NATIONAL BANK OF POLAND
W O R K I N G PA P E R
No. 105
The interdependences of central bank’s forecasts
and economic agents inflation expectations.
Empirical study
Magdalena Szyszko
Warsaw 2011
Magdalena Szyszko – Wyższa Szkoła Bankowa w Poznaniu, Banking and Financial Market Department,
e-mail: [email protected]
This research project was conducted under the NBP Economic Research Committee’s open competition for research projects
to be carried out by the NBP staff and economists from outside the NBP and was financed by the National Bank of Poland.
Design:
Oliwka s.c.
Layout and print:
NBP Printshop
Published by:
National Bank of Poland
Education and Publishing Department
00-919 Warszawa, 11/21 Świętokrzyska Street
phone: +48 22 653 23 35, fax +48 22 653 13 21
© Copyright by the National Bank of Poland, 2011
http://www.nbp.pl
Contents
Contents
1. Introduction
3
6
2. Modern monetary policy and inflation expectations
5
9
3. The idea of inflation forecast targeting
8 12
4. Inflation forecast targeting in practice
15 19
5. Empirical study on the forecast and inflation expectations
25 29
5.1 Inflation forecast results
25 29
5.2 Inflation expectations
27 31
5.3 Versions of the analysis
30 34
5.4 Measures applied
34 38
5.5 Results and comments
35 39
6. How the results can be interpreted
45 49
7. Conclusion
4953
WORKING PAPER No. 105
2
3
List of tables, diagrams and charts
List of tables:
1. Inflation forecasting in Czech Republic, Hungary, Poland and Romania
16 20
2. Inflation forecast targeting implementation
19 23
3. Data availability and time extent
29 33
4. Versions of association tested for balances of answers
31 35
5. Versions of association tested for level and directions of expectations
33 37
6. The associations between the forecasts results and the balances of the answers
36 40
7. The associations between policy path and balances for the CNB
38 41
8. The Spearman rank and Kendall tau correlation coefficients
39 43
9. Chi-square test for independence - the Czech Republic
41 45
10. Chi-square test for independence - Hungary
42 46
11. Chi-square test for independence - Poland
43 47
List of diagrams:
1. The associations presented quarterly
32 36
2. Statistics and range of associations tested
35 39
List of charts:
37 41
1. Central path and balances for the Czech Republic
4
52
N a t i o n a l
B a n k
o f
P o l a n d
Magdalena Szyszko
Abstract
The Interdependences of Central Bank’s Forecasts
Magdalena Szyszkoand Economic Agents Inflation Expectations.
Empirical study
The Interdependences of Central Bank’s Forecasts
and
Economic
Agents
Inflation
Key words: inflation
forecasts,
inflation
forecast
targeting,Expectations.
inflation expectations
Empirical study
JEL: E52, E58
Abstract
Key words: inflation forecasts, inflation forecast targeting, inflation expectations
JEL: E52, E58
This paper focuses on the associations between the inflation forecasts of the central
bank
and inflation expectations of the households. The first part is of a descriptive nature. It
Abstract
gives the theoretical background of modern monetary policy focusing on the role of
expectations.
It alsofocuses
presents
of inflation
forecast
Then the of
framework
of
This paper
onthe
theidea
associations
between
thetargeting.
inflation forecasts
the central
the
forecast
targeting inoffour
theThe
Czech
Poland
andIt
bankinflation
and inflation
expectations
the countries:
households.
firstRepublic,
part is of aHungary,
descriptive
nature.
Romania
is presented.
The empirical
part of monetary
the study policy
is an attempt
to find
associations
gives the theoretical
background
of modern
focusing
on the
role of
Magdalena Szyszko
between
the inflation
forecaststheresults
and
inflation
expectations
of consumers
derived onof
the
expectations.
It also presents
idea of
inflation
forecast
targeting.
Then the framework
basis
of surveys.
The targeting
theory gives
sound
background
for theRepublic,
existenceHungary,
of such relationships.
the inflation
forecast
in four
countries:
the Czech
Poland and
The Interdependences of Central Bank’s Forecasts
The
interdependences
in several
The last
of thetopaper
focuses on the
Romania
is presented. are
Thetested
empirical
part ofways.
the study
is anpart
attempt
find associations
and
Economic
Inflation
Expectations.
results and
between
theconclusions.
inflation
forecasts
resultsAgents
and inflation
expectations
of consumers derived on the
Empirical
study
basis of surveys. The theory gives sound
background
for the existence of such relationships.
The interdependences are tested in several ways. The last part of the paper focuses on the
I would
likeconclusions.
to
thank an
anonymous
Referee
and the
participants
of the
seminar held at the
results
and
Key
words:
inflation
forecasts,
inflation
forecast
targeting,
inflation
expectations
NBP
in September
for helpful comments and suggestions that contributed to this paper and
JEL: E52,
E58
my further research.
IAbstract
would like to thank an anonymous Referee and the participants of the seminar held at the
NBP in September for helpful comments and suggestions that contributed to this paper and
my further
Thisresearch.
paper focuses on the associations between the inflation forecasts of the central
bank and inflation expectations of the households. The first part is of a descriptive nature. It
gives the theoretical background of modern monetary policy focusing on the role of
expectations. It also presents the idea of inflation forecast targeting. Then the framework of
the inflation forecast targeting in four countries: the Czech Republic, Hungary, Poland and
Romania is presented. The empirical part of the study is an attempt to find associations
between the inflation forecasts results and inflation expectations of consumers derived on the
1
basis of surveys. The theory gives sound background for the existence of such relationships.
The interdependences are tested in several ways. The last part of the paper focuses on the
results and conclusions.
WORKING PAPER No. 105
1
5
Introduction
1. Introduction
1. Introduction
According to the broadly accepted consensus (so-called New Neoclassical Synthesis,
1
NNS), shared
by central
academics,
modern
monetary
should have
a
According
to the bankers
broadly and
accepted
consensus
(so-called
Newpolicy
Neoclassical
Synthesis,
forward-looking
The fact
today’s situation
is decisivepolicy
for future
results
NNS), shared by character.
central bankers
andthat
academics,
modern monetary
should
haveof
a the
economy,
including
price development,
inflation
expectations
and inflation
forward-looking
character.
The fact that implies
today’s that
situation
is decisive
for future
results of the
forecasts
an important
role in monetary
policy.
importance
of forward-looking
economy,play
including
price development,
implies
that The
inflation
expectations
and inflation
components
monetary
policy
results
also from
theThe
factimportance
that monetary
authorities can affect
forecasts
playinan
important
role in
monetary
policy.
of forward-looking
economy
with
components
in lags.
monetary policy results also from the fact that monetary authorities can affect
There
a few discussion areas on the role of inflation expectations and forecasts in
economy
withare
lags.
monetary
policy.
of them show
a close
between
these variables.
First ofin
There
are Some
a few discussion
areas
on therelationship
role of inflation
expectations
and forecasts
all,
inflation
expectations
influence
therelationship
behavior ofbetween
economic
agents
(pricing
monetary
policy.
Some of should
them show
a close
these
variables.
First of
decisions),
inflationshould
forecasts
or policy
announcement
itself
should
have an impact
all,
inflationwhile
expectations
influence
thepath
behavior
of economic
agents
(pricing
on
expectations.
latter forecasts
will be evaluated
in this paper.
inflation
decisions),
whileThe
inflation
or policyempirically
path announcement
itselfSecondly,
should have
an impact
expectations
and The
inflation
being theempirically
most important
variables
in a
on expectations.
latterforecasts,
will be evaluated
in thisinformation
paper. Secondly,
inflation
forward-looking
a decision-making
process of
the monetary
policy in a
expectations and analysis,
inflation support
forecasts,
being the most important
information
variables
committees.
The analysis,
expectations
are aa reliable
predictor process
of futureofinflation.
The forecast
forward-looking
support
decision-making
the monetary
policy shows
the most probable
developmentare
ofathe
economy
consistent
withinflation.
the central
bank’s
view
on the
committees.
The expectations
reliable
predictor
of future
The
forecast
shows
economy
structuredevelopment
and the transmission
mechanism.
Thirdly,
authorities
wantontothe
the
most probable
of the economy
consistent
withmonetary
the central
bank’s view
influence
inflation expectations
and inflation
forecasts
as this helps
to govern
price want to
economy structure
and the transmission
mechanism.
Thirdly,
monetary
authorities
development.
influence inflation expectations and inflation forecasts as this helps to govern price
Therefore, the importance of inflation expectations and inflation forecasts of central
development.
banks isTherefore,
not questionable.
Several of
authors
focus
on this issue
theoretical
and
the importance
inflation
expectations
andshowing
inflationitsforecasts
of central
empirical
aspects.
Some ofSeveral
them focus
on the
nature
of inflation
expectations
(Forsells,
banks
is not
questionable.
authors
focus
on this
issue showing
its theoretical
and
Kenny 2004)
and their
in inflation
dynamics
alsoofininflation
New Member
States (Gábriel
empirical
aspects.
Somerole
of them
focus on
the nature
expectations
(Forsells,2009,
Benkovskis
Kenny 2004)2008).
and their role in inflation dynamics also in New Member States (Gábriel 2009,
In this
paper, inflation expectations and inflation forecasts are analyzed jointly. One
Benkovskis
2008).
argues that
inflation
to help and
to shape
inflation
expectation
of thejointly.
economic
In this
paper,forecasts
inflation ought
expectations
inflation
forecasts
are analyzed
One
agent. On
other hand,
inflation
serveinflation
as an input
into the inflation
forecast.
argues
thatthe
inflation
forecasts
oughtexpectations
to help to shape
expectation
of the economic
The paper
contributes
to ainflation
broader expectations
analysis of the
forward-looking
attitude
of theseforecast.
central
agent.
On the
other hand,
serve
as an input into
the inflation
banks
which
implement
directanalysis
inflationoftargeting
(DIT) as wellattitude
as inflation
forecast
The paper
contributes
toboth
a broader
the forward-looking
of these
central
targeting
(IFT).
banks which
implement both direct inflation targeting (DIT) as well as inflation forecast
The
paper also concerns the discussion on transparency of monetary policy which is
targeting
(IFT).
an immanent
part of
DIT
strategy.
forecasts’
should
increase policy
the monetary
The paper
also
concerns
theThe
discussion
ondisclosure
transparency
of monetary
which is
6
an immanent part of DIT strategy. The forecasts’ disclosure should increase the monetary
N a t i o n a l
B a n k
o f
P o l a
3
3
n d
agent. On the other hand, inflation expectations serve as an input into the inflation forecast.
The paper contributes to a broader analysis of the forward-looking attitude of these central
Introduction
banks which implement both direct inflation targeting (DIT) as well as inflation forecast
targeting
(IFT).
policy
predictability,
which limits the volatility of the interest rates. It should also anchor
paper alsomore
concerns
the discussion
transparency
of monetary
policy which is
inflationThe
expectations
precisely
- foster theoncentral
bank’s control
over markets’
an immanent part
of DITthere
strategy.
The forecasts’
should
the monetary
expectations.
However,
the optimal
level of disclosure
transparency
is notincrease
unambiguously
policy predictability,
which
limits
the interest
It should
alsoit anchor
determined.
The research
does
not the
givevolatility
a simpleofanswer
to therates.
question
whether
is sufficient
3 should
inflation
more
precisely
- fosterpath
the
central also
bank’s
over markets’
be control
revealed.
to
reveal expectations
the forecast, or
whether
the policy
1
expectations.
However,
there
levelbetween
of transparency
not unambiguously
This paper
focuses
on the optimal
relationship
inflationisforecast
and inflation
determined.
research does
not give
simple
to theconducted.
question whether
it is sufficient
expectationsThe
of households
measured
onathe
basisanswer
of surveys
Two hypotheses
are
to reveal the
forecast,
orassumes
whether that
the policy
path should alsobetween
be revealed.
presented.
The
first one
the interdependences
the inflation forecast
Thisconsumers
paper focuses
on the
relationshipexist.
between
inflation
forecast and
and qualitative
inflation
results and
inflation
expectations
There
are theoretical
expectations
of households
onhypothesis
the basis ofimplies
surveys
conducted.
Twoof
hypotheses
are
existence. measured
The second
that
the existence
the
premises
of such
presented.
The
first one
the interdependences
between thedepends
inflationneither
forecast
associations
between
theassumes
inflationthat
forecasts
and inflation expectations
on
and qualitative
results
consumers
exist. There
are theoretical
detailedand
solutions
in theinflation
field ofexpectations
forecasting inflation
applied
by central banks
nor on their
premisestowards
of suchinflation
existence.
The second
hypothesis
implies that the existence of the
attitude
forecast
targeting
implementation.
associations
between the
inflation
and inflation
expectations
dependsthe
neither
The territorial
scope
of the forecasts
research covers
four European
economies:
Czechon
detailed
in Poland
the fieldand
of Romania.
forecastingAll
inflation
by central
nor on their
Republic,solutions
Hungary,
of themapplied
are countries
withbanks
derogation,
attitude
towards
inflation
forecast
targeting
implementation.
implementing
DIT
strategy
and using
interest
rate as the main monetary policy instrument on
four European
economies:
theforecast
Czech and
The territorialand
scope
of the research
covers
the macroeconomic
operational
level. Their
central
banks produce
inflation
Republic,
and Romania.
All ofFour
themcentral
are countries
with derogation,
use
it as anHungary,
input in aPoland
decision-making
process.
banks covered
by the study claim
implementing
and using interest
ratethe
as acceptance
the main monetary
policy
instrument on
It implies
of inflation
expectations’
that
they acceptDIT
thestrategy
NNS frameworks.
the
macroeconomic
and operational level.
Their
banks
produce
inflation
forecast and
importance
and the forward-looking
character
ofcentral
inflation.
They
also use
inflation
use
it as an input
in a decision-making
Four central banks
covered byinthe
study
claim
expectations
in formalized
forecasting process. Nevertheless,
the differences
these
central
they
accept are
the substantial
NNS frameworks.
impliesthe
thecharacter
acceptance
of inflation
that
banks
practices
and theyItimpose
of the
research:expectations’
a case study,
importance
and cases
the forward-looking
character precedes
of inflation.
They also use
inflation
limited
in some
by the data availability,
a comparative
analysis.
expectations
in formalized
process.
Nevertheless,
theforecasts
differences
in these
central
The central
banks inforecasting
question started
to publish
inflation
between
2001
and
banks
practices
are substantial
they impose
character
of the
a case study,
2005. For
each central
bank theand
research
startingthe
point
is the date
ofresearch:
the first forecast
precedes
a comparative
analysis.
limited
in1some
by the data
availability,
. The cases
examination
period
finishes with
the end
of 2010.
relevance
centralisbanks
in question
started
to publish
inflation
forecasts
between
2001of
and
The paper
organized
as follows:
Section
2 focuses
on the
role and
importance
2005. Forexpectations
each centralinbank
the research
starting
is the
date ofthe
theidea
firstof
forecast
inflation
inflation
modern
monetary
policy;point
Section
3 shows
1
. The examination
period
the end
2010.
relevance
forecast targeting;
the fourth part
of finishes
the paperwith
presents
theofbackground
of inflation forecasting
Thecountries
paper is organized
follows:
SectionSection
2 focuses
on the role
importanceof
ofthe
in the four
covered byasthe
examination;
5 describes
theand
methodology
Section
shows the idea
of inflation
inflation and
expectations
modern monetary
research
the resultsinthemselves;
Sectionpolicy;
6 presents
the 3interpretation
of the
empirical
forecastand
targeting;
fourth
part
of the paper presents the background of inflation forecasting
results
the finalthe
part
draws
conclusions.
in the four countries covered by the examination; Section 5 describes the methodology of the
1
The Czech National Bank (CNB) and the National Bank of Hungary (NBH) published the forecast for the first
research
andthethe
resultsBank
themselves;
Section
presents
interpretation
of the (NBR)
empirical
time in 2001,
National
of Poland (NBP)
- in62004
and thethe
National
Bank of Romania
- in 2005.
results and the final part draws conclusions.
4
1
The Czech
National
WORKING
PAPER No.
105
Bank (CNB) and the National Bank of Hungary (NBH) published the forecast for the first
time in 2001, the National Bank of Poland (NBP) - in 2004 and the National Bank of Romania (NBR) - in 2005.
7
relevance . The examination period finishes with the end of 2010.
The paper is organized as follows: Section 2 focuses on the role and importance Introduction
of
inflation expectations in modern monetary policy; Section 3 shows the idea of inflation
forecast targeting; the fourth part of the paper presents the background of inflation forecasting
in the four countries covered by the examination; Section 5 describes the methodology of the
research and the results themselves; Section 6 presents the interpretation of the empirical
results and the final part draws conclusions.
1
1
The Czech National Bank (CNB) and the National Bank of Hungary (NBH) published the forecast for the first
time in 2001, the National Bank of Poland (NBP) - in 2004 and the National Bank of Romania (NBR) - in 2005.
4
8
N a t i o n a l
B a n k
o f
P o l a n d
Modern monetary policy and inflation expectations
2. Modern monetary policy and inflation expectations
The New Neoclassical Synthesis is a broadly accepted consensus on monetary policy
and its impact on economy. It integrates the classical elements with Keynesian approach
(imperfect competition and nominal rigidities) and Real Business Cycles models. The NNS
frameworks were described in literature (Mankiv 1990, Goodfriend, King 1997, Galí 2002).
2
The conclusions on monetary policy impact on the economy are as follows (Goodfriend, King
1997):
-
monetary policy actions can have an important influence on the real economy,
which can persist over several years due to gradual adjustment of individual prices
and general price level; this monetary nonneutrality is caused by the presence of
rigidities in the economy,
-
there is little long-run trade-off between inflation and real activity, even if price
adjustment is costly,
-
eliminating inflation brings significant gains because of increased transaction
efficiency and elimination of relative price distortion,
-
credibility plays an important role in understanding monetary policy.
Inflation dynamics is directly connected with modern transmission mechanism
perception. The NNS models emphasize the forward-looking nature of inflation. This
property is the result of a price setting model which is in line with Calvo model. The model
assumes that the prices cannot be reset continuously and price revision is non-synchronous.
Each price setter is allowed to change its price at the random chosen moment. The probability
of this change in period t is independent of the moment of the last price change and it is also
independent across the firms. A firm knows that the change of price may not be possible in
the following period. That is why a company is assumed to set its price taking into account
the expected average price and the state of the market (excess demand) in the future (Calvo
1983). Firms re-setting their prices nowadays recognize that the prices they choose are likely
to remain effective for more than one period. Such firms will find it optimal, when making
their current pricing decisions, to take into consideration their expectations regarding future
cost and demand conditions. Since changes in the aggregate price level are (by definition) a
consequence of current pricing decisions inflation must be an important forward-looking
component. That property appears to be clearly reflected in the so-called New Keynesian
Phillips curve (Galí 2002), formally the New Phillips Curve, which is presented by the
inflation equation given below:
WORKING PAPER No. 105
5
9
Modern monetary policy and inflation expectations
t = Et{t+1}+ kxt + t
where:
t – current inflation figure,
Et{t+1}- inflation expected for the next period,
xt – output gap,
t – random disturbance.
2
The output gap is directly connected with the marginal costs in economy2. The
average markup is the ratio of price level to marginal cost for an average firm. It depends on
the productivity and the cost of labor and materials. Because of the fact that price adjustment
is costly and is not possible at any time, firms consider changing their product prices only if
demand and cost conditions threaten to compress or elevate actual markups significantly and
persistently with reference to flexible-price profit-maximizing levels. Firms consider
increasing products prices if the marginal cost moves above trend because labor productivity
falls below its trend, or if strong aggregate demand increases the intensity of resource
utilization and thereby causes wage or materials costs to rise relative to trend. Conversely,
firms consider cutting product prices if labor productivity rises relative to trend, or if weak
aggregate demand relaxes resource utilization and thereby causes wage or materials costs to
fall relative to trend. To sustain low inflation, monetary policy must manage aggregate
demand while taking aggregate productivity into account, so that marginal cost rises at the
targeted rate of inflation — then firms will raise product prices at the targeted rate of inflation
because they are confident that doing so will keep actual markups at flexible-price profitmaximizing markups (Goodfriend 2007).
The inflation equation shows also the importance of inflation expectations. In a
standard transmission model an aggregate demand channel is accompanied with the
expectations channel that allows the central banker to affect inflation expectation, which, in
turn, affects inflation both directly and indirectly (via wage and price-setting behavior) as
inflation equation shows (Svensson 1998). Knowing that price readjusting may not be
possible in the next period, firms are taking into account not only the current level of
aggregate demand but also the expected one. It brings the necessity to shape market
expectations in the way in which interest rates, inflation and income for the following periods
2
The output gap concept in NNS models differs form traditional approach. Output gap is proportional to
deviations of the real marginal cost form steady state. In other words, it is the deviation of output form its
equilibrium level in the absence of nominal rigidities.
10
6
N a t i o n a l
B a n k
o f
P o l a n d
Modern monetary policy and inflation expectations
will be consistent with the central bank goals. If it is possible for a central bank to affect
expectations, this should be an important tool of stabilization policy. The fact that prices may
remain unchanged for some time naturally implies that firms must be forward-looking,
assessing not just the current economic environment but also the outlook for the future. Not
only expectations about policy matters, but very little else matters. The ability of a central
bank to influence expenditure, and hence pricing decisions, is crucially dependent upon its
2
ability to influence market expectations regarding the future path of overnight interest rate,
and not merely their current level (Woodford 2003).
