Int Adv Econ Res (2012) 18:1–14

DOI 10.1007/s11294-011-9328-x

The Bank Lending Channel and Monetary Policy Rules:

Evidence from European Banks

Nicholas Apergis & Effrosyni Alevizopoulou

Published online: 12 October 2011

# International Atlantic Economic Society 2011

Abstract There exist two main channels of the monetary transmission mechanism:
the interest rate and the bank lending channel. This paper focuses on the latter, which
is based on the central bank’s actions that affect loan supply and real spending. The
supply of loans depends on the monetary policy indicator, which, in most studies, is
the real short-term interest rate. The question investigated in this paper is how the
operation of the bank lending channel changes when this short-term indicator is
allowed to be endogenously determined by the target rate the central bank sets
through a monetary rule. We examine the effect that a rule has on the bank lending
channel in European banking institutions spanning the period 1999–2009. The
expectations concerning inflation and output affect the decision of the central bank
for the target rate, which, in turn, affect private sector’s expectations —commercial
banks— by altering their loan supply.

Keywords Monetary policy rules . Bank lending channel . European banks . GMM
methodology

JEL G21 . E52 . C33

Introduction

The monetary transmission mechanism is a powerful tool, since through this the
monetary authorities can affect the economy with their decisions. This mechanism
consists of various channels through which monetary policy is conducted, but the
two main ones are: the interest rate channel (money view) and the credit channel
(credit view). As far as the former is concerned, monetary policy changes affect
aggregate demand through interest rates, whereas the latter accommodates the

N. Apergis (*) : E. Alevizopoulou

Department of Banking & Financial Management, University of Piraeus, 80 Karaoli & Dimitriou,
18534 Piraeus, Greece
e-mail: napergis@unipi.gr
2 N. Apergis, E. Alevizopoulou

transmission of policy decisions by altering the availability and supply of loans

(Hernando and Pages 2001). One of the sub-channels of the credit view is the bank
lending one, which “stems from financial market incompleteness and relies on
imperfect substitutability” (Gambacorta 2005).
The alteration in banks’ loan supply is caused by changes in their reserves, which
is initially affected by the decisions of central banks concerning the course of the
interest rate. The goal of this paper is to investigate the effect on the operation of the
bank lending channel, in case that central banks replace their primary monetary
policy instrument they usually use with the target rate, the estimation of which is
based on a set of macroeconomic variables. In other words, this paper investigates
the impact of the bank lending channel on the economy when different types of
interest rate rules are used as the monetary policy indicator. The formulation of these
rules depends on data, the timing of which differs: they may be lagged, current, or
forecasts and then the results are compared to one another. The empirical findings
display that the bank lending channel operates better in the case when forward-
looking rules are considered as the monetary policy indicator.
The rest of the paper is organized as follows: The following section reviews the
literature concerning the bank lending channel and interest rate rules, whereas the
next section analyses the data. The following section refers to the methodology used
both for the estimation of the different types of rules and the lending channel. The
final two sections present the results and conclusions, respectively.

Literature Review

The Bank Lending Channel

Concerning the bank lending channel, this study investigates how the decisions of
the monetary authorities are transmitted and whether there is any impact of these
decisions on economic activity. The operation of this channel mostly depends on the
supply of loans and the factors that determine their course. In particular, a restrictive
monetary policy leads to a reduction in bank reserves and deposits and, consequently,
to a fall in loan supply. Therefore, businesses and consumers, who depend on bank
lending, reduce their purchases of durable goods and purchases of capital for
investment and, hence, output is also affected in a negative way (Golodniuk 2006).
The opposite occurs in the case of an expansionary monetary policy.
There are three necessary conditions for this channel to have economic power.
First, firms should not be perfectly indifferent to the different types of finance. They
should be dependent on bank loans and not be able to replace losses of bank loans—
due to decreases in loan supply by the monetary authorities—with other types of
finance (Oliner and Rudenbusch 1995). If firms are indifferent between the two
types of financing, then the decrease in supply of loans does not affect them at all.
Second, the central bank should be capable of affecting the supply of loans through
the changes it imposes on the volume of reserves. For instance, in the case of a
restrictive monetary policy, banks must not be capable of offsetting the decrease in
funds from deposits by raising funds from other sources (Oliner and Rudenbusch
1995). The third condition that should be satisfied is that there must be some
Monetary Policy Rules 3

