Djusting To Leaner Times: 5 A R Ceps M P G
Djusting To Leaner Times: 5 A R Ceps M P G
ISBN 92-9079-443-7
Chapter 1. How did Euroland get caught in the slow-growth trap? ...1
1.1 Setting the scene: Why is demand so anaemic in Euroland? .............1
Investment....................................................................................1
Consumption ................................................................................2
1.2 What can policy do under these circumstances?..............................5
Structural policy.......................................................................... 5
Fiscal policy ............................................................................... 5
Monetary policy .......................................................................... 6
The euro and transatlantic relations ............................................. 8
Annex. The housing channel of monetary policy in the US...................11
References .......................................................................................98
List of Figures
1.1 Productivity slowdown in the euro area .......................................3
1.2 EU forecast vs actual GDP .........................................................6
1.3 EU CPI forecast vs actual...........................................................7
2.1a Unemployment rates, working-age population (15-64 years old)..15
2.1b Employment rates, working-age population (15-64 years old) .....16
2.2 Weight of the population aged 55-64 in working-age population .20
2.3 Beveridge curve for the euro area..............................................26
3.1 Fiscal policy after the asset bubble ............................................40
3.2 Japan vs. Sweden: Fiscal policy after the asset bubble .................41
4.1 Actual and perceived inflation...................................................43
4.2 Perceived inflation and actual inflation for hotel and
restaurant services....................................................................43
4.3 Perceived inflation and purchasing intentions .............................44
4.4 The ECB follows consensus views on growth ............................48
4.5 Euroland three-month rate and the calibrated Taylor rate ............51
4.6 Euroland three-month rate and the estimated Taylor rate.............53
4.7 Euroland relative unit labour costs (ULCs) rose, despite the
economic downturn… ..............................................................55
4.8 … which meant inflation was slow to fall...................................56
4.9 Because the Federal Reserve could cut rates faster…..................56
4.10 … the US economy received a big boost from fiscal policy. ........57
4.11 No deflation risk according to index-linkers...............................59
4.12 Wage growth remains sticky .....................................................61
4.13 Inflation down, expectations down sharply.................................62
4.14 Money growth is strong ............................................................63
4.15 Output gap to widen .................................................................64
4.16 Recent developments in line with history...................................67
4.17 Private credit growth down, but largely as a correction ...............69
4.18 Price stability: A ‘clarification’, not a ‘change’ ..........................78
4.19 Money, pr ivate loan growth and asset price inflation in the
eurozone ..................................................................................80
4.20 Money growth and inflation for G4............................................81
4.21 Money growth and consumer price inflation in Euroland.............81
List of Tables
1.1 Long-term growth expectations (%) ............................................3
1.2 Growth rates and permanent income............................................4
1.3 The importance of international trade (2002) ................................8
1.4 Real effective exchange rates ......................................................9
2.1a Youth (15-24) employment status in the EU and in the US .........18
2.1b Employment rates (%) of adult workers in the EU
and in the US...........................................................................19
2.2 Forecasted unemployment rates and employment rates in
EU member states, 2010 and 202...............................................21
2.3a Employment rates (%) by education levels, men
(25-54 years old) ......................................................................24
2.3b Employment rates (%) by education levels, women
(25-54 years old) ......................................................................25
4.1 Revised ECB strategy versus pure infla tion targeting..................83
List of Boxes
2.1 The German agenda, 2010 ........................................................27
3.1 The basic formal framework for the tax-smoothing argument ......32
3.2 Should large-scale econometric models serve as guides to
policy-makers? ........................................................................35
4.1 Proactive Federal Reserve – sclerotic ECB?...............................55
4.2 Where Germany lacks competitiveness......................................70
5.1 Rules for determining exchange rate policy for the euro area ......87
PREFACE
T
his is already the fifth annual report issued by the CEPS
Macroeconomic Policy Group. As has been our practice for
several years now, we have brought together a distinguished group
of economists to produce a thorough analysis of the key challenges facing
economic policy-makers in the EU.
For the last two years, our reports have emphasised the dismal
productivity record as the key reason for the disappointing growth
performance (and probably prospects) of Europe. It is no source of
satisfaction that this point of view was again so strongly validated during
2002-03. In this year’s report we analyse the implications of the
productivity slowdown for fiscal policy. Our conclusion is quite simple:
Whatever its causes, slower productivity means fewer resources available
for redistribution. This applies to governments as well.
True to our mission as ‘ECB watchers’, we also analyse at length the
effectiveness of monetary policy, and in particular the results of the
strategic review recently announced by the European Central Bank. We
might be less strident in our criticism than some of our colleagues, but
this is the outcome of a careful analysis, which finds it hard to fault the
ECB on fundamental issues.
The Group is grateful for the comments from participants attending an
off-the-record seminar organised by Deutsche Bank in Frankfurt, in
particular Lorenzo Bini-Smaghi from the Italian Treasury, Bernd
Fitzenberger from the University of Mannheim, Otmar Issing from the
ECB and Robert Price from the OECD.
Leonor Coutinho, a Marie Curie Research Fellow at CEPS, provided
excellent research assistance and important ideas of her own. All remaining
errors are ours.
The work of the CEPS Macroeconomic Policy Group would not have
been possible without the continuing support from our main sponsor,
Deutsche Bank, London, and, more recently Tudor Investments. I wish to
thank them once more for their material and financial contributions.
Daniel Gros
Director
i
ADJUSTING TO LEANER TIMES
5TH ANNUAL REPORT OF THE
CEPS MACROECONOMIC POLICY GROUP
EXECUTIVE SUMMARY
Macroeconomic Issues
Euroland is currently caught in a slow-growth trap. But, since the
fundamental factors restraining growth in Euroland are structural,
demand management cannot achieve a lot. This applies, mutatis
mutandis, to both fiscal and monetary tools policy.
Fiscal policy has to accept the constraints of low growth and ageing. Tax
cuts without a credible commitment to restrain future expenditure would
be useless. We conclude that:
Governments should concentrate on pension reform and cuts in age-
related entitlements. Only after these measures have been achieved
should tax cuts be considered. In order to prepare for the pension bomb
that will hit Europe starting from about 2010, it makes sense for
governments to start aiming now for budgets that are balanced, or in
small surplus, over the cycle.
Monetary policy can only aim at allowing the eurozone to achieve its
meagre growth potential (1.5-2%) and guard against deflation. The
European Central Bank (ECB) seems to accept this challenge, but its
strategy could be improved. We conclude that:
The monetary pillar in its old form (comparing M3 growth against a
reference value) is useless. The new ‘financial stability pillar’ should
include an analysis of the evolution of financial structures in a broad
sense to find out whether disequilibria in balance sheets or indebtedness
present a risk to economic stability.
The E(M)U Constitution
Managing the euro. The exchange rate is important for an economy, like
that of Euroland, whose exports amount to about 20% of GDP, and the
unwelcome deflationary impact of the recent appreciation of the euro
underscores the importance of the external value of the euro.
The Intergovernmental Conference that will soon start to finalise the new
Constitutional Treaty for the EU should make the President of the
Eurogroup a “Mr Euro”, who would be empowered to present the
official view on the external value of the euro.
iii
EXECUTIVE S UMMARY
Fiscal policy
Our analysis in Chapter 3 suggests that fiscal policy has already reached
the limit in most member countries with cyclically adjusted deficits in all
large member countries close to 3% and in some cases clearly beyond.
Our calculations suggest that if Europe is to keep public finances under
control despite the rapid ageing of its population, cyclically adjusted
deficits need to be kept well below 3%, probably close to the Stability
and Growth Pact (SGP) goal of balance.
