic hiking cycle of each country. The results are as follows.
First, there is heterogeneity across different types of loans and deposits (Figure 1). We find that the pass-
through is on average highest for loans to NFCs, followed by time deposits, and weakest for overnight
(O/N) deposits. The response of mortgage rates to changes in the policy rate lies in between them. In
part, this heterogeneity in pass-through could be linked to differences in the maturity of the underlying
loans and deposits, whereby a longer maturity can weaken the link between policy rates and bank rates.
For instance, compared to NFC loans, which often have shorter maturities, it is plausible that mortgage
rates respond less to policy rate changes, as term premia and other factors affect longer-maturity loans
(the maturity of mortgages differs significantly across countries – see Section 4.5). These findings are
broadly consistent with the existing literature. The relatively weak pass-through to deposit rates,
especially overnight deposit rates, is consistent with the notion that high switching costs give banks
market power in deposit markets but runs counter the hypothesis that longer duration weakens the link
between policy and bank rates (see Polo, 2021, for a discussion). Messer and Niepmann (2023) also
document that pass-through to O/N deposit rates has been weaker than to time deposit rates in the post-
pandemic cycle.
Second, there is significant heterogeneity across countries, both in the dispersion of the pass-through
across different rates and the level of the pass-through (Figure 2). Some of these cross-country
differences can reflect differences in financial sector concentration, liquidity, and profitable lending
opportunities relative to savings as we discuss below. In addition, public policies that cap lending rates or
impose floors on deposit rates could also play a role. For instance, Hungary introduced temporary
mortgage interest rate caps, with the stated rationale of helping households with rising debt service
payments (Valderrama, 2023).
7
7
The IMF's integrated macroprudential policy (iMaPP) database contains comparable cross-country data on the use of such tools.
However, at the time of writing, the most recent data refers to 2021 and does not contain measures that were introduced since
the beginning of the post-pandemic hiking cycle.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
10
Figure 1: Bank Interest Rate Betas
Figure 2: Bank Interest Rate Betas by Country
Sources: ECB; Haver analytics; country authorities; and IMF staff calculations.
Note: Ratio of cumulative change in interest rates relative to cumulative changes in the policy rate from month of first rate
increase to three months after last rate increase (or the first rate cut or the end of the sample, whichever is shorter) in the post-
pandemic tightening cycle. Figure 1 represents the unweighted average of up to 30 countries. Countries with a shaded area for
the range of interest rate betas in Figure 2 indicate that at least one bank rate is not available. Thus, Figure 2 does not imply a
cross-country ranking of the combined pass-through to all bank rates, not all of which are available for all countries. For
example, for Switzerland, the NFC and household time deposit rates, the aggregate mortgage rate and the aggregate NFC loan
rate are not available, but evidence suggests that the pass-through to flexible rate mortgages has been strong (see Annex Table
A2 for details on data). For Norway, the end month has been shifted back one month in order to have more than one rate
available.
Third, there are differences across customer types (or sectors): the pass-through to NFC deposit rates
generally exceeds the pass-through to household deposit rates (Figures 3 and 4). This finding holds for
nearly all countries, which could suggest greater market power of credit institutions vis-à-vis households
relative to NFCs and reflect lower stickiness of NFC deposits.
Figure 3: Overnight Deposit Rate Betas
Figure 4: Time Deposit Rate Betas
Note: Ratio of cumulative change in O/N and time deposit interest rates relative to cumulative changes in the policy rate from
month of first rate increase to three months after last rate increase (or the first rate cut or the end of the sample, whichever is
shorter) in the post-pandemic tightening cycle.
4.2 The pass-through has been weaker this time, except to NFC loan rates
We now compare the magnitude of the pass-through during the post-pandemic tightening cycle to
previous tightening cycles. To this end, for each country, we first identify the previous tightening cycle
0 0.2 0.4 0.6 0.8 1
HH O/N
deposits
NFC O/N
deposits
Consumer
loans
Mortgages
HH time
deposits
NFC time
deposits
NFC loans
HU
CZ
NO*
SE*
RS*
SK
RO
PL*
IT
AL*
UK
AT
FR
DE
BE
LT
PT
ES
EE
LV*
IE*
FI
SI
LU
NL*
GR*
MT*
CY
MK*
CH*
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
9.5
10.5
11.5
12.5
13.5
14.5
15.5
16.5
17.5
18.5
19.5
20.5
21.5
22.5
23.5
24.5
25.5
26.5
27.5
28.5
29.5
-0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4
AT
BE
CH
CY
DE
ES
FI
FR
GR
IE
IT
LU
MT
NL
PT
SE
UK
AL
CZ
EE
HU
LT
LV
PL
RO
SI
SK
0
0.1
0.2
0.3
0.4
0.5
0.6
-0.1 0 0.1 0.2 0.3 0.4 0.5 0.6
NFC O/N deposits
HH O/N deposits
Non-CESEE CESEE
AT
BE
CY
DE
ES
FI
FR
GR
IT
LU
MT
NL
PT
UK
CZ
EE
HU
LT
LV
PL
RO
SI
SK
0
0.2
0.4
0.6
0.8
1
1.2
1.4
0 0.2 0.4 0.6 0.8 1 1.2
NFC time deposits
HH time deposits
Non-CESEE CESEE
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
11
with the largest policy rate increase covered by our data, referred to as ‘comparator cycle’. In the case of
the euro area, Poland, Romania, Sweden, and the UK, the comparator cycle covers the 2005-2008 pre-
GFC monetary policy tightening. For some countries, there is either no data available for any previous
hiking cycles (e.g., Albania and Switzerland), or the comparator cycle is different (e.g., Czech Republic,
Hungary, and Serbia); see Annex Table A1 and A4 for details. The post-pandemic tightening cycle has
been generally more rapid and larger in magnitude than the comparator cycles. The average cumulative
policy rate increases in the post-pandemic cycle amounted to 595bps, compared to 231bps in the
comparator cycles. In addition, the pace of the increase averaged around 35bps per month, compared to
15bps per month during the comparator cycles.