imperfections in the adjustment of the aggregate price level. The imperfect

adjustment in prices is necessary because monetary policy would have no impact
if prices could adjust by the same percentage every time money supply changes
(Golodniuk 2006).
The bank lending channel literature has focused on investigating the presence of a
channel in different economies or in a group of countries and, more specifically,
examines whether the effect on lending responses differs depending on the strength
of a bank, which is also determined by variables, such as asset size, capitalization
and liquidity. The empirical evidence supports the idea that well-capitalized and
liquid banks are less affected by monetary policy changes than banks that have low
capital and liquidity ratios. As far as size is concerned, the majority of studies show
that small banks do not seem to be more sensitive to monetary policy shocks than
large banks (Peek and Rosengren 1995; Gambacorta 2005; Golodniuk 2006). In
another strand of literature, however, empirical studies find that large banks, in
combination with high capitalization ratios, are less responsive to monetary policy
shocks (Kishan and Opiela 2000).
As far as the empirical procedure is concerned, there has been an important
problem to overcome. More precisely, the problem is detected in the difficulty to
identify whether the effects on output are due to either shifts in loan demand or shifts
in loan supply. The fact that both output and bank loans decrease after a negative
change in monetary policy does not lead to the conclusion that this is due to changes
in loan supply (Oliner and Rudenbusch 1995; Brissimis et al. 2001). By contrast,
such changes may be the result of a shift in loan demand. For instance, a tight
monetary policy leads to an increase in interest rates and, consequently, in higher
costs, which does not favor investments, leading to a fall in loan demand and,
therefore, in the volume of loans. For this problem to be resolved, the literature has
focused not on the use of aggregate data, but on the analysis of microeconomic data
of firms and banks (Kashyap et al. 1993).

Monetary Policy Rules

Over the last decades, a shift from discretionary policy to rule-based monetary policy
has been evident. Hence, central banks adopted monetary policy rules, in which the
main instrument was a short-term interest rate that was adjusted according to a set of
macroeconomic variables. The first paper to thoroughly investigate the operation and
effectiveness of monetary policy rules was that by Taylor (1993), who showed that
monetary policy in the U.S. could be described by a simple rule, in which the short-
term interest rate was close to the actual decision-making process and was adjusted
according to two major monetary policy objectives: price stability and economic
growth. More precisely, the target rate—that is the target for the federal funds rate—
adjusts according to four main variables: the first variable concerns the long-run
equilibrium real interest rate, the second one is the current inflation rate, whereas the
third and the fourth factor concern the inflation and output gap, respectively.
Analytically, the inflation gap is the deviation of the inflation rate from the target
inflation. In the case where inflation is larger than its target, the rule recommends an
increase in the interest rate, whereas the opposite occurs when the inflation gap is
negative. This term represents the goal of the central bank to achieve price stability.
4 N. Apergis, E. Alevizopoulou

The output gap is based on the difference between real GDP and potential GDP.
Respectively to the inflation gap, when the output gap is positive, Taylor rule
recommends the increase of the interest rate, whereas when real GDP is below its
potential level, then the interest rate should decline. The fourth factor represents the
promotion of the maximum sustainable growth. As Kozichi (1999) points out, the factor
concerning the output gap plays a forward-looking role, implying that a positive gap
signals potential increases of inflation in the future. Therefore, adjustments of the
interest rate vis-à-vis the output gap reflect policy actions, which are designed to
preempt an anticipated rise in inflation. The form of the Taylor rule yields:
 »
it ¼ p t þ i þ0:5»ðp t  p Þ þ 0:5»yt ð1Þ

where it is the target interest rate, πt is the inflation rate, i is the long-run equilibrium
interest rate assumed to be equal to 2%, π* is the target for inflation, which is also
assumed to be 2% percent, and yt is the output gap.
This simple rule by Taylor (1993) seems to have accurately interpreted the
monetary policy actions by the Federal Reserve and the historical analysis, as
mentioned in Orphanides (2003), suggests that this rule serves as a “useful
organizing device for interpreting past policy decisions and mistakes, but adoption
of the Taylor-type framework for policy analysis is not insurance that past policy
mistakes would not have occurred.”
Nevertheless, the Taylor rule has certain limitations: the first concerns timing,
because central banks do not know the contemporaneous output and inflation gap
when setting the interest rate for a given time. This problem can be solved by using
lagged output and inflation gap data, i.e., in this case rules are referred as backward-
looking rules, or by replacing the current measures of variables with forecasts i.e.,
the forward-looking rules case. Another limitation of the Taylor’s classic formulation
is in the weights that represent the policy responsiveness to deviations of inflation
and output from their targets and which are strictly specified. Finally, it has been
observed that central banks have the tendency to smooth movements of interest
rates, which may also be incorporated into the rule (Kozichi 1999).
The classic Taylor rule can be interpreted narrowly, i.e., in its algebraic form as
mentioned above, or in a broader way. As Orphanides (2003) notes, the broad
interpretation allows for a degree of flexibility to overcome the limitations of the
classic framework. Taylor himself emphasizes that the rule can be viewed as a
monetary policy program, which is used to attain the fundamental policy objectives
and should not be used as a mechanical formula, but rather as a guiding principle for
monetary authorities.
As already mentioned, in addition to the classic Taylor rule, there are two other
main types of rules: backward- and forward-looking rules. There is a debate on
which of these rules fits historical data more accurately and gives better outcomes.
Taylor (1999) argues that forecasting rules may have one possible advantage over
the simple rule proposed in 1993, which is the incorporation of additional variables
in addition to inflation and output that may be useful for the forecast. However,
when applying these forecasting rules, and particularly the rules suggested by
Haldane and Batini (1999), to the interest rate setting by the European central bank,
simple rules dominate over the forecast inflation rule. Taylor continues with his
Monetary Policy Rules 5