The fundamental challenge for fiscal-policy makers in Euroland is to
accept the constraints from a low-growth environment compounded by an
ageing population. The constant hope (or rather illusion) of a rapid return
to strong growth has been one key factor in the present difficulties for
public finances. Almost invariably budgetary plans rely on assumptions
about long-term growth prospects (2.5-3%) that appear unrealistic in the
iv
ADJUSTING TO LEANER TIMES
Monetary policy
If the current weakness derives from the bursting of a bubble, monetary
policy might have a role to play. The bursting of the expectations of the
Lisbon bubble might not be, per se, a problem for monetary policy. But the
exuberance that accompanies a bubble usually leads to the build-up of severe
balance sheets problems that must be taken into account by monetary policy,
whose main task then becomes to ensure the stability of the financial system
(in a wider sense, not just the banking system). In practice, this means that
the ECB should be vigilant against the danger of deflation because there
exists a pronounced asymmetry. As the experience of Japan shows, even a
small dose of deflation can have very high costs whereas the costs of a small
dose of controlled inflation are likely to be quite limited (this was recognised
by the Chairman of the US Federal Reserve when he underlined the danger
of an ‘unwelcome further fall in inflation’ in his May statement). Hence we
discuss in Chapter 4 whether there exists a danger of deflation and what can
be done about it.
In its strategic review, the ECB has implicitly re-affirmed its determination
to fight against deflation (should the danger arise). We fully agree with
stance and with other aspects of the revised strategy.
Our main criticism of the new strategy is that it represents an opportunity
missed to clarify the nature of the monetary ‘pillar’. It should be clear by
now that under the present circumstances it is not useful to just look at M3
growth and conclude that potential inflationary pressures exist because M3
has grown more quickly than a certain reference value (which anyway has –
for good reasons – been ignored for some time now). We would argue that
the monetary pillar should be interpreted more broadly as a ‘financial
stability’ pillar. The pursuit of financial stability requires the analysis of a
broad range of issues. For example, the present issue is not primarily how the
ECB should react if another financial market price bubble were to arise. Nor
is it so much whether the euro-area banking system (or even that of any
member country) is on the brink of collapse. Rather, an important issue at
present is whether balance sheets of the non-financial sector are
overstretched to the point where firms are cutting investment to achieve the
lower debt/equity ratios demanded by financial markets in the post-bubble
environment (with potentially adverse consequences for price stability).
Hence we would argue that the ECB should explain that it will look in the
monetary pillar at the evolution of financial structures in a broad sense to
find out whether disequilibria in balance sheets or indebtedness present a risk
to economic stability.
v
CHAPTER 1
HOW DID EUROLAND GET CAUGHT
IN THE SLOW-GROWTH TRAP?
Investment
A prolonged period of weak investment demand is to be expected after
the bursting of a bubble. This is true for two inter-related reasons:
i) A bubble usually leads to over-investment. When the bubble bursts,
there is likely to be an excess of installed capacity. It is thus natural that
there is little need for new investment. This mechanism can explain the
persistence of weakness for a decade in the Japanese economy where
investment rates were running at over 30% during the 1980s. In the US
there was also an investment boom, which led to an increase in the
investment-to-GDP ratio of around 5 percentage points during the 1990s.
Recent data on capacity utilisation for the US continue to run at
historically low levels, indicating that some capital overhang might exist.
It seems that this is not the case in Euroland, however. At most, one can
speak of a ‘boomlet’ as the investment/GDP ratio rose by about 1
percentage point above the longer-run average of around 21% of GDP.
Moreover, the data on capacity utilisation for Euroland do not indicate an
unusual amount of excess capacity. The value reported for the first
quarter of 2003 was very close to the long-run average.
1
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
ii) During a bubble, the corporate sector often takes risks with balance
sheets. When an important stock imbalance is built up during the
ascending phase, when capital appears cheap, it often takes a long time
for firms to get back to more normal ratios. This is usually achieved by
cutting back on investment so as to conserve cash flow for a
strengthening of the equity base. This effect seems to have operated in
Euroland where one can observe that the corporate sector is trying to cut
back on debt levels as documented more fully below in Chapter 4.
Consumption
The most surprising aspect of the current ‘soft patch’ is the weakness of
consumption demand. In Euroland, private consumption demand has
grown by only about 1% per annum over the last two years, and might be
close to zero during 2003. By contrast, demand in the US has kept
growing at close to 3% per annum until recently.
Since Euroland’s households are not really over-indebted, this weakness
of consumption is not easy to explain a priori. 1 However, this weakness
of consumption becomes straightforward to understand if one considers
the radical revision of growth expectations that has taken place since the
bursting of the ‘Lisbon bubble’. At the special European Council of
Lisbon in early 2000, the Heads of State and Government of the EU had
solemnly promised to make the EU the ‘most competitive economy’ by
2010, setting inter alia precise numerical targets for employment rates.
Productivity growth at the time seemed satisfactory so that growth rates
in excess of 2.5% seemed within reach. However, as we now know, these
expectations were bitterly disappointed: growth has been anaemic as
productivity growth has plummeted (see Figure 1.1).
1
According to ECB statistics and national sources for in 2001, the ratio of gross
household debt outstanding to gross disposable income was 75.7% for the euro
area, while for the US and Japan it was considerably higher, at 104% and
109.6%, respectively. The figures for 2002 are not yet available.
2
ADJUSTING TO LEANER TIMES
2.0%
1.5%
1.0%
0.5%
0.0%
1975 1978 1981 1984 1987 1990 1993 1996 1999 2002
*The euro area refers to the weighted average of its three largest members (Germany,
France and Italy).
Source: Deutsche Bank London.
3
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
Interest rate
Gross rate 0.03 0.04 0.05 0.06
0.01 155 139 131 127
0.015 206 166 150 141
0.02 309 208 175 159
0.025 618 277 210 182
0.03 416 263 212
* Permanent income from an income stream starting with one unit and growing at the rate
indicated in the first column, in per hundred of today’s income. These values would be
equivalent to selling an asset whose return increases at this growth rate and then investing
the proceeds in a fixed income bond yielding the interest rates indicated in the first
column.
Source: Own calculations.
4
ADJUSTING TO LEANER TIMES
For example, if expected growth rates fall from 2.5% to 2% per annum
(as suggested by the Consensus Forecasts expectations shown above and
which is much less than the deterioration experienced by the EU during
the 1990s) and if this change is expected to be permanent, the permanent
income of the average household would be reduced by about 20%. One
could thus expect that consumption demand would follow by a similar
proportion. If the slowdown is only temporary, the fall in permanent
income, and hence consumption, would be smaller, but still substantial.
For example, a productivity slowdown of the magnitude used so far that
lasted ten years would still reduce permanent income by over 5%. These
illustrative calculations show that even small shifts in expectations of
future growth rates can potentially have a strong impact on consumption.
Fiscal policy
Our analysis in Chapter 3 suggests that fiscal policy has already reached
the limit in most member countries with cyclically adjusted deficits in all
large member countries close to 3% and in some cases clearly beyond.
Our calculations suggest that if Europe is to keep public finances under
control, despite the rapid ageing of its population, cyclically adjusted
deficits need to be kept well below 3%, and probably close to the SGP
goal of balance.
5
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
5 % yoy
0
Forecast GDP
-1 Actual GDP
-2
1989 1991 1993 1995 1997 1999 2001 2003
Monetary policy
If the current weakness derives from the bursting of the bubble, monetary
policy might have a role to play. Severe balance sheets problems in the
corporate sector can be alleviated by monetary policy, whose main task
then becomes to ensure the stability of the financial system (in a wider
sense, not just the banking system). In practice this means that the ECB
6
ADJUSTING TO LEANER TIMES
5
EU Forecast
4.5
EU actual
4
3.5
3
2.5
2
1.5
1
0.5
0
1989 1991 1993 1995 1997 1999 2001 2003
7
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
The ECB has recently confirmed its two-pillar strategy. We provide our
evaluation, which is much more positive than most, but also suggest
some improvements.