This analysis suggests that across all countries, pass-through as measured by interest rate betas was
generally weaker during the post-pandemic tightening cycle than in the comparator cycles. Put differently,
more and faster tightening was accompanied by weaker pass-through per percentage point increase in
the policy rate. The interest rate betas during the post-pandemic cycle expressed as percent of the
corresponding betas in the comparator cycle are shown in Figure 5. These results remain broadly
unchanged if we only consider EA-12 economies (referred to as ‘euro area’ in the charts, which have the
same cycles and data coverage). These findings are broadly robust to assuming that anticipation effects
imply that pass-through at least for some bank rates begin up to six months before the actual policy rate
hikes (Annex Figure A1). They also apply to most individual countries (Figure 6). For instance, the pass-
through to household O/N rates and mortgage rates has been generally smaller in the post-pandemic
tightening cycle.
Figure 5: Pass-Through in Post-Pandemic
Relative to the Comparator Cycle
Figure 6: Pass-Through to O/N Deposit and
Mortgage Rates
Note: Figure 5 shows the unweighted average of the ratio of interest rate betas in the post-pandemic period relative to the interest
rate betas in the comparator cycle. The interest rate betas are defined as the cumulative change in interest rates relative to
cumulative changes in the policy rate from the month of the first rate increase to three months after last rate increase (or the first
loosening, or the end of sample, whichever is shorter). Figure 6 shows interest rate betas of the post-pandemic and the
comparator cycles. Both Figures only contain countries that have comparator cycles as per Annex Table A1.
Next, to broaden the analysis beyond the single largest previous cycle, we compare the pass-through
during the post-pandemic tightening cycle to all other previous tightening cycles using regression
analysis. We follow the exchange rate pass-through literature (Burstein and Gopinath, 2014), which
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Loans O/N
deposits
Time
deposits
Mortgages O/N
deposits
Time
deposits
NFC HH
Europe Euro Area
AT
CY
CZ
DE
ES
FI
FR
GR
HU
IE
IT
LU
MT
NL
PL
PT
RO
SE
SI
AT
BE
CZ
DE
ES
FI
FR
GR
IE
IT
NL
PT
RS
UK
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
-0.2 0 0.2 0.4 0.6 0.8 1 1.2 1.4
Post-pandemic tightening cycle
Comparator tightening cycle
Household O/N deposits
Mortgages
Pass-through
smaller in the
past-pandemic
Pass-through
larger in the past-
pandemic
tightening cycle
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
12
Messer and Niepmann (2023) recently applied to deposit rates, and estimate the following equation in a
panel setting to assess how changes in policy rates are associated with changes in bank rates:
,
(1)
where
denotes an interest rate of country i in month t,
is a country fixed effects,
denotes
changes in the policy rate, with the superscript T (L) referring to rate hikes (cuts);
is a dummy
variable that is one for the post-pandemic tightening cycle. This allows for different pass-through for policy
rate hikes and cuts, implying that
is the pass-through for loosening and
is the pass-through for
tightening (both for lag k), and for a different pass-through in the post-pandemic tightening cycle (
.
In contrast to the interest rate betas used so far, this regression controls for confounding factors. We
include the contemporaneous and K lags of the 12-month log change in industrial production (IP) and
core consumer price inflation (CPI) to control for changes in deposit and credit demand. We estimate the
regression separately for NFC loan rates, mortgage rates, as well as time and O/N deposit rates of NFCs
and households. Standard errors are clustered at the country level. The estimated pass-through based on
Eq. (1) is broadly robust to including the lagged levels and lagged changes of the bank interest rates, as
well as year fixed effects; see Annex Figure A2.
8
In Figure 7, we report the long-run pass-through of the different rates separately for the post-pandemic
cycle and all earlier tightening periods. The long-run pass-through is defined as the cumulative effect of a
policy rate change over K consecutive periods and given by
(and by
for the post-
pandemic tightening cycle). We set K=6 but our results are similar with longer lags.
The results confirm that the pass-through is significantly weaker in the post-pandemic cycle than in
previous tightening cycles, both in the full sample and among EA-12 countries, except for NFC loans in
the full sample and NFC time deposits in both samples. The declines are large: On average across both
samples, the differences in the estimated pass-through between previous tightening cycles and the post-
pandemic cycle are 0.2 for HH and NFC O/N deposits and 0.4 for HH time deposits and mortgages, and
all statistically significant at the ten percent level.
9
A difference in pass-through of 0.2 implies that of a
given policy rate hike, 20 percent less is passed on to bank interest rates. For NFC loans, the differences
are not significant at conventional levels in the full sample.
8
Including the lagged change of the interest rate somewhat increases the long-run pass-through to lending rates.
9
The relatively large decline for pass-through to mortgage rates is partly caused by the very low pass-through to mortgage rates in
some countries in which mortgage rates increased considerably before the ECB started its post-pandemic tightening cycle.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
13
Figure 7. Estimated Policy Pass-Through: Post-pandemic Tightening vs. Previous Tightening
Cycles
Note: Pass-through estimated based on Eq. (1). The differences between solid bars are statistically significant at least at the 10
percent level.
4.3 The pass-through has been slower this time, except to NFC loan rates
Another aspect of the pass-through is how fast changes in policy rates translate into changes in lending
and deposit rates. To analyze the speed, we first estimate impulse response functions for policy rate
hikes (month-over-month increase in monetary policy rates) using the local projection model proposed by
Jordá (2005). As in the previous subsection, we compare the post-pandemic and all previous tightening
cycles and again separately include one variable for positive shocks (i.e., policy rate hikes) to account for
tightening episodes, and a separate variable for negative shocks (i.e., policy rate cuts) to account for
loosening episodes. We estimate the following equation:
(2)
in which i, t, and h refer to the country, month, and the horizon of the projection, respectively,
refers to
policy rate hikes and
refers to policy rate cuts. As in the previous section, the regressions assess how
changes in policy rates are associated with changes in bank rates and include controls for the year-over-
year change in industrial production,
, year-over-year change in core inflation,
, and country
fixed-effects
In addition, we include controls for policy changes that are in between the time of the
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
14
change and the time of the cumulative response,
and the first lag of the month-over-month
change in the monetary policy rate,
.