argument by stating that in reality, forecasts are based on current and lagged data and
hence, forward-looking rules are based on current and lagged data as well.
By contrast, there is strong doubt over the use of past values, because as Greenspan
(1997) points out, the observation of past macroeconomic behavior and its use in the
formation of rules embodies a notion that the future will be like the past, while forecasts
play an important role in the way monetary policy responses and this is not reflected in
the standard Taylor rule (Meyer 2002). Hence, the formulation of monetary policy
should be treated as a forward-looking process, which ought to take into consideration
all available information to form adequate policy rules (Orphanides 2003).
This approach is also followed by Clarida et al. (1998, 2000) who replace current
and recent outcomes of output and inflation with forecasts of these variables. The
same procedure applied by Fourcans and Vranceanu (2004) for the European central
bank’s interest rate rule; their results indicate that the response of the interest rate to
deviations of inflation from its target is stronger if based on future inflation than on
current inflation.

Data Description

Interest Rate Rule Data Description

For the estimation of the interest rate rule, quarterly data from Datastream and
Bloomberg databases were derived. It must be noted, at this point, that we estimate
interest rate rules for three different economies: a European group including Austria,
Belgium, Finland, France, Germany and the Netherlands, i.e., countries which use
the euro as a common currency; Denmark; and the United Kingdom, two countries
with their own currency which follow the monetary policy of the euro zone, but still
maintain some degree of autonomy as far as their central bank is concerned. For
each country, the consumer price index is used to measure inflation, while GDP is
detrended by using the Hodrick-Prescott filter to obtain a measure of the output gap.
The short-term interest rate is the one that the European Central Bank (ECB) uses for
the main refinancing operations, whereas for Denmark the discount rate of
Danmarks Nationalbank is used, and for the United Kingdom the bank rate of
Bank of England is used. The period under examination spans the period 1999 to
2009. We should mention at this point that for the construction of the Euro-group,
we make use of weights for each country, for the estimation of which we followed a
similar approach recommended by the International Monetary Fund.

Bank Lending Channel Data Description

As far as the bank lending channel is concerned, quarterly data of total loans is used
as the dependent variable and is obtained from the BankScope database spanning the
period 2000 to 2009. In particular, a balanced sample of 771 European commercial
and savings banks is used. The European countries are Austria, Belgium, Finland,
France, Germany, and the Netherlands, which are used as a group, whereas Denmark
and the United Kingdom are examined in a separate fashion. Table 1 presents the
countries and the number of banks in each of them.
6 N. Apergis, E. Alevizopoulou

We also obtain short-term interest rates, which are used as a proxy for monetary
policy, from the Bloomberg database. GDP values and inflation rates for each
country are obtained from Datastream to control for demand effects, i.e., to isolate
changes in total loans, which are caused by movements in loan demand.

The Econometric Approach

The Interest Rate Rule

The methodology used in the first part of the analysis is mostly based on the one by
Clarida et al. (1998) and, therefore, the notation used here resembles theirs to a great
degree. The first assumption concerns price and wage rigidities, because in this way,
it is feasible for the monetary policy to affect real activity in the short run. What also
should be noted is that central banks’ monetary policy indicator is the short-term
interest rate. This allows the central bank to choose the level of the interest rate from
period to period and conduct policy. More specifically, central banks vary the
nominal interest rate—the target rate—so as to effectively affect the real interest rate.
This target rate is estimated as follows:

i»t ¼ i þ b½Eðp tþ1 =Ωt Þ ¼ p » þ gxt ð2Þ


where, i* is the target interest rate, i is the long-run equilibrium nominal interest
rate, πt+1 is the inflation rate between periods t and t+1, and xt is the output gap,
which is the difference between output and its potential level. Furthermore, E is the
expectation operator and Ωt is the information set at time t, when central banks set
the target for the interest rate. As we can observe from this equation, the target rate
depends both on expected inflation and the output gap.
As Clarida et al. (1998, 2000) consider, the target for the ex-ante real interest rate
is iit  it  Eðp tþ1 =Ωt Þ. Therefore, the above equation turns out to be:
ii»t ¼ ii þ ðb  1Þ½Eðp tþ1 =Ωt Þ  p » þ gxt ð3Þ