Regarding this theoretical background, the modern monetary policy strategy focuses
on expectations stabilization. One potential benefit from a successful implementation of
inflation targeting is the anchoring of expectations with its stabilizing effect on
macroeconomic activity. Failing to anchor expectations might result in undesired fluctuations
and economic instability (Eusepi, Preston, 2007). Well-anchored expectations help obviously
in achieving inflation target with less volatility in the real sphere.
The short overview on modern monetary policy and its role in the economy shows that
the expectation channel is the most important transmission channel. Simultaneously, it is the
most difficult to be modeled formally. The first, and the most important question is how to
support shaping private sector expectations. Policy commitment and proper communication
with the market, including inflation forecast publishing, are a possible choice in this field.
The fact that the frameworks of monetary policy are being rethought nowadays is
worth noticing. The NNS does not close the discussion on monetary policy. The associations
of the inflation forecast and expectations may be influenced by drawbacks of this framework,
which is described in the last section of this paper.
WORKING PAPER No. 105
7
11
The idea of inflation forecast targeting
3. The idea of inflation forecast targeting
The inflation forecast is an immanent part of a fully-fledged direct inflation targeting
strategy (DIT). A central bank produces its own forecasts, uses them in decision-making
process and reveals them. Inflation targeting, which is perceived to be the best monetary
policy strategy by the consensus of NNS, may become inflation forecast targeting (IFT),
where inflation forecast is an intermediate objective of monetary policy. One argues that
subscribing the function of an intermediate target to the inflation forecast simplifies
implementing and monitoring monetary policy (Svensson 1996). The idea of IFT has been
3
broadly described in the literature. The main threads of this discussion are presented below.
Inflation forecast targeting is a simple rule of monetary policy. The central bank’s
inflation forecast for the period of inflation forecast targeting horizon becomes an
intermediate target. Hence the instrument should be set so as to make the inflation forecast
equal to the inflation target. If the inflation target is above (below) the target, the main rate of
the monetary policy should be raised (lowered). Following this rule is claimed to be the best
central bank’s practice. Ex post inflation may differ from the targeted level because of
forecast errors (Svensson 1996). The decision-making procedure of IFT is repeated by the
monetary committee at any decision point. The forecasts and interest rate level are brought up
to date if it is necessary. This is why IFT becomes a dynamic optimization procedure.
Two version of IFT may be discussed. The first one assumes that inflation forecast
targeting horizon is equal to the inflation control lag (the shortest period when a central bank
can affect inflation). Only one choice will be optimal for the monetary policy if the central
bank has only one argument in its loss function (inflation). Within such a strict DIT strategy,
the central bank in period t should choose a sequence of current and future main rates so as to
minimize the loss function (the expected squared deviations of the future inflation rate form
the inflation target). Instead of minimizing the expected squared deviations of the future
inflation rate from the inflation target, the central bank can minimize the squared deviation of
the current forecast within the forecast-targeting horizon from the inflation target (Svensson
1996).
The other version occurs when inflation forecast targeting horizon is longer that the
inflation control lag. Right now the policy path is not unambiguously determined. The policy
can become more contradictory (or lax) sooner or later. This is why the additional restriction
is put on the policy – constant interest rate condition. The interest rate is now set at the rate
which, on the assumption that it is kept constant throughout the forecast-targeting horizon,
12
8
N a t i o n a l
B a n k
o f
P o l a n d
The other version occurs when inflation forecast targeting horizon is longer that the
inflation control lag. Right now the policy path is not unambiguously determined. The policy
The idea of inflation forecast targeting
can become more contradictory (or lax) sooner or later. This is why the additional restriction
is put on the policy – constant interest rate condition. The interest rate is now set at the rate
which, on the assumption that it is kept constant throughout the forecast-targeting horizon,
will ensure that the inflation forecast is on target. If the inflation forecast at the forecasttargeting horizon is not on target, assuming that the prevailing interest rate level is given, the
8
interest rate is exactly adjusted to correct this (Leitemo 2006). Under forecast targeting, the
central bank first constructs conditional inflation, output-gap, and interest-rate forecasts
corresponding to alternative feasible policies, and then chooses the preferred scenario
according to the specified loss function (Svensson, Woodford 2002). This procedure assumes
that the central bank has more than one goal, however, price stability can be a priority.
The central bank inflation forecast can be perceived as a quite good intermediate
3
target. It is by definition a current variable which is the most correlated with the final goal, it
is more controllable than the final goal, and it can be made more observable. It can also be
made very transparent and facilitate the central bank’s communication with the public
(Debelle 1997, Svensson 1996). Moreover, inflation forecast integrates a broad set of
historical and current data as well as expectations. It means that one variable – the forecast –
includes the idea of analyzing various data, simultaneously giving the simplicity of following
an intermediate target commitment.
Inflation forecast as an intermediate target may fulfill internal and external functions.
The former is connected with supporting decision procedures of a monetary committee.
Inflation forecast targeting gives a simple rule of monetary policy, which has already been
described. However, the IFT should be treated as the rule of thumb (closed loop policy). The
central bank is not obliged by any legal act to follow this rule. At any decision point it
analyzes the forecast as well as other data. The consequence of following the rule depends on
various factors of mostly qualitative nature which are further presented.
The external functions of the forecast are connected with a growing importance of
transparency and other qualitative aspects of the monetary policy. These factors are in turn
connected with the importance of expectations of economic agents for the effectiveness of the
monetary policy. The main reason why the forecasts are revealed is connected with the
inflation expectation shaping. Central banks can affect economy via expectations of future
policy actions. They can directly influence short-term interest rate and they seek to influence
longer-term rates. Shaping expectations of future policy actions is the best way to achieve it.
While making decisions, economic agents are taking into consideration the whole expected
interest rate path. Publishing inflation forecast may help to guide longer-term expectations.
Inflation forecast can also anchor expectations when the inflation target is temporarily missed.
It can serve as such a temporary anchor, especially in situations where the target is missed
because of shocks that are out of control of the central bank. Anticipated course of inflation,
13
WORKING PAPER No. 105
9
While making decisions, economic agents are taking into consideration the whole expected
The idea of inflation forecast targeting
interest rate path. Publishing inflation forecast may help to guide longer-term expectations.
Inflation forecast can also anchor expectations when the inflation target is temporarily missed.
It can serve as such a temporary anchor, especially in situations where the target is missed
because of shocks that are out of control of the central bank. Anticipated course of inflation,
showed by a credible central bank may limit the expectations’ growth (Skoepa, Kotlán
2003). Publication of the forecast gives also the9 central bank a chance to ex post explanation
and justification of its actions. However, the forward-looking dimension of revealing the
forecasts seems more important.
Nowadays, the pros of publishing forecasts seem to prevail. Formal models, as well as
empirical studies (Chortareas, Stasavage, Sterne 2002) show that publishing forecasts could
improve macroeconomic outcomes (reduction of the inflation bias, lower inflation rate, loss
3
function minimalization). The central bank accepts the importance of expectations in its
policy and finds the forecasts a useful tool of shaping these expectations.
The majority of studies show advantages of revealing the forecasts. However, this
does not mean that there are no drawbacks. A central bank can be bound more tightly by the
publication of an inflation forecast than is actually warranted by the quality of that forecast. In
practice a forecast cannot amount to a complete summary of all the information relevant to
monetary policy decisions. This is mainly because there is no universally optimal forecasting
method. The specification of the macroeconometric model used for forecasting contains a
number of degrees of freedom. Moreover, some assumptions also have to be made regarding
the exogenous variables (Remsperger, Worms 1999). Another problem is that
macroeconometric models cannot adequately capture structural changes. Recognizing
transmission mechanism and access to the long enough and reliable set of data can also be
problematic.
These problems concern first of all the economies and central banks in transitory
periods. The four countries covered by the examination first abandoned central planning.
Then, having finished some stage of disinflation process and after implementing a traditional
strategy for some time, they adopted the DIT strategy. Finally, they started to produce and
reveal inflation forecasts. These changes, taking place during quite a short period, meant
enhanced structural changes (entrance to the European Union meant it as well) and new
demands for the monetary policy with adopting new frameworks. This is why, even if the
central banks see and accept the advantages of using inflation forecast in the monetary policy,
they may postpone implementing of IFT. European Central Bank’s example shows, that even
a central bank that implements modern monetary policy strategy may not accept the forecast
as the most valuable input in the analysis.
There are theoretical papers that show that this knowledge transparency might be
socially harmful if the central bank possesses private information about real shocks.
14
N a t i o n a l
B a n k
o f
P o l a n d
central banks see and accept the advantages of using inflation forecast in the monetary policy,
they may postpone implementing of IFT. European Central Bank’s example shows, that even
The idea of inflation forecast targeting
a central bank that implements modern monetary policy strategy may not accept the forecast
as the most valuable input in the analysis.
There are theoretical papers that show that this knowledge transparency might be
socially harmful if the central bank possesses private information about real shocks.
(Gersbach 2003). There are others that show that transparency is beneficial up to a certain
extent (Cecchetti Krause 2002, Jensen 2001).
10
Despite these augments disclosure of inflation
forecast is broadly claimed to be the
accepted level of transparency. Central banks publish their own forecast in different ways
(numerical results, fan chart with probability distribution, risk assessment, description of the
situation). However, at the beginning, usually just after making decision of DIT
3
implementation, they may prefer implicit IFT. It means that the forecast is not revealed as
long as the central bank is not satisfied with its reliability, however, this very forecast is
discussed by the monetary committee during the decision process.
This conclusion does not close the discussion on the desired transparency level. The
term inflation forecast includes unconditional forecast as well as conditional forecast
(projection). The latter implies assumption that the central bank does not change the level of
its interest rate throughout the forecast horizon. The differences between conditional and
unconditional forecasts and their implications for the central bankers open another discussion.
And again, no unambiguous solution is elaborated.
Research on the advantages of conditional or unconditional forecast does not give
simple results, it usually depends on the assumptions that lay beneath the model. Some of it
shows that only the publication of the projections reduces uncertainty about inflation target
and enhances the central bank to built its reputation. In addition, it gives the central bank
greater flexibility to respond to shocks in the economy (Geraats 2001). Other research, on the
other hand, gives contradictory results, showing that unconditional forecasts include all the
information from which the policy target can be inferred (Tarkka, Mayes, 1999).
The choice of the unconditional forecast opens the possibility to disclose the most
likely future evolution of the central bank’s interest rate. It can be done in a descriptive way
(qualitatively) or quantitatively (on the fan chart). There is an empirical analysis that shows
that the communication of future policy intentions, either quantitative or qualitative, improves
the ability of market participants to predict monetary policy decisions (Ferrero, Secchi, 2007).
The central bank can go further and publish also a policy path which is the sequence
of current and expected settings of the policy rate, consistent with achieving its goals.
Formulating and communicating a policy path is the next frontier in monetary policy
transparency (Khan 2007). Neither theory nor formal models show a sound and unambiguous
justification for publishing a policy path. However, several pros of publishing policy path are
connected with the context of shaping expectations.
WORKING PAPER No. 105
15
The central bank can go further and publish also a policy path which is the sequence
of current and expected settings of the policy rate, consistent with achieving
its goals.
The idea of inflation forecast targeting
Formulating and communicating a policy path is the next frontier in monetary policy
transparency (Khan 2007). Neither theory nor formal models show a sound and unambiguous
Firstly, communicating a policy path may help demonstrate central banker’s
justification for publishing a policy path. However, several pros of publishing policy path are
commitment to long-term goals, showing the way in which they can be achieved. If such a
connected with the context of shaping expectations.
commitment helps to built longer-term expectations, temporary shocks on inflation are less
Firstly, communicating a policy path may help demonstrate central banker’s
likely to transform into persistently higher expectations. As a result, the policy path disclosure
commitment to long-term goals, showing the way in which they can be achieved. If such a
helps to enforce optimal policy (Khan 2007, Archer
2005).
11
commitment helps to built longer-term expectations,
temporary shocks on inflation are less
Secondly, the policy path announcement can also contribute to shaping short-term
likely to transform into persistently higher expectations. As a result, the policy path disclosure
expectations. Announcing inflation target should first of all anchor medium and longer term
helps to enforce optimal policy (Khan 2007, Archer 2005).
expectations, which is consistent with the monetary policy horizon. When the inflation is
Secondly, the policy path announcement can also contribute to shaping short-term
above the target, the short-term expectations can be above the central bank desired level. The
expectations. Announcing inflation target should first of all anchor medium and longer term
policy path should anchor them more explicitly than just the forecast of inflation. It limits the
expectations, which is consistent with the monetary policy horizon. When the inflation is
near-term concern about monetary policy actions. It allows the market to price more
above the target, the short-term expectations can be above the central bank desired level. The
efficiently financial assets, both short-term and long-term ones. Announcing the policy path
policy path should anchor them more explicitly than just the forecast of inflation. It limits the
may give the policymakers greater leverage over the long-term interest rates (Khan 2007).
near-term concern about monetary policy actions. It allows the market to price more
Moreover, disclosing a policy path explicitly may enforce the forecasts quality as well
efficiently financial assets, both short-term and long-term ones. Announcing the policy path
as it improves the central bank’s accountability. Without the information on the future path of
may give the policymakers greater leverage over the long-term interest rates (Khan 2007).
the economy only ex post evaluation of the central bank is possible (Mishkin 2004). Due to
Moreover, disclosing a policy path explicitly may enforce the forecasts quality as well
the lags in the monetary policy, the monetary policy committee term may be over when its
as it improves the central bank’s accountability. Without the information on the future path of
actions influence the economy. The arguments in favor of publishing a policy path are also
the economy only ex post evaluation of the central bank is possible (Mishkin 2004). Due to
supported by a theoretical analysis.
the lags in the monetary policy, the monetary policy committee term may be over when its
Despite the arguments that show that the policy path publishing can bring benefits,
actions influence the economy. The arguments in favor of publishing a policy path are also
minority of the central banks – inflation targeters - have decided to do so. There are a few
supported by a theoretical analysis.
reasons why.
Despite the arguments that show that the policy path publishing can bring benefits,
First of all, they do not have such a path (Khan 2007). While the policymakers may
minority of the central banks – inflation targeters - have decided to do so. There are a few
have a sense of direction of the future policy rate changes, in most cases they do not reach
reasons why.
agreement on an explicit policy path. The problem occurs not only in the area of choosing the
First of all, they do not have such a path (Khan 2007). While the policymakers may
monetary policy rule, which can be quite controversial,3 but also because of the inconsistency
have a sense of direction of the future policy rate changes, in most cases they do not reach
of the decision making procedures with giving an explicit policy path for the period longer
agreement on an explicit policy path. The problem occurs not only in the area of choosing the
that the interval between monetary policy committee meetings. It is not possible, and
monetary policy rule, which can be quite controversial,3 but also because of the inconsistency
reasonable, to make binding decision on an interest rates path instead of a one-step change.
of the decision making procedures with giving an explicit policy path for the period longer
There is empirical research that confirms that the forecasts of interest rates had little or no
that the interval between monetary policy committee meetings. It is not possible, and
3
reasonable, to make binding decision on an interest rates path instead of a one-step change.
16
3 Usually the Taylor–type rule is applied as the monetary policy committee reaction function. It has several
There
is empirical
confirms
that the forecasts
of interest
hadwas
little
or no
disadvantages
that are research
described that
in (Taylor
Interest…1999).
Despite these
pitfalls rates
the effort
taken
to make
Taylor-type rules operational. The Taylor-type rules in forward-looking version are broadly applied in central
banks’ formal models. They can be quite good prescriptive tools if the models are well micro-founded.
3 Usually the Taylor–type rule is applied as the monetary policy committee reaction function. It has several
disadvantages that are described in (Taylor Interest…1999). Despite these pitfalls the effort was taken to make
Taylor-type rules operational. The Taylor-type rules in forward-looking version are broadly applied in central
12 tools if theNmodels
banks’ formal models. They can be quite good prescriptive
micro-founded.
a t i oare
n awell
l
B
a n k
o f
P o l a n d
agreement on an explicit policy path. The problem occurs not only in the area of choosing the
monetary policy rule, which can be quite controversial,3 but also because of the inconsistency
The idea of inflation forecast targeting
of the decision making procedures with giving an explicit policy path for the period longer
that the interval between monetary policy committee meetings. It is not possible, and
reasonable, to make binding decision on an interest rates path instead of a one-step change.
There is empirical research that confirms that the forecasts of interest rates had little or no
informational value when the horizon exceeded two quarters (six months), although they were
good
in the
quarterrule
andisreasonable
in the
following
(Goodhart,
Bin
Lim, 2011).
3 Usually
the next
Taylor–type
applied as the
monetary
policyone
committee
reaction
function.
It has several
disadvantages that are described in (Taylor Interest…1999). Despite these pitfalls the effort was taken to make
Moreover, rules
the risk
or uncertainty
temporarily
connected
with the
forecast
as well
as with
Taylor-type
operational.
The Taylor-type
rules in
forward-looking
version
are broadly
applied
in central
banks’ formal models. They can be quite good prescriptive tools if the models are well micro-founded.
policy path can be substantial.
Secondly, the public may perceive a given policy path as a commitment. Market
12 conditions and economic outlook more
participants will adjust their behavior to the new
3
sluggishly. They can feel cheated by the central bank if its subsequent decisions on the
interest rate are different from the announced policy path. From the central bank’s point of
view, the latter may create limited willingness to change a policy path which was once
announced, even if the forecast changes. Such an approach may limit the flexibility of the
central bank (Khan 2007). There is also another problem connected with perceiving a policy
path as a commitment. It can influence credibility negatively if the public does not understand
conditional nature of such a path (Woodford 2005).
A theoretical analysis shows that gains from disclosing the path of the future interest
rate by central banks that have already engaged in publishing macroeconomic projections may
be lower than those achieved after embarking on disclosing macro-projections and
macromodels on the basis of which these projection were made. Weighted against the fears of
revealing the future interest rate path, this may explain the reluctance of certain central banks
to push their transparency framework that far (Brzoza–Brzezina, Kot, 2008).
However, to underline the ambiguity of the results described above, the results of the
other research – empirical one this time - is worth mentioning. The study (Ferrero, Sccheci
2007) shows that even for a very transparent central bank (Reserve Bank of New Zealand),
the publication of the expected interest rate path has a significant impact on market
expectations. Moreover, the change in market interest rates in the period between two
publications of the interest rate path was similar to the revision of the published path, thus
suggesting that market operators well understood the conditionality of the central bank’s
projections.
The discussion on transparency and its optimal level means nowadays deliberations on
central bank’s forecast relevance as well as publishing other information connected with
forecasting procedures (assumptions, model, errors, risk assessment, policy path).
Transparency is beneficial when it serves to simplify communication with the public and
helps generate support for central banks to conduct the monetary policy optimally with an
appropriate focus on long-run objectives. However, it can complicate the communication
process and weaken support for a central bank focus on long-run objectives (Mishkin 2004).
WORKING PAPER No. 105
17
The discussion on transparency and its optimal level means nowadays deliberations on
central bank’s forecast relevance as well as publishing other information connected with
The idea of inflation forecast targeting
forecasting procedures (assumptions, model, errors, risk assessment, policy path).
Transparency is beneficial when it serves to simplify communication with the public and
helps generate support for central banks to conduct the monetary policy optimally with an
appropriate focus on long-run objectives. However, it can complicate the communication
process and weaken support for a central bank focus on long-run objectives (Mishkin 2004).
The optimal level of transparency cannot be easily determined. The theoretical discussion on
13practices in this field are not unique and their
this issue is still in progress. The central banks’
choices are justified differently. However, the transparency level of monetary policy is largely
aligned. The public knows the main goal, instruments and main premises of the decision
making process, including forecasts. Moreover, the monetary authority explains the decision
3
which has just been made. The differences are mainly connected with the intentions of future
actions demonstration.
The countries covered by the examination also implement different solutions in this
field, which is described in the next section.
18
N a t i o n a l
B a n k
o f
P o l a n d
Inflation forecast targeting in practice
4.
4. Inflation
Inflation forecast
forecast targeting
targeting in
in practice
practice
This
This section
section examines
examines practical
practical aspects
aspects of
of inflation
inflation forecasting
forecasting procedures
procedures and
and the
the
related
related institutional
institutional framework
framework in
in four
four central
central banks:
banks: the
the Czech
Czech National
National Bank
Bank (CNB),
(CNB), the
the
National
National Bank
Bank of
of Hungary
Hungary (NBH),
(NBH), the
the National
National Bank
Bank of
of Poland
Poland (NBP)
(NBP) and
and the
the National
National Bank
Bank
of
of Romania
Romania (NBR).
(NBR). They
They all
all declare
declare the
the acceptance
acceptance of
of the
the NNS
NNS frameworks,
frameworks, including
including the
the
importance
of inflation
inflation expectations
expectations and
and the
the role
role of
of forecast
forecast in
in the
the monetary
monetary policy.
policy. Table
Table 11
importance of
presents the
the main
main facts
facts on
on forecasting
forecasting systems
systems and
and procedures
procedures in
in the
the countries
countries covered
covered by
by the
the
presents
study.
study.