10
Standard errors are clustered by country.
Figure 8 presents impulse response functions (IRFs) of interest rates to monetary policy hikes in the post-
pandemic tightening cycle and in previous tightening episodes using panel local projections. The IRFs in
Figure 8 are broadly consistent with the previous section in the sense that the pass-through in the post-
pandemic cycle has been weaker than in previous tightening cycles for most interest rates. By contrast
(but in line with the earlier results), the pass-through to NFC loan and NFC time deposit rates is similar.
These IRFs imply that the pace of pass-through to rates of has been slower in the post-pandemic cycle
compared to the earlier tightening episodes, with some caveats and exceptions. Eyeballing the pace of
pass-through from the slopes of the IRFs suggests that for NFC loans and time deposits, the pass-
through was slow but over time caught up to some extent to the pass-through in previous tightening
cycles. With respect to mortgage and O/N deposit rates, the pass-through was slower and weaker, with
no catch-up over time.
We then use a simple quantitative measure of the pace of pass-through, namely the peak response
divided by the time it takes to reach the peak. The peak response is defined as the largest significant
response within the six months after the hike (where significance is defined as within one standard
deviation). Figure 9—partially based on the IRFs in Figure 8—and Figure 10—based on country-by-
country impulse response functions—confirm these results in a more formal way for the whole sample
and the euro area. Annex Figure A3 includes a country-by-country comparison between the pace of pass-
through in the post-pandemic hiking cycle and previous tightening episodes.
10
Methodologically, both equations 1 and 2 yield the same pass-through under some conditions.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
15
Figure 8. Cumulative Responses of Interest Rates to Monetary Policy Rate Hikes
Notes: The solid lines show the responses to the changes in the monetary policy rates derived from panel local projection. The
dotted lines show the 68 percent confidence interval (+/- 1 standard deviation).
0.0
0.5
1.0
1.5
0 1 2 3 4 5 6
Responses of NFC loan rates
(percentage points)
Post-pandemic tightening episode
0.0
0.5
1.0
1.5
2.0
2.5
0 1 2 3 4 5 6
Responses of mortgage rates
(percentage points)
Previous tightening episodes
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
0 1 2 3 4 5 6
Responses of NFC time deposit rates
(percentage points)
0.0
0.5
1.0
1.5
0 1 2 3 4 5 6
Responses of HH time deposit rates
(percentage points)
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0 1 2 3 4 5 6
Responses of NFC ON deposit rates
(percentage points)
0.0
0.1
0.2
0.3
0.4
0 1 2 3 4 5 6
Responses of HH ON deposit rates
(percentage points)
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
16
Figure 9: Pace of Pass-through to Interest Rates: Panel Local Projection
Note: This figure Is based on the panel local projection responses shown in Figure 8.
Figure 10. Average Pace of Country-Specific Pass-Through to Interest Rates
Note: This graph takes the simple average of the country-specific results gathered through impulse response functions. The pace
of the transmission is measured as the ratio of the magnitude of the highest response to the time of to get that response. ***, **, *
indicate that the difference between post-pandemic tightening episode and earlier tightening episode is statistically significant at
99 percent, 95 percent, and 90 percent confidence intervals, respectively.
0
0.1
0.2
0.3
0.4
Europe EA-12 Europe EA-12 Europe EA-12 Europe EA-12 Europe EA-12 Europe EA-12
Loans O/N deposits Time deposits Mortgages O/N deposits Time deposits
NFC HH
Post-pandemic tightening episode Previous tightening episodes
0
0.1
0.2
0.3
0.4
0.5
Europe EA-12 Europe EA-12 Europe EA-12 Europe EA-12 Europe EA-12 Europe EA-12
Loans O/N deposits Time deposits Mortgages O/N deposits Time deposits
NFC HH
Post-pandemic tightening episode Previous tightening episodes
*
**
**
*** ***
***
***
*
**
***
**
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
17
4.4 Lower banking competition, ample household and NFC deposits, and high
banking liquidity are sometimes associated with lower deposit rate pass-through
This subsection empirically examines two potential determinants of monetary policy pass-through
econometrically: banking sector competition (captured via financial sector concentration) and the volume
of household and NFC deposits relative to loans (captured by the loan-to-deposit ratio) in our full sample.
In a separate cross-section analysis, we explore the role of liquidity more broadly.
Conceptually, it seems plausible that more concentrated banking sectors could lead to lower monetary
policy pass-through to deposit rates during tightening cycles. When banks compete less, deposit rates
may increase less than the policy rates. At the same time, more concentrated banking sectors could be
associated with higher pass-through to loan rates, again a result of lower competition. The Herfindahl
index—a measure of concentration—of total assets of credit institutions suggests that bank concentration
in Europe has been increasing modestly over time (Figure 11). Increasing financial sector concentration
may have hence contributed to the lower pass-through to deposit rates and higher pass-through to
lending rates during the post-pandemic tightening cycle, but we do not find any supporting evidence in
this regard.
In addition, the volume of household and NFC deposits in combination with profitable lending
opportunities (proxied by the loan-to-deposit ratio, LTD) determine incentives for banks to compete for
deposits and hence pass-through to deposit rates. However, there is no clear theoretical link between
loan-to-deposit ratios and the pass-through of loan rates. The LTD has fallen over time (Figure 11),
consistent with evidence on post-pandemic excess savings (McGregor et al., 2022), and the sluggish
recovery which undermine lending opportunities and loan volumes (see Lane, 2023b, for evidence of
declining loan flows).