Exports Imports
US Eurozone US Eurozone
Current account
12.2% 24.2% 17.0% 23.4%
transactions, % of GDP
Goods trade, % of GDP 6.5% 15.0% 11.2% 13.1%
Goods and services trade,
9.3% 19.7% 13.5% 17.6%
% of GDP
Sources: Own calculations based on ECB and Bureau of Economic Analysis data.
What could result from the combination of a weak dollar and the
relatively high degree of openness of the eurozone? The first point to note
here is that the ‘euro’ is not just the mirror image of the ‘dollar’.
8
ADJUSTING TO LEANER TIMES
EU-US bilateral trade links are very important in themselves. But the
dollar/euro rate is not necessarily the most important single exchange rate
for Euroland. For the eurozone, trade with the UK is slightly more
important than trade with the US. Likewise, for the United States, trade
with Canada alone is more important than trade with Euroland. What
matters for Euroland is the ‘euro’ (the effective exchange rate of the
euro), not the dollar (i.e. the bilateral rate). Do these two move together in
reality? The answer is not straightforward; the ‘dollar’ and the ‘euro’
have a strong tendency to move in a mirror image fashion in the short
run, but important deviations from this pattern are possible.
The short-term correlation between the bilateral dollar/euro exchange rate
and measures of the effective exchange rate of Euroland has in the past
been rather high – at over 80% (the precise value depends on the exact
measure of the effective exchange rate chosen). This suggests that the
two move almost always in a similar direction. But by how much? In the
past, a useful rule of thumb had been that only about one-half of any
change in the bilateral dollar/euro rate translated into a change of the
effective exchange rate of the euro area (whether in nominal or real terms
does not really matter in this context as price levels move much more
slowly than do exchange rates).
More recently, however, this relationship seems to have changed. Table
1.4 shows the real effective exchange rates as calculated by the ECB.
This table shows that since the start of 2003, the euro has appreciated in
effective terms by about 15%, whereas the dollar has depreciated only by
about one half as much (7%). This shows that many other currencies
(including Sterling and the yen) have moved with the dollar, thus pushing
all the adjustment on the eurozone. If one looks at a longer period, this
tendency becomes even more apparent: the euro is now, in effective
terms, back to the (relatively high) level attained in 1995 by the synthetic
euro, but at the same time the dollar is also still much stronger (again in
effective terms) than in 1995.
9
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
The recent pattern (if it persists), in which the currencies of the UK, Asia
and emerging markets are effectively pegged to the dollar, has two
implications:
i) Most of the counterpart for the current account adjustment of the US
would be forced on the eurozone; and
ii) A further large move in the bilateral dollar/euro rate will be needed
before the dollar can get even close to a level that would produce a
sizeable adjustment in the US current account.
These considerations suggest that a further strong appreciation of the euro
cannot be ruled out. As argued in Chapter 4, we regard such a
development as the main potential source of deflation for the euro area.
Given this potential threat from the external side and the increase in the
volatility of the euro’s exchange rate, it thus become even more important
than before to clarify who is responsible for the external value of the
euro. As Chapter 5 shows, the present arrangements are not satisfactory
in this respect. We discuss ways in which it would be possible to create
the position of a ‘Mr Euro’, i.e. a person who could provide foreign
exchange markets with an official view should there be major movements
in the exchange rate. The proposals contained in the draft Constitutional
Treaty agreed by the Convention on the Future of Europe in July were
disappointing in this respect. We hope that the Intergovernmental
Conference, which will start in October of this year, will address this
issue and find a solution, which could be similar to the one found for the
newly created position of Foreign Minister of the EU.
10
ADJUSTING TO LEANER TIMES
Annex
The Housing Channel of Monetary Policy in the US
11
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
13
CHAPTER 2
LABOUR MARKETS AND STRUCTURAL REFORMS
I
n its 2001 annual report (Gros et al., 2001), the CEPS
Macroeconomic Policy Group expressed pessimistic views on the
structural situation of labour markets of EU member countries,
despite acknowledging some improvements in indicators such as
unemployment and employment rates. It was then argued that the
reduction of unemployment and the rise of employment were not
affecting all population groups equally and that, although there were
some reforms aimed at improving the functioning of labour markets,
these were done in a piecemeal fashion, targeted at new entrants in the
labour market and, hence, did not imply fundamental changes on which
to permanently base future improvements. Moreover, in last year’s report
(Gros et al., 2002a), the Group expressed some concerns about the
disappointing performance of labour productivity in most EU countries
and called for more aggressive supply-side reforms.
These pessimistic views are usually received with some scepticism. It is
often argued that there have been evident structural improvements in EU
labour markets during the last five years, with employment growing at an
average annual growth rate of 1.5% (1.6% in the euro area) and the
unemployment rate falling by 2.6 percentage points (2.8 percentage
points in the euro area) during the 1997-2001 period. The fact that this
strong employment growth took place without significant inflationary
pressures is indicative of a fall in the equilibrium rate of unemployment
and suggests that the gains in employment and participation are due to
structural factors and, hence, sustainable in the long-run. Moreover,
despite the deceleration in growth in 2001-02, the unemployment rate in
the EU has ‘only’ risen to 8.1% (8.8% in the eurozone), according to the
latest estimates from EUROSTAT. As for productivity growth, the
slowdown is usually perceived as a temporary phenomenon, needed to
facilitate the rise of employment through an increase in the ‘employment
content of growth’.
14
ADJUSTING TO LEANER TIMES
20
18
16
14
12
(%)
10
0
EU US Germany France Italy Spain UK
1990 1993 2001
15
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
80
Lisbon Objectives
Lisbon's (2010)
Objective (2010)
70
Source: Employment in Europe 2002, European Commission.
60
50
(%)
40
30
20
10
0
EU US Germany France Italy Spain UK
1990 1993 2001
With regard to the employment rate, there is also some good news and
some bad news. The good news is that throughout the 1990s, there was a
clear upward trend in most EU countries, so that it increased by almost 4
p.p. in the whole EU, and by almost 1 p.p. in Germany between 1993 and
2001 vs a rise of 1.3 p.p. in the US in the same period. And in this case,
the change seems to be the result of a structural trend: when comparing
1990 and 2001, we also observe a common increase in employment-
population ratios in all of the other four largest EU member states (by 2.4
p.p. in France, 0.9 p.p. in Italy, 0.2 p.p. in the UK and 7.1 p.p. in Spain).
There is bad news too, however. First, despite this increase, employment
rates in western and southern Europe remain too low in particular,
relative to the ambitious objectives established by the European Council
in its summit at Lisbon (2000) and subsequently clarified in other
summits (Stockholm and Barcelona).2 And by looking at the experience
2
The Lisbon European Council established the following targets by 2010: the
average aggregate employment rate on the EU should be as close as possible to
70%, the average female employment rate on the EU should be higher than 60%
for women. The Stockholm European Council in 2001 set intermediary targets of
67% for the aggregate employment rate, and 57% for the female employment
16
ADJUSTING TO LEANER TIMES
of the last decade there is a serious risk that these goals cannot be
achieved, more so when the enlargement of the EU will bring the
accession countries which, on average, have lower employment rates than
the average employment rate of the EU.