The implementation
implementation of
of DIT
DIT strategy
strategy is
is aa precondition
precondition of
of the
the IFT
IFT implementation
implementation44..
The
Direct
Direct inflation
inflation targeting
targeting strategy
strategy can
can be
be implemented
implemented in
in various
various ways.
ways. It
It means
means that
that within
within
4
the
strategy, the
the central
central bank
bank can
can operate
operate in
in different
different
the set
set of
of standard
standard features
features of
of such
such aa strategy,
frameworks and
and implement
implement diverse
diverse solutions
solutions at
at an
an operational
operational level.
level. This
This brings
brings
frameworks
consequences for
for IFT.
IFT. This
This is
is why
why the
the starting
starting point
point for
for the
the analysis
analysis focuses
focuses on
on selected
selected
consequences
aspects of
of monetary
monetary policy
policy strategy.
strategy.
aspects
One remarkable
remarkable feature
feature for
for the
the monetary
monetary policy
policy implementation
One
implementation is
is the
the choice
choice of
of
exchange rate.
rate. However,
However, the
the
exchange
exchange rate
rate regime.
regime. Fully-fledged
Fully-fledged DIT
DIT implies
implies aa floating
floating exchange
intermediate solutions
solutions are
are also
also acceptable.
acceptable. Such
Such aa temporary
temporary situation
situation may
may reveal
reveal
intermediate
inconsistency in
in the
the MPC
MPC behavior.
behavior. If
If the
the central
central bank
bank declares
declares that
that the
the forecast
forecast is
is an
an
inconsistency
important premise
premise in
in aa decision
decision making
making process
process and
and the
the forecast
forecast is
is revealed,
revealed, the
the public
public will
will
important
understand that
that the
understand
the decision
decision on
on interest
interest rates
rates is
is not
not in
in line
line with
with the
the forecast.
forecast. It
It undermines
undermines the
the
declaration
on forecast
forecast importance
importance and
and lowers
lowers its
its significance
significance as
as aa tool
tool that
that helps
helps shaping
shaping
declaration on
inflation
inflation expectations.
expectations. Such
Such an
an ambiguous
ambiguous situation
situation occurred
occurred in
in Hungary
Hungary and
and Romania.
Romania.
In
In Hungary
Hungary DIT
DIT was
was declared
declared to
to be
be the
the official
official strategy
strategy of
of the
the NBH
NBH in
in 2001.
2001.
However,
However, aa fixed
fixed exchange
exchange rate
rate with
with the
the fluctuation
fluctuation band
band of
of 15%
15% around
around the
the central
central parity
parity
an about
about 7-year7-yearwas
was introduced
introduced in
in October
October 2001
2001 and
and abandoned
abandoned in
in February
February 2008.
2008. It
It meant
meant an
period of
of coexistence
coexistence of
of fixed
fixed exchange
exchange rate
rate with
with inflation
inflation targeting.
targeting. Despite
Despite the
the NBH
NBH
period
declaration
declaration of
of DIT
DIT implementation,
implementation, its
its strategy
strategy was
was quite
quite eclectic.
eclectic. It
It had
had numerous
numerous features
features of
of
DIT
DIT (numerically
(numerically set
set inflation
inflation target
target as
as the
the priority,
priority, inflation
inflation forecast,
forecast, independence
independence and
and
accountability
accountability of
of the
the central
central bank,
bank, communication
communication with
with the
the public).
public). At
At the
the same
same time
time the
the
often reacted
reacted to
to the
the exchange
exchange rate
rate of
of forint
forint fluctuations,
fluctuations, which
which meant
meant that
that
Bank
Bank Board
Board quite
quite often
the forecast
forecast was
was neglected.
neglected.
the
4
4
Inflation
Inflation forecast
forecast can
can be
be produced
produced even
even within
within another
another monetary
monetary policy
policy strategy
strategy frameworks
frameworks but
but it
it therefore
therefore
fulfill
different
tasks.
The
role
of
forecast
as
the
intermediate
objective
is
therefore
negligible.
fulfill different tasks. The role of forecast as the intermediate objective is therefore negligible.
WORKING PAPER No. 105
15
15
19
20
N a t i o n a l
- QPM: Quarterly Projection
Model up to May 2008,
- g3 from August 2008
39 forecasts
unconditional from July 2002
policy path disclosure form
2008
first, 6 quarters, then up to 8
quarters
4-6 quarters
4-6quarters
- Analytical Scheme up to the
end of 2003,
- QPM: Quarterly Projection
Model from February 2004,
- DELPHI (Dynamic
Econometric Large-scale
Prognosticator of Hungarian
Inflation) from May 2010
first, 6 quarters, then 8 quarters
conditional (projection)
38 projections
4-6 quarters
- MAPM: model for medium
term analysis and projection
- New Analytical Scheme and
MSMI (Small Structural
Inflation Model) up to
February 2005,
- ECMOD up to February
2008,
- NECMOD from June 2008
8 quarters
conditional (projection)
22 projections
quarterly:
February, May, August,
November
forecast of inflation (fan chart)
from August 2005
2005 (August)
Romania
5-7 quarters
first, 2 years, then up to 3 years
conditional (projection)
quarterly up to the end of 2007
(schedule of publication was
changed a few times),
then 3 times per year:
February, June, October
22 projections
forecast of inflation and GDP,
fan chart (inflation, GDP)
forecast of inflation and GDP,
fan chart (inflation, form
February GDP)
quarterly:
February, May, August,
November
from August 2004
1999
from August 2001
2001 (June)
Poland
MPC and forecasting
Discrete involvement
Discrete involvement
No involvement
Iterative involvement
procedure
Source: own study based on Inflation Reports of the Czech National Bank, the National Bank of Hungary, the National Bank of Poland and the National Bank of Romania.
Main model
Transmission horizon
Forecast horizon
Conditionality
Set of data
Forecast frequency
quarterly:
January, April, July, October;
from 2008: February, May,
August, November
from April 2001
forecast of inflation and GDP,
fan chart (inflation, GDP, policy
path for 2008 and exchange rate
path from 2009)
Forecast disclosure
Output
1998
DIT introduction
Hungary
4
The Czech Republic
Table 1. Inflation forecasting in Czech Republic, Hungary, Poland and Romania
Inflation forecast targeting in practice
B a n k
o f
P o l a n d
Inflation forecast targeting in practice
In Romania, a floating exchange rate regime was functioning before the DIT
implementation. The Romanian case was the example of the existence of the discrepancies
between de jure and de facto regime, so called “fear of floating”. It describes the situation
where regimes de jure are more flexible than de facto, as the country follows unofficial
exchange rate target usually in the period of large macroeconomic imbalance and low foreign
reserves to adopt an official peg (Nerlich 2002). After the DIT introduction in 2005, the
exchange rate regime became more floating, however, NBR still reacted sometimes to the
exchange rate changes.
In the Czech Republic and Poland direct inflation targeting was introduced in the late
nineties. Before its introduction, Czech koruna had already operated in a managed floating
regime. The forecast disclosure started in August 2001. Since 2001 DIT strategy in the Czech
4
Republic has been a fully-fledged version.
In Poland a floating exchange rate was introduced in 2000 after a short period of DIT
implementation. At that time the inflation forecast was not published. In Poland the interval
between the DIT introduction and the first forecast disclosure was quite long, as the first
projection was published in August 2004.
The forecasting procedures in the countries covered by the study evolved over time.
The main forecasting tool – the model – was first a quite simple tool that integrated all the
relevant data. It also provided consistent frameworks for the analysis (Analytical Schemes,
MAPM). Structural changes in national economies (i.e. caused by the access to the European
Union), in global economy, as well as the development of econometrics resulted in changes of
the main forecasting tools. The most developed tool used in the countries covered by the
study is the g3 model of the Czech National Bank. The g3 is a Dynamic Stochastic General
Equilibrium Model. DSGE models belong to the fourth generation of macromodels describing
the monetary policy impact on the economy. They are perceived as the most advanced
methodological approach in macromodeling. However, one argues that they have numerous
drawbacks (for further description see: Tovar 2008). All models used for forecasting and
policy analysis, regarding their complexity and scale, capture monetary transmission
mechanism that lies within the frameworks of New Neoclassical Synthesis.
Nevertheless, a forecast is not the outcome of an automatic modeling procedure.
Experts’ involvement is necessary as forecasting is an iterative process. Experts’ opinions
prevail in the short term. They better capture temporary shocks occurring in the economy. On
the other hand, models describe longer term relations in a better way. This is why the central
WORKING PAPER No. 105
17
21
Inflation forecast targeting in practice
bank forecast is usually a combined forecast: main models’, experts’ and sometimes forecast
produced with an additional tool.
Experts involved in forecasting procedures are central bank staff. However, the
expression central bank forecast does not specify who the author of the forecast is. The
forecast may incorporate the view of the staff and the MPC members. This situation occurs in
the Czech Republic, Hungary and Romania. In Romania the Board Members are even
involved iteratively. It means consultations between the staff and the Board on subsequent
stages of the forecasting round. The forecast produced with an iterative involvement of the
MPC members incorporates to a large extent their view on the future course of the economy.
And they are the ones who make decisions on interest rates, which means that these decisions
are tailored on the basis of their opinion. As a consequence, the forecast is a less objective
4
premise of a decision making process. On the other hand, a decision should be in line with the
forecast message as it simplifies the communication with the market.
In Poland the Monetary Policy Council members are not involved in the forecasting
process. They have no impact on the forecast which is produced by the staff and discussed
during the MPC meeting. No-involvement case also creates a problem as the MPC members
may not share the view of the staff on the economic development. As a result, they will not
react in the way suggested by the forecast. It may dampen the forecasts’ impact on the
expectations of economic agents.
Taking into consideration the alternatives described above, one might argue that
discrete involvement of monetary policy committee constitutes a satisfactory compromise. In
fact, it does not. The involvement of MPC members seems to have no impact on the
consequence in IFT implementation. The National Bank of Hungary implemented the same
solution as the Czech National Bank (discretionary involvement). The first one does not
follow the message of the forecast, while the second one does it with a surprising consistency
even in a turbulent period for the financial markets (Szyszko 2009). It seems that there are
some other, qualitative aspects which seem to be decisive. One of them could be the belief of
the MPC that the forecast is the best supporting tool of a decision-making process or the
pressure from the central bank’s staff which has equipped the MPC with such a tool.
The second part of the descriptive analysis focuses on inflation forecast targeting
implementation in the countries in question. Table 2 presents a summary of the comparison
of the IFT implementation. Four features are considered: formal declaration on the importance
of inflation forecasts, consistency of the decision of the MPC with the inflation forecast result,
decision timing and finally – the way in which the decisions on interest rate were justified.
22
18
N a t i o n a l
B a n k
o f
P o l a n d
The second part of the descriptive analysis focuses on inflation forecast targeting
implementation in the countries in question. Table 2 presents a summary of the comparison
Inflation forecast targeting in practice
of the IFT implementation. Four features are considered: formal declaration on the importance
of inflation forecasts, consistency of the decision of the MPC with the inflation forecast result,
decision timing and finally – the way in which the decisions on interest rate were justified.
Table 2. Inflation forecast targeting implementation
18
The Czech
Republic
Declaration on
the forecast role
in monetary
policy
Consistency in
IFT
implementation
Timing of
decision-making
The forecast in
decision
explanation
Source: own work
the forecast is of
greatest relevance
in decisionmaking
Hungary
partial input in
decision-making
process;
previously:
intermediate
objective
Poland
Romania
partial input in
decision-making
process
partial input in
decision-making
process
4 times the MPC
did not follow the
message of the
forecast; each time
it was due to
exogenous factors
and clearly
explained;
low, numerous
decisions were not
in line with the
forecast result,
they were
explained by the
current economic
situation
the MPC decisions
were in line with
the message of the
forecast; however
a flexible approach
to input in
decision-making
process dominated
just after the
forecast is made
main factor, even
in the months
when the new
forecast was not
revealed
wait and see
position
wait and see
position
the decisions were
in line with the
message of the
forecast, except for
3 cases which
were explained by
current economic
situation; in other
cases - flexible
approach to input
in decision-making
process
wait and see
position
one of the
numerous factors,
sometimes
neglected
one of the
numerous factors,
sometimes
neglected
one of the
numerous factors,
sometimes
neglected
4
Forecast disclosure is usually a part of the communication strategy. Forecast appears
in the Inflation Report. This document presents a broader context of the monetary policy in
order to provide the public with information relevant for the decision-making process.
Inflation forecast targeting is not dependable on detailed solutions on forecasting procedures
and forecasts disclosure. As it was shown in table 1 and described above, the systems of
forecasting inflation in the Czech Republic, Hungary, Poland and Romania differ. Regardless
of the details, revealing the forecast by the central bank is always perceived as a part of a
communication strategy that should help to build transparency and hence anchor expectations.
This is also the reason why central banks give the rationale for their decision. Usually central
banks explain also why the forecast is made and published. This declaration should show the
importance of the forecast as the rationale for interest rate adjustments.
The Czech National Bank declares on its website that the forecast for inflation at the
“monetary policy horizon” (about 12–18 months ahead) is of greatest relevance to the
decision-making on the current interest rate settings. In its monetary policy decision-making
the CNB Bank Board assesses the latest CNB forecast and evaluates the risks of non23
WORKING PAPER No. 105
19
banks explain also why the forecast is made and published. This declaration should show the
importance of the forecast as the rationale for interest rate adjustments.
Inflation forecast targeting in practice
The Czech National Bank declares on its website that the forecast for inflation at the
“monetary policy horizon” (about 12–18 months ahead) is of greatest relevance to the
decision-making on the current interest rate settings. In its monetary policy decision-making
the CNB Bank Board assesses the latest CNB forecast and evaluates the risks of nonfulfilment of this forecast. Based on these considerations the Bank Board then votes on
whether and how to change the settings of monetary policy instruments. By changing these
19
instruments the central bank seeks to offset excessive inflationary or disinflationary pressures
which are deviating future inflation from the inflation target or from the tolerance band
around this target. The first declaration on the forecast importance was given in the CNB
Monetary Strategy Document that was published in 1999. It was stated there that the key
source material for the CNB Bank Board's decisions would be the CNB's macroeconomic
forecast. The central bank of the Czech Republic gave quite a clear picture of the forecast role
in the monetary policy.
4
The National Bank of Hungary, at the beginning of DIT implementation, declared
explicitly that inflation forecast played the role of the intermediate target (Monetary Policy in
Hungary, 2002). This was quite an unambiguous statement, showing a strong involvement in
IFT. In 2010 new information appeared on the website: In taking its policy decisions, the
Monetary Council takes into account the baseline path of the NBH staff's projection, the
uncertainty around it and risk scenarios that are judged to be relevant. However, decisionmaking is not a mechanical exercise, as the Monetary Council may give consideration to
other aspects, in addition to the NBH staff's projection. The latest version demonstrates much
less importance of the inflation forecast in a decision making process. At the same time it is
more consistent with the behavior of the Hungarian Monetary Council.
In Poland the rationale for projections and their publication was given for the first time
in 2004, when the first projection was published. It was limited to the short note in the
Inflation Report saying that the projection is one of the inputs to the Monetary Policy
Council’s decision-making process. Monetary Policy Strategy document (Monetary Policy
beyond 2003) was published in 2003 – before the first projection disclosure. The projections
and their role in the monetary policy, however, were not mentioned. Starting form 2005,
Monetary Policy Council refers to the projections in Monetary Policy Guidelines. In the latest
ones it was stated that monetary policy is pursued under uncertainty which excludes strict
control of economic processes. This uncertainty means that while taking decisions related to
monetary policy it is necessary to take into account all available information relevant for
inflation developments, rather than the results of inflation projection only. Models used by
central banks to forecast inflation may be imperfect in adequately reproducing behavior of
the economy if only because of its ongoing structural changes. In addition, it is not possible to
adopt a simple policy rule which could be known ex ante to market participants (Monetary
24
N a t i o n
a l
B a nopinion
k
o f the
P o l
Policy Guidelines 2010). This declaration shows that in Monetary
Council’s
a n d
control of economic processes. This uncertainty means that while taking decisions related to
monetary policy it is necessary to take into account all available information relevant for
Inflation forecast targeting in practice
inflation developments, rather than the results of inflation projection only. Models used by
central banks to forecast inflation may be imperfect in adequately reproducing behavior of
the economy if only because of its ongoing structural changes. In addition, it is not possible to
adopt a simple policy rule which could be known ex ante to market participants (Monetary
Policy Guidelines 2010). This declaration shows that in Monetary Council’s opinion the
projection is not a primary input in a decision making process. It is taken into consideration
20
but has an equal position as the other data. Moreover,
it is emphasized that this projection
could temporarily be of little importance (rationale for decision-making during the financial
crisis).
The National Bank of Romania presents on the Bank’s website only one quite a
laconic piece of information on inflation forecast importance. It says that the forecasts
generated by the model are a very important input in the NBR Board's policy decision
making. There is neither monetary policy strategy document nor yearly published information
4
on monetary policy assessment. However, just before implementation of the new monetary
frameworks in 2005 an effort was made to ensure completion of the models for the analysis
and short and medium-term forecasting of inflation. They were to be used as an input for the
monetary policy decision-making. Annual Reports confirm the declared role of the inflation
forecast.
It is remarkable that the central banks underline first of all the internal function of the
forecast: supporting the decision-making of the MPC. The theory as well as empirical
research focus on external functions – shaping inflation expectations. Forecast disclosure
accompanied with the declaration that it will be quite an important input in a decision-making
process certainly opens the possibility of fulfilling external functions of the forecast - the
central bank can guide the market indirectly via expected instruments adjustments.
The declaration on the forecast importance is not sufficient for the market participants,
however, it can serve as the starting point. If the forecast is to be taken into consideration
while making decisions by market participants, the central bank has to prove that it really
implements IFT. The authorities’ declarations, including monetary authorities, may not be
reflected in their behavior. Therefore, publishing the forecast will be of little importance for
shaping market expectations. Bearing that in mind the assessment of central bank’s credibility
is quite difficult. It would mean the answer to the question whether inflation forecast is really
the input for the decision-making process. This would be the research of mostly qualitative
nature, having numerous limitations:
-
there is no possibility to measure formal declaration on forecast importance; main
input in a decision-making process, greatest relevance to the decision-making, not
automatic exercise – these are purely descriptive expressions,
-
the latest declaration of the HNB and the NBP as well as the CNB and the NBR
WORKING PAPER No. 105
has almost the same meaning from the linguistic point of view; however, it does
not mean that the CB’s intentions were the same,
25
the input for the decision-making process. This would be the research of mostly qualitative
nature, having numerous limitations:
Inflation forecast targeting in practice
-
there is no possibility to measure formal declaration on forecast importance; main
-
input in a decision-making
process,
greatest
relevance
to the decision-making,
assessment
of IFT should show
a broader
context
of communication
with the not
automatic
exercise
these arewere
purely
market;
it is
not the–situation
thedescriptive
rule of IFTexpressions,
was not implemented that is
-
the latest declaration
of the
HNB
the NBP asofwell
the CNB IFT
and is
theclosed
NBR
important
but the reason
why
and and
the frequency
suchasbehavior;
has almost
samemeans
meaning
fromallowance
the linguistic
point
of view;the
however,
does
loop
policy,the
which
ex ante
for not
following
rule in ait specific
-
not mean no
thatautomatic
the CB’s procedure
intentions is
were
same, but those situations where the
situation,
everthe
possible;
assessment
of IFT should
a broader
context
of communication
with the
IFT
is not followed
shouldshow
be rare
and well
accounted
for,
21 rule of IFT was not implemented that is
market;
it isthere
not the
situation
the
sometimes
is no
simple were
solution
given by the message of the forecast; it is
important
why
the frequency
ofthe
such
behavior;
is closed
connected but
withthe
thereason
attitude
of and
the central
banks to
DIT
strategy;IFT
they
loop
policy,rather
whichflexible
means inflation
ex ante allowance
not means
following
the rulethat
in a there
specific
implement
targeting; for
which
in practice
is
situation,
nofluctuation
automatic procedure
is ever
possible;
butthe
those
situations
where the to
an
accepted
band around
inflation
target;
CB shows
commitment
-
4
IFTmain
is notgoal
followed
should be rareitand
well
accountedthat
for,the inflation level that
the
but simultaneously
sends
a message
sometimes
there or
is below
no simple
solution
given
by the message
of theespecially
forecast; itwhen
is
would be above
the target
may
be temporarily
accepted,
connected
the attitude
of (sometimes
the central banks
DIT strategy;
they not
it is due to with
exogenous
shocks
the listtoofthe
caveats
is presented);
implement
flexible
inflation
targeting;
which means for
in practice
focusing onrather
economic
shock
supports
output stabilization;
the IFT that there is
an
accepted fluctuation
band around
target; the
CBisshows
commitment
to
implementation
this flexibility
meansinflation
that sometimes
there
more than
one option
thethe
main
goal
but simultaneously
it sends aas
message
inflation
level that
of
MPC
decision
that can be perceived
being inthat
linethe
with
the message
of the
would be above or below the target may be temporarily accepted, especially when
forecast.
it is due
to exogenous
(sometimes
theexamination
list of caveats
is presented);
not
Because
of the
limitations shocks
presented
above, the
of IFT
implementation
focusing
on economic
shock
supports output
for of
thethe
IFT
may not lead
to unambiguous
results.