Figure 11. Financial Market Characteristics: Post-Pandemic Relative to Comparator Cycle
(ratio)
Note: The bars on the left shows the ratio of the average Herfindahl index in the post-pandemic cycle to that in the comparator
cycle for the full sample and euro area countries.
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
EUR EA EUR EA
HerfindahI index Loan-to-deposit ratio
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
18
To estimate the impact of financial sector concentration and liquidity empirically, we use Eq. (1) from
Section 4.2, but with the interaction term now indicating whether a given country has a high financial
sector concentration or loan-to-deposit ratio. Countries are classified every year into two groups based on
year-specific medians. We estimate the same equation as before:
,
(3)
where variables are defined as above but the dummy
is now one when the country characteristic
(concentration or loan-to-deposit ratio) is above the median. To compare the pass-through across groups,
we compare
to
.
The results suggest that high concentration (and hence low competition) lowers pass-through to some
deposit rates. This holds both in the full sample and among EA-12 economies, but not all differences
across all rates and both samples are statistically significant at the 10 percent level (Figure 12).
Differences in the pass-through to NFC deposit rates, for example, are not significant at conventional
levels in the EA-12 sample. The results for pass-through to loans are either non-conclusive or far from
statistically significant, indicating that competition is a less important driver of lending rates.
A higher loan-to-deposit ratio is mostly associated with stronger pass-through to some deposit rates.
While the differences are less pronounced, three of them are statistically significant at the 10 percent
level, with a higher statistical significance in the euro area (Figure 13); differences in the pass-through to
time deposit rates are not significant in either sample. The finding that loan-to-deposit ratios seem to play
a greater role in explaining pass-through in the euro area would be consistent with the generally more
sluggish recovery and fewer profitable loan opportunities in the euro area.
11
Figure 12. Financial Sector Concentration and
Pass-Through
Figure 13. Loan-to-Deposit Ratio and Pass-
Through
Note: Pass-through estimated based on Eq. (3). The differences between solid bars are statistically significant at least at the 10
percent level.
11
In Annex Figure A4, we show that there is some correlation between financial sector concentration as well as LTD levels and
cross-country differences in pass-through in the post-pandemic tightening cycle.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
19
Liquidity more generally can also lead to weak pass-through. More liquid banks require less stable
funding (of which deposits are part) to support the same volume of lending, implying that higher bank
liquidity could induce banks to compete less for deposits, hence undermining pass-through to deposit
rates. By contrast, the effects of liquidity on pass-through to loan rates are ambiguous. More liquid banks
may compete in credit markets more, as they have a higher capacity to lend, or may compete less when
bank liquidity increases in times when banks are less willing to lend (or when lending demand is weak).
High bank liquidity – as seen during the post-pandemic tightening cycle – could therefore also explain
weak pass-through to deposit rates, not least because the post-pandemic tightening cycle takes place in
times of high bank liquidity. Bank liquidity is driven by several factors, including regulatory changes (such
as the introduction of liquidity coverage ratio (LCR) and the net stable funding ratio (NSFR)), bank risk-
management improvements in the post-GFC period, and past QE which has reduced private holdings of
government debt and raised bank deposits (which has contributed to excess reserves in the banking
system, defined as bank deposits at central banks minus minimum required reserves).
In Figure 14 we explore simple correlations between bank liquidity and monetary policy pass-through
based on cross-country data for the euro area. The analysis confirms a statistically significant negative
association between bank liquidity—as captured by the LCR—and NFC O/N deposit betas, consistent
with the theoretical predictions. However, the relationship with household deposit betas is negative but
not significant, and the relationship with bank excess reserves is statistically insignificant at conventional
levels (even if corrected for outliers).
12
The analysis shows no correlation between bank liquidity and loan
rate betas (neither for NFC nor for household loans), consistent with the ambiguous theoretical
predictions.
Another argument why pass-through has been weaker in the post-pandemic tightening relates to the fact
that in some countries, the post-pandemic tightening cycle started from zero or negative rates. As Abadi
et al. (2023) argue, pass-through could be lower around the zero-lower bound. In a robustness check, we
again estimate differences between the post-pandemic tightening cycle and previous cycles, in analogy to
Eq. (1), but we include two dummies for tightening in the regression model - one capturing increases from
positive policy rates and one capturing increases from zero or negative rates. While the differences in
pass-through become smaller, the effects are almost negligible quantitatively. We therefore conclude that
the negative or zero starting values weakened pass-through, but that the effects were relatively small in
the current setting (see Annex Figure A5).
12
While excess reserves do not affect monetary policy pass-through to interest rates based on country-level data, recent analysis
based on bank- and loan-level data by Fricke et al. (2023) suggests that the transmission of monetary policy tightening differs
across banks depending on their level of excess reserves.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
20
Figure 14: Correlation between interest rate betas and the liquidity coverage ratio as well as
excess reserves
Note: Excess reserves are defined as bank deposits at central banks minus minimum required reserves (currently 1%)
and are expressed in percent of total euro area MFI assets.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
21
4.5 Pass-through to rates of existing mortgages weakened over time as the share
of fixed rate mortgages increased, with heterogeneous effects across countries
In this subsection, we focus on pass-through to mortgage rates. Mortgages are the largest part of household
debt in Europe, varying from about half of household debt in Hungary to more than 90 percent in the
Netherlands. In this subsection, we focus on how pass-through to the rates of existing mortgages (which is
determined by the share of flexible-rate mortgages and the average duration of fixed-rate mortgages)
13
varies
and transmits monetary policy differently across countries.
First, since the GFC, the share of flexible-rate mortgages in new mortgages has decreased in many European
countries, which reduces pass-through to the rates of outstanding mortgages, everything else equal (Figure
15). However, even fixed-rate mortgages are subject to repricing when the fixed term expires, implying that
pass-through decreases with higher average duration of fixed-rate mortgages. Based on limited available
information on average duration of household mortgages, there is large cross-country variations. For example,
the majority of mortgage rates are fixed for periods longer than ten years in Belgium, whereas in the UK,
mortgages have much lower maturities (but are still classified as fixed rate). Such variations also imply very
different pass-through to the rates of outstanding mortgages (Figure 16).