Secondly, almost all the rise in employment rates is explained by the
higher participation rates of the adult female population, while high youth
unemployment rates and low employment rates among older workers
have not shown any sign of improvement. In the case of young workers
(see Table 2.1a), there is no significant reduction in unemployment rates
in four of the five EU countries considered (the exception is Spain), while
youth employment rates are decreasing due to increasing enrolment in
higher levels of education. For the prime aged (25-54) male population,
employment rates have not shown any significant change between 1990
and 2001, while for the older (55-64) population, they have fallen in
Germany, France, and, especially, Italy (see Table 2.1b). While the
increases in female participation and female employment are to be
welcomed, they cannot be the basis for further and future rises in the
overall employment rate, particularly in a demographic scenario in which
the weight of the older workers in the labour force is expected to increase
noticeably over the coming decades, as shown in Figure 2.2, which shows
EUROSTAT’s data regarding the latest national population forecasts.3 As
can be seen, in the whole EU the share of older persons (55-64) in the
working-age population will rise by about 5 percentage points in the next
15 years or so. According to these forecasts, the member states whose
working-age population will age more rapidly are Belgium, France, Spain
and Finland.
rate by 2005, and introduced a new target of 50% for workers aged 55 to 64 by
2010. In the Barcelona summit the European Council pledged to transform
Europe into ‘the most dynamic and knowledge-based economy of the world,
capable of sustainable economic growth while providing more and better jobs
and greater social cohesion’.
3
See EUROSTAT (2001).
17
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
Germany* 8.0 7.8 8.1 9.4 10.3 8.3 51.7 53.7 49.8 46.5 48.6 44.0
France 19.2 15.7 22.9 19.5 17.6 21.8 34.4 38.0 31.0 29.5 33.3 25.7
Italy 28.1 22.4 35.3 28.1 24.9 32.0 29.8 34.0 25.4 26.3 30.4 22.1
Spain 28.1 22.4 35.3 21.5 16.6 27.9 32.0 38.4 25.6 33.1 39.2 26.9
UK 10.4 11.5 9.2 11.9 13.2 10.3 66.4 70.4 62.3 56.9 59.5 54.3
* Data cited for year 1990 are actually drawn from 1993 due to data availability.
Sources: European Commission, Employment in Europe 2002, and OECD, Employment Outlook, (several years) (for the US).
18
ADJUSTING TO LEANER TIMES
Table 2.1b Employment rates (%) of adult workers in the EU and in the US
Population aged 25-54 Older workers (55-64)
Males Females Males Females
1990 2001 1990 2001 1990 2001 1990 2001
EU* 85.8 87.3 60.4 66.8 47.9 48.6 24.2 28.8
US 89.1 87.9 70.6 73.5 65.2 65.8 44.0 51.6
* Data cited for year 1990 are actually drawn from 1993 due to data availability.
Sources: European Commission, Employment in Europe 2002, and OECD, Employment Outlook, (several years) (for the US).
19
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
26%
24%
22%
20%
18%
16%
14%
12%
y
y
ia
d
ce
om
e
ga
EU
k
um
en
n
al
an
an
an
nd
ec
ur
ar
tr
ai
an
It
tu
ed
us
gd
Sp
gi
el
bo
nl
rla
re
or
Fr
el
en
w
er
Ir
Fi
A
in
G
he
B
K
G
D
xe
et
d
Lu
te
ni
U
2000 2010 2020
4
See Shimer (2001) and Jimeno (2003).
5
These calculations are taken from Jimeno (2003) who considers 18 population
groups distinguishing gender, age (15-24, 25-54 and 55-64) and three
educational levels (low, medium and high). The weights of each group in the
20
ADJUSTING TO LEANER TIMES
first horizon is relevant since that is the deadline for the fulfilment of the
targets set at the Lisbon summit. The second horizon is also relevant
since at that time population ageing will start to accelerate even more in
many EU countries, thereby putting more stress on pension systems.
total population are taken from EUROSTAT’s data regarding the latest national
population forecasts by five-year age groups. The age-gender-education specific
employment and unemployment rates are those prevailing in 2001.
21
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
For these calculations, there are two relevant demographic trends. First,
as the share of youth population in the working-age population decreases,
the aggregate employment rate will rise. But at the other extreme of the
age distribution, the composition effects have the opposite sign. As the
shares of older individuals in the working-age population and in the
labour force increase, the aggregate unemployment rate will rise and the
aggregate employment rate will fall. In principle, the facts that the
educational levels of the younger cohorts of Europeans are higher than
those of previous generations and that unemployment (employment) rates
are decreasing (increasing) in education may compensate for the rise of
the relative weights in the population and in the labour force of the older
individuals.
As for the unemployment rate, the calculations show that the changing
age composition of the labour force would have a rather minor effect on
the aggregate unemployment rate, which would remain at about 7.4% in
the EU-15. In some countries, like Spain, where the baby boom took
place later than in the other EU-15 member states, this composition effect
would produce a negative trend in the aggregate unemployment rate that
would fall by about 1 percentage point in both countries between 2000
and 2020. With regard to the aggregate employment rate, the conclusion
is more negative. As a result of population ageing, the aggregate
employment rate would fall by about 1 percentage point between 2001
and 2010, and 1.4 percentage points between 2001 and 2020. Looking at
the longer horizon, Germany, France, Italy, Luxembourg, Austria, the
Netherlands and Portugal are the member states where the fall in the
employment rate would be higher as a result of the changing composition
of the labour supply.
6
See, for instance, European Commission (2003).
7
Recent estimates of the NAIRU can be found in Pisani-Ferry (2003) for France;
Bentolila and Jimeno (2003) for Spain; Franz (2003) and Berthold and Fehn
22
ADJUSTING TO LEANER TIMES
there are good reasons to believe that the fall in inflation observed since
1994 has a lot to do with the change in the monetary policy regime, but
not much with labour market reforms. Finally, even taking for granted
that the reduction in the NAIRU observed in several countries is
permanent, further improvements need a different policy approach which
gives more momentum to effective employment policies.
There are additional signals on the lack of a substantive, fundamental
improvement in the functioning of European labour markets. One is the
lack of progress in some labour market institutions that are key for
maintaining low levels of structural unemployment. It is true that tax
reforms have reduced the tax wedge, especially for unskilled workers,
and that temporary employment contracts have been liberalised,
promoting job creation, especially for young workers, and enhancing
labour market flexibility. But it is also true that the cost of supporting
private employment through social security contributions and tax
incentives may be rather high (for instance, in the case of France, Pisani-
Ferry, 2003, places estimates close to 2%), that the employment rates of
the low-skilled workers have not significantly risen, that the gap with the
employment rate of skilled workers remains too high (see Table 2.3a and
2.3b) and that a higher proportion of temporary employment does not
necessarily imply lower structural unemployment. 8 Moreover, the
regulation of collective bargaining, in particular, and the overall legal
framework affecting wage-setting, in general, has remained untouched, as
has also happened with institutions affecting the demand and supply of
labour from older workers, such as the pension systems and some
elements of the employment protection legislation that make expelling
older workers from the labour force the easiest way to cope with labour
force adjustment within firms. As for active labour-market policies, the
European Employment Strategy launched around the so-called
Luxembourg process has resulted in a rise in the expenditures in active
labour market policies, but the effectiveness of these programmes
remains a mystery. In very few cases has there been a rigorous evaluation
of the impact of these programmes on the employment rates and wages of
the participating individuals, while the battery of experimental studies
from other countries (e.g. the US) shows that very few programmes pass
a cost-benefit test.
(2003) for Germany; and Fabiani and Mestre (2001) for the whole of the euro
area.
8
For more on this, see the collection of papers from the Symposium on
Temporary Employment published in the 2002 volume of The Economic
Journal.
