Inflation Targeting:
Czech Experience
Kateřina Šmídková*
Executive Director
Research Department
Czech National Bank
OECD-CCBS/Bank of England
Paris, February 2007
*) The views expressed are those of the author, and do not necessary represent those of the CNB.
1
1998-2007
After nearly ten years of targeting
inflation in the CNB,
there are certainly lessons to
share
What to Learn from the Czech
Experience?
2
• One should think about IT in three
steps
– certain preconditions should be met to
introduce IT successfully (but it is
important to note that one learns
swimming only by jumping into water)
– several components should be designed
with care (do not focus only on target or
forecasting model)
– certain outcomes should be observed
relatively soon (others in medium term)
• One should view IT as a strategy in
motion
– IT needs modifications to reflect new
theoretical views and concepts that
develop every 3-5 years (think SDGE
models)
– IT needs modifications to reflect new
views on best practices (think
transparency)
– IT needs modifications to reflect new
external challenges (think global low
inflation)
3
IT in Three Steps: Preconditions
• important to have prior to IT
• externally:
– institutional set-up (independence
accountability)
– well-developed financial markets
• internally
– support of the Board
– core team of educated experts
• not so important to have
• externally:
– perfect set-up (who owns target may be
clarified in the introductory period)
– deep financial markets (some will develop
under IT only)
• internally
– large team of experts (they will come)
– fashionable complicated forecasting
model (it will be developed later)
4
IT in Three Steps: Components
• Target specification
– time horizon (CNB: 1Y, 3Y, infinite)
– interval versus point (CNB: points in time,
continuous interval, point with bands)
– which price index (CNB: net inflation, CPI)
• Caveats
– initially no caveats (net inflation)
– now ex ante caveats
– no formal explanatory letter
• Decision-Making Framework
– framework for discussion (model forecast by
staff)
– analysis of risks (staff and Board) e.g.
alternative scenarios
– actual decision (Board votes, often different
views)
• Communication (transparency)
– decision (press conference)
– reasons (minutes)
– details about forecast (Inflation report)
– details about decision (internal protocol)
5
Target specification
December 1997 April 2001
November 1998 December 2005
April 2000
6
Caveats
The following exceptions ex ante
(caveats/escape clauses) from
achieving the inflation target are
used by the CNB:
-major deviations in world prices of
energy
- major deviations in agricultural
prices
-major changes in indirect taxes
- major changes in regulated prices
In these cases, the CNB does not
respond to the primary impacts of
the shock.
7
Decision-making Framework
8
IT in Three Steps: Outcomes
• primary external outcome: low
inflation (low volatility could be
aimed at) and stable expectations
• secondary external outcome: good
economic performance
• primary internal outcome:
institution changes (staff improves
their modeling know-how, Board
focuses more on price stability)
• secondary internal outcome:
various processes can change
(pressure on well-maintained web
to comply with transparence
9
I/1
9
-2
0
2
4
6
8
10
12
14
16
III 9 8
/1
99
I/1 8
9
III 9 9
/1
99
I/2 9
0
III 0 0
/2
00
I/2 0
0
III 0 1
/2
CNB REPO
00
I/2 1
0
III 0 2
/2
00
I/2 2
0
III 0 3
/2
00
Net inflation
I/2 3
0
III 0 4
/2
00
I/2 4
0
III 0 5
/2
Primary Outcome
00
CPI inflation
I/2 5
0
III 0 6
/2
10
00
6
-10
-5
0
5
10
15
20
25
30
35
I/1998
III/1998
I/1999
III/1999
I/2000
III/2000
Deregulations (%)
I/2001
III/2001
I/2002
GDP (%)
III/2002
I/2003
III/2003
I/2004
III/2004
I/2005
III/2005
Secondary Outcome
I/2006
III/2006
11
CZK/EURO (in crowns, RA)
0
5
10
15
20
25
30
35
40
IT in Motion:
Current Challenges
– CNB is improving the forecasting
model (from QPM, used since 2002,
to SDGE model)
– CNB is improving transparency
over time (faster publishing of
minutes, new publishing of
description of interest rate path)
– CNB considers level of the target
(global low inflation, and also due
to the expected euro adoption after
2010)
12