Table 1 Number of banks in each country

Country Number of banks

Austria 72
Belgium 17
Finland 2
France 99
Germany 487
Netherlands 3
Denmark 62
United Kingdom 29
Total 771
Monetary Policy Rules 7


In this case, ii is the long-run equilibrium real interest rate. In an economy where
inflation targeting is the key, β plays an important role. If β>1, then the target rate
will adjust as a response to a rise in expected inflation, whereas, if β<1, the nominal
target rate also adjusts, albeit in a smaller degree, compared to expected inflation.
Therefore, the real interest rate will not increase satisfactorily, but on the contrary it
will decrease. As far as γ is concerned, according to economic theory, it is expected
to be positive. These coefficients of the policy rule indicate the weights that central
banks set on inflation and output gaps and how the monetary policy alters in
response to their changes.
Another assumption that is made is that the real interest rate should adjust to its
target and, hence, the rule should incorporate interest rate smoothing, which takes
the form:
»
it ¼ ð1  rÞit þ rit1 þ ut ð4Þ

where 0≤ρ≤1 is the degree of interest rate smoothing and ut is an exogenous random
shock, which follows an i.i.d process.
Additionally, the following definition must be made, so as to be able to estimate
the equation of the rule:
 »
a  i bp ð5Þ

Using the above definition in the first Eq. (2), we obtain:

»
it ¼ a þ bEðp tþ1 =Ωt Þ þ gxt ð6Þ
Next, if we incorporate Eq. (6) into Eq. (4), yields:
it ¼ ð1  rÞa þ ð1  rÞbEðp tþ1 =Ωt Þ þ ð1  rÞgxt þ rit1 þ ut ð7Þ
The next step is to rewrite the above equation in terms of realized variables:
it ¼ ð1  rÞa þ ð1  rÞbp tþ1 þ ð1  rÞgxt þ rit1 þ "t ð8Þ

where, "t  ð1  rÞ½bp tþ1  Eðp tþ1 =Ωt Þ þ ut : a linear combination of the
forecast errors of inflation and the exogenous random shock ut. As Clarida et al.
(1998, 2000) indicate, ut is a vector of variables that are available when the central
bank sets the interest rate target and it is orthogonal to εt, that is:
Eð"t =uÞ ¼ 0 ) Eðit  ð1  rÞa  ð1  rÞbp tþ1  ð1  rÞgxt  rit1 =ut Þ ¼ 0 ð9Þ

The parameters we need to estimate are α, β, γ, and ρ, which are estimated using
the Generalized Method of Moments (GMM) methodology. The instrument list
contains lagged values of inflation, the output gap, and interest rates.
The model by Clarida et al. (1998, 2000) provides a relation between the target
inflation and the long-run equilibrium real interest-rate, by using the coefficients β
and α. In other words, it is now feasible to estimate π*:

» ii a
p ¼ ; ð10Þ
b1
8 N. Apergis, E. Alevizopoulou

which becomes obvious from the given relations:

 »
a  i bp and
ð11Þ
  »
i ¼ ii þp
In accordance with Eq. (3), combined with the fact that backward interest rate
rules are nested in this equation, rules in this case are formed as follows:
i»t ¼ i þ ðb  1Þb
p t1 þ gxt1 ð12Þ
where, bp t1 ; xt1 are the lagged inflation and the output gap, respectively.
As with the forward rule, this one is also rearranged so as to obtain the rule for the
real target rate:
ii»t ¼ ii þ ðb  1Þb
p t1 þ gxt1 ð13Þ
Interest rate smoothing is also incorporated into the model and after the
adjustment mechanism, the rule takes the following form:
it ¼ ð1  rÞa þ ð1  rÞb b
p t1 þ ð1  rÞgxt1 þ rit1 ð14Þ
The GMM methodology is also used for the estimation of the parameters in the
backward case. Finally, we adjust the classic Taylor rule to the European data and
find the interest rate target, by adding the interest rate smoothing process to the rule
and using current data (Taylor-type rule). The estimating procedure yields:
it ¼ ð1  rÞa þ ð1  rÞb b
p t þ ð1  rÞgxt þ rit1 : ð15Þ