23
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
24
ADJUSTING TO LEANER TIMES
Notes: Vacancy data cover around 64% of the euro area. Calculation excludes
France, Ireland and Italy.
Sources: Eurostat (LFS), NCBs, BIS, ECB calculations.
26
ADJUSTING TO LEANER TIMES
The policy conclusion following from this state of affairs is clear. Given
the demographic projections for the current decade, the EU must create
about 15 million jobs between 2002 and 2010 to achieve the 70% target
for the employment rate set in Lisbon. And given the ageing of the labour
force, many of these jobs will have to be filled by older workers who
currently are leaving employment too early. Thus, more fundamental
reforms are needed to improve the functioning of European labour
markets. In France and Austria, these reforms are currently dealing with
the pension system. In Germany there are positive signs of a political will
to proceed with fundamental reforms (see Box 2.1 on Agenda 2010). But
as the fierce opposition to these reforms shows, the task will not be easy,
particularly when the cyclical situation is not too buoyant. But the
reforms are highly needed and delay would make them only more
harmful. Moreover, these reforms may enhance growth but only in the
long run. This calls for a well designed, gradualist, time-consistent reform
programme with a proper timing of the measures to be introduced and
coordination with demand policies – exactly what has not been done
during the last upswing of the business cycle.
9
Similar conclusions are obtained from Beveridge curves plotted with OECD
data (see OECD, 2001).
27
HOW DID EUROLAND GET CAUGHT IN THE SLOW -GROWTH TRAP?
28
CHAPTER 3
FISCAL POLICY DURING TOUGH TIMES:
PREPARE FOR EVEN LEANER YEARS AHEAD OR SPEND
YOUR WAY OUT OF A CYCLICAL SLUMP?
29
FISCAL POLICY DURING TOUGH TIMES
GDP. The data for 2003 are likely to be even worse. The headline
deficit is now expected to go above 4% of GDP, with the structural
deficit probably about the same level as in 2002, i.e. around 3.2% of
GDP. The excessive deficits of Germany are thus clearly not due to
the business cycle, but to a structural weakness of fiscal policy.
ii) A second implication of lower potential growth pertains to the
sustainability of debt levels. If potential growth is as low as 1.5% and
if the ECB achieves an average inflation rate of 1.5%, the maximum
allowable deficit to keep public debt at 60% of GDP is only 1.8% of
GDP (not the 3% as assumed under Maastricht parameters). Again,
the German example is instructive in this respect. If Germany were to
continue with its structural deficit of close to 3%, its debt-to-GDP
ratio would soon start to rise and would eventually stop only at 100%
of GDP.
The very low productivity growth in Europe thus imposes some hard
constraints on fiscal policy that have not been sufficiently recognised so
far. Policy-makers should face up to this problem and stop blaming an
anonymous global business cycle.
It is interesting to note that it is mainly the large countries that have a
problem with fiscal policy. The three ‘large’ eurozone countries (France,
Germany and Italy) are currently violating or close to violating their
commitments, whereas most of the small countries (with the notable
exception of Portugal) have been able to stick to their commitments. The
reason for this is quite clear. The eight ‘virtuous’ small eurozone
countries were able to cut expenditure on average by around 1.5% of
GDP over the last three years, whereas the three large members and the
sinner Portugal were not able to manage even one-third of this. It is thus
not surprising that the deficits are under control in the smaller eurozone
countries. It seems that the body politic of the smaller countries has been
quicker to realise the merit of meeting their obligations under the
Stability Pact.
Box 3.1 The basic formal framework for the tax-smoothing argument
The key assumption is that due to ageing of the EU population, desired public
spending will go up in about 20 years. (Supplementary assumption: new
steady state starts already in 2020.)
Standard model: Social loss is increasing in tax rate (= tax take as % of GDP),
denoted by t, and is increasing in deviation of public expenditure, g (again %
of GDP) from target, G.
Time is divided into two periods:
• Period 1 (runs from present to 2020)
• Period 2 (runs from 2020 to infinity, new steady state).
The core is the usual quadratic social welfare loss function:
(1) Social loss = [at1 2 + (g 1 – G1 )2 ] + (1+d)-1[at2 2 + (g 2 – G2 )2 ]
Where a denotes the weight of taxes in social loss and d is the inter-temporal
discount factor.
The objective for policy-makers is to minimise this social loss subject to the
inter-temporal budget constraint:
(2) t2 = g 2 + (1+r)(g 1 – t 1 )
where r denotes the interest rate on public debt. Minimisation implies:
(3) 2at1 = (1+d)-1[2a(1+r)t 2 ], or t 1 = [(1+r)/ (1+d)] t 2
(3)’ (g 1 – G1 ) = [(1+r)/ (1+d)] (g 2 – G2 )
This yields the well known result that if the interest rate equals the discount
rate (i.e. assuming d=r):
(4) t1 = t 2 = t and (g 1 – G1 ) = (g 2 – G2 )
32
ADJUSTING TO LEANER TIMES
34
ADJUSTING TO LEANER TIMES
10
There might be a reason for this state of affairs: Somewhat surprisingly, the
data one would need to estimate directly the effectiveness of fiscal policy simply
does not exist for many countries. After a long and careful search for data, we
were able to perform rigorous statistical tests for four OECD countries (CN,
GER, UK and US). These are the only countries for which data are available at
the required frequency.
35
FISCAL POLICY DURING TOUGH TIMES
the future at a higher rate than the rate of time preference; and constrained
agents, who do not have access to credit markets and are therefore obliged to
consume all their disposable income in each period. The ‘Blanchard-Yaari’
assumption effectively shuts off Ricardian Equivalence, and introduces a role for
changes in taxes in affecting the consumption even of unconstrained individuals.
For this group of individuals, a permanent increase in government consumption
causes a fall in consumption (if they do not discount the future too much), as the
future increase in taxation causes their wealth to fall. For constrained individuals,
future taxes are irrelevant and an increase in government consumption causes an
increase in private consumption. Of course, the overall effect depends on the
relative proportion of constrained and unconstrained individuals.
Let us take the models of the IMF and the European Commission as examples. It
turns out that a key difference in the specification of the private sector behaviour
of MULTIMOD and QUEST II is precisely in the share of constrained agents
they assume: QUEST II assumes a very small share, 30% in all countries;
MULTIMOD a much larger share, ranging from about 50% in Germany and the
UK to 75% in Canada. As a result, an increase in government spending financed
by a future increase in taxes causes a fall in private consumption in QUEST II,
but an increase in MULTIMOD (see European Commission, 1997 and Masson,
Symansky and Meredith, 1990).
While there are other differences in the specification of the consumption function
and in the simulation scenarios, the share of constrained individuals most likely
plays an important role in these simulation outcomes. Considering that we have
so little evidence on this parameter, it would seem rather dangerous to base any
policy conclusion on such flimsy foundations.
Recent empirical research of the CEPS MPG on the impact of fiscal
policy on demand has shown that the impact of discretionary fiscal policy
on GDP since 1980 has been close to zero. This is particularly true for the
impact effect. But in those countries where one finds after this essentially
zero initial response, that there is evidence that the output response builds
up over time, one finds that even the impact is negative (Perotti, 2002).
One result common to most countries is that the effect on private GDP is
negative. This implies that even if one takes into account that an increase
in government expenditure by definition increases the public part of
GDP, there cannot be any ‘multiplier’ effect as is often assumed. In fact,
in the post-1980 period, the response of private consumption seems to be
small and there is evidence of a negative response of private investment.