Figure 15. Share of Flexible-Rate Mortgages
(percent)
Figure 16. Maturity of Fixed-Rate Mortgages
(percent, 2023Q2)
Sources: ECB; National Central Banks; The European Mortgage Federation; and IMF staff calculations
Note: In the left panel, we used 2012, 2022 or the closest year with data. In the right panel, data are not available for other types of
mortgages in Greece. Data for the Netherlands are as of 2023Q1.
Second, the extent of pass-through to the rates of outstanding mortgages, and their likely effects for monetary
policy transmission differ significantly across countries. As shown in Figure 17, while mortgage costs are more
responsive to increases in policy rates in many central, eastern and southeastern countries, the relatively low
share of households with mortgages limits the impact on the whole economy, softening monetary transmission.
On the other hand, despite the high share of households with mortgages in countries like the Netherlands, the
low pass-through mitigates monetary transmission. There are also countries with a low share of households
with mortgages and low pass-through, in which policy rate changes have arguably the least effects.
13
Flexible rate mortgages are defined as mortgages with an interest fixation of less than a year, whereas fixed rate mortgages are
defined as mortgages with interest fixation of above one year.
AT
BE
BG
HR
CY
CZ
DK
EE
FI
FR
DE
GR
HU
IE
IT
LV
LT
LU
MT
NL
PL
PT
RO
SK
SI
ES
SE
NO
UK
IS
0
10
20
30
40
50
60
70
80
90
100
0 20 40 60 80 100
Share of flexible
-rates in 2022
Share of flexible-rates in 2012
0
20
40
60
80
100
120
Finland
Portugal
Sweden
Italy
Romania
Greece
Poland
Denmark
Spain
Netherlands
Germany
Ireland
UK
Czech Republic
Belgium
Hungary
Up to 1Y 1-5Y 5-10Y >10Y
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
22
Third, for some countries, strong pass-through to rates of outstanding mortgages in combination with high stock
of mortgages can imply large aggregate changes in household debt service costs. In Figure 18, we quantify the
aggregate changes in debt service costs in the post-pandemic cycle which are a function of the magnitude of
the pass-through and the stock of outstanding mortgages. We proxy the changes in annual interest payments
for mortgages by the change in the interest rate on existing mortgages multiplied by the stock of outstanding
mortgages at the beginning of the cycle. The variation across euro area economies is large, with households in
Portugal having experienced an increase of interest costs of more than 1.2 percent of GDP annually based on
this approximation. However, as explained in Section 2, these effects cannot be equated to the aggregate
output effects of monetary policy.
Figure 17. Pass-through and Share of Households
with Mortgages (2021-23)
Figure 18. Changes in annual mortgage
service costs due to ECB policy hikes
(percent of 2022 GDP and relative to 2022M7
mortgage stock)
Sources: ECB, Haver Analytics, national authorities, and IMF calculations.
Note: In the right panel, the ending date is August 2023.
5. Conclusions
The post-pandemic hiking cycle has been unprecedented in terms of the cumulative magnitude and pace of the
policy rate increases. Yet, in this paper, we find compelling early evidence that the pass-through (relative to the
change in policy rates) has been smaller and slower compared to the past, with some nuances across rates,
samples, and methodologies. One exception are NFC loan rates, for which we do not find robust evidence of
slower or weaker pass-through, in line with previous evidence; another exception relates to the pass-through to
NFC time deposits, which is not significantly weaker. In addition, our impulse response functions suggest that
the initially weak pass-through of NFC loan and time deposit rates caught up with historical averages over time.
Using regression analysis, we find some evidence that higher financial sector concentration, the volume of
deposits relative to loans, and liquidity have contributed to weak pass-through to deposit rates. Finally, we
show that the pass-through to the interest rate of existing mortgages can have potentially vastly different
implications for monetary policy transmission depending on the volume of mortgage debt. Given that pass-
through to interest rates is an important aspect of monetary policy transmission, insights into the strength and
speed of pass-through to bank interest rates can help inform monetary policy decisions. In the context of the
post-pandemic tightening cycle, slower pass-through implies longer lags in monetary policy transmission than
in the past, but ultimately the same policy effects. Conceptually, anticipation effects (implying that some of the
AT
BE
CY
DE
ES
FI
FR
IE
IT
LU
MT
NL
NO
PT
SE
UK
AL
EE
HU
LT
LV
PL
RO
RS
SI
SK
-0.2
0
0.2
0.4
0.6
0.8
1
0 10 20 30 40 50 60 70
Betas of rates of outstanding mortgages
Share of HHs with mortgages
Non-CESEE CESEE
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
PRT
FIN
EST
ESP
LUX
CYP
LTU
AUT
LVA
ITA
SVN
SVK
IRL
NLD
BEL
DEU
FRA
MLT
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
23
pass-through occurred before the actual increase in policy rates) and negative or zero initial rates at the start of
the post-pandemic cycle could have added to the weaker pass-through, but our results point to small effects.
Future research could revisit our results once more data becomes available to account for any potential lags in
pass-through in the post-pandemic tightening cycle and further test the robustness of our results, including by
using alternative measures of concentration. In addition, future research could expand our work by comparing
the pass-through between hiking and loosening cycles, and analyzing the pass-through to NFC loan rates in a
more granular way. A natural complement to our analysis is a more comprehensive analysis of the effects of
monetary policy changes in conjunction with deposit and loan volumes, similarly to Lane (2023a, b), which
together with our analysis of pass-through would provide a more comprehensive perspective on monetary
policy transmission. For instance, Lane (2023c) shows that loan volumes in the euro area have weakened
sharply starting from the end of 2022. In addition, some countries in our sample are partially euroized, implying
that the pass-through to deposits and loans in domestic currency matters less. Future research could study the
determinants and effects of pass-through in these countries which could be interdependent with the interest
rate on euro denominated deposits and loans; see Chen et al. (2024) for a discussion on monetary policy
transmission in partially euroized European countries.