The Bank Lending Channel

The second part of the econometric approach investigates the presence of the bank
lending channel. The baseline equation takes the following form:
X X X
ΔlnLkt ¼ ak þ 8ΔlnLkt1 þ bj Δlnrktj þ d j ΔlnGDPktj þ wj p ktj þ "kt
j j j

ð16Þ
with k=1, ….8, t=1, ..T, where k denotes the country, Lkt denotes the loans of
country k at year t, rkt denotes the monetary policy indicator, GDPkt denotes the
GDP of country k at year t, πkt denotes the inflation in country k at year t, and εkt
denotes the error term.
It must be noted, that the monetary policy indicator takes two forms: the first
one concerns the short-term interest rate ad hoc, and the second one is set by the
central bank through an interest rate rule. The focus of this paper is on examining
how the loan growth reacts when the monetary policy indicator is not the usual
short-term interest rate obtained in an ad hoc manner from the data, but the interest
rate target and comparing the results to those when conventional monetary policy
indicators are used.
In Eq. (16), the growth rate of a country’s lending (ΔlnL) is regressed on GDP
growth rates (ΔlnGDP) and on inflation rates (π) to control for country-specific loan
Monetary Policy Rules 9

demand changes due to macroeconomic activity. In other words, we have to isolate

shifts in total loans caused by movements in loan demand. The introduction of these
two variables is important because it isolates the monetary policy indicator, which is
the short-term interest rate. Additionally, we include lagged values of the dependent
variable, because lagged loans affect current loans in an environment where a stable
relationship is established between the bank and the customer. In other words, the
bank acquires informational monopoly over a client and, hence, it is extremely costly
for a customer to change a bank because the services of the new bank will be more
expensive, since it needs to collect information about the new customer (Golodniuk
2006). Monetary policy can also affect lending with lags, due to long-term contractual
commitments. According to the theory of the bank lending channel, the coefficient of
the interest rate must be negative to imply that loans fall after a monetary tightening.
The model is estimated using the GMM estimator, suggested by Arellano and Bond
(1991), while only statistically significant lags are used in the estimation.

Empirical Analysis

Interest Rate Rules Results

The results for the interest rate channels are presented in Table 2 for the Euro-group,
in Table 3 for Denmark, and in Table 4 for the U.K., with each of these tables
showing the results for all types of rules.
The top line of Table 2 reports the estimates of the coefficients for the backward-
looking rule. According to the principle of Taylor rule, the key result is the
coefficient of β, i.e., of the inflation gap, which is 1.6753, implying that the central
bank will raise interest rates to 68 basis points, if inflation rises by one percentage
point. The coefficient of the output gap is positive as well, though it is very small
compared to the coefficient of inflation gap. From the same table we draw the
conclusion that the central bank puts substantial weight to the smoothing parameter,
whereas the J-statistic implies that we cannot reject the over identifying restrictions
of the model. The second line of Table 2 indicates the response of the target rate to
the inflation and the output gap in the case of the Taylor-type rule. The difference
with the previous results is that the β coefficient is less than one, implying that the
central bank, following a Taylor-type rule, fails to sufficiently raise interest rates.
The coefficient of the output gap is also small (0.0009), but positive as expected.

Table 2 Interest rate results for Eurogroup

Eurogroup

α β Γ ρ Prob J-statistic R-squared Adj. R-squared

BWD 0.0189 1.6753 0.0004 0.6828 0.1955 0.1092 0.8336 0.8208

TAYLOR 0.0251 0.5228 0.0009 0.4791 0.1740 0.1212 0.9229 0.9167
FWD 0.0225 1.1558 0.0016 0.0245 0.9966 0.0013 0.5833 0.5521
10 N. Apergis, E. Alevizopoulou

Table 3 Interest rate results for Denmark

Denmark

α β Γ ρ Prob J-statist R-squared Adj. R-squar

BWD 0.0185 2.1909 0.0034 0.6534 0.6913 0.0304 0.8009 0.7873

TAYLOR 0.0267 0.5987 0.0070 0.6279 0.1045 0.1282 0.8161 0.8036
FWD 0.0202 1.4376 0.0028 0.6552 0.0717 0.1557 0.9050 0.8981