It would be rash to argue that our results show that fiscal policy has
absolutely no impact on demand, or that this impact is always necessarily
negative. We would emphasise once more, however, that one cannot
discuss the potential use of fiscal policy today without taking into
36
ADJUSTING TO LEANER TIMES
account the nature of the present slowdown. If it is not only, perhaps not
even mainly cyclical, one must be even more careful than usual in the use
of fiscal policy as a counter-cyclical tool. Given that the slowdown of
potential growth has brought fiscal policy in the large member countries
closer to the limit of sustainability, as illustrated above, forward-looking
consumers will be even more likely to react negatively to ‘deficits as far
as the eye can see’ than if the slowdown were purely cyclical.
Our scepticism concerning the usefulness of the effectiveness of fiscal
policy under current circumstances is confirmed by recent polls in
Germany in which the following question was asked: Do you believe that
anticipating the tax cut by one year will be good for the conjunctural
situation? Only 34% of respondents replied ‘yes’; whereas 58% said that
they did not think that anticipating the tax cut would improve matters.
11
Since the European system of national accounts has been changed recently, the
longer range data are not totally comparable. But for the period for which data
are available for both the old and the new system, the difference is minor.
37
FISCAL POLICY DURING TOUGH TIMES
38
ADJUSTING TO LEANER TIMES
had to confront the first oil crisis with a debt-to-GDP ratio of about twice
the average of the rest of the EU.
It is sometimes even argued that one should re-interpret Maastricht as
saying that public spending on capital investment should entirely be left
out of the computations for the deficit. This goes much further than the
golden rule mentioned above. Under current circumstances, it would
amount to condoning deficits up to 5.4% of GDP (3% plus public capital
expenditure of 2.4% of GDP, as mentioned above). Proponents of the
idea that public spending on capital is good and therefore should not
count under the Maastricht rules for excessive deficits usually do not put
it this way because they realise that this would be unacceptable.
Nevertheless, it would be the logical implication of such a position.
As an aside, we note that the German Constitution actually contains the
golden rule. In Germany, the government is not allowed to borrow more
bonds than needed for financing public investment. However, this has not
prevented the German government from running up deficits above 4%
(justified, ex post, by a serious economic disequilibrium). Moreover, as
we have argued while the distinction between current and capital account
expenditures is an important one, and there is indeed a stronger case for
financing capital expenditures through debt, we do not think, at this point,
that allowing the Maastricht target to be increased for capital
expenditures would be appropriate.
12
Many thanks to Spencer Glendon of Wellington Management, Boston for
pointing this out. The vertical axis was also unified to make all three surpluses at
the peak comparable.
39
FISCAL POLICY DURING TOUGH TIMES
US. For Japan we now have over 10 years of post-bubble data, which
shows a continuing deterioration of the fiscal accounts. The US and
Europe have so far spent only a couple of years on the way down. The
deterioration of the US fiscal accounts to date has been even stronger
than that experienced by Japan in the early 1990s. This might be due to
the costs associated with the aftermath of the 9/11 attacks and the war in
Iraq. But whatever the reasons, this development does not bode well for
the future if the US follows the remainder of the Japanese trajectory.
What has been the result of a decade of increasing fiscal deficits in
Japan? There is no visible benefit as growth has not revived, but the
accumulated debt is now so large (standing close to 150% of GDP) that it
cannot be serviced if growth picks up again and interest rates return to
more normal levels. (Public debt in Japan amounts to about 5 times the
annual revenues of the Japanese government. This seems to be the more
appropriate metric if one takes the current ratio of revenues-to-GDP as
indicating the maximum that Japanese taxpayers are willing to sustain.
By contrast, public debt in the eurozone is equivalent to ‘only’ 1.5 times
annual public sector revenues.)
Forecast
0.0
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
-2.0
-4.0
-6.0
Japan (1990)
US (2000) adj
Source: European Commission (2003), Public Finances in EMU 2003, May and 2001
international financial statistics of the IMF.
40
ADJUSTING TO LEANER TIMES
but Sweden operated a sharp correction about five years after the bubble
had burst. This shows that it is possible to rein in public finances even in
a difficult environment. In Sweden growth started almost immediately
after public finances were brought under control. This might have been
helped by sharp devaluation. But growth subsequently remained strong
even as the exchange rate regained some of the terrain it had lost.
In Sweden, the public sector is about twice as large a percentage of GDP
as in Japan. It is thus not surprising that its deficits (and surpluses) have
been about twice as large as Japan’s. Figure 3.2 shows the evolution of
public sector balances in Sweden and the same concept for Japan, but
with different scales: the scale for Japan is compressed to one-half that of
Sweden. This graph suggests a surprising similarity until about year 6,
after the bursting of the bubble when the fiscal adjustment in Sweden
turned out to be permanent, whereas Japan’s deficits widened again.
Figure 3.2 Japan vs. Sweden: Fiscal policy after the asset bubble
10
Deficit, % of GDP Point 0 marks the peak year of asset prices for each 4.0
country (1989 or 1990). Note the difference in right and
left hand scales (2x).
5
2.0
0 0.0
-7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9 10
-2.0
-5
-4.0
-10
-6.0
-15
-8.0
Sweden (1989, left
scale)
Japan (1990, right scale)
-20 -10.0
Source: European Commission, General Government Data, Part II: Tables by Series,
spring 2001 and 2001 international financial statistics of the IMF.
41
CHAPTER 4
AN ASSESSMENT OF ECB POLICY
I
n this chapter we assess monetary policy developments in the euro
area. Section 1 briefly reviews the policy decisions since publication
of the last MPG Report in June 2002. In section 2, we present new
evidence on an ECB interest rate reaction function based on the Taylor
rule. Section 3 addresses the widely discussed question whether the
eurozone is at risk of falling into deflation, and section 4 concludes the
chapter with an evaluation of the ECB’s monetary policy strategy.
42
ADJUSTING TO LEANER TIMES
70 6
Perceived inflation (lhs)
60
Actual inflation, % yoy (rhs) 5
50
4
40
30 3
20
2
10
1
0
-10 0
1991 1993 1995 1997 1999 2001 2003
2.0 0
1.5 -10
1996 1997 1998 1999 2000 2001 2002 2003
43
AN ASSESMENT OF ECB POLICY
50
-20
40
-18
30
-16
20
-14
10
0 -12
-10 -10
1991 1993 1995 1997 1999 2001 2003
As the German government reviewed its fiscal policy outlook after its re-
election in September 2002, a debate began across Europe about the
merits of the Stability and Growth Pact (SGP). Several government
officials from the euro area and the European Commission questioned the
usefulness of the pact, and the EU Commission President Romano Prodi
called the pact “stupid”. As fiscal deficit estimates in the larger countries
rose and fiscal discipline risked being undermined by the criticism of the
SGP, the ECB issued an unusual statement on October 24, in which it
threw its weight behind the pact. The Council presented four theses:
1) The principle of budgetary discipline enshrined in the treaty and the
Stability and Growth Pact are indispensable for Economic and
Monetary Union (EMU).
2) The Stability and Growth Pact has been successful in promoting
sound public finances and fiscal convergence.
3) The Stability and Growth Pact is in the interest of member states.
4) The Stability and Growth Pact supports price stability. In conclusion,
it admonished governments, stating that “Respecting the provisions
of the treaty and the full implementation of the Stability and Growth
Pact remain fundamental to monetary union and to each individual
member state. Full compliance with the fiscal framework will also
send an important message to accession countries.”
44
ADJUSTING TO LEANER TIMES
5 December 2002
Towards the end of 2002, the economy continued to weaken on the back
of rising oil prices and declining confidence prior to the Iraq war. After a
controversial Council debate on 7 November, the ECB decided to leave
rates unchanged, but gave strong hints that a rate cut was in the pipeline.