References
Abadi, J., Brunnermeier, M., and Koby, Y. “The Reversal Interest Rate.” American Economic Review, 113 (8):
2084-2120.
Auclert, A. "Monetary Policy and the Redistribution Channel." American Economic Review, 109, no. 6 (2019):
2333-2367.
Bernhofer, D., and van Treeck, T. (2013). “New Evidence of Heterogeneous Bank Interest Rate Pass-Through
in the Euro Area.” Economic Modelling, 35, 418-429.
Binici, M., 2023. “Interest Rate Pass-Through in Malta”, IMF Selected Issues Paper, (forthcoming).
Borio, C., Gambocarta, L., and Hofmann, B. "The Influence of Monetary Policy on Bank
Profitability." International Finance 20, no. 1 (2017): 48-63.
Burstein, A., and Gopinath, G. "International Prices and Exchange rates." Handbook of International Economics
Vol. 4. Elsevier, 2014. 391-451.
Byrne, D. and Foster, S. 2023, “Transmission of Monetary Policy: Bank Interest Rate Pass-Through in Ireland
and the euro area,” Economic Letter, Vol. 2023, No. 3.
Cabezon, E., and Kovachevska, S., 2024, “Interest Rate Pass-Through in North Macedonia.” IMF Selected
Issues Paper (forthcoming).
Chen, W., Jamaludin , F., Misch, F., Pienkowski, A., Wang, M., and Zhu , Z., 2024. “Monetary Policy
Transmission in Euroized Countries.” The Experience of Emerging European Economies, Unpublished
manuscript.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
24
de Bondt, G. J. (2005). “Interest Rate PassThrough: Empirical Results for the Euro Area.” German Economic
Review, 6(1), 37-78.
Egert, B., Crespo-Cuaresma, J., and Reininger, T. (2007). “Interest Rate Pass-Through in Central and Eastern
Europe: Reborn from ashes merely to pass away?” Journal of Policy Modeling, 29(2), 209-225.
Fricke, D., Greppmair, S., and Paludkiewicz, K. (2023). “Excess Reserves and Monetary Policy Tightening.”
Available at SSRN.
Goodfriend, M., and G. King, R. "The New Neoclassical Synthesis and the Role of Monetary Policy." NBER
macroeconomics annual 12 (19fr97): 231-283.
Gregor, J., and Melecký, M. (2018). “The Pass-Through of Monetary Policy Rate to Lending rates: The role of
Macro-Financial Factors.” Economic Modelling, 73, 71-88.
Gregor, J., Melecký, A., and Melecký, M. (2021). “Interest Rate PassThrough: A MetaAnalysis of the
Literature.” Journal of Economic Surveys, 35(1), 141-191.
Guindos, L.D. (2023). “The Inflation Outlook and Monetary Policy in the Euro Area.” First Annual Conference of
the Central Bank of Cyprus. Limassol, October 4.
Hansen, N. J. H., and Welz, P. (2011). "Interest Rate Pass-Through During the Global Financial Crisis: The
Case of Sweden.”
Itamar, D., Savov, A., and Schnabl, P. "The Deposits Channel of Monetary Policy." The Quarterly Journal of
Economics 132, no. 4 (2017): 1819-1876.
Jorda, O., 2005, “Estimation and Inference of Impulse Responses by Local Projections,” American Economic
Review, Vol. 95, No. 1, March, pp. 161-182.
Kashyap, Anil K., and Stein, J.C., (2000). “What Do a Million Observations on Banks Say About the
Transmission of Monetary Policy?” American Economic Review 90, no. 3: 407-428.
Kho, S., 2023. “Deposit Market Concentration and Monetary Transmission: Evidence from the Euro Area.” DNB
Working Paper, Series No. 790
Kiyotaki, N., and Moore, J. (1997) “Credit cycles.” Journal of Political Economy, 105(2), 211-248.
Lane, P.R. (2023a). “Panel Discussion on Banking Solvency and Monetary Policy.” NBER Summer Institute
2023 Macro, Money, and Financial Frictions Workshop. Cambridge, MA, July 12.
Lane, P.R. (2023b). “Speech on Monetary Policy Transmission in the Euro Area.” Banque de France-CEPR-
ECB Conference on Monetary Policy Challenges for European Macroeconomies. Paris, September 26.
Lane, P.R. (2023c). “Keynote Speech on Monetary Policy Tightening and the Financing of Firms.” Enterprise
Ireland Summit 2023. Dublin, April 19.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
25
Lane, P.R. (2023d). “The Banking Channel of Monetary Policy Tightening in the Euro Area.” NBER Summer
Institute 2023 Macro, Money, and Financial Frictions Workshop. Cambridge, MA, July 12.
Lettau, M., and Ludvigson S. C., 2004, "Understanding Trend and Cycle in Asset Values: Reevaluating the
Wealth Effect on Consumption." American Economic Review 94, no. 1 (2004): 276-299.
Lopez-Quiles, C. (2021). “Interest Rate Pass-Through in the Euro Area.” Essays in financial economics.
Chapter 3. European University Institute. PhD Thesis.
McGregor, T., Suphaphiphat, N., and Toscani, F. (2022). “Europe’s Consumers are Sitting on 1 Trillion Euros in
Pandemic Savings.” IMF blog.
Messer, T. and Niepmann, F., 2023, “What Determines Passthrough of Policy Rates to Deposit Rates in the
Euro Area?” FED Notes¸ Board of Governors of the Federal Reserve System, July 28.
Mishkin, F. S., (1996) “The Channels of Monetary Transmission: Lessons for Monetary Policy.” NBER Working
Paper Series. Working Paper 5464.
Polo, A. (2021). “Imperfect pass-through to deposit rates and monetary policy transmission.”, Staff Working
Paper No. 933, Bank of England.