Qualitative
analysisstabilization;
of the practice
four central
implementation
thisresults
flexibility
meansbriefly
that sometimes
is more than
one 2.
option
banks in question
can give the
presented
below andthere
summarized
in table
the MPC
decision
that can
bewas
perceived
as being in
the message Itof the
Theof
Board
of Czech
National
Bank
very consistent
in line
IFT with
implementation.
forecast.
followed the
message of the forecast. When it did not - it was clearly explained why. Usually
5 examination of IFT implementation
Because
of the limitations
above,
the
. If there was a need for the interest rate
the reason
lied beyond
the impact presented
of the central
bank
may
not CNB
lead to
results.
analysiswas
of the
central
change,
didunambiguous
not hesitate to
do soQualitative
when the forecast
the practice
most up of
to the
datefour
without
banks
question
give theMoreover,
results presented
briefly below
andthe
summarized
in table
taking in
wait
and seecan
position.
the communication
with
public focused
on 2.
the
Board
of and
Czech
BankBalance
was very
IFT
implementation.
It
forecast,The
policy
path
riskNational
assessment.
of consistent
the risk forinthe
forecast
was presented
followed
the message
of thethe
forecast.
When itdid
didnot
notappear.
- it was clearly explained why. Usually
even
in those
months when
new forecast
5
was aCouncil.
need for For
the several
interest rate
the reason
lied beyond
thewas
impact
of the by
central
bank . If there
Different
attitude
presented
the Hungarian
Monetary
change,theCNB
didclaimed
not hesitate
to do
so when
theintermediate
forecast wastarget.
the most
up same
to date
without
years,
NBH
that the
forecast
is its
At the
time
the
taking waitwas
andneglected
see position.
Moreover,
with
the public
focusedinterest
on the
projection
as the
input forthe
thecommunication
decision-making
process.
The official
forecast, policy path and risk assessment. Balance of the risk for the forecast was presented
5
even
in those
months
when
theforecast
new forecast
appear.
In April
2003 the
message
of the
suggesteddid
thatnot
there
should be no change of interest rates. The rates
were cut in June. The CNB explained that the wrong assumption on time and extent of direct tax change was
attitude
was presented
bythe
thechanges.
Hungarian
Monetary
Council.
several
made. In Different
the meantime
the government
postponed
Similar
situation took
place inFor
January
2006. The
exchange
rate
appreciation
was
stronger
that
expected.
It
meant
more
tight
monetary
condition
than
years, the NBH claimed that the forecast is its intermediate target. At the same time the planned
without interest rate rise.
projection was neglected as the input for the decision-making process. The official interest
22
26
N a t i o n a l
n k
o f
o l a n d
In April 2003 the message of the forecast suggested that there should
be no change Bof ainterest
rates.P The
rates
were cut in June. The CNB explained that the wrong assumption on time and extent of direct tax change was
made. In the meantime the government postponed the changes. Similar situation took place in January 2006. The
exchange rate appreciation was stronger that expected. It meant more tight monetary condition than planned
5
followed the message of the forecast. When it did not - it was clearly explained why. Usually
the reason lied beyond the impact of the central bank5. If there was a need for the interest rate
Inflation forecast targeting in practice
change, CNB did not hesitate to do so when the forecast was the most up to date without
taking
wait and
see mainly
position.
Moreover,
the communication
with Sometimes
the public focused
on the
rates changes
were
caused
by exchange
rate fluctuations.
other current
forecast,were
policy
path and
assessment.
Balance
thean
risk
for the forecast
factors
decisive.
Therisk
projection
seemed
to be of
only
additional
input inwas
the presented
decisioneven in those
months
when the new
forecast
did not
appear.
making
process.
The flexibility
which
is allowed
by fluctuation
band around the inflation
Different
attitude
was
presented
thechange
Hungarian
Council.
For several
target gives
freedom
not to
change
rates by
or to
themMonetary
in the direction
suggested
by the
years, theThe
NBH
claimedBank
that of
theHungary
forecast often
is its intermediate
the same
time the
changed themtarget.
in theAt
direction
opposite
to the
forecast.
National
6
projection
was neglected
the input
for the
decision-making
process.
The
. Such
behavior
was inconsistent
with
theofficial
stronginterest
message suggested
by theasforecast
rates changes
mainly caused
by exchange
ratethe
fluctuations.
other current
declaration
onwere
the projections
importance.
In 2010
declarationSometimes
was changed.
5
In Aprilwere
2003decisive.
the message
ofthat
the
forecast
suggested
thatbethere
be
no change
of interest
rates.
The rates
factors
The
projection
seemedCouncil
to
only
an
additional
input
in the
decisionSimultaneously,
it seems
the
Monetary
hasshould
been
slowly
changing
its attitude
were cut in June. The CNB explained that the wrong assumption on time and extent of direct tax change was
making
process.
Thetheflexibility
which
is allowed
byStarting
fluctuation
around
the
inflation
made. In
the
meantime
government
postponed
the changes.
Similar situation
took
place
in January
2006. the
The
since
managed
floating
exchange
rate introduction.
fromband
February
2008
only once
exchange rate appreciation was stronger that expected. It meant more tight monetary condition than planned
target
notwith
to change
rates ormessage.
to change them in the direction suggested by the
withoutgives
interest
rate rise.
wasfreedom
not
in line
the forecast
decision
often changed
the direction
opposite
to the
forecast.InThe
National
Bank of Hungary
Poland
the Monetary
Policy Council
proved athem
ratherineclectic
approach
to choosing
6
22
. Such
inconsistent
with
strong
message
by the forecast
the
input suggested
of a decision-making
process.
Itsbehavior
decisionswas
were
generally in
linethe
with
the message
4
declaration
on the
projections
importance.
In when
2010 there
the declaration
wasband
changed.
of
the forecast.
It might
be inferred
then that
is a tolerance
around inflation
Simultaneously,
it seems
that inflation
the Monetary
has level
been slowly
changing its attitude
horizon is
target,
deviation of
projected
from Council
the targeted
within transmission
since
exchange
ratechoose
introduction.
from is
February
2008 only once the
often managed
within thefloating
band. The
MPC can
whetherStarting
the reaction
worth producing.
decision was
notare
in two
line important
with the forecast
message.that the projection was not of primary
However,
there
issues proving
In Poland
the Monetary
Policy Council proved
rather eclectic
approach
to choosing
importance:
the moment
of the decision-making
and itsaexplanation.
Firstly,
the projection
the input
of a decision-making
process.
decisions
were
in line
with the
message
was
produced
quarterly and from
2008, Its
3 times
a year.
Justgenerally
after it was
produced
it was
the
of therelevant.
forecast.During
It mightthe
benext
inferred
thatmeetings,
when there
a tolerance
band around
inflation
2 or then
3 MPC
theisprojection
became
stale. But
at the
most
is
target,
deviation
of projected
inflation
from thewhether
targetedthe
level
within transmission
same time
there was
a possibility
of assessing
assumptions
were right.horizon
When the
often within
the band.
The MPC can
reaction
is worth
MPC
implements
IFT consistently
it choose
usually whether
makes a the
decision
in line
with producing.
the forecast
However,instead
there are
two important
provingThe
thatrationale
the projection
of primary
message
of taking
wait andissues
see position.
behindwas
thenot
decision
also did not
importance:
moment of the decision-making and its explanation. Firstly, the projection
focus
on the the
projection.
was produced
quarterly
andoffrom
2008,had
3 times
a year.
Just as
after
was produced
it was
the
attitude
theitNational
Bank of
Poland.
The National
Bank
Romania
similar
2 oranalysis
3 MPC was
meetings,
became
stale.
at the
most
relevant.
During point
the next
However,
the starting
of the
a littlethe
bit projection
different, what
should
beBut
underlined
same time
there was
possibility
assessing
whether In
thePoland
assumptions
were right.
When the
while
comparing
the acentral
banksofunder
examination.
the forecast
was revealed
for
MPC
implements
consistently
it usually
makes a decision
in line withthe
thecentral
forecast
the
first
time in theIFT
sixth
year of DIT
implementation.
As a consequence,
bank had
messagefinished
instead implementing
of taking wait new
and see
position.frameworks.
The rationaleSimultaneously,
behind the decision
also did not
market
already
operational
focus on thegot
projection.
participants
used to the interest rate policy and its signaling effects. In Romania the
Romania
had similar
the National
Bank of The
Poland.
forecastThe
wasNational
disclosedBank
just of
after
the introduction
of attitude
the newas
monetary
frameworks.
However, the
starting
of thepolicy
analysis
was
a little
different,
what
should
be underlined
operational
level
of thepoint
monetary
was
still
beingbit
adjusted.
The
impact
of NBR
while comparing the central banks under examination. In Poland the forecast was revealed for
6
16 times between August 2001 and the end of 2010 the message of forecast was ignored. The rationale for
the
first time in the sixth year of DIT implementation. As a consequence, the central bank had
interest rate change referred sometimes directly to the speculative attacks on Hungarian forint. Usually there was
no reference
to the implementing
forecast in the rationale
for the decision.
The other Simultaneously,
reasons were underlined
(risk premium
already
finished
new operational
frameworks.
market
change, demand pressure, imbalance of the Hungarian economy).
participants got used to the interest rate policy and its signaling effects. In Romania the
forecast was disclosed just after the introduction
23 of the new monetary frameworks. The
operational level of the monetary policy was still being adjusted. The impact of NBR
WORKING PAPER No. 105
6
16 times between August 2001 and the end of 2010 the message of forecast was ignored. The rationale for
27
message instead of taking wait and see position. The rationale behind the decision also did not
focus on the projection.
Inflation forecast targeting in practice
The National Bank of Romania had similar attitude as the National Bank of Poland.
However, the starting point of the analysis was a little bit different, what should be underlined
while comparing the central banks under examination. In Poland the forecast was revealed for
the first time in the sixth year of DIT implementation. As a consequence, the central bank had
already finished implementing new operational frameworks. Simultaneously, market
participants got used to the interest rate policy and its signaling effects. In Romania the
forecast was disclosed just after the introduction of the new monetary frameworks. The
operational level of the monetary policy was still being adjusted. The impact of NBR
measures on short term interest rates was assessed as unsatisfactory. This was the first reason
6
16 times between August 2001 and the end of 2010 the message of forecast was ignored. The rationale for
why the
Bank
Board
decisions
were
not to
inthe
accordance
the
the forecast.
Thewas
interest
rate
change
referred
sometimes
directly
speculativewith
attacks
onmessage
Hungarianof
forint.
Usually there
no reference to the forecast in the rationale for the decision. The other reasons were underlined (risk premium
seconddemand
reasonpressure,
was connected
eclecticeconomy).
approach to the data analysis. Sometimes the
imbalancewith
of theanHungarian
change,
Board simply decided that current data are more important. Between August 2005 and the end
23 direction 3 times. It also took wait and see
of 2010 the Board reacted in the counter-forecast
4
position and the forecast was not the central point in decision rationale.
This section gave the background for empirical study. Simultaneously, it was the
starting point for verifying the second hypothesis: the central banks under examination apply
different forecasting rules, proving at the same time different approach to the decision-making
input. Market participants can quite easily asses the importance of the forecast in decision
making procedures as well as its reliability. Then they decide whether the forecast should be
taken into consideration while shaping the expectations. The next section presents the results
of a quantitative study on the relationships between inflation forecast and expectations.
However, it can be stated as the summary of this section, that the existence of this kind of
relationships depends strongly on qualitative aspects, including the central bank’s attitude and
consistency in IFT implementation. In some countries preconditions for these relationships
are fulfilled while in others – they are not. Preliminary conclusion that can be drawn here
implies that the relations between inflation forecasts and expectations should be the strongest
in the case of the CNB and the weakest in the case of Hungarian central bank.
28
N a t i o n a l
B a n k
o f
P o l a n d
Empirical study on the forecast and inflation expectations
5. Empirical study on the forecast and inflation expectations
5. Empirical study on the forecast and inflation expectations
The existence of associations between the inflation forecast and inflation expectations
The existence of associations between the inflation forecast and inflation expectations
of households is empirically verified in this section. This section describes the data,
of households is empirically verified in this section. This section describes the data,
methodological issues and finally, it presents the results of the study. However, a few general
methodological issues and finally, it presents the results of the study. However, a few general
remarks need to be presented beforehand.
remarks need to be presented beforehand.
First of all, the data availability limits the research perspective. Unique information on
First of all, the data availability limits the research perspective. Unique information on
forecasts and expectation, except balances, for the four countries in question simply does not
forecasts and expectation, except balances, for the four countries in question simply does not
exist. Each central bank presents the forecast differently. Only the information that is
exist. Each central bank presents the forecast differently. Only the information that is
published may influence expectations. A similar problem – limited accessibility of the time
published may influence expectations. A similar problem – limited accessibility of the time
series - concerns inflation expectations, measuring of which changed over time.
series - concerns inflation expectations, measuring of which changed over time.
For each country two sets of variables are analyzed: one on the forecast message and
For each country two sets of variables are analyzed: one on the forecast message and
the other – on inflation expectations of economic agents (first of all – consumers, business
the other – on inflation expectations of economic agents (first of all – consumers, business
expectation case is included for the Czech Republic). Because of the reason described above
expectation case is included for the Czech Republic). Because of the reason described above
they do not cover exactly the same time series. The idea was to use the data that has similar
they do not cover exactly the same time series. The idea was to use the data that has similar
informational value. However, in certain cases, especially for the Czech Republic, additional
informational value. However, in certain cases, especially for the Czech Republic, additional
information is used (policy path). It results form the fact that the divergences in forecasting
information is used (policy path). It results form the fact that the divergences in forecasting
procedures, in the way of publishing forecast and in the expectations measurement exist.
procedures, in the way of publishing forecast and in the expectations measurement exist.
The associations between the forecast results and expectation are tested in a few areas.
The associations between the forecast results and expectation are tested in a few areas.
Obviously, the correlation does not bring any information on the causality. The theoretical
Obviously, the correlation does not bring any information on the causality. The theoretical
underpinnings suggest that the forecast should influence expectations, but the correlation
underpinnings suggest that the forecast should influence expectations, but the correlation
refers to any statistical relationships. However, a priori no result can be expected, and
refers to any statistical relationships. However, a priori no result can be expected, and
moreover – cannot be indicated as a proper one. It can be only assumed that there should be a
moreover – cannot be indicated as a proper one. It can be only assumed that there should be a
statistically important relationship between these two variables. Even the existence a of
statistically important relationship between these two variables. Even the existence a of
negative correlation can be interpreted: for example, when the forecast shows that inflation is
negative correlation can be interpreted: for example, when the forecast shows that inflation is
above the target but the goal is exceeded less than the last forecast suggested, the expectations
above the target but the goal is exceeded less than the last forecast suggested, the expectations
may fall. Another example is connected with a conditional forecast: when the central path is
may fall. Another example is connected with a conditional forecast: when the central path is
above the target and the society understands the conditionality of the forecast, it will expect
above the target and the society understands the conditionality of the forecast, it will expect
the main rate to rise and its inflation expectation may fall.
the main rate to rise and its inflation expectation may fall.
5
5.1 Inflation forecast results
5.1 Inflation forecast results
There is no unique information on inflation forecast that is accessible in the four
There is no unique information on inflation forecast that is accessible in the four
countries. To make the data comparable and to enable the use of the longest time series
countries. To make the data comparable and to enable the use of the longest time series
WORKING PAPER No. 105
25
25
29
Empirical study on the forecast and inflation expectations
possible, the information on the forecast result is processed to present forecast result in a
qualitative way. The relation between the inflation forecast result and the inflation target at
the horizon of t+4Q is taken into consideration. It is the most coherent with the way in which
the expectations of economic agents are measured in surveys (which will be described
further). The suggested way of coding inflation forecast results is also coherent with the way
how the forecast is presented in the media. Consumers do not usually read Inflation Reports
or central banks’ documents. They learn the forecast results indirectly. Analytics present the
message of the forecast referring to the relation of the forecast and the central bank goal.
Two cases were distinguished. The first one – and simpler – assumes three
possibilities:
-
the forecast is below the inflation target;
-
it is at the inflation target level;
-
it is above the inflation target.
This version (henceforth referred to as 3 possibilities case) does not account for the
5
situation when the central path shows that the target will be missed but at the same time the
inflation rate will remain within the accepted fluctuation band and the situation when central
path is below or above lower or upper boundary of the fluctuation band. Meanwhile, the
inflation targeters usually implement flexible inflation targeting. This means that they accept
fluctuation around the targeted level of inflation. It limits output fluctuations giving the
central bank possibility not to react in case of certain shocks (usually predefined and
published). This is why for the second case the message of the forecast is simplified to five
possible positions of the central path of the forecast in relation to the inflation target at the
horizon t+4Q (henceforth referred as 5 possibilities case):
-
it is below the upper boundary of the fluctuation band;
-
it is below the inflation target but within the fluctuation band;
-
it is at the target level;
-
it is below the lower boundary of the fluctuation band;
-
it is below the inflation target but within the fluctuation band.
The other approach could also be tested: the direction of the change in the central path.
However, it has a remarkable drawback in comparison with the first approach that is
connected with the nature of forecasting tools. The forecasting models are general equilibrium
models. In the long run inflation as well as the other economic variables return to equilibrium
level, which means – in the case of inflation – the targeted level of it. Usually, when
transmission horizon starts the shocks that influence the economy at the starting point of the
30
26
N a t i o n a l
B a n k
o f
P o l a n d
connected with the nature of forecasting tools. The forecasting models are general equilibrium
Empirical study on the forecast and inflation expectations
models. In the long run inflation as well as the other economic variables return to equilibrium
level,
which
means
– in the
inflation
targeted by
level
it. Usually,
when
forecast
horizon
vanish.
Thecase
newofones
cannot– the
be captured
theofmodel.
As a consequence,
transmission
horizonafter
starts4-6
thequarters
shocks returns
that influence
the economy
thethe
starting
pointthat
of the
the
path of inflation
to the inflation
level.atFor
countries
miss
forecast
vanish.
The
new
cannot
the model. As a consequence,
the
targethorizon
it means
usually
that
theones
central
pathbe
ofcaptured
inflation by
falls.
26isthe
the pathThe
of inflation
after 4-6 quarters
returns to
inflation
For thethe
countries
other shortcoming
of this approach
that
it does level.
not capture
relation that
of miss
the target
it means
thethe
central
pathisofsupposed
inflation to
falls.
central
path
and theusually
target. that
When
inflation
rise, it can rise from the level
Thelower
otherboundary
shortcoming
of fluctuation
this approach
is that
it does
not capture
relation
below the
of the
band
as well
as from
the levelthe
above
the of
target or
central
andboundary.
the target.The
When
the inflation
is supposedshould
to rise,be
it different
can rise from
thecases.
level
even thepath
upper
impact
on the expectations
in both
below the
boundary
of theis fluctuation
band
as of
well
from the
level above
target or
Thelower
additional
solution
applied in the
case
theasCzech
National
Bank. the
It produces
even
the upper boundary.
impact
ontogether
the expectations
should path
be different
in both cases.
an
unconditional
forecast. The
It means
that
with the central
of the forecast,
a policy
The additional
solution
is applied
of the
Czech
National
Bank.
producesof
path is described
in Inflation
Reports.
Thisinisthe
thecase
reason
why,
for this
country,
theItdirection
an
unconditional
forecast.
It the
means
thatfollowing
together with
the centralispath
the forecast,
a policy
change
of the policy
path in
period
the disclosure
alsooftaken
into
path
is described in Inflation Reports. This is the reason why, for this country, the direction of
consideration.
change The
of the
policy
in the period
the different
disclosure
is also taken
into are given
length
ofpath
the sample
is alsofollowing
different for
countries.
The details
consideration.
in
table 3.
The length of the sample is also different for different countries. The details are given
5
in table
3. expectations
5.2
Inflation
5.2 Inflation
expectations
The theoretical
discussion on monetary policy and its impact on the economy shows
the priority of shaping the expectations in the actions of the central bank. It has already been
The theoretical
discussion
onasmonetary
policy
and its of
impact
the economy
emphasized
that they can
be treated
an unbiased
predictor
futureoninflation
as wellshows
as an
the
priority
expectations
in However,
the actionsinflation
of the central
bank. It especially
has alreadythose
beenof
indicator
of of
theshaping
central the
bank’s
credibility.
expectations,
theybecan
be treated
as an unbiased
inflation
as well as an
aemphasized
household,that
cannot
directly
observed.
There arepredictor
two waysofoffuture
measuring
inflation
7
indicator of the
central
bank’s
credibility.
However,
inflationsurveys
expectations,
especially
expectations.
. Theof
The
first one
consists
of conducting
economic
on a regular
basisthose
a household,
be directly
There
aregive
two the
ways
of measuring
inflationinflation
survey
is of acannot
qualitative
nature.observed.
Consumers
do not
exact
level of expected
7
8
. The
expectations.
Thequestions
first oneon
consists
of conducting
economic
surveys on a9.regular
basis
but
they answer
their inflation
perception
and expectations
Then the
answers
survey
is of in
a qualitative
do not
give the is
exact
level of expected inflation
can
be used
two ways. nature.
First ofConsumers
all the balance
of answers
calculated:
but they answer questions on their inflation perception8 and expectations9. Then the answers
can= be
in two
ways.