In its Monthly Bulletin of November, the bank pointed to rising
uncertainty as a major reason for the “hesitant pace of economic
expansion and current lacklustre confidence”, and concluded that it was
“difficult, at this juncture, to predict the timing and strength of the
economic upswing, both in the euro area and globally”. On 5 December,
after almost a year of unchanged rates, the Council cut the ECB lending
rates by 50 basis points (bp). Such a move had been widely expected, and
only the size of the cut had been subject to intense debate. In the event,
market participants widely agreed that both the ECB’s preparation of the
move as well as the decision itself had been fully appropriate. This was in
stark contrast to several episodes in earlier years, when the ECB had been
heavily criticised for failing to adequately communicate the motivation
and timing of interest rate decisions.
In the run-up to the decision, several ECB Council members had
emphasised the role of the economic forecasts issued by the staff of the
European System of Central Banks (ESCB) in providing a basis for the
move. These forecasts had been available to the Council before its 5
December meeting, and they were subsequently published in the
December Monthly Bulletin. Compared to the projections published in
June, the December forecasts cut expectations for GDP growth in 2003
by one percentage point to 1.1%-2.1%, but left expectations for inflation
virtually unchanged. Thus, to the extent that the rate cut was motivated
by forecast revisions – as suggested by Council members – it reflected
the downgrading of growth expectations. Moreover, it is interesting to
note that the 2:1 ratio between the changes in expected GDP growth and
policy rates corresponded well with the policy recommendation given by
a standard Taylor rule. We will return to this point later.
45
AN ASSESMENT OF ECB POLICY
6 March 2003
Although the ECB had concluded after the December rate cut that “key
ECB interest rates [had] now reached a very low level by historical
standards”, renewed disappointment about the outlook for economic
growth triggered a further cut of 25 basis points in early March. Like the
decision before, this move had also been clearly signalled in advance.
ECB President Wim Duisenberg, at a press conference held after a G7
meeting on 22 February, indicated that the staff’s growth forecast had
been revised down further. Markets again took this as a strong hint for a
rate cut on 6 March, which the ECB indeed delivered. In the March
Monthly Bulletin , concerns about growth were again given as the main
reason for the cut: “The decision to lower the key ECB interest rates
reflects the Governing Council’s assessment that the outlook for price
stability over the medium-term has improved in recent months, owing in
particular to the subdued pace of economic growth and the appreciation
of the exchange rate of the euro.” When responding to questions at the
press conference after the March 6th Council meeting, ECB President
Duisenberg indicated that the staff’s internal growth forecast for 2003
had again been revised down by about ½% to around 1%. Press reports
published later suggested that the inflation forecast had not changed.
Again, the cut of 25 basis points was consistent with the downward
revision to growth, in the framework of a simple Taylor rule.
5 June 2003
During the run-up to the Iraq war, the ECB expressed its hope that a swift
and successful conclusion of the conflict would reduce uncertainty,
restore business and consumer confidence and lay the foundation for
recovery in the second half of 2003. Despite the military success of allied
forces and the subsequent fall in oil prices, business confidence declined
in the aftermath of the war. This may have been related to concerns about
the consequences of political disagreements between the US and parts of
the EU about the Iraq war; to renewed fears about the external
environment triggered by the outbreak of the SARS virus in China; and,
to dissatisfaction with domestic economic policies (especially the slow
pace of structural reform). Renewed disappointment about the economic
performance and concern about the effects of the sharp rise of the euro
led markets to expect a further monetary easing in early summer of this
year. Continuing pressure to reduce structural budget deficits and the
possibility of an even easier monetary policy stance in the US also
supported expectations of another rate cut by the ECB.
46
ADJUSTING TO LEANER TIMES
47
AN ASSESMENT OF ECB POLICY
3.0
2003 DB
2.5 2002
forecasts
2.0
0.5
0.0
J M M J S N J M M J S N J M
2001 2002 2003
48
ADJUSTING TO LEANER TIMES
of the cyclically neutral rate, the output gap, and the difference between
the actual and targeted inflation rate. A modified version of his original
rule is given by:
49
AN ASSESMENT OF ECB POLICY
13
Equating the real interest rate to the real growth rate of output can be
rationalised with theories of inter-temporal maximisation of utility from
consumption and steady-state economic growth. Combining the Ramsey and
Solow growth models, Solow has shown that under certain simplifying
assumptions, the real equilibrium interest rate can be set equal to the steady state
growth rate of output (which is equal to the sum of population growth and the
rate of technical progress, see R.M. Solow, 1969). For this equality to hold,
however, the economy must grow at its steady state rate, and production and
social utility functions must be of a certain form.
14
Following the recent clarification of the definition of price stability, the point
estimate of the ECB’s desired inflation rate is, over the medium-term, likely to
shift up to 1.75%. At the same time, after several years of very low growth, the
estimate of potential growth may well ease to 1.75%. However, the nominal
neutral rate would remain at 3.5% after these adjustments.
15
Given the well known end-of-period estimation errors of HP filters, we used
data estimated with the filter only until the fourth quarter of 2000, and
50
ADJUSTING TO LEANER TIMES
which we also assume to be 1.5%. For the weights of the output gap and
the difference between actual and targeted inflation (inflation gap), we
have used 0.6 and 0.4, respectively. These weights have tended to
produce a somewhat better fit of the Taylor rates to actual rates than the
usual weight of 0.5 for each variable.
Quarterly averages of three-month rates calculated with this formula and
actual rates are given in Figure 4.5 below. There are two periods in which
there are clear deviations of actual rates from the theoretical ones: the
spring of 1999 and the autumn of 2001. In both cases, the deviations can
be explained by exceptional events. During the spring of 1999, the ECB
(somewhat belatedly) reacted to fears of deflation triggered by the long-
term capital management (LTCM) crisis; in the autumn of 2001, the ECB
cut rates more aggressively than suggested by economic fundamentals in
the wake of the terrorist attacks in the US on September 11. But overall,
the chart suggests that we may have set the neutral rate too high.
Figure 4.5 Euroland three-month rate and the calibrated Taylor rate
5.5
5.0
4.5
4.0
3.5
2.0
1999 2000 2001 2002 2003
Our earlier discussion of the ECB’s performance over the last year
concluded that the bank is increasingly using its staff forecasts for
monetary policy decisions. This suggests that we should perhaps use the
extrapolated the series for the first quarter of 2001 to the first quarter of 2003. In
the fourth quarter of 2000, potential growth was estimated at 2.1%. We kept this
estimate for the first and second quarters of 2001, and lowered it to 2.0% for the
remainder of the period (i.e. until the first quarter of 2003).
51
AN ASSESMENT OF ECB POLICY
16
We found that economists’ one-year advance forecasts were highly correlated
with present economic conditions during the 1990s (with a correlation
coefficient of 0.9 for inflation and 0.8 for GDP growth).
52
ADJUSTING TO LEANER TIMES
5.5
5.0
4.5
4.0
3.5
2.0
1999 2000 2001 2002 2003
17
We also estimated the Taylor equation in first differences, to cross check the
level estimates. The results supported the Taylor framework for an interest rate
reaction function for the ECB. Specifically, we obtained:
di3m = -0.01 + 0.72 dogap + 0.30 digap; adjR2 = 0.49; DW = 1.57
(-0.14) (3.22) (1.04)
53
AN ASSESMENT OF ECB POLICY
economic cycles, which would probably require two decades of data. The
difference between the observed neutral rate and a theoretically plausible
level implies that our estimated Taylor formula will probably not be
stable over time. As we add new data in future years, the parameters are
likely to change, with the implied neutral rate likely to increase.