Saborowski, C., and Weber, S. (2013). “Assessing the Determinants of Interest Rate Transmission Through
Conditional Impulse Response Functions1.” IMF Working Papers, 2013(023).
Sander, H., and Kleimeier, S. (2006). “Convergence of Interest Rate Pass-Through in a wider Euro
zone?” Economic Systems, 30(4), 405-423.
Serhan, C., Cui, L., Engler, P., Wang, M., und Yao, J. 2023. “Monetary Transmission Channels in Europe“.
unpublished manuscript.
Sørensen, C., and Werner, T. (2006). “Bank Interest Rate Pass-Through in the Euro Area: A Cross Country
Comparison” (No. 0580). European Central Bank.
Valderrama, L., Gorse, P., Marinkov, M., and Topalova, P., (2023). “European Housing Markets at a Turning
Point Risks: Household and Bank Vulnerabilities, and Policy Options.” IMF Working Paper, 23/76
Wang, Y., Whited, T. M., Wu, Y., and Xiao, K. (2022). “Bank Market Power and Monetary Policy Transmission:
Evidence from a structural estimation,” The Journal of Finance, 77(4), 2093-2141.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
26
Annex I. Data Description
Table A1: Post-pandemic Hiking Cycle and Comparator Hiking Cycle
14
Country
Start
End
The number of
policy rate hikes
Length of the
cycle of hikes
Cumulative policy
rate change in bps
Pace
(bps/month)
Albania
Mar-22
Mar-23
6
12
250
21
Czech Republic
Aug-17
Feb-20
9
30
220
7
Jun-21
Jun-22
9
12
675
56
Hungary
Jun-03
Nov-03
2
5
600
120
Jun-21
Sep-22
15
15
1240
83
North
Macedonia
May-16
May-16
1
1
75
75
Apr-22
Jun-23
12
14
475
34
Norway
Sep-18
Sep-19
4
12
100
8
Sep-21
Jun-23
11
21
375
18
Poland
Apr-07
Jun-08
8
14
200
14
Oct-21
Sep-22
11
11
665
60
Romania
Nov-07
Aug-08
7
9
325
36
Oct-21
Jan-23
11
15
575
38
Serbia
Oct-10
Apr-11
6
6
350
58
Apr-22
Jul-23
15
15
550
37
Sweden
Jan-06
Sep-08
13
32
325
10
May-22
Jul-23
7
14
375
27
Switzerland
Jun-22
Jun-23
5
12
250
21
United
Kingdom
Aug-06
Jul-07
5
11
125
11
Dec-21
Jun-23
13
18
490
27
Euro Area
Dec-05
Jul-08
9
31
225
7
Jul-22
Jul-23
9
12
425
35
Average
Previous comparator
6
15
255
35
Post-pandemic cycle
11
14
533
38
Note: A comparison episode to the post-pandemic tightening cycle is selected from the past tightening periods based on the largest
cumulative policy rate change. The start date of each cycle is defined as the month of the first-rate hike, and the end date is chosen
as the month of the last rate hike. We restrict the cycles to periods in which at least one rate is available.
14
In some countries, the end of the post-pandemic tightening cycle is not covered by our bank interest rate data. Albania resumed
monetary policy tightening in November 2023, and Norway resumed monetary policy tightening in December 2023.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
27
Table A2: Country Coverage and Longest Time Span for the Bank Interest Rates
Country
Longest time span
Start End
HH time
deposits
NFC
time
deposits
HH O/N
deposits
NFC
O/N
deposits
Consumer
loans
NFC
loans
Mortgages
Albania
Dec-18
Aug-23
✓
✓
✓
✓
Austria
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Belgium
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Cyprus
Jan-03
Aug-23
✓
✓
✓
Czech
Republic
Jan-04
Aug-23
✓
✓
✓
✓
✓
✓
✓
Estonia
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Finland
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
France
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Germany
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Greece
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Hungary
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Ireland
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Italy
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Latvia
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Lithuania
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Luxembourg
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Malta
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Netherlands
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
North
Macedonia
Jan-15
Aug-23
✓
✓
✓
✓
✓
✓
✓
Norway
Dec-13
Aug-23
✓
✓
✓
✓
✓
✓
✓
Poland
Jan-04
Aug-23
✓
✓
✓
✓
✓
✓
Portugal
Jan-03
Aug-23
✓
✓
✓
✓
Romania
Jan-07
Aug-23
✓
✓
✓
✓
✓
✓
✓
Serbia
Sep-10
Aug-23
✓
✓
✓
✓
✓
✓
✓
Slovak
Republic
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
✓
Slovenia
Jan-03
Aug-23
✓
✓
✓
✓
✓
Spain
Jan-03
Aug-23
✓
✓
✓
✓
✓
✓
Sweden
Aug-05
Aug-23
✓
✓
✓
✓
✓
✓
✓
Switzerland
Jun-17
Aug-23
✓
✓
✓
United
Kingdom
Jan-03
Jul-23
✓
✓
✓
✓
✓
✓
✓
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
28
Table A3: Excluded Countries in Comparisons of Pass-Through Over Time
Bank rate
Excluded countries in comparisons of pass-through over time
HH time deposit rate
Albania, Greece, Serbia, Lithuania, Norway, Latvia, North Macedonia
NFC time deposit rate
Albania, Ireland, Finland, Greece, Lithuania, Luxembourg, Latvia, North
Macedonia, Malta, Serbia, Norway, Sweden
HH overnight deposit rate
Albania, Greece, Serbia, Lithuania, Latvia, North Macedonia
NFC overnight deposit rate
Albania, Greece, Serbia, Lithuania, Latvia, North Macedonia
NFC loan rates
Albania, Greece, Hungary, Lithuania, Luxembourg, Latvia, Malta, North
Macedonia, Poland, Sweden
HH mortgage rates
Albania, Greece, Hungary, Lithuania, Luxembourg, Latvia, Malta, North
Macedonia, Poland, Sweden
HH time deposit rate
Albania, Greece, Serbia, Lithuania, Norway, Latvia, North Macedonia
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
29
Table A4: Variable Definitions and Sources
Variables
Definition
Frequency
Data sources
Monetary policy
rate
Policy rate
We use main refinancing
operations (MRO) for ECB policy
rate and short-term interest rate
for non-euro area countries
Monthly
ECBnational
central banks
Bank interest rates
HH O/N deposits
Interest rate on overnight