First of all the balance of answers is calculated:
B
(PPused
+ ½P)
− (½M
+ MM)
BThe
= (PP
+ ½P) − (½M + MM)
second one consists of deriving inflation expectations from the prices of financial assets, which is not
7
subject of this paper.
The question is: How do you think that consumer prices have developed over the last 12 months? And the
7
answers
to chose
They
a lot, risen
moderately,
slightly,
about
the which
same, fallen,
The second
onefrom:
consists
of have…risen
deriving inflation
expectations
fromrisen
the prices
of stayed
financial
assets,
is not
don't know.
subject
of this paper.
98
The question
comparison
with that
the past
12 months,
dodeveloped
you expectover
that the
consumer
will And
develop
question is:
is:By
How
do you think
consumer
priceshow
have
last 12prices
months?
the
in
the
next
12
months?
And
the
answers
to
chose
from:
They
will…increase
more
rapidly,
increase
at
the
same
answers to chose from: They have…risen a lot, risen moderately, risen slightly, stayed about the same, fallen,
rate, increase
don't
know. at a slower rate, stay about the same, fall, don't know.
9
The question is: By comparison with the past 12 months, how do you expect that consumer prices will develop
in the next 12 months? And the answers to chose from: They will…increase more rapidly, increase at the same
rate, increase at a slower rate, stay about the same, fall,27don't know.
8
31
WORKING PAPER No. 105
27
survey is of a qualitative nature. Consumers do not give the exact level of expected inflation
9
but they answer questions on their inflation perception8 and expectations
. Then the answers
Empirical study on the forecast and inflation expectations
can be used in two ways. First of all the balance of answers is calculated:
B = (PP + ½P) − (½M + MM)
where PP denotes the percentage of respondents who chose the option “the most positive”
7
The second one consists of deriving inflation expectations from the prices of financial assets, which is not
(consumer
subject
of thisprices
paper. have risen a lot for the inflation perception and they will increase more
8
The question
is: How inflation);
do you thinkPthat
consumer
prices havepositive”
developedanswer
over the(consumer
last 12 months?
And the
rapidly
for expected
denotes
“moderately
prices
answers to chose from: They have…risen a lot, risen moderately, risen slightly, stayed about the same, fallen,
don't know.
have
risen moderately for inflation perception and they will increase at the same rate for
9
The question is: By comparison with the past 12 months, how do you expect that consumer prices will develop
expected
M - the
theanswers
percentage
of respondents
who chosemore
the “moderately
negative
in the next inflation);
12 months? And
to chose
from: They will…increase
rapidly, increase
at the same
rate, increase at a slower rate, stay about the same, fall, don't know.
option” (consumer prices stayed about the same for inflation perception and they will stay
about the same for expected inflation); MM - the percentage of respondents who chose the
27
option “very negative” (consumer prices have fallen for inflation perception and they will fall
in next 12 months for expected inflation) (The Joint… 2007). That balance values range from
−100, when all the respondents choose the most negative option to +100, when all the
respondents choose the most positive one.
This kind of survey is made monthly by the European Commission. It covers the new
5
member states as well: the Czech Republic from 1995, Hungary from February 1993, Poland
and Romania from May 2001. The balance of answers presented in Business and Consumer
Surveys Results does not directly measure the inflation expectations, thus it cannot be
interpreted in a straightforward way. When it is positive it means that the number of
respondents who expected prices to increase more rapidly over the next 12 months than in the
past exceeded the number of those who expected prices to remain the same or increase more
slowly that in the past. When the indicator (balance) of expected inflation turns negative, it
suggests that the number of respondents who expect prices not to rise over the next 12 months
is higher than the number of those who expect prices to remain the same or increase more
rapiddly than in the past. This is how the balance has been presented and interpreted by the
Czech National Bank in its Inflation Reports since April 2007.
The balances from Business and Consumer Surveys are directly used as the source of
information on inflation expectations. Despite the drawbacks described above (they sill do not
measure directly inflation expectations as well as they are not easily interpreted) they have
quite an important advantage. The same data with the time series longer than time series for
inflation forecast are accessible for all the countries covered by the examination. At the same
time, balances can represent roughly the perception of expected inflation by households.
Two different approaches, except for balances, to expectations are considered: the
level of expectations (derived form qualitative surveys or central banks surveys with
quantitative question) and the direction of its change.
32
N a t i o n a l
28
B a n k
o f
P o l a n d
Empirical study on the forecast and inflation expectations
The structure of the answers to the surveys’ question can be quantified using Carlson–
Parkin The
method
of quantification
(described
in Łyziak
2003). can
Thebe
quantification
is done
in two
structure
of the answers
to the surveys’
question
quantified using
Carlson–
The structure
of theinflation
answersrate
to the
surveys’ question
can be perceived
quantified using Carlson–
steps. First,
theofperceived
is quantified,
then using
Parkin
method
quantification
(described
in Łyziak 2003).
Thethe
quantificationinflation,
is done inthe
two
Parkin method
of quantification
(describedQuantified
in Łyziakdata
2003).
The
quantification
is done
inbeen
two
expected
inflation
can
be
also
quantified.
on
inflation
expectations
have
steps. First, the perceived inflation rate is quantified, then using the perceived inflation, the
steps. First,
perceived
inflation rate
is These
quantified,
thenalso
using
perceived
inflation, the
available
forthe
Poland
1992.
datadata
are
an the
input
to the research.
expected
inflation
canmonthly
be also since
quantified.
Quantified
on inflation
expectations
have been
expected
inflation
can
be
also
quantified.
Quantified
data
on
inflation
expectations
Howeverfor
there
is nomonthly
obligation
of expectations
oninput
the national
level. have
The been
available
Poland
since
1992. These quantification
data are also an
to the research.
available
for may
Poland
monthly
since 1992.
Thesebased
data are
an inputquestion:
to the research.
central bank
asses
expectations
in surveys
on also
quantitative
household
However
there
is no
obligation
of expectations
quantification
on the national level.
The
However
there
is
no
obligation
of
expectations
quantification
on
the
national
level.
The 12sector respondents
are asked
about their
inflation
expectations
at chosen
horizon,
usually
central
bank may asses
expectations
in surveys
based
on quantitative
question:
household
central horizon.
bank may
asses
expectations
in surveys based
on quantitative
question:inhousehold
month
The
CNB
andabout
the NBH
information
on
expectations
way12up
sector
respondents
are
asked
their collected
inflation expectations
at chosen
horizon, this
usually
sector
respondents
are asked unsatisfactory
about their inflation
expectations
at chosen
horizon,tousually 12to 2007.
This procedure
for theinformation
reasons why
is not applied
month
horizon.
The CNBwas
and the NBH collected
onitexpectations
in this way up
month
horizon.
The
CNB
and
the
NBH
collected
information
on
expectations
in target
this way
households:
awareness
of current economic
as well
of applied
inflation
is up
to
2007. Thistheir
procedure
was unsatisfactory
for thesituation
reasons why
it isasnot
to
to 2007. Qualitative
This procedure
was unsatisfactory
the reasons
why it is not applied
to in purely
limited.
questions
easiereconomic
for for
consumers.
The
measured
households:
their awareness
ofare
current
situation
asexpectations
well as of inflation
target is
households: way
theirare
awareness
ofThis
current
economic
situation
as wellexpectations
as of inflation
target is
quantitative
volatile.
is
why
this
way
of
measuring
was
limited. Qualitative questions are easier for consumers. The expectations measuredabandoned.
in purely
limited.
Qualitative
questions
are easier in
forthe
consumers.
The expectations
measured
in
purely
No
national
survey
has
been
conducted
Czech
Republic
and
in
Hungary.
The
European
quantitative way are volatile. This is why this way of measuring expectations was abandoned.
quantitative
way
volatile.
This
is why in
this way of Report
measuring only
expectations
was abandoned.
Commission
dataare
are
presented
in a descriptive
manner,
No
national survey
hasused
beenand
conducted
in theInflation
Czech Republicbut
and in Hungary.
The European
No
national
surveyforhas
been conducted
in theinflation.
Czech Republic
andthe
in direction
Hungary. of
The
using
the balance
perceived
and expected
However,
theEuropean
change
Commission
data are used
and presented
in Inflation Report
but only
in a descriptive
manner,
Commission
data are used
presented inisInflation
Report
but only in a descriptive manner,
in
inflation
expectations
of and
the households
explicitly
presented.
using
the balance
for perceived
and expected inflation.
However,
the direction of the change
using the balance for perceived and expected inflation. However, the direction of the change
in inflation expectations of the households is explicitly presented.
Table
3. Data
availability
andhouseholds
time extentis explicitly presented.
in
inflation
expectations
of the
The Czech Republic
Hungary
Poland
Inflation
forecast
Table 3. Data availability and time extent
Central
of availability
Table
3.path
Data
time extent
The Czech and
Republic
Hungary
Poland
inflation in
The
Czech
Republic
Hungary
Poland
Inflation
forecast
2001Q2-2010Q4
2001Q3-2010Q4
2004Q3-2010Q4
relation
Inflation forecast
Central
pathtoof
inflation
target
Central
path
inflation
in of
2001Q2-2010Q4
2001Q3-2010Q4
2004Q3-2010Q4
Policy
path
2002Q3-2010Q4
x
x
inflation
relation
toin
2001Q2-2010Q4 Inflation
2001Q3-2010Q4
expectations 2004Q3-2010Q4
relation
to
inflation
target
Balances
inflation
target
Policy
path
2002Q3-2010Q4
x
x
2001Q2-2010Q4
2001Q3-2010Q4
2004Q3-2010Q4
(consumers
only,
Policy path
2002Q3-2010Q4 Inflation
x
x
expectations
monthly)
Inflation expectations
Balances
2001Q2-2007Q2
Balances
2001Q2-2010Q4
2001Q3-2010Q4
2004Q3-2010Q4
(consumers
only,
August 2004(consumers,
2002Q2-2007Q3
2001Q2-2010Q4
2001Q3-2010Q4
2004Q3-2010Q4
(consumers
only,
monthly)
Expected inflation
December 2010
quarterly)
(consumers,
monthly)
2001Q2-2007Q2
(quantified)
(consumers,
August
20042001Q2-2010Q4
quarterly)
2001Q2-2007Q2
(consumers,
2002Q2-2007Q3
monthly)
August
2004Expected inflation
December
2010
(business,
quarterly)
(consumers,
2002Q2-2007Q3
quarterly)
(consumers,
December
2010
Expected
inflation
(quantified)
(consumers,
Direction
of
(consumers,
quarterly)
2001Q2-2010Q4
quarterly)
2001Q2-2010Q4
(consumers,
(quantified)
monthly)
changes in
2001Q3-2010Q4
2004Q3-2010Q4
2001Q2-2010Q4
quarterly)
(business,
quarterly)
(consumers,
monthly)
expectations
(consumers)
(consumers)
(business,
quarterly)
Direction
of
business)
2001Q2-2010Q4
(monthly)
Direction
changes
inof
2001Q3-2010Q4
2004Q3-2010Q4
2001Q2-2010Q4
(consumers,
Source:
own study.
changes
in
2001Q3-2010Q4
2004Q3-2010Q4
expectations
(consumers)
(consumers)
(consumers,
business)
expectations
(consumers)
(consumers)
(monthly)
business)
Source:(monthly)
own study.
Source: own study.
WORKING PAPER No. 105
29
29
29
5
Romania
Romania
Romania
2005Q3-2010Q4
2005Q3-2010Q4
x
2005Q3-2010Q4
x
2005Q3-2010Q4
x
2005Q3-2010Q4
2005Q3-2010Q4
x
x
x
x
x
x
33
Empirical study on the forecast and inflation expectations
The
The CNB
CNB still
still conducts
conducts quarterly,
quarterly, quantitative
quantitative surveys
surveys among
among companies.
companies. This
This is
is why
why
The CNB still conducts quarterly, quantitative surveys among companies. This
10 is why
the
Only for
for
the examination
examination for
for the
the Czech
Czech Republic
Republic covers
covers also
also business
business inflation
inflation expectations
expectations10.. Only
the examination for the Czech Republic covers also business inflation expectations10. Only for
the
the Czech
Czech case
case two
two groups’
groups’ expectations
expectations are
are covered.
covered.
the Czech case two groups’ expectations are covered.
For
For Romania
Romania no
no information
information except
except for
for balances
balances calculated
calculated in
in the
the EC
EC surveys
surveys is
is
For Romania no information except for balances calculated in the EC surveys is
available.
available. The
The national
national surveys
surveys only
only refer
refer to
to financial
financial market
market expectations.
expectations.
available. The national surveys only refer to financial market expectations.
Table
Table 33 presents
presents details
details on
on expectations
expectations and
and forecasts
forecasts that
that were
were examined
examined for
for each
each
Table 3 presents details on expectations and forecasts that were examined for each
country
country with
with their
their time
time extent.
extent. Three
Three approaches
approaches to
to expectations
expectations are
are applied:
applied:
country with their time extent. Three approaches to expectations are applied:
-- unique
unique data
data delivered
delivered by
by EC
EC surveys
surveys on
on balances
balances of
of answers
answers on
on expected
expected inflation,
inflation,
- unique data delivered by EC surveys on balances of answers on expected inflation,
the
the national
national data
data on
on quantified
quantified expectations
expectations of
of households
households (and
(and business
business for
for the
the
the national data on quantified expectations of households (and business for the
Czech
Czech Republic),
Republic),
Czech Republic),
-- the
the direction
direction of
of expectation
expectation change
change that
that was
was described
described in
in Inflation
Inflation Reports.
Reports.
- the direction of expectation change that was described in Inflation Reports.
The
The surveys
surveys results,
results, quantified
quantified or
or presented
presented as
as the
the balance
balance of
of answers,
answers, have
have also
also
The surveys results, quantified or presented as the balance of answers, have also
another
another shortcoming.
shortcoming. They
They reflect
reflect feelings
feelings of
of consumers
consumers rather
rather than
than their
their real
real behavior
behavior driven
driven
another shortcoming. They reflect feelings of consumers rather than their real behavior driven
by
by expectations.
expectations. The
The latter
latter is
is better
better reflected
reflected in
in financial
financial market
market assets
assets change.
change. Regardless
Regardless of
of
by expectations. The latter is better reflected in financial market assets change. Regardless of
this
this drawback,
drawback, surveys
surveys are
are commonly
commonly used
used to
to asses
asses expectations
expectations by
by central
central banks.
banks.
this drawback, surveys are commonly used to asses expectations by central banks.
5
5.3
5.3 Versions
Versions of
of the
the analysis
analysis
5.3 Versions of the analysis
Table
and 55 present
present pairs
pairs of
of variables
variables whose
whose relationships
relationships are
are tested
tested for
for each
each
Table 44 and
Table 4 and 5 present pairs of variables whose relationships are tested for each
country. Table
Table 44 presents
presents the
the relations
relations tested
tested for
for balances
balances of
of answers.
answers. In
In the
the case
case of
of the
the Czech
Czech
country.
country. Table 4 presents the relations tested for balances of answers. In the case of the Czech
Republic
Republic the
the associations
associations for
for the
the central
central path
path are
are additionally
additionally tested.
tested. The
The remaining
remaining pairs
pairs are
are
Republic the associations for the central path are additionally tested. The remaining pairs are
the
the same
same for
for each
each country.
country.
the same for each country.
The
The question
question on
on lags
lags appears
appears at
at this
this point.
point. The
The forecast
forecast is
is produced
produced with
with lower
lower
The question on lags appears at this point. The forecast is produced with lower
frequency
surveys on
on expectations
expectations are
are made.
made. From
From the
the theoretical
theoretical point
point of
of view,
view, the
the
frequency than
than the
the surveys
frequency than the surveys on expectations are made. From the theoretical point of view, the
forecast should
should be
be one
one of
of the
the most
most decisive
decisive signals
signals affecting
affecting inflation
inflation expectations.
expectations. However,
However,
forecast
forecast should be one of the most decisive signals affecting inflation expectations. However,
this signal
signal is
this
is issued
issued only
only at
at aa low
low frequency
frequency and
and can
can therefore
therefore become
become stale
stale over
over time.
time. As
As aa
this signal is issued only at a low frequency and can therefore become stale over time. As a
result,
result, financial
financial market
market participants
participants are
are likely
likely to
to increasingly
increasingly base
base their
their views
views on
on private
private
result, financial market participants are likely to increasingly base their views on private
11. In
information.
may increase
increase (Ehrmann,
(Ehrmann, Sondermann
Sondermann 2009)
2009)11
. In
information. In
In this
this case,
case, market
market volatility
volatility may
information. In this case, market volatility may increase (Ehrmann, Sondermann 2009)11. In
the
the light
light of
of these
these study
study results,
results, the
the influence
influence of
of inflation
inflation forecast
forecast on
on expectations
expectations should
should be
be
the light of these study results, the influence of inflation forecast on expectations should be
10
10 The research focuses on consumers expectations. Having
The research focuses on consumers expectations. Having
10 them and compare with the results for households.
use
time
time series
series on
on business
business expectations,
expectations, it
it was
was worth
worth to
to
The
research
focuses
on
consumers
expectations.
Having
time
series
on
business
expectations,
it
was
worth to
use
them
and
compare
with
the
results
for
households.
11
11 This
study
reaction
of
interest
use
them
and focuses
compareon
with
the results
households.
This
study
focuses
on
reaction
of UK
UKforshort-term
short-term
interest rates
rates to
to the
the Bank
Bank of
of England’s
England’s inflation
inflation report
report and
and to
to
11
macroeconomic
announcements
on
the
one
hand,
and
to
the
interrelationship
between
the
two
on
the
other
hand.
This
study
focuses
on
reaction
of
UK
short-term
interest
rates
to
the
Bank
of
England’s
inflation
report
to
hand.
macroeconomic announcements on the one hand, and to the interrelationship between the two on the otherand
The
on
on
stylized
model
macroeconomic
announcements
and
the
release
the
macroeconomic
announcements
one hand,
andthat
to the
interrelationship
between the two
other of
hand.
The authors
authors proved
proved
on the
the basis
basison
onthe
stylized
model
that
macroeconomic
announcements
and on
thethe
release
of
the
inflation
report
clearly
leadbasis
to aaonreduction
reduction
in market
market
volatility. This
This suggests
suggests
that public
public
signals
serve
to
The
authors
proved
on the
stylized model
that macroeconomic
announcements
and the
release
of the
inflation
report
clearly
lead
to
in
volatility.
that
signals
serve
to
homogenize
the
information
sets
of
market
participants,
thus
reducing
the
role
of
private
information.
inflation
report
clearly
lead
to
a
reduction
in
market
volatility.
This
suggests
that
public
signals
serve
to
homogenize the information sets of market participants, thus reducing the role of private information.
homogenize the information sets of market participants, thus reducing the role of private information.
34
30
30
30
N a t i o n a l
B a n k
o f
P o l a n d
Empirical study on the forecast and inflation expectations
the
the strongest
strongest when
when the
the lag
lag is
is quite
quite short.
short. Two
Two lag
lag options
options (1M
(1M and
and 2M)
2M) are
are tested
tested wherever
wherever it
it
12
the
strongest
when
the
lag
is
quite
short.
Two
lag
options
(1M
and
2M)
are
tested
wherever
it
12
is
Due to
to the
the timing
timing of
of the
the survey
survey and
and the
the inflation
inflation forecast
forecast publications,
publications, no-lag
no-lag
is possible
possible .. Due
is
possible12.not
Due to the timing of the survey and the inflation forecast publications, no-lag
version
version was
was not tested.
tested.
version was not tested.
Table
Table 4.