The central bank smoothens interest rates as well. The last line of this table reports
the results for the forward-looking rule, with the Taylor principle (β coefficient
greater than unity) to prevail, since β is equal to 1.1558, and hence, by following a
forward-looking rule, the central bank accomplishes to raise nominal as well as real
rates. The coefficient for the output gap is positive, albeit small, whereas the
smoothing parameter does not play a very important role. The J-statistic in the case
of the Taylor-type and the forward-looking rule indicates that used instruments are
valid.
Next, we consider the results for the three rules concerning Denmark (Table 3),
which are quite similar to those for the Euro-group. Analytically, we begin with the
estimates for the backward-looking rule and specifically with the coefficient of the
inflation gap, which equals 2.1909 and satisfies the Taylor principle. This is not the
case, though, with the Taylor-type rule, where β is 0.5987, whereas the coefficient of
the inflation gap for the forward-looking rule is greater than unity and specifically is
equal to 1.4376. The γ coefficient of the output gap is positive in all types and the
same occurs with the interest rate smoothing parameter ρ, which is quite large in all
cases, revealing that the central bank puts a significant weight to past values of
interest rates. The test for over identifying restrictions indicates that the instruments
are valid again in all models.
Table 4 reports the results for the U.K., in which it is obvious that the coefficient
of the inflation gap is greater than unity in the case of the backward-looking rule,
whereas it does not satisfy the Taylor-principle for the rest types of rules. The
output gap coefficient is positive and the same occurs with the smoothing
parameter, indicating that the Bank of England puts substantial effort for smoothing
interest rates. Finally, in all models the validity of the instruments is indicated by
the J-statistic.

Table 4 Interest rate results for the United Kingdom

UK

α β γ ρ Prob J-statist R-squared Adj. R-squar

BWD 0.0322 2.4081 0.0001 0.9205 0.2141 0.0933 0.8688 0.8599

TAYLOR 0.0430 0.5795 0.0002 0.6755 0.4451 0.0557 0.8923 0.8850
FWD 0.0334 0.5882 0.0001 0.8942 0.1394 0.1247 0.8930 0.8849
Monetary Policy Rules 11

Bank Lending Channel Results

As long as the bank lending channel is concerned, the results for the three economies
are presented in Tables 5, 6, and 7, respectively. This part of the analysis investigates
first, the presence of the bank lending channel when four alternatives of the
monetary policy indicator are considered: the official interest rate of the central
bank; the target rate derived from the backward-looking rule; the Taylor-type rule;
and the forward-looking rule; and second, in what case the lending channel seems to
be stronger.
The entries in Table 5 report the coefficients of the variables in Eq. (16) and their
corresponding p-values estimated for the Euro-group by applying four different
indicators for the estimation of the bank lending channel. In columns 1 and 2, we
applied the interest rate of ECB in the model for the estimation of the bank lending
channel. This model in which we use the central bank rate will hereafter be referred
as model I. In columns 3 and 4, the target rate that is derived from the backward-
looking rule is the monetary policy indicator (model II), whereas in model III, which
is presented in columns 5 and 6, the target rate from the Taylor–type rule is selected
as the appropriate indicator. Finally, in columns 7 and 8 we use the interest rate from
the forward-looking rule (model IV).
The coefficients of the monetary policy indicator, showing the long-run effects of
the decisions of monetary policy on lending, have the expected negative sign in all
four models as it is obvious from the first line in Table 5 that they are statistically
significant. This suggests the reduction in loan growth as a result of the increase in
interest rates, either in the case of ECB or in the case of the target rate that is set by
the different types of rules. What should be noted at this point, though, is the
differential responses of the loan growth to the one percentage increase of the
different interest rates. Hence, if we consider the absolute values of the coefficients,
it is apparent that the coefficient of the target rate, derived from any of the three
types of rules, is larger than the coefficient of the central bank rate. Additionally,
when the target rate of the forward-looking rule is used as the policy indicator, its
coefficient is the largest one among all models under study, indicating that the bank
lending channel is stronger in this case. The second and third lines in Table 5 show

Table 5 The bank lending results for Eurogroup

Dependent variable: annual Monetary policy Monetary policy Monetary policy Monetary policy
growth rate of lending indicator: CB indicator: indicator: Taylor indicator:
rate (I) backward rule (II) rule (III) Forward rule (IV)

Coef. Prob Coef. Prob Coef. Prob Coef. Prob

Monetary policy coef.