household deposits
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
NFC O/N deposits
Interest rate on overnight non-
financial corporation
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
HH time deposits
Interest rate on household time
deposits with agreed maturity for
new business
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
NFC time deposits
Interest rate on non-financial
corporation deposits with agreed
maturity for new business
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
Consumer loans
Interest rate on consumer loans
with agreed maturity for new
business
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
NFC loans
Interest rate on loans to non-
financial corporation for new
business
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
Mortgages
Interest rate on new mortgages,
all maturity
Monthly
ECB: MIR-MFI
Interest Rate
Statistics;
national central
banks
Financial sector
characteristics
Concentration
We use Herfindahl index of total
assets of credit institutions as a
proxy for banking sector
concentration
Annual
ECB
LTD
loan-to-deposits ratio as proxy for
volume of deposits relative to
profitable lending opportunities
Monthly, Quarterly
(EA19, Romania)
ECB, national
central banks
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
30
Annex II. Additional Charts and Results
Figure A1: Pass-Through in Post-Pandemic Relative to the Comparator Cycle
(Pass-through computed starting from six months before first rate hike)
Note: Interest rate betas during the post-pandemic cycle expressed as percent of the corresponding betas in the comparator cycle
as in Figure 5, but assuming that anticipation effects imply that pass-through at least for some bank rates begin up to six months
before the actual policy rate hikes
Figure A2: Estimated Pass-Through with Different Types of Fixed Effects
Note: This figure covers the entire sample for all periods and does not include any interaction terms.
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Loans O/N deposits Time
deposits
Mortgages O/N deposits Time
deposits
NFC HH
Europe Euro Area
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
31
Figure A3. Estimated Pace of Pass-Through by Country
Note: Pace of pass-through estimated based on country-by-country impulse response functions. The pace of pass-through which is
the peak response divided by the time it takes to reach the peak.
Austria
Belgium
Czech Republic
Estonia
Finland
France
Germany
Ireland
Italy
Netherlands
Norway
Portugal
Romania
Serbia
Slovak Republic
Slovenia
Spain
United Kingdom
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Previous tightening episodes
Post-pandemic tightening episode
Pace of pass-through to NFC loan rates
(pp/month)
Austria
Belgium
Cyprus
Czech Republic
Estonia
Finland
France
Germany
Ireland
Italy
Norway
Portugal
Romania
Slovak Republic
Slovenia
Spain
United Kingdom
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Previous tightening episodes
Post-pandemic tightening episode
Pace of pass-through to mortgage rates
(pp/month)
Austria
Belgium
Cyprus
Czech Rep.
Estonia
Finland
France
Germany
Hungary
Ireland
Italy
Luxembourg
Netherlands
Norway
Poland
Romania
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Previous tightening episodes
Post-pandemic tightening episode
Pace of pass-through to NFC time deposit
rates
(pp/month)
Austria
Belgium
Cyprus
Czech Republic
Estonia
Finland
France
Germany
Hungary
Ireland
Italy
Luxembourg
Malta
Netherlands
Poland
Portugal
Romania
Slovenia
Spain
Sweden
United Kingdom
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
Previous tightening episodes
Post-pandemic tightening episode
Pace of pass-through to HH time deposit
rates
(pp/month)
Austria
Belgium
Cyprus
Czech Rep.
Estonia
Finland
France
Germany
Hungary
Italy
Luxembourg
Malta
Netherlands
Norway
Poland
Romania
Slovak Republic
Spain
Sweden
UK
0
0.1
0.2
0.3
0.4
0 0.1 0.2 0.3 0.4
Previous tightening episodes
Post-pandemic tightening episode
Pace of pass-through to NFC O/N deposit
rates
(pp/month)
Austria
Belgium
Cyprus
Czech Republic
Estonia
Finland
France
Germany
Hungary
Ireland
Italy
Luxembourg
Malta
Netherlands
Norway
Poland
Portugal
Romania
Slovenia
Spain
Sweden
United Kingdom
0
0.05
0.1
0.15
0.2
0.25
0 0.05 0.1 0.15 0.2 0.25
Previous tightening episodes
Post-pandemic tightening episode
Pace of pass-through to HH O/N deposit
rates
(pp/month)
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
32
Figure A4: Post-Pandemic Interest Rate Betas and Country Characteristics
Note: Pass-through calculated as the ratio of cumulative change in interest rates relative to cumulative changes in the policy rate
from month of first rate increase to three months after last rate increase (or the first rate cut or the end of the sample, whichever is
shorter) in the post-pandemic tightening cycle.
IMF WORKING PAPERS
Monetary Policy Pass-Through to Interest Rates
INTERNATIONAL MONETARY FUND
33
Figure A5: Estimated Policy Pass-Through: Post-pandemic Tightening vs. Previous Tightening Cycles
(controlling for different pass-through at zero or negative policy rates)
Note: Figure shows differences between post-pandemic and previous tightening cycles as estimated in analogy to Eq. (1) but with
inclusion of two dummies for tightening in the regression model - one capturing increases from positive policy rates and one
capturing increases from zero or negative rates.
-0.45
-0.4
-0.35
-0.3
-0.25
-0.2
-0.15
-0.1
-0.05
0
NFC time
deposits
HH time
deposits
NFC O/N
deposits
HH O/N
deposits Mortgages
Consumer
loans
Baseline
Increase from positive policy rate
Monetary Policy Pass-Thorough to Interest Rates: Stylized Facts from 30 European Countries
Working Paper No. WP/2024/009