4. Versions
Versions of
of association
association tested
tested for
for balances
balances of
of answers
answers
Table 4. Versions of association tested for
balances
of
answers
Quarterly:
Quarterly:
Central
/balances
Central path
path (5
(5 possibilities)
possibilities)Quarterly:
/balances 1M
1M lag
lag
1M
Central
path
(5
possibilities)
/balances
2M
Central path (5 possibilities) /balances 2M lag
lag
Central
1M
(3
possibilities) /balances
/balances 2M
1M lag
lag
Central path
path (5
(3 possibilities)
Central
path
(3
possibilities)
/balances
1M
lag
2M
Central path (3 possibilities) /balances 2M lag
Central
2M lag 1M
(5
possibilities)/
average
Central path
path (3
(5 possibilities)
possibilities)//balances
average balance
balance
1M lag
lag
Central
path
(5
possibilities)/
average
balance
1M
lag
balances
2M
Central path (5 possibilities)/ average balances 2M lag
lag
Central
2Mlag
lag
(3
balance
Central path
path (5
(3 possibilities)/
possibilities)/ average
average balances
balance 1M
1M
lag
Central
balances
2M
lag
Central path
path (3
(3 possibilities)/
possibilities)/ average
average balance
balances1M
2Mlag
lag
Central
path(3
(3possibilities)
possibilities)//balances
average balances
Policy
1M
Policy path
path
(3
possibilities)
/balances
1M lag
lag 2M lag
2M lag
Policy
Policy path
path (3
(3 possibilities)
possibilities) /balances
/balances 1M
2M
lag
Policy
(3 possibilities)/
possibilities) /balances
2M lag 1M
average
path (5
(5
possibilities)/
average balance
balance
1M lag
lag
Policy path
Policy
balances1M
2Mlag
lag
Policy path
path (5
(5 possibilities)/
possibilities)/ average
average balance
balances
2M
lag
Monthly:
Policy path (5 possibilities)/Monthly:
average balances 2M lag
Central path
path repeated
repeated (case
(case with
with
possibilities)/balances 1M
Monthly:
Central
55 possibilities)/balances
1M lag
lag
5
Central
2M
Central path
path repeated
repeated (case
(case with
with 55 possibilities)/balances
possibilities)/balances 1M
2M lag
lag
Central
1M lag
Central path
path repeated
repeated (case
(case with
with 533 possibilities)/balances
possibilities)/balances 2M
1M
lag
Central
2M
Central path
path repeated
repeated (case
(case with
with 33 possibilities)/balances
possibilities)/balances 1M
2M lag
lag
Central
pathrepeated/
repeated (case
with1M
3 possibilities)/balances
2M lag
balances
1M
lag
Policy path
path
repeated/
balances
lag
Policy
Policy
2M
Policy path
path repeated/
repeated/ balances
balances 1M
2M lag
lag
Policy
path
repeated/
balances
2M
lag
Italics
relations tested
tested only
only for
for the
the Czech
Czech Republic
Republic
Italics shows
shows relations
Italics shows
relations tested only for the Czech Republic
Source:
own
Source:
own study
study
Source: own study
Quarterly,
Quarterly, (which
(which is
is presented
presented in
in diagram
diagram 1),
1), the
the forecast
forecast result
result (5
(5 or
or 33 possibilities)
possibilities) is
is
Quarterly, (which is presented in diagram 1), the forecast result (5 or 3 possibilities) is
confronted
confronted with:
with:
confronted
with:
-- the
the balance
balance that
that was
was calculated
calculated on
on the
the basis
basis of
of the
the next
next month
month survey,
survey,
-- the
the balance
balance that
that was
was calculated
calculated on
on the
the basis
basis of
of the
the survey
next month
survey,
after
- the balance that was calculated on the basis of the survey that
that took
took place
place 22 months
months after
- the
balance
calculated on the basis of the survey that took place 2 months after
the forecast
forecast that
was was
revealed,
the
was
revealed,
the
was
revealed,
the forecast
average of
of
balance
that was
was calculated
calculated on
on the
the basis
basis of
of the
the next
next month
month survey
survey and
and
-- the
average
balance
that
- the
of balance
that
was calculated
on the basis of the next month survey and
twoaverage
subsequent
balances
(quarterly
average),
two
subsequent
balances
(quarterly
average),
two
subsequent
balances
(quarterly
average),
the average
average
of balance
balance
that
was calculated
calculated
on the
the basis
basis of
of survey
-- the
of
that
was
on
survey that
that took
took place
place 22
- months
the average
of balance
that was
calculated on
the basis of survey
that took placeaverage).
2
months after
after the
the forecast
forecast was
was revealed
revealed and
and two
two subsequent
subsequent balances
balances (quarterly
(quarterly average).
months after the forecast was revealed and two subsequent balances (quarterly average).
12
12
It
the existence
existence of
of the
the associations
It seems
seems reasonable
reasonable to
to check
check the
associations with
with longer
longer lag
lag for
for Poland
Poland starting
starting form
form 2008,
2008,
12
when
the
frequency
of
the
forecast
was
lowered.
However,
the
time
series
would
be
quite
short
here,
which
is
when
the
frequency
of
the
forecast
was
lowered.
However,
the
time
series
would
be
quite
short
here,
which
is
It seems reasonable to check the existence of the associations with longer lag for Poland starting form 2008,
why
the
idea
was
dropped.
why
was dropped.
whenthe
theidea
frequency
of the forecast was lowered. However, the time series would be quite short here, which is
why the idea was dropped.
WORKING PAPER No. 105
31
31
31
35
Empirical study on the forecast and inflation expectations
The first and the second options check whether there is an association between the
first and
theand
second
options check
whether
thereofisanswers)
an association
between
the
inflationThe
forecast
result
the expected
inflation
(balance
covering
only one
inflationOne
forecast
result
andlag
theisexpected
(balance
answers)
covering
only
survey.
or two
month
assumed.inflation
The third
and theoffourth
options
assume
thatone
the
survey. One
oraffects
two month
lag is assumed.
Thetothird
the fourth
optionsinflation
assume that
forecast
result
the structure
of answers
the and
question
on expected
up tothe
the
13
forecast
result
the structure
of answers
on expected
inflation up to the
time
when
the affects
new forecast
is revealed.
Again to
thethe
lagquestion
is assumed
.
13
time when the new forecast is revealed. Again the lag is assumed .
Diagram 1. The associations presented quarterly
Diagram 1. The associations presented quarterly
Balances in
months:
Balances in
months:
Forecast A
revealed in month t
Forecast A
revealed in month t
t+1
t+1
Average
Average
of months
of months
t+1 –t+1
t+3– t+3
Forecast B
revealed in month t+3
Forecast B
revealed in month t+3
t
Average
Average
of months
of months
t+2 –t+2
t+4– t+4
5
t
t+2
t+2
t+3
t+3
t+4
t+4
t+5
t+5
Source: own study.
Source: own study.
With the balances of the answers given with lower frequency that the forecast result
With the
balancesisofalso
thetested
answers
given
with result
lower is
frequency
the two
forecast
resultit
the alternative
approach
– the
forecast
repeated that
for the
(or three,
the alternative
is also tested
– the forecast
result
two (orthe
three, it
depends
on theapproach
forecast frequency)
subsequent
months
andisisrepeated
assumedfor
to the
influence
depends separately
on the forecast
frequency)
subsequent
monthsalso
andreferred
is assumed
influence
the
answers
in each
month (repeated
version,
to astomonthly
approach).
answersThe
separately
in each
month (repeated
version,
referred
to aswere
monthly
approach).
time series
on balances
of answers
on the also
survey
questions
the same
for four
TheAstime
on balances
of answers
ondirection
the survey
were
the same
countries.
the series
time series
on the level
and the
of questions
expectations
differ,
the for four
countries. Astested
the time
the level
and the12direction
of expectations
the
associations
alsoseries
coveron
various
versions:
for the Czech
Republic,differ,
4 for Hungary,
20
associations
tested5).also cover various versions: 12 for the Czech Republic, 4 for Hungary, 20
for
Poland (table
for Poland (table 5).
13
I.e. July forecast result is confronted with the quarterly average balances calculated for August, September and
October
(1 month lag) or calculated for September, October and November. It means the assumption that the
13
I.e. July forecast result is confronted with the quarterly average balances calculated for August, September and
households need at least a month to read the message of the forecast.
October (1 month lag) or calculated for September, October and November. It means the assumption that the
households need at least a month to read the message of the forecast.
36
32
32
N a t i o n a l
B a n k
o f
P o l a n d
Empirical study on the forecast and inflation expectations
The
The restriction
restriction on
on lags
lags is
is placed
placed due
due to
to surveys
surveys frequency.
frequency. In
In the
the Czech
Czech Republic
Republic and
and
The restriction on lags is placed due to surveys frequency. In the Czech Republic and
Hungary
Hungary the
the survey
survey was
was held
held aa month
month before
before the
the forecast
forecast disclosure.
disclosure. As
As aa consequence,
consequence, in
in
Hungary the survey was held a month before the forecast disclosure. As a consequence, in
each
case
where
the
lag
is
not
described,
2M
lag
is
considered.
Still,
for
Poland
having
each case where the lag is not described, 2M lag is considered. Still, for Poland having
each case where the lag is not described, 2M lag is considered. Still, for Poland having
monthly
monthly observation
observation on
on expectation,
expectation, aa similar
similar approach
approach as
as with
with the
the balances
balances is
is applied.
applied.
monthly observation on expectation, a similar approach as with the balances is applied.
Table
and directions
directions of
of expectations
expectations change
change
Table 5.
5. Versions
Versions of
of association
association tested
tested for
for level
level and
Table 5. Versions of association tested for level and directions of expectations change
CNB
CNB
(12)
CNB
(12)
(12)
NBH
NBH
(4)
NBH
(4)
(4)
NBP
NBP
(20)
NBP
(20)
(20)
Association
Association tested
tested
Central
path
(case
with
5
possibilities)/direction
of
Association
tested expectations
Central path (case with 5 possibilities)/direction
of consumers
consumers
expectations change
change
business
expectations
Central
path
(case
with
5
possibilities)/direction
of
consumers
expectations
change
Central path (case with 5 possibilities)/direction of business expectations change
change
consumers
expectations
change
Central
expectations
change
Central path
path (case
(case with
with 533 possibilities)/direction
possibilities)/direction of
of business
consumers
expectations
change
business
expectations
change
Central
path
(case
with
3
possibilities)/direction
of
consumers
expectations
change
Central path (case with 3 possibilities)/direction of business expectations change
path/direction
of
consumers
expectations
change
Policy
of
business
expectations
change
Central
path
(case
with
3
possibilities)/direction
Policy path/direction of consumers expectations change
Policy
change
consumers
expectations
change
Policy path/direction
path/direction of
of business
business expectations
expectations
change
Central
path
(case
with
5
possibilities)/consumers
Policy
path/direction
of
business
expectations
change
Central path (case with 5 possibilities)/consumers expectations
expectations (limited
(limited to
to 2007Q2)
2007Q2)
possibilities)/
business
expectations
Central
path
(case
with
5
possibilities)/consumers
expectations
(limited
to
2007Q2)
Central path (case with 5 possibilities)/ business expectations
possibilities)/consumers
expectations (limited
(limited to
to 2007Q2)
2007Q2)
Central
business expectations
Central path
path (case
(case with
with 533 possibilities)/
possibilities)/consumers
expectations
possibilities)/business
expectations
Central
path
(case
with
3
possibilities)/consumers
expectations
(limited
to
2007Q2)
Central path (case with 3 possibilities)/business expectations
Policy
(limited
(case with 3expectations
possibilities)/business
expectations
Centralpath/
path consumers
Policy
path/
consumers
expectations
(limited to
to 2007Q2)
2007Q2)
Policy
path/
business
expectations
consumers
expectations
(limited
to
2007Q2)
Policy path/ business expectations
Central
path
(case
with
5
possibilities)/direction
Policy
path/
business
expectations
Central path (case with 5 possibilities)/direction of
of consumers
consumers expectations
expectations change
change
Central
path
(case
with
3
possibilities)/direction
of
consumers
expectations
5
Central path (case with 3 possibilities)/direction of consumers expectations change
change
Central
(limited
possibilities)/direction ofexpectations
consumers expectations
change
Central path
path (case
(case with
with 553 possibilities)/consumers
possibilities)/consumers
expectations
(limited to
to 2007Q3)
2007Q3)
expectations
(limited
to
2007Q3)
3
Central
path
(case
with
5
possibilities)/consumers
Central path (case with 3 possibilities)/consumers expectations (limited to 2007Q3)
possibilities)/direction
consumers
change
Central
(limited to 2007Q3)
Central path
path (case
(case with
with 355 possibilities)/consumers
possibilities)/direction of
ofexpectations
consumers expectations
expectations
change (1Mlag)
(1Mlag)
Central
path
(case
with
5
possibilities)/direction
of
consumers
expectations
change
(1Mlag)
Central path (case with 5 possibilities)/direction of consumers expectations change (2Mlag)
(2Mlag)
Central
path
(case
with
3
possibilities)/direction
of
consumers
expectations
change
(1Mlag)
5
(2Mlag)
Central path (case with 3 possibilities)/direction of consumers expectations change (1Mlag)
(2Mlag)
Central
path
(case
with
3
possibilities)/direction
of
consumers
expectations
change
(1Mlag)
Central path (case with 3 possibilities)/direction of consumers expectations change (2Mlag)
Central
path
(case
with
5
possibilities)/consumers
expectations
(1Mlag)
of
consumers
expectations
change
(2Mlag)
3
possibilities)/direction
Central path (case with 5 possibilities)/consumers expectations (1Mlag)
(2Mlag)
Central
Central path
path (case
(case with
with 55 possibilities)/consumers
possibilities)/consumers expectations
expectations (1Mlag)
(2Mlag)
3
(1Mlag)
Central
path
(case
with
5
possibilities)/consumers
expectations
(2Mlag)
Central path (case with 3 possibilities)/consumers expectations (1Mlag)
(2Mlag)
Central
path
(case
with
3
possibilities)/consumers
expectations
(1Mlag)
Central path (case with 3 possibilities)/consumers expectations (2Mlag)
5
possibilities)/
average
consumers
expectations
(1Mlag)
Central
path
(case
with
3
possibilities)/consumers
expectations
(2Mlag)
Central path (case with 5 possibilities)/ average consumers expectations (1Mlag)
with
5
possibilities)/
average
(2Mlag)
consumers
expectations
(1Mlag)
Central
path
(case
Central path (case with 5 possibilities)/ average consumers expectations (2Mlag)
(1Mlag)
Central
Central path
path (case
(case with
with 533 possibilities)/
possibilities)/ average
average consumers
consumers expectations
expectations (2Mlag)
(1Mlag)
(2Mlag)
Central
path
(case
with
3
possibilities)/
average
consumers
expectations
(1Mlag)
Central path (case with 3 possibilities)/ average consumers expectations (2Mlag)
repeated
(case
with
5
possibilities)/direction
of
consumers
expectations
change (1Mlag)
(1Mlag)
Central
path
(case
with
3
possibilities)/
average
consumers
expectations
(2Mlag)
Central path repeated (case with 5 possibilities)/direction of consumers expectations change
expectations
change
(1Mlag)
Central
path
repeated
(case
with
3
possibilities)/direction
of
consumers
5
Central path repeated (case with 3 possibilities)/direction of consumers expectations change (1Mlag)
Central
(1Mlag)
Central path
path repeated
repeated (case
(case with
with 553 possibilities)/direction
possibilities)/direction of
of consumers
consumers expectations
expectations change
change (2Mlag)
(2Mlag)
Central
path
repeated
(case
with
3
possibilities)/direction
of
consumers
expectations
change
(2Mlag)
5
Central path repeated (case with 3 possibilities)/direction of consumers expectations change (2Mlag)
Central
(1Mlag)
possibilities)/direction
consumers expectations
change (2Mlag)
Central path
path repeated
repeated (case
(case with
with 535 possibilities)/consumers
possibilities)/consumersofexpectations
expectations
(1Mlag)
Central
path
repeated
(case
with
3
possibilities)/consumers
expectations
(1Mlag)
5
Central path repeated (case with 3 possibilities)/consumers expectations (1Mlag)
Central
(1Mlag)
Central path
path repeated
repeated (case
(case with
with 355 possibilities)/consumers
possibilities)/consumers expectations
expectations (2Mlag)
(2Mlag)
Central
path
repeated
(case
with
3
possibilities)/consumers
expectations
(2Mlag)
5
Central path repeated (case with 3 possibilities)/consumers expectations (2Mlag)
Central path
repeated (case with 3 possibilities)/consumers expectations (2Mlag)
Source:
Source: own
own study
study
Source: own study
5
The
The Hungarian
Hungarian case
case presents
presents the
the basic
basic scope
scope of
of the
the research.
research. The
The associations
associations between
between
The Hungarian case presents the basic scope of the research. The associations between
the relation
relation of
of the
the central
central path
path and
and the
the inflation
inflation target
target (described
(described as
as central
central path)
path) in
in two
two
the
the relation of the central path and the inflation target (described as central path) in two
versions (5
(5 and
and 33 possibilities)
possibilities) are
are tested
tested with
with inflation
inflation expectations
expectations (levels
(levels up
up to
to 3Q2007,
3Q2007,
versions
versions (5 and 3 possibilities) are tested with inflation expectations (levels up to 3Q2007,
direction of
of the
the change).
change). The
The case
case of
of the
the Czech
Czech Republic
Republic covers
covers also
also additional
additional cases:
cases:
direction
direction of the change). The case of the Czech Republic covers also additional cases:
associations of
of policy
policy path
path and
and expectations
expectations as
associations
as well
well as
as business
business expectations.
expectations. In
In Poland
Poland
associations of policy path and expectations as well as business expectations. In Poland
WORKING PAPER No. 105
33
33
33
37
Empirical study on the forecast and inflation expectations
having the time series of monthly frequency was confronted with the forecast that was made
quarterly or 3 times a year. This is why in Polish case a few approaches were applied. They
were the same as while testing relations for balances of answers.
The poorest access to the data for Romania means that only the correlation of the
inflation forecast and balances of answers was tested.
5.4 Measures applied
The correlation between inflation forecast and expectations measures a degree of
monotonic association between these two variables. This is one of the simplest statistical
relations, however, here it is imposed by the range of data availability that also implies that
only non-parametric correlation measures can be used. The diagram with the statistical
measures of association is presented below. A range of measures, depending on the character
of time series – qualitatively expressed forecast results are accompanied by qualitative
5
(directions of change) or quantitative (levels, balances) information on expectations - is used.
The Spearman and Kendall correlation coefficients are suitable for the situation where the
data are in the form of ranks or are on ordinal scale. A chi-square test and related statistics can
be used when data are enumerative (counts or frequencies). The statistical hypotheses that are
to be tested are also presented in diagram 2.
Under a null hypothesis there is no monotonic association between inflation forecast
results and expectations or the two variables are independent for  = 0.05. It should be noticed
that inflation forecast results as well as inflation expectations are expressed in terms described
in the previous part of this section. It does not necessarily mean their levels.
The Spearman rank correlation coefficient is the basic measure of associations that can
be used here regardless of the way in which the expectations are expressed. It is the only
measure applied in the case of testing relations between inflation forecast and balances of
answers. The rationale for using other measures in the case of levels of expectations and the
direction of changes is poorer time series quality.
38
34
N a t i o n a l
B a n k
o f
P o l a n d
Empirical study on the forecast and inflation expectations
Diagram 2. Statistics and range of associations tested
Statistics
The Spearman Rank
Correlation Coefficient (rs)
Kendall Tau
Correlation Coefficient ()
H0: s = 0
H 0:  = 0
H1: s  0
H1:   0
Chi-Square Test for
Independence
H0: the inflation forecast and
inflation expectations are
independent of each other.
H1: the inflation forecast and
inflation expectations are
dependent.
Forecast expressed as the
relation to the goal.
Expectations – levels.
Forecast expressed as the
relation to the goal.
Expectations – balances.
5
Both variables of qualitative
nature (ordinal scale):
− forecast – relation to the goal,
− expectations – directions.
Source: own study.
5.5 Results and comments
The results are presented in tables 6-11. They are divided into 3 groups. The starting
point is testing associations between the inflation forecast results and expectations expressed
as the balances of answers on survey questions. As the same time series presented in Business
and Consumer Surveys Results by the European Commission are available for the four
countries, this time the Romanian case can be covered as well as the three others. This
approach offers greater comparability of results. The Spearman rank correlation coefficient is
given it table 6.
Regardless of the broad range of the associations tested, none of them proved to be
statistically important for the significance level of 5% in the case of the CNB. The null
hypothesis is thus not rejected. This result is quite interesting considering preliminary
assumption that the association here should be quite strong because of consistency of the
CNB in IFT implementation. Chart 1 shows the time series for central path for 3 possibilities
(left scale) and quarterly average of balances with 1M lag (right scale). The inconsistency
between the time series appears to be the strongest from 2002 to 2004. A few explanations
can be given. First of all, the introduction of unconditional forecast in mid-2002 might have
changed the perception of the forecast. However, an almost 3-year-period seems to be too
35
long for learning how to interpret the forecast. The second explanation is connected with the
WORKING PAPER No. 105
way in which the CNB formulated the main goal up to the end of 2005: only the range for
39
The Czech Republic
Quarterly
Central path (5 possibilities)
39
/balances 1M lag
Central path (5 possibilities)
39
/balances 2M lag
Central path (3 possibilities)
39
/balances 1M lag
Central path (3 possibilities)
39
/balances 2M lag
Central path (5 possibilities)
39
/average balance 1M lag
Central path (5 possibilities)
39
average balances 2M lag
Central path (3 possibilities)
39
/average balance 1M lag
Central path (3 possibilities)
39
/average balances 2M lag
Monthly
Central path repeated (case with 5
117
possibilities)/balances 1M lag
Central path repeated (case with 5
117
possibilities)/balances 2M lag
Central path repeated (case with 3
117
possibilities)/balances 1M lag
Central path repeated (case with 3
117
possibilities)/balances 2M lag
Statistically important associations are bolded.
Source: own calculations.