Δrkt-1 −0.0065 0.0978 −1.7479 0.1023 −4.5315 0.0244 −10.305 0.0019
GDPgrowth
ΔGDPkt-1 0.1255 0.0000 0.1073 0.0001 0.2054 0.0004 0.5733 0.0002
Inflation
πkt-1 0.4654 0.5671 1.5589 0.0813 −1.6357 0.2164 −1.4503 0.4584
12 N. Apergis, E. Alevizopoulou

Table 6 The bank lending results for Denmark

Dependent variable: annual Monetary policy Monetary policy Monetary policy Monetary policy
growth rate of lending indicator: CB indicator: indicator: Taylor indicator:
rate (I) backward rule (II) rule (III) Forward rule (IV)

Coef. Prob Coef. Prob Coef. Prob Coef. Prob

Monetary policy coef.

Δrkt-1 −0.0195 0.0000 −1.2572 0.0000 −14.3559 0.0000 −5.2648 0.0000
GDPgrowth
ΔGDPkt-1 0.5763 0.0000 0.6298 0.0000 0.3776 0.0000 0.4463 0.0000
Inflation
πkt-1 0.2084 0.0000 0.3319 0.0000 0.6294 0.0000 0.5041 0.0000

the coefficients and their corresponding p-values for GDP growth and inflation rate,
respectively, for the four models. The coefficients of GDP growth are positive and
statistically significant in all models, but the coefficient of the inflation rate is
positive and statistically significant only in the case of the backward rule.
The results for the bank lending channel in Denmark are reported in Table 6. As
before, the entries of Table 6 indicate the coefficients and their respective p-values of
the variables in Eq. (16), again with the difference in the monetary policy indicator.
Therefore, columns 1 and 2 represent model (I), in which the rate of Danmarks
Nationalbank is used, columns 3 and 4 show the results of model (II), in which the
target rate of the backward-looking rule is selected as the policy indicator. Model
(III), with the Taylor-type rule rate as the indicator, covers columns 5 and 6, whereas
the forward-looking rule is applied in model (IV), the results of which are the entries
in columns 7 and 8. The operation of the bank lending channel is apparent,
independently of the policy indicator used, but it is weaker when the central bank
rate is selected as the policy indicator. Unlike the previous results, the rate derived
from the Taylor-type rule gives the largest coefficient in absolute values, with the

Table 7 The bank lending results for the United Kingdom

Dependent variable: annual Monetary policy Monetary policy Monetary policy Monetary policy
growth rate of lending indicator: CB indicator: indicator: Taylor indicator:
rate (I) backward rule (II) rule (III) Forward rule (IV)

Coef. Prob Coef. Prob Coef. Prob Coef. Prob

Monetary policy coef.

Δrkt-1 −0.16731 0.0000 −1.4423 0.0539 −3.4616 0.0000 −3.8001 0.0000
GDPgrowth
ΔGDPkt-1 0.2969 0.0000 0.3973 0.0000 0.2764 0.0000 0.3530 0.0000
Inflation
πkt-1 0.5643 0.0000 0.9081 0.0000 0.5641 0.0000 0.2619 0.0000
Monetary Policy Rules 13

forward-looking rate following. In Denmark, both the coefficients of GDP growth

and the inflation rate are positive and statistically significant in all four models.
Finally, Table 7 presents the results for the estimation of the bank lending channel
for the U.K., with the four models indicating the different rates applied as the
monetary policy indicator. As in the case of Euro-group, model (IV), in which the
target rate is derived from the forward-looking rule, has the largest coefficient in
absolute terms. The coefficient of the target rate derived from any rule is greater than
its counterpart of the central bank rate, indicating the stronger operation of the bank
lending channel. The coefficients of the inflation rate and GDP growth are again
positive and statistically significant.
Overall, the empirical analysis indicates that the bank lending channel operates
better if interest rate rules are taken into consideration for the estimation of the
channel and, especially, if the target rate that is derived from the forward-looking
rule is used as a monetary policy indicator. Central banks should take that into
account so that their decisions can be more effective in terms of having an impact on
real economy.

