Memorandum by Professor Seppo Honkapohja,
Professor of Economics, Academy Professor, University of Helsinki
1. In response to the Call for Evidence,
2 May 2003, I would like to give comments on selected issues among
those listed in the Call. My comments will be limited to the monetary
policy strategy of the ECB, ie the first focus of the inquiry.
2. The two-pillar strategy has been subject
to fair amount of debate and discussion ever since the start of
the ECB. These pillars have been seen as to some extent contradictory
since the ECB has a unique primary objective of price stability.
It appears that the first pillar of giving prominent role to money
has not been followed so closely, since the ECB seems to have
missed M3 target growth for relatively long periods.
These deviations seem fairly large relative to those of the inflation
from its target (close to 2 per cent). One explanation might be
that use of M3 as the measure of monetary aggregate has been somewhat
problematic in the last couple of years, since with low interest
rates there may have been substitution between components of M3
and non-money items in investors' portfolios.
3. The discussion about the two-pillar strategy
has been fairly critical of the ECB practice, at least from academic
viewpoints. However, it should be pointed out that some recent
research into monetary policy has discovered that a money-supply
based strategy can sometimes be useful, but its role should be
limited to extreme situations of high inflation or deflation.
(This is sometimes called the zero-bound problem for interest
rate setting.) Indeed, the research suggests that the latter seems
particularly worrisome when monetary policy is conducted on the
basis of interest rate setting. While this research is currently
perhaps too limited to form a basis for policy recommendation,
this concern is beginning to influence current thinking about
conduct of monetary policy. For example, recent background papers
prepared in the ECB in the recent assessment of ECB strategy suggest
this. It is also reflected in the recent declaration of the ECB
council that it will try to keep the inflation rates close to
2 per cent in the medium term.
4. This has also been subject to a significant
amount of debate and discussion. Perhaps the main concern has
been the fact that a low inflation target for EMU average might
lead to deflation or close to deflation outcome for some of the
richer EMU countries. This is due to the different levels of economic
development among EMU countries with inflation tending to be somewhat
higher in countries with lower level of GDP per capita but relatively
faster economic growth (the so-called Balassa-Samuelson effect).
These problems might well become more acute in the future enlargement
when some new member countries will want to join the monetary
union as well.
5. The appropriate level of the inflation
target should apparently be based on two central considerations.
First, the risks of deflation for an EMU member country are a
concern so that the level of the target should not be too low.
It can be argued that the current below 2 per cent is on the low
end of the possibilities from this view point. Second, there could
be a floor level for the inflation target to avoid the possibilities
of a wider than country-specific deflation. The recent amendment
to the strategy (see paragraph 3 above) reflects this concern
and indeed the earlier ECB formulation of inflation between 0
and 2 per cent in the medium term carried more risks in this respect.
6. The ECB has been relatively imprecise
about (i) the level of the inflation target and (ii) about the
period over which the target is to be reached. The recent evaluation
has given more information about the target level, ie that it
is now close to 2 per cent, but there has not been similar clarification
of the phrase "medium term". This might perhaps be worth
clarifying more in the future. However, it should be noted that
in any case numerical targets will usually not be met precisely.
Moreover, the ECB faces a "rolling window" ie the horizon
of a medium term will shift correspondingly forward as the time
passes. At any given moment of the time the horizon is to be used
for making current decisions about the level of policy instruments
so as to achieve the target over a specified horizon ("medium
4 June 2003
2 See The Monetary Policy of the ECB, ECB Frankfurt
See eg the Monthly Bulletin of the ECB for the record. Back
See <au0,2>www.ecb.int<xu for the recent ECB council
decision and the background papers. Back