N
1.36
0.98
1.45
0.13
0.09
0.13
0.71
0.12
1.07
0.39
0.06
0.10
0.63
0.16
0.03
0.10
0.57
0.09
0.42
0.18
0.03
0.07
0.81
0.13
rs
N a t i o n a l
0.148
0.326
0.176
0.286
0.484
0.695
0.532
0.678
0.872
0.571
0.860
0.423
p
B a n k
0.50
0.49
0.47
113
113
113
36
0.51
113
0.50
0.53
38
38
0.50
38
0.51
0.52
38
38
0.51
38
0.54
0.51
38
38
rs
N
5.66
5.94
6.10
6.17
3.44
3.55
3.83
3.73
3.46
3.62
3.57
3.59
t(N2)
0.000
0.000
0.000
0.000
0.001
0.001
0.000
0.000
0.001
0.000
0.001
0.000
p
Poland
Hungary
40
Variables:
t(N2)
5
Table 6. The associations between the forecasts results and the balances of the answers
0.25
0.16
0.22
0.32
22
22
22
22
77
77
77
77
22
22
0.34
0.27
0.37
0.30
0.31
0.30
0.32
0.21
22
22
rs
N
3.16
2.45
3.43
2.77
1.46
1.41
1.52
1.51
1.01
0.73
1.15
0.98
t(N2)
0.002
0.016
0.000
0.007
0.160
0.174
0.143
0.145
0.322
0.471
0.261
0.340
p
0.15
0.56
0.12
0.46
22
22
22
22
65
65
65
65
22
22
0.24
0.32
0.30
0.37
0.28
0.39
0.36
0.60
22
22
rs
N
1.96
2.73
2.51
3.14
1.28
1.89
1.71
2.32
0.55
2.99
0.69
3.34
t(N-2)
0.054
0.008
0.014
0.002
0.214
0.073
0.103
0.030
0.590
0.007
0.495
0.003
p
Empirical study on the forecast and inflation expectations
o f
P o l a n d
Romania
hypothesis is thus not rejected. This result is quite interesting considering preliminary
assumption that the association here should be quite strong because of consistency of the
Empirical study on the forecast and inflation expectations
CNB in IFT implementation. Chart 1 shows the time series for central path for 3 possibilities
(left scale) and quarterly average of balances with 1M lag (right scale). The inconsistency
between the time series appears to be the strongest from 2002 to 2004. A few explanations
can be given. First of all, the introduction of unconditional forecast in mid-2002 might have
changed the perception of the forecast. However, an almost 3-year-period seems to be too
long for learning how to interpret the forecast. The second explanation is connected with the
way in which the CNB formulated the main goal up to the end of 2005: only the range for
inflation was given. This is not consistent with expressing inflation forecast result in 5
possibilities, which would explain the lack of relations. On the other hand, it is consistent with
3 possibilities and none of these options proved to be statistically important as well.
Chart 1. Central path and balances for the Czech Republic







5

























































Source: own work.
Table 7. The associations between policy path and balances for the CNB
Table 7. The associations between policy path and balances for the CNB
Quarterly
Quarterly
Policy path / balances t+1
Policy
Policy path
path // balances
balances t+1
t+2
Policy
path
/
t+2
Policy path / balances
average balances
t+1
Policy
path
/
average
balances
Policy path / average balances t+1
t+2
Policy
path
/
average
balances
t+2
Monthly
37
Monthly
Policy path repeated/ balances t+1
Policy
Policy path
path repeated/
repeated/ balances
balances t+1
t+2
Policy
path
repeated/
balances t+2
Statistically important associations are bolded.
Statistically
associations are bolded.
Source: ownimportant
calculations.
Source: own calculations.
N
N
rs
rs
t(N-2)
t(N-2)
p
p
34
34
34
34
34
34
34
34
0.57
0.57
0.30
0.30
0.38
0.38
0.34
0.34
3.88
3.88
1.78
1.78
2.32
2.32
2.05
2.05
0.000
0.000
0.084
0.084
0.027
0.027
0.048
0.048
102
102
102
102
0.40
0.40
0.36
0.36
4.36
4.36
3.82
3.82
0.000
0.000
0.000
0.000
The third explanation is connected with the nature of unconditional forecast. In the
The third explanation is connected with the nature of unconditional forecast. In the
case of unconditional forecast it is the policy path that may guide market expectations
case of unconditional forecast it is the policy path that may guide market expectations
directly. The Spearman rank correlation coefficients for policy path and balances are
WORKING
PAPER
No. Spearman
105
directly.
The
rank correlation coefficients for policy path and balances are
presented in table 7. The coefficients are statistically important in 5 out of 6 cases and their
presented in table 7. The coefficients are statistically important in 5 out of 6 cases and their
strength is moderate. It means that the consumers in Czech Republic take into consideration
41
Policy path repeated/ balances t+1
Policy path repeated/ balances t+2
Statistically important associations are bolded.
Source: own calculations.
102
102
0.40
0.36
4.36
3.82
0.000
0.000
Empirical study on the forecast and inflation expectations
The third explanation is connected with the nature of unconditional forecast. In the
case of unconditional forecast it is the policy path that may guide market expectations
directly. The Spearman rank correlation coefficients for policy path and balances are
presented in table 7. The coefficients are statistically important in 5 out of 6 cases and their
strength is moderate. It means that the consumers in Czech Republic take into consideration
the policy path, not the forecast itself, which is in line with the central bank’s intentions.
However, the coefficients are positive. In the case of the Czech Republic the decisions where
in line with the policy path. Considering at least a month lag, it would mean that the results
show the correlation between the decision of the MPC and the balances of the answers.
However, the policy path explanation is not the only possible one. Previously
described options consisting in searching for possible explanation for the period 2002-2004
are also valuable in the light of the additionally tested relations. The same pair of variables
was tested for the case of the CNB for the period 2005-2010. Here the relations between
5
inflation forecast results and balances exist and have moderate strength, whereas for the
policy path they are even stronger (around 0.7).
In Hungary the monotonic association between the inflation forecast results and the
balances of answers to the survey questions on expected inflation exists. The Spearman rank
correlation coefficients are positive, but the strength of association is moderate, around 0.5 for
every pair of variables. Again the result is surprising, regarding the practice of the HNB. This
time the explanation seems to be simpler: the consumers may not perceive the projection as
the tool of deriving the next step of the monetary authorities but rather as an indicator of the
future inflation. When the forecast is above the target the expectations (here balances) rise.
The households do not expect the central bank to dampen inflation.
In Poland, only the associations of variables analyzed monthly are statistically
important, however, they are quite week. The strength is consistent with the practice of the
NBP. The forecast is the decisive input neither in a decision-making process nor in
38frequency of the forecast was reduced.
communication of its rationale. Moreover, the
Regardless of the sound explanation of the central bank of such a step, it manifests that the
forecast is probably loosing its importance in Polish monetary policy.
In Romania three out of four associations given in monthly cycle and 3 out of 8
relations in quarterly cycle are statistically important (positive coefficients of weak or
moderate strength). These 3 relations assume 1 month lag. Here the interpretation is limited
because of the reasons that were already described: the NBR operated in different market
condition than the other banks.
42
The next step in quantitative analysis is testing the existence of interdependences
N a t i o n a l
B a n k
o f
P o l a n d
between inflation forecast results and the level of inflation expectations or their direction of
change. Table 8 presents the Spearman and Kendal coefficients. The second one can be
moderate strength). These 3 relations assume 1 month lag. Here the interpretation is limited
Empirical study on the forecast and inflation expectations
because of the reasons that were already described: the NBR operated in different market
condition than the other banks.
The next step in quantitative analysis is testing the existence of interdependences
between inflation forecast results and the level of inflation expectations or their direction of
change. Table 8 presents the Spearman and Kendal coefficients. The second one can be
interpreted in the terms of probability of having concordant pairs. Both of them indicate the
existence (or not) of monotonic association between the two variables.
Table 8. The Spearman rank and Kendall tau correlation coefficients
Variables:
N
rs
The Czech Republic
Central path (case with 5 possibilities)/consumers
24
0.418
expectations
Central path (case with 5 possibilities)/direction
39
-0.019
of consumers expectations change
Central path (case with 5 possibilities)/business
39
0.277
expectations
Central path (case with 5 possibilities)/direction
39
0.505
of business expectations change
Central path (case with 3 possibilities)/consumers
24
0.398
expectations
Central path (case with 3 possibilities)/direction
39
-0.046
of consumers expectations change
Central path (case with 3 possibilities)/business
39
0.232
expectations
Central path (case with 3 possibilities)/direction
39
0.478
of business expectations change
Policy path/ consumers expectations
19
0.258
Policy path/direction of consumers expectations
34
0.084
change
Policy path/ business expectations
34
0.587
Policy path/direction of business expectations
34
0.507
change
Hungary
Central path (case with 5 possibilities)/consumers
22
0.539
expectations
Central path (case with 5 possibilities)/direction
38
0.136
of consumers expectations change
Central path (case with 3 possibilities)/consumers
22
0.600
expectations
Central path (case with 3 possibilities)/direction
0.097
3938
of consumers expectations change
Poland
Central path (case with 5 possibilities)/consumers
22
0.396
expectations (1Mlag)
Central path (case with 5 possibilities)/direction
22
0.609
of consumers expectations change (1Mlag)
Central path (case with 5 possibilities)/consumers
22
0.439
expectations (2Mlag)
Central path (case with 5 possibilities)/direction
22
0.071
of consumers expectations change (2Mlag)
Central path (case with 3 possibilities)/consumers
22
0.227
expectations (1Mlag)
Central path (case with 3 possibilities)/direction
WORKING PAPER No. 105
22
0.704
of consumers expectations change (1Mlag)
Central path (case with 3 possibilities)/consumers
22
0.294
expectations (2Mlag)
t(N-2)
prs

p
2.157
0.042
0.352
0.016
-0.113
0.911
-0.007
0.944
1.754
0.088
0.204
0.067
3.559
0.001
0.436
0.000
2.034
0.054
0.324
0.026
-0.278
0.783
-0.035
0.753
1.453
0.155
0.173
0.119
3.311
0.002
0.420
0.000
1.103
0.285
0.184
0.269
0.475
0.638
0.075
0.532
4.100
0.000
0.499
0.000
3.331
0.002
0.454
0.000
2.865
0.010
0.430
0.005
0.823
0.416
0.120
0.288
3.355
0.003
0.493
0.001
0.582
0.564
0.091
0.418
1.931
0.068
0.314
0.041
3.436
0.003
0.553
0.000
2.183
0.041
0.353
0.021
0.317
0.755
0.067
0.662
1.041
0.310
0.179
0.244
4.430
0.000
0.656
0.000
1.376
0.184
0.238
0.121
5
43
Central path (case with 5 possibilities)/direction
of consumers expectations change (1Mlag)
Central path (case with 5 possibilities)/consumers
expectations (2Mlag)
Central path (case with 5 possibilities)/direction
of consumers expectations change (2Mlag)
Central path (case with 3 possibilities)/consumers
expectations (1Mlag)
Central path (case with 3 possibilities)/direction
of consumers expectations change (1Mlag)
Central path (case with 3 possibilities)/consumers
expectations (2Mlag)
Central path (case with 3 possibilities)/direction
of consumers expectations change (2Mlag)
Central path (case with 5 possibilities)/average
consumers expectations (1Mlag)
Central path (case with 5 possibilities)/average
consumers expectations (2Mlag)
Central path (case with 3 possibilities)/average
consumers expectations (1Mlag)
Central path (case with 3 possibilities)/average
consumers expectations (2Mlag)
Central path repeated (case with 5
possibilities)/consumers expectations (1Mlag)
Central path repeated (case with 5
possibilities)/direction of consumers expectations
change (1Mlag)
Central path repeated (case with 5
possibilities)/consumers expectations (2Mlag)
Central path repeated (case with 5
possibilities)/direction of consumers expectations
change (2Mlag)
Central path repeated (case with 3
possibilities)/consumers expectations (1Mlag)
Central path repeated (case with 3
possibilities)/direction of consumers expectations
change (1Mlag)
Central path repeated (case with 3
possibilities)/consumers expectations (2Mlag)
Central path repeated (case with 3
possibilities)/direction of consumers expectations
change (2Mlag)
Statistically important associations are bolded.
Source: own calculations.
5
22
0.609
22
0.439
22
3.436
0.003
0.553
0.000
Empirical study on the forecast and inflation expectations
2.183
0.041
0.353
0.021
0.071
0.317
0.755
0.067
0.662
22
0.227
1.041
0.310
0.179
0.244
22
0.704
4.430
0.000
0.656
0.000
22
0.294
1.376
0.184
0.238
0.121
22
0.101
0.456
0.653
0.093
0.543
22
0.396
1.931
0.068
0.314
0.041
22
0.439
2.183
0.041
0.353
0.021
22
0.227
1.041
0.310
0.179
0.244
22
0.294
1.376
0.184
0.238
0.121
77
0.378
3.534
0.001
0.294
0.000
76
0.002
0.015
0.988
0.003
0.973
77
0.376
3.519
0.001
0.300
0.000
75
-0.059
-0.508
0.613
-0.051
0.519
77
0.241
2.150
0.035
0.196
0.012
76
0.069
0.591
0.556
0.062
0.427
77
0.257
2.306
0.024
0.217
0.005
75
-0.010
-0.085
0.932
-0.009
0.905
Taking into consideration the Spearman rank correlation coefficient for the Czech
Republic, 5 out of 12 associations are statistically important, 2 out of 4 for Hungary and 9 out
of 20 for Poland. The Kendall tau correlation coefficient gives similar results for the
significance level of 5% (for the Czech Republic40one more relation is statistically important).
However, the strength of associations is not imposing: a relative indication of the degree of
association between inflation forecast result and inflation expectation does never approach 1,
and rarely approaches 0.7. Once it attains 0.704 (the association of the central path, case with
3 possibilities and the direction of consumers expectations change with1M lag for Poland) . It
is the strongest relation, considering both measures of rank correlation. The coefficients that
are statistically important are positive, which means that inflation forecast results and inflation
44
expectations are changing in the same direction.
N a t i o n a l
B a n k
Table 9. Chi-square test for independence - the Czech Republic
o f
P o l a n d
association between inflation forecast result and inflation expectation does never approach 1,
and rarely approaches 0.7. Once it attains 0.704 (the association of the central path, case with
Empirical study on the forecast and inflation expectations
3 possibilities and the direction of consumers expectations change with1M lag for Poland) . It
is the strongest relation, considering both measures of rank correlation. The coefficients that
are statistically important are positive, which means that inflation forecast results and inflation
expectations are changing in the same direction.
Table 9. Chi-square test for independence - the Czech Republic
2
df
p
Statistics:
Central path (case with 5 possibilities)/direction of consumers expectations change
Pearson's 2
9.3426
df=8
p=0.314
2
 of the highest credibility
9.2685
df=8
p=0.320
Cramér's V
0.3461
Central path (case with 5 possibilities)/direction of business expectations change
Pearson's 2
12.23724
df=8
p=0.140
2of the highest credibility
14.63492
df=8
p=0.066
Cramér's V
0.39609
Central path (case with 3 possibilities)/direction of consumers expectations change
Pearson's 2
4.871379
df=4
p=0.300
2
 of the highest credibility
5.026195
df=4
p=0.284
Cramér's V
0.24990
Central path (case with 3 possibilities)/direction of business expectations change
Pearson's 2
df=4
p=0.046
9.686688
2of the highest credibility
df=4
p=0.025
11.07218
Cramér's V
0.35240
Policy path/direction of consumers expectations change
Pearson's 2
0.64866
df=4
p=0.957
2of the highest credibility
0.65117
df=4
p=0.957
Cramér's V
0.09766
Policy path/direction of business expectations change
Pearson's 2
df=4
p=0.010
13.18307
2of the highest credibility
df=4
p=0.008
13.67578
Cramér's V
0.44030
Statistically important associations are bolded.
Source: own calculations.
5
The calculation presented in table 8 is consistent with those presented in table 6. For
the Czech Republic 5 associations are statistically important but only one of them represents
households expectations. Business expectations were not represented in EC surveys. Here the
results suggest that there is an interdependence between inflation forecast result and
expectations of companies. At the same time no relation is found for the consumers
expectations except for the central path with 5 possibilities and the levels of consumers
expectations. There are two interdependences for Hungary, which is not in contradiction to
41
previously obtained results – here the expectation comes form own survey of the HNB. For
Poland 4 version with repeated forecast are statistically important but the relation is of low
strength which confirms the results obtained for balances. 5 possibilities case for inflation
forecast is also correlated with average of the inflation expectations for 1M and 2M lag. Two
other pair of variables (central path with 5 possibilities/direction of expectations change,
central path with 3 possibilities/ direction of expectations change) are also statistically
important.
WORKING PAPER No. 105
The last scope of the empirical part concerns the context of the time series of the
lowest quality: inflation forecast and expectations are now expressed on ordinal scale. The
45
Poland
4 version
with
repeated
forecast
are statistically
the relation
is of low
previously
obtained
results
– here
the expectation
comesimportant
form ownbut
survey
of the HNB.
For
Poland 4 version with repeated forecast are statistically important but the relation is of low
strength
confirms
the results
obtained
for balances.
5 possibilities
for inflation
Poland 4which
version
with repeated
forecast
are statistically
important
but thecase
relation
is of low
Empirical study on the forecast and inflation expectations
strength which confirms the results obtained for balances. 5 possibilities
case for inflation
forecast is
also confirms
correlatedthe
with
average
of thefor
inflation
expectations
for 1M
and
lag. Two
strength
which
results
obtained
balances.
5 possibilities
case
for2M
inflation
forecast is also correlated with average of the inflation expectations for 1M and 2M lag. Two
other
pairisof
variables
(central
withof5 the
possibilities/direction
of expectations
change,
forecast
also
correlated
with path
average
inflation expectations
for 1M and 2M
lag. Two
other pair of variables (central path with 5 possibilities/direction of expectations change,
centralpair
path
3 possibilities/
direction
expectations change)ofare
also statistically
other
ofwith
variables
(central path
with 5of
possibilities/direction
expectations
change,
central path with 3 possibilities/ direction of expectations change) are also statistically
important.
central
path with 3 possibilities/ direction of expectations change) are also statistically
important.
The last scope of the empirical part concerns the context of the time series of the
important.
The last scope of the empirical part concerns the context of the time series of the
lowest quality:
and expectations
arethe
now
expressed
ordinal
scale.
The
The lastinflation
scope offorecast
the empirical
part concerns
context
of theontime
series
of the
lowest quality: inflation forecast and expectations are now expressed on ordinal scale. The
interdependences
between
them have
already beenare
tested.
the alternative
is used:
lowest quality: inflation
forecast
and expectations
nowHere
expressed
on ordinalstatistic
scale. The
interdependences between them have already been tested. Here the alternative statistic is used:
chi-square
test forbetween
independence
and Cramér's
V astested.
the measure
of interdependence.
interdependences
them have
already been
Here the
alternative statistic is used:
chi-square test for independence and Cramér's V as the measure of interdependence.
chi-square test for independence and Cramér's V as the measure of interdependence.
Table 10. Chi-square test for independence - Hungary
Table 10. Chi-square
test for independence
- Hungary
2
df
p
Statistics
Table
10.
Chi-square
test
for
independence
-ofHungary
2
df expectations change
p
Statistics
Central path
(case with 5 possibilities)/direction
consumers
25.174706
df expectations
p
2Statistics
Centralpath
(case with 5 possibilities)/direction
of consumers
change
Pearson's
df=8
p=0.738
2
Central
(case with 5 possibilities)/direction
of consumers expectations
change
Pearson's
path
5.174706
df=8
p=0.738
2of
the highest
credibility
6.675948
p=0.571
2
Pearson's
2
5.174706
df=8
p=0.738
Cramér's
of the highest
credibility
6.675948
df=8
p=0.571
V
0.26093
2
Cramér's
of
the highest
6.675948
df=8
p=0.571
Vpath credibility
0.26093
Central
(case with 3 possibilities)/direction
of consumers expectations
change
2
Cramér's
V
0.26093
Central
path
(case
with
3
possibilities)/direction
of
consumers
expectations
change
Pearson's 
2.401389
df=4
p=0.662
2
Central
(case with 3 possibilities)/direction
of consumers expectations
change
Pearson's
path
2.401389
df=4
p=0.662
2of
the highest
credibility
3.390794
p=0.494
2
Pearson's
2
2.401389
df=4
p=0.662
Cramér's
of the highest
credibility
3.390794
df=4
p=0.494
V
0.17775
2

of
the
highest
credibility
3.390794
df=4
p=0.494
Source:
ownVcalculations.
Cramér's
0.17775
Cramér's
0.17775
Source:
ownVcalculations.
Source:
own
calculations.
Six relations are tested for the Czech Republic (table 9). For the significance level of
5
Six relations are tested for the Czech Republic (table 9). For the significance level of
5%, a null
says
thatfor
inflation
expectations
independent
inflation forecast
Sixhypothesis
relations are
tested
the Czech
Republicare
(table
9). For theofsignificance
level of
5%, a null hypothesis says that inflation expectations are independent of inflation forecast
results.
Thehypothesis
hypothesissays
assuming
that theexpectations
direction of are
the independent
change of inflation
expectations
5%, a null
that inflation
of inflation
forecast of
results. The hypothesis assuming that the direction of the change of inflation expectations of
households
not depend
on the
relation
of the central
path of inflation
forecast
to the of
results. The does
hypothesis
assuming
that
the direction
of the change
of inflation
expectations
households does not depend on the relation of the central path of inflation forecast to the
inflation
target
the transmission
horizon
cannot
be path
rejected.
As a result,
the measures
households
doeswithin
not depend
on the relation
of the
central
of inflation
forecast
to the
inflation target within the transmission horizon cannot be rejected. As a result, the measures
of interdependence:
Cramér's
V is not statistically
important.
The same
true for
inflation
target within
the transmission
horizon cannot
be rejected.
As a isresult,
thethe
measures
of interdependence: Cramér's V is not statistically important. The same is true for the
following
pairs of variables:
path
(case withimportant.