Conclusions

Interest rate rules seem to be an interesting aspect of monetary policy, since they can
be considered a convenient way to investigate the behavior of central banks. Equally
important is the bank lending channel, the operation of which accommodates the
passing through of monetary authorities’ decisions to the economy by altering the
loan supply by banks.
In this paper, three types of rules were estimated, the difference of which lies on
the timing of data, i.e., using lagged data for the backward-looking rule, current for
the Taylor-type rule, and forecasts for the forward-looking rule. This procedure
was performed to examine the interest rate rules for three economies: the Euro-
group, which consists of European countries with common currency; Denmark;
and the U.K. over the period 1999–2009. The estimates from this part of the
analysis are used in the second part, in which the presence of the bank lending
channel in the same economies is investigated under four scenarios concerning the
interest rate used as a monetary policy indicator. These alternatives refer to the
central bank rate and the target rates derived from the backward-looking, the
Taylor-type, and the forward-looking rule.
The results indicate that the lending channel is apparent in all cases, but what
differs is the degree of responsiveness of loan growth to a change in the monetary
policy indicator. Thus, the bank lending channel seems to be stronger when target
rates are applied as indicators rather than when the central bank rates are used
directly from data instead and, specifically, when the target rate is derived from the
forward-looking rules—in this case it seems to cause the greatest response of loan
supply. This may have useful implications for the effectiveness of monetary policy,
since it seems easier for the monetary authorities to pump out liquidity into the
economy, when the target rate from the rules and, especially, the forward-looking
rule is used as the monetary policy indicator. A possible explanation is that this type
of rule incorporates inflationary expectations that seem to affect the decisions for the
14 N. Apergis, E. Alevizopoulou

target rate and, hence, monetary policy. The latter seems to highly guide, through its
actions and announcements, the private sector’s expectation, which in this case is
banks. Therefore, banks are responsible to alter their supply of loans according to the
rules, making monetary policy decisions more effective.

Acknowledgement The authors wish to thank the participants of the Topics in Macroeconomics session
at the 71st Atlantic Economic Association meetings in Athens, March 2011, for their valuable comments
and suggestions on an earlier draft of this paper. Needless to say, the usual disclaimer applies.

References

Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an
application to employment equations. Review of Economic Studies, 58, 277–297.
Brissimis, S. N., Kamberoglou, N. C. & Simigiannis, G. T. (2001). Is there a bank lending channel of
monetary policy in Greece? Evidence from bank level data. European Central Bank, Working Paper
Series, Working Paper No. 104.
Clarida, R., Gali, J., & Gertler, M. (1998). Monetary policy rules in practice: Some international evidence.
European Economic Review, 42, 1033–1067.
Clarida, R., Gali, J., & Gertler, M. (2000). Monetary policy rules and macroeconomic stability: Evidence
and some theory. Quarterly Journal of Economics, 115, 147–180.
Fourcans, A., & Vranceanu, R. (2004). The ECB interest rate rule under the Duisenberg presidency.
European Journal of Political Economy, 20, 579–595.
Gambacorta, L. (2005). Inside the bank lending channel. European Economic Review, 49, 1737–1759.
Golodniuk, I. (2006). Evidence on the bank lending channel in Ukraine. Research in International
Business and Finance, 20, 180–199.
Greenspan A. (1997). Rules versus discretionary monetary policy. Remarks at the 15th Anniversary
Conference of the Center for Economic Policy Research at Stanford University. Stanford, CA.
Haldane, A., & Batini, N. (1999). Forward-looking rules for monetary policy. In J. B. Taylor (Ed.),
Monetary policy rules. Chicago: University of Chicago Press.
Hernando, I. and Pages, J. M. (2001). Is there a bank lending channel of monetary policy in Spain?.
European Central Bank, Working Paper Series, Working Paper No. 99.
Kashyap, A. K., Stein, J. C., & Wilcox, D. W. (1993). Monetary policy and credit conditions: Evidence
from the composition of external finance. American Economic Review, 83, 78–98.
Kishan, R. P., & Opiela, T. P. (2000). Bank size, bank capital and the bank lending channel. Journal of
Money, Credit, and Banking, 32, 121–141.
Kozichi, N. (1999). How useful are Taylor rules for monetary policy?. Economic Review, Federal Reserve
Bank of Kansas, 5–33.
Meyer, L. H. (2002). Rules and discretion: Remarks at the Owen Graduate School of Management.
Nashville: Vanderbilt University.
Oliner, S. D., & Rudenbusch, G. D. (1995). Is there a bank lending channel for monetary policy? Federal
Reserve Bank of San Francisco. Economic Review, 2, 1–22.
Orphanides, A. (2003). Historical monetary policy analysis and the Taylor rule. Journal of Monetary
Economics, 50, 983–1022.
Peek, J. & Rosengren, E. S. (1995). Bank lending and the transmission of monetary policy. In: J. Peek &
E. S. Rosengren (Eds.), Is bank lending important for the transmission of monetary policy?, Federal
Reserve Bank of Boston Conference Series, No 39, 47–68.
Taylor, J. B. (1993). Discretion versus policy rules in practice. Carnegie-Rochester Conference Series on
Public Policy, 39, 195–214.
Taylor, J. B. (1999). The robustness and efficiency of monetary policy rules as guidelines for interest rate
setting by the European central bank. Journal of Monetary Economics, 43, 655–679.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.