5 possibilities)/direction
business
of interdependence:
Cramér'scentral
V is not
statistically
The same is trueoffor
the
following pairs of variables: central path (case with 5 possibilities)/direction of business
expectations
change,
central path
(case
3 possibilities)/direction
of consumers
following pairs
of variables:
central
pathwith
(case
with 5 possibilities)/direction
of business
expectations change, central path (case with 3 possibilities)/direction of consumers
expectations change, policy
expectationsof
change.
centralpath/direction
path (case withof3consumers
possibilities)/direction
consumers
expectations change, policy path/direction of consumers expectations change.
For the
case ofpolicy
central
path (case with
3 possibilities)/direction
of business
expectations
change,
path/direction
of consumers
expectations change.
For the case of central path (case with 3 possibilities)/direction of business
expectations
change
and
policypath
path/direction
business expectations of
change,
for the
For the
case of
central
(case with 3ofpossibilities)/direction
business
expectations change and policy path/direction of business expectations change, for the
significance
of and
5%, policy
null hypothesis
can be
at the same
time –for
alternative
expectations level
change
path/direction
ofrejected,
businessand
expectations
change,
the
significance level of 5%, null hypothesis can be rejected, and at the same time – alternative
significance level of 5%, null hypothesis can be rejected, and at the same time – alternative
hypothesis saying that inflation expectations of business depend on the relation of inflation
42
forecast to the inflation target can be accepted.
42 However, the interdependences are quite
42
weak.
Table 11. Chi-square test for independence - Poland
46
2
df
p
Statistics
Central path (case with 5 possibilities)/direction of consumers expectations change
(1Mlag)
2
Pearson's 
13.20000
df=8
p=0.105
2of the highest credibility
16.48927
df=8
N a t i o n a l
B a n k
o p=0.035
f
P o l
Cramér's V
0.54772
Central path (case with 5 possibilities)/direction of consumers expectations change
(2Mlag)
a n d
forecast to the inflation target can be accepted. However, the interdependences are quite
Empirical study on the forecast and inflation expectations
weak.
Table 11. Chi-square test for independence - Poland
2
df
p
Statistics
Central path (case with 5 possibilities)/direction of consumers expectations change
(1Mlag)
2
Pearson's 
13.20000
df=8
p=0.105
2of the highest credibility
16.48927
df=8
p=0.035
Cramér's V
0.54772
Central path (case with 5 possibilities)/direction of consumers expectations change
(2Mlag)
Pearson's 2
6.678571
df=8
p=0.571
2of the highest credibility
7.849231
df=8
p=0.448
Cramér's V
0.38959
Central path (case with 3 possibilities)/direction of consumers expectations change
(1Mlag)
Pearson's 2
df=4
p=0.015
12.32000
2
 of the highest credibility
df=4
p=0.004
15.30433
Cramér's V
0.52915
Central path (case with 3 possibilities)/direction of consumers expectations change
(2Mlag)
Pearson's 2
3.775794
df=4
p=0.437
2
 of the highest credibility
5.112453
df=4
p=0.275
Cramér's V
0.29293
Central path repeated (case with 5 possibilities)/direction of consumers expectations
change (1Mlag)
Pearson's 2
5.730035
df=8
p=0.677
2
 of the highest credibility
5.418060
df=8
p=0.712
Cramér's V
0.19415
Central path repeated (case with 5 possibilities)/direction of consumers expectations
change (2Mlag)
Pearson's 2
6.237344
df=8
p=0.620
2
 of the highest credibility
5.935405
df=8
p=0.654
Cramér's V
0.203917
Central path repeated (case with 3 possibilities)/direction of consumers expectations
change (1Mlag)
Pearson's 2
1.468189
df=4
p=0.832
2
 of the highest credibility
1.427284
df=4
p=0.839
Cramér's V
0.098280
Central path repeated (case with 3 possibilities)/direction of consumers expectations
change (2Mlag)
2
Pearson's 
4.046806
df=4
p=0.399
2of the highest credibility
3.840728
df=4
p=0.427
Cramér's V
0.164252
Statistically important associations are bolded.
Source: own calculations.
5
Table10 presents the results for Hungary. Only two possible dependences were tested.
None of them proved to be statistically important. The similar outcome is for Poland (table
11) where only pair of variables: central path (case with 3 possibilities)/direction of
43
consumers expectations change (1Mlag) is depended.
The chi-square test for independence and measures derived from it offer the possibility
to test associations between variables that are expressed in an enumerative way. However,
WORKING PAPER No. 105
here the results should be interpreted with caution – in several cases theoretical (expected)
frequencies of certain events occurrence are lower than 5. The calculations above present only
47
Table10 presents the results for Hungary. Only two possible dependences were tested.
None of them proved to be statistically important. The similar outcome is for Poland (table
Empirical study on the forecast and inflation expectations
11) where only pair of variables: central path (case with 3 possibilities)/direction of
consumers expectations change (1Mlag) is depended.
The chi-square test for independence and measures derived from it offer the possibility
to test associations between variables that are expressed in an enumerative way. However,
here the results should be interpreted with caution – in several cases theoretical (expected)
frequencies of certain events occurrence are lower than 5. The calculations above present only
a rough picture of the interdependence between inflation forecast and expectations.
A few general conclusions can be drawn for cross-country comparison. 5 cases version
does not seem to reflect the association better than 3 cases version of the central path relation
to the goal. The same is true for lags. However, it should be noticed that in the case of the
Czech Republic and Hungary (levels and directions) the lag of 2M was imposed. The balances
of the answers to the survey questions are here the time series that provide the greater
comparability. In this case no version proved to be more suitable for 4 countries. However,
the analysis of expectations in terms of the direction of change proved to be the poorest
solution.
5
The Czech Republic case included also business expectations. Here the statistically
important relationships exist for 4 pairs of variables (levels and directions) whereas there is
only 1 for consumer expectations (taking into consideration the same set of variables). Better
understanding of the forecast message and higher reliability of surveys held among companies
are possible explanations here.
Testing the associations of the inflation forecast results and the expectations gave no
unambiguous results. This conclusion should lead to a broader economic interpretation that
will be given in the next section.
44
48
N a t i o n a l
B a n k
o f
P o l a n d
How the results can be interpreted
6. How the results can be interpreted
Insofar as the measures applied allow, it can be stated that the theoretical
underpinnings pointing to the inflation expectations formation being the most important
function of inflation forecast are not empirically confirmed. If they were, there should be an
association between the two variables. The interdependences are not statistically important or
the strength of association is at the most moderate. The empirical study resulted in no
unambiguous conclusions despite a broad range of measures applied and various pairs of
variables tested. Simultaneously, the study of the associations between the inflation forecast
and inflation expectations has a strong justification in theory. There are several reasons why
the study results may seem unsatisfactory with reference to theory.
First of all, the theory itself does not seem satisfactory for all researchers. It is
grounded in the NNS framework which was subject to criticism even before the global crisis
erupted in 2007. The reproaches inter alia are: weak empirical evidence, unrealistic Calvo
pricing model, and difficulty in obtaining parameters in New Keynesian Models (Toedter
2008). One argues that this theory does not determine the price level or the inflation rate
6
(Cochrane 2007). There are two alternative views of the central bankers and academics on the
future development of the central banking discussed. The first one - continuity view – does
not reject thoroughly the NNS. The central bank, while acknowledging the changes that have
occurred, interprets them essentially as unusual “shocks” in the context of a fairly stable
macroeconomic environment or model of the economy. By contrast, in the second, or newenvironment view, the central bank sees recent economic developments as related to and
reinforcing one another. Taken jointly, they would reflect a qualitative change in the
macroeconomic environment. Such a change is not so easily captured by economic models
and can point to a more significant modification to traditional policy guides. Such a central
bank is more likely to adjust its beliefs regarding the operation of the economy and to make
corresponding changes in its policy approach (Borio, English, Filardo 2003). The discussion
on monetary policy framework was enforced by the crisis that revealed some deficiencies in a
prevailing theory and monetary frameworks. The new consensus in the monetary policy is
being developed. It is broadly agreed that the way monetary policy transmission is described
in macroeconomic models needs to be fundamentally reworked. Transmission is
oversimplified, especially regarding various channels related to financial institutions (Frait,
Komárková, Komárek, 2011).
WORKING PAPER No. 105
45
49
How the results can be interpreted
The empirical research does not also confirm the assumption that the economic agents
form their expectations rationally. That assumption lies beneath the NNS framework. The
research actually proves that the expectations formation is of a hybrid nature: partly adaptive
and partly forward-looking (Gerberding 2001, Forsells, Kenny 2002). It confirms that the
inflation forecast cannot stand for the most important factor in expectations shaping process.
The composition of expectations is reflected in Hybrid New Keynesian Phillips Curve.
Simultaneously rational expectations hypothesis is being weaken taking the form of adaptive
learning hypothesis (Cierkoski 2008).
The drawbacks of the transmission models are directly connected with the main focus
of this study. They account for the second reason why the results do not confirm the theory.
If the model does not capture economic situation properly, the forecast is biased, of which the
monetary policy committee is aware. It does not perceive the forecast as the main input to the
decision-making process. This behavior may be intensified temporarily – especially in
turbulent periods, the MPC may underline the inadequacy of the model. The public realizes
the inconsistency of the MPC and does not take into consideration the inflation forecast while
making economic decisions.
6
The third reason why the results are not unambiguous is an extension of the second
one. The MPC may ignore the forecast message for any reason. Not only because it perceives
the model as inadequate. It may also rely on current issues, especially on exchange rate
fluctuations as the main inflation determinant while making decisions on interest rates. The
consequence in inflation forecast targeting is a necessary precondition of fulfilling external
function (expectations shaping). The three countries covered by the examination did not prove
to be consistent.
In this context it is necessary to focus for a while on the Czech Republic case. The
Czech National Bank really intended to target inflation forecast. Moreover, the forecast was
exposed in communication with the markets. One can assume that the interdependences
between the inflation forecast result and inflation expectations should be the strongest in the
case of the Czech Republic. In fact they are not. There are no statistically important
associations between the inflation forecast results and balances of the answers to the question
on expected inflation regardless of the version of the analysis. Chi-square test does not
confirm the dependence of the inflation forecast result and the expectations as well. The
Spearman and Kendall tau rank correlation coefficients are statistically important only once in
the case of households expectations. Only most of the relations regarding business
expectations are statistically important. The lack of associations for this country can be
50
46
N a t i o n a l
B a n k
o f
P o l a n d
on expected inflation regardless of the version of the analysis. Chi-square test does not
confirm the dependence of the inflation forecast result and the expectations as well. The
How the results can be interpreted
Spearman and Kendall tau rank correlation coefficients are statistically important only once in
the case of households expectations. Only most of the relations regarding business
expectations are statistically important. The lack of associations for this country can be
explained with the policy path announcement that should guide market expectations directly.
The versions with the central path should show46negative correlation coefficient assuming that
the public understand the forecast (rising interest rates will dampen economic activity and
expected inflation should decrease). In fact the associations proved to be positive.
Here the fourth reason why the results are not satisfactory appears: the understanding
of the central bank forecast and its actions by the public. Revealing forecast means only
increasing the institutional transparency of the central banker. It does not necessarily mean
that the forecast, especially the projection, is properly interpreted. The way of communicating
monetary policy stance, the decisions on interest rates as well as revealing inflation forecast
are also important. The forecast is not published as the single information. It appears at the
same time as the Inflation Report that includes more information on the central bank’s view
on the economy. It can influence the expectations as well as the forecast does. Moreover, the
information published can be perceived as more decisive for the MPC, which has been
already described. The simultaneity of information publishing is also important: assuming that
the conditionality of projection is understood, any change of the interest rate makes the
6
projection invalid. Here the coefficients are positive. It means that the forecast is not read in
terms of probable future policy action. It may be connected with the understanding of the
forecast as well as with the central bank credibility.
The fifth explanation is connected with the measures of expectations. Surveys results
regardless of their shortcomings, are commonly used by the central banks and referred to in
the Inflation Reports. No associations between quantitative surveys on households
expectations and inflation forecasts could be understood – the surveys were perceived as
unreliable and this is the reason why they were abandoned in the Czech Republic and
Hungary. However, if only methodological issues lied beneath the results, the
interdependences of the forecast results and inflation expectations for Poland, where
consistent data on expectations based are available, should be relatively stronger.
The sixth reason for not finding sound associations of the two variables may be
connected with the territorial scope of the research. The time series for the countries covered
by the examination are relatively short. The central banks still develop forecasting procedures.
The forecast has been revealed for a few years (especially in Poland and Romania) only. The
MPCs are not always consistent in IFT implementation, which discourages the public to take
the forecast into account.
The hypotheses in the introduction are not confirmed. The first one because of not
unambiguous results and the second one - stating that the existence of the interdependences
51
WORKING PAPER No. 105
47
by the examination are relatively short. The central banks still develop forecasting procedures.
The forecast has been revealed for a few years (especially in Poland and Romania)
only. The
How the results can be interpreted
MPCs are not always consistent in IFT implementation, which discourages the public to take
the forecast into account.
The hypotheses in the introduction are not confirmed. The first one because of not
unambiguous results and the second one - stating that the existence of the interdependences
between the inflation forecasts and inflation expectations does not depend on detailed
47
solutions in the field of forecasting inflation applied
by central banks – because the firs one is
not confirmed. The study covers four countries that were chosen due to certain similarities. In
spite of them, there are numerous divergences, even on the strategy level (exchange rate
regime) not mentioning the effectiveness of monetary policy. Forecasting tools and
procedures differ as well. The Czech central bank was the only one to produce unconditional
forecasts. The central banks seem to put different attention to the message of the forecast.
None of applied solutions proved to support the relationship between the inflation forecast
and expectations.
6
52
N a t i o n a l
B a n k
o f
P o l a n d
Conclusion
7. Conclusion
This paper contributes to the literature on modern monetary policy strategy and its
implementation. It focuses on the relation of the inflation forecast results and inflation
expectations. The expectations of households derived from surveys are taken into
consideration. There is a sound theoretical background showing the importance of inflation
expectations in the modern monetary policy as well as the usefulness of the inflation forecast
in shaping the expectations. These theoretical underpinnings accompanied with the behavior
of the central banks (producing and revealing inflation forecasts) led to the presentation of
two hypotheses. The first one assumes that the associations between the inflation forecast
results and consumers inflation expectations exist. In order to verify the first hypothesis the
review of the literature and the practice of the central bank is made. The crucial part of the
study is of an empirical nature – statistically important interdependences between several
pairs of variables are verified. Such relations were not found or they were of weak or
moderate strength. The results were similar regardless of the country. This is why the second
hypothesis assuming that no solution in the field of forecasting inflation may be perceived as
more appropriate may not be verified.
Regardless of the barriers of the study, it can be the first step in the direction to
7
measure this kind of relations. There are also a few possible extensions of the study:
extending the territorial scope of the research to countries that implement IFT for a longer
period of time, quantifying surveys data, and, finally, alternative coding of inflation forecast
message. The analysis could also take into consideration the credibility of the central bank
and ex post errors of the forecast.
WORKING PAPER No. 105
49
53
References
References:
1.
Batini N., Haldane A., Monetary Policy Rules and Inflation Forecast, w: Bank of England Quarterly
Bulletin February 1999, Bank of England, London 1999.
2.
Benkovskis K., The Role of Inflation Expectations in the New EU Member States: a Survey Based Results,
Czech Journal of Economics and Finance, 58, 2008, no. 7-8.
3.
Brzoza–Brzezina M., Kot A., The Relativity Theory Revised: Is Publishing Interest Rates Forecast Really so
Valuable?, National Bank of Poland Working Papers No. 052/2008.
4.
Calvo G., Staggered Prices in Utility-Maximizing Frameworks, Journal of Monetary Economics 12/1983.
5.
Cecchetti S.G., Krause S., Central Bank Structure, Policy Efficiency, and Macroeconomic Performance:
Exploring Empirical Relationships, The Federal Reserve Bank of St. Louis Working Paper Series,
July/August 2002.
6.
Chortareas G., Stasavage D., Sterne G., Does it Pay to Be Transparent? International Evidence form
Central Bank Forecasts, The Federal Reserve Bank of St. Louis, St. Louis 2002.
7.
Cierkoski M., Adaptacyjne uczenie si a modelowanie i ocena polityki pieninej, Bank i Kredyt 6/2008.
8.
Debelle G., Inflation Targeting in Practice, Working Paper No. 35, International Monetary Fund,
Washington, D.C. 1997.
9.
Ehrmann M., Sondermann D., The Reception of Public Signals on the Financial Markets. What if Central
Bank Communication Become Stale? ECB Working Paper Series No. 1077, Frankfurt am Main, 2009.
10. Eusepi S., Preston B., Central Bank Communication and Expectations Stabilization, NBER Working Paper
Series No. 13259, Cambridge 2007.
11. Ferrero G., Secchi A., The Announcement of Future Policy Intentions, Third Conference on Money,
Banking and Finance: Monetary Policy Design and Communication, Bank of Italy, September 2007.
12. Forsells M., Kenny G., The Rationality of Consumers’ Inflation Expectations: Survey-Based Evidence for
the Euro Area, ECB Working Papers Series, working Paper No. 163, 2002.
13. Forsells M., Kenny G., Survey Expectations, Rationality and the Dynamics of Euro Area Inflation, in
Journal of Business Cycle Measurement and Analysis, Vol 1, No. 1, 2004.
14. Gábriel P., Households inflation expectations and inflation dynamic, MNB Working Papers 12/2010.
15. Galí J., New Perspective on Monetary Policy, Inflation and Business Cycles, NBER Working Paper Series
No. 8767, Cambridge 2002.
16. Geraats M. P., Why Adopt Transparency? The Publication of the CentralBank Forecast, ECB Working
Paper No. 41, Frankfurt am Main 2001.
17. Gerberding Ch., The Information Content of Survey Data on Expected Price Developments for Monetary
Policy, Discussion paper 9/01Economic Research Centra of the Deutsche Bundesbank, 2001.
18. Gersbach H., On the Negative Social Value of Central Banks’ Knowledge Transparency, Economics of
Governance 4, 2003.
19. Goodfriend M., King R. G., The New Neoclassical Synthesis and the Role of Monetary Policy, Working
Paper No. 5, Federal Reserve Bank of Richmond, Richmond 1998.
20. Goodfriend M., How the World Achieved the Consensus on Monetary Policy, NBER Working Paper Series
No. 13580, Cambridge 2007.
21. Goodhart Ch.A.E., Bin Lim W., Interest Rate Forecast – a Pathology, International Journal of Central
Banking, June 2011.
22. Jansen H., Optimal Degrees of Transparency in Monetary Policymaking, Discussion Paper 04/01, Economic
Research Centre of the Deutsche Bundesbank, January 2001.
23. Khan G., Communicating a Policy Path: the Next Frontier in Central Bank Transparency? Federal Reserve
Bank of Kansas City, Economic Review Q1 2007.
54
50
N a t i o n a l
B a n k
o f
P o l a n d
References
24. Leitemo K., Open-Economy Inflation Forecast Targeting, w: German Economic Review, Vol. 7/2006.
25. Łyziak T., Consumer Inflation Expectations in Poland, ECB Working Paper Series, Working Paper no. 287,
2003.
26. Mankiw N.G., A Quick Refresher Course in Macroeconomics, NBER Working Paper Series No. 3256,
Cambridge 1990.
27. Mishkin F.S., Can Central Bank Transparency Go too Far? NBER Working Paper No. 10829, Cambridge
2004.
28. Monetary Policy Guidelines for years 2004-2011, NBP.
29. Monetary Policy in Hungary, MNB, Budapest 2002.
30. Nerlich C., Exchange Rate Strategies of EU Accession Countries: Does the Exchange Rate Policy Matters?
ECB, 2002.
31. Remsperger H., Worms A., Transparency in Monetary Policy, CFS Working Paper No. 16, Frankfurt am
Main 1999.
32. Skoepa M., Kotlán V., Assessing Future Inflation in Inflation Targeting: Forecasts or Simulations, w:
Monetary Policy in Changing Environment, BIS Paper No. 19, Basel 2003.
33. Svensson L.E.O., Inflation Forecast Targeting: Implementing and Monitoring Inflation Targets, NBER
Working Paper No. 5797, Cambridge 1996.
34. Svensson L.E.O., Inflation Targeting As a Monetary Policy Rule, NBER Working Paper No. 6790,
Cambridge 1998.
35. Svensson L. E. O., Woodford M., Implementing Optimal Policy through Inflation – Forecast Targeting,
CEPR Discussion Paper No. 4229, London 2002.
36. Tarkka J., Mayes D., The Value of Publishing Official Central Bank Forecasts, Bank of Finland Discussion
Papers No. 22, Helsinki 1999.
37. Taylor Interest Rate and Monetary Condition Index, Deutche Bundesbank Montly Report, April 1999,
Deutche Bundesbank, Frankfurt am Main 1999.
38. The Joint Harmonised EU Programme of Business and Consumer Surveys, User Guide, European
Commision, July 2007.
39. Tovar C.E., DSGE Models and Central Banks, Economic Discussion Paper, October 2008.
40. Woodford M., Interest and Prices. Foundations of Theory of Monetary Policy, Princeton University Press,
2003.
WORKING PAPER No. 105
51
55