Against this background, any interest rate forecast based on the estimated
Taylor equation can only be indicative. At present, based on consensus
growth and inflation forecasts (and assuming continuing stability of the
equation), our estimated Taylor equation suggest that the refi rate may
fall to 1.5% by the end of this year and remain there until the second half
of 2004.
Even after four-and-half years of operation, the ECB is still occasionally
criticised for not communicating clearly enough or for being obsessed
with inflation and neglecting ‘growth’. Our review of recent decisions
and estimation of an ECB interest rate reaction function suggest that this
criticism is wrong. Earlier flaws in communications appear to have been
corrected and hick-ups can largely be avoided. In our view, the ECB has
made considerable progress in this regard over the last one to two years.
Moreover, our analysis shows that the ECB can be very well explained by
using a standard Taylor rule. The Council seems to give equal weight to
the output and inflation gaps, and to lean towards interest-rate smoothing.
All this is astonishingly similar to the behaviour of other central banks at
present and in the past. There is no support for the accusation that the
ECB would be an overly zealous inflation-fighter and not responsive to
real economic developments. On the contrary, the ECB seems to go
almost out of its way to follow a well balanced monetary policy. If there
is a risk, then it is that continuous, biased criticism pushes the ECB too
far away from the focus on preventing inflation, towards stimulating
demand and worrying about deflation. We certainly would not want to
claim that the probability of deflation is zero. But as we will explain in
more detail later, the risk of deflation in the euro area is fairly small and
tends to be overestimated by some financial market participants.
54
ADJUSTING TO LEANER TIMES
-1
-2
-3
-6
1992 1994 1996 1998 2000 2002
55
AN ASSESMENT OF ECB POLICY
2.0 2.5
1.5 2.0
1.0 1.5
0.5 1.0
1998 1999 2000 2001 2002 2003
Figure 4.9 Because the Federal Reserve could cut rates faster…
8
Euro Zone
6 US
-2
1991 1993 1995 1997 1999 2001 2003
56
ADJUSTING TO LEANER TIMES
Figure 4.10 … the US economy received a big boost from fiscal policy
2.0 United States
restrictive
1.5 Euro area
1.0
0.5
0.0
-0.5
-1.0
-1.5
Fiscal impulse (change in the structural
-2.0 budget deficit) as a % of GDP
-2.5
1986 1988 1990 1992 1994 1996 1998 2000 2002
central banks and governments in the early 1980s had ended in a Pyrrhic
victory? Concerns began to rise when the Asian and Russian crises in
1997-98 increased financial market instability and the burst of the
Internet bubble in 2000 caused stock prices to fall on a worldwide scale.
An increasing number of financial market participants now fear that the
virus will spread from Japan to the rest of the world. In this section, we
discuss whether these fears are warranted. Our main conclusion is that
there is little risk of Euroland following Japan into deflation. Deflation is
not fate. It is a sequence of serious economic policy mistakes that pushes
an economy into deflation and keeps it there. We are quite hopeful that
the European economic policy-makers will avoid such a sequence of
mistakes.
Before we explain our view in more detail, we need to clearly define the
subject of our analysis. We focus on the malign form of deflation (or
‘corrosive deflation’, according to Alan Greenspan), which is
characterised by a fall in the general price level and aggregate demand.
This is what occurred during the Great Depression, and what has afflicted
Japan in recent years. There is also a benign form of deflation,
characterised by a falling price level and rising demand. This may occur
when technical progress or an improvement in the terms of trade causes a
fall in prices that stimulates real demand.
Does benign deflation raise the risk of malign deflation? Not
necessarily. 18 In an environment of benign deflation, where productivity
growth is high or terms of trade improve, economic growth is robust and
real interest rates are positive (see Figure 4.11). Unless the price level
falls at a very fast pace – which is unlikely in normal circumstances –
positive real interest rates require positive nominal rates. Moreover,
government revenues are likely to grow at a healthy rate while cyclical
effects dampen spending growth. Hence, both monetary and fiscal
policies have room for manoeuvre in case of a negative demand shock.
Consumers are accustomed to a falling price level and hence unlikely to
suddenly change their behaviour and hoard liquidity when a negative
demand shock occurs.19 All this suggests that there is little reason to
worry about benign deflation, and that we may focus on its malign form
as follows.
18
We measure deflation here as the probability of malign deflation.
19
This could be different if falling prices raised consumer debt in real terms to
unsustainable levels. But there are no signs of an excessive consumer debt
burden in Euroland.
58
ADJUSTING TO LEANER TIMES
2.5
2.0
1.5
1.0
0.5
2000 2001 2002 2003
59
AN ASSESMENT OF ECB POLICY
Long before the beginning of EMU, a fairly well integrated market for
tradable goods, services and capital had emerged in the European Union.
The project launched in the early 1980s to create a Single Market by 1992
extended trade and capital markets integration with some success into
previously uncovered areas, such as financial and professional services.20
The creation of EMU in 1999 and the introduction of euro notes and
coins in 2002 have fostered integration further and induced (pre-tax)
price conversion of tradable goods and services. Divergence of inflation
rates among countries has not disappeared, but it has narrowed, compared
to pre-EMU conditions.
In this environment, we would expect prices of tradable goods and
services to be largely determined at the Euroland level instead of at the
country level. In these circumstances, deviations of inflation rates in a
particular country from the common inflation rate would largely reflect
changes in prices for non-tradable goods and services. Thus, for a large
number of companies, the Euroland market and the Euroland price level
are much more important than the national or regional markets and the
price levels calculated for these markets. Only companies that exclusively
supply to national or regional markets depend entirely on the supply-and-
demand conditions that exist there. By the same token, the prices of a
large number of consumer goods and services are determined at the
Euroland level, and only the prices of locally offered goods and services,
notably housing, exclusively reflect supply-and-demand conditions in the
regional and national markets.
As long as the common inflation rate for tradable goods and services
prices remains positive, low or even negative inflation in an EMU
member country would be indicative of a depreciation of the real
exchange rate of this particular country within EMU, rather than of
macroeconomic deflation. Depreciation would improve the country's
competitiveness, allowing it to gain market share and increase
production.
Hence, we find Euroland indicators of prices and price expectations, of
money and credit growth, and of the real economy as well as the
exchange rate more appropriate data for assessing deflation risks in
Euroland than national or regional data. National asset price indicators
and the health of national financial sectors are important, to the extent
that they have a bearing on Euroland inflation and financial conditions.
Thus, our approach differs from that of the IMF in that we regard the euro
20
Thus, in 2002, total Euroland exports (the sum of intra- and extra Euroland
exports) amounted to 36% of GDP.
60
ADJUSTING TO LEANER TIMES
area as the relevant economic entity for the analysis of deflation risk.
Recent inflation and monetary and real economy developments in
Euroland do not show any symptoms of deflation.
i) Prices and price expectations
Recent price developments and the near-term outlook give little reason to
worry about deflation in Euroland. Headline consumer price inflation has
come off its peak of almost 3.5% recorded in May 2001 and is presently
hovering around 2%. Despite lower oil prices, it is expected to ease only
moderately in the near term. Moreover, core inflation, which abstracts
from food and energy price developments, has remained close to 2% and
is expected to ease only slowly in the course of the next 12 months. One
reason for the continuously high level of core inflation is the second-
round effects of earlier food and energy price increases, which are
presently working their way through the economy. Another reason is last
year’s acceleration of wage growth, which is still exerting upward
pressure on unit labour cost growth (see Figure 4.12). The year-on-year
increase in hourly compensation accelerated to 7% in the fourth quarter
of last year from 6% in the same period of the preceding year. In view of
the sharp slowdown of productivity growth, unit labour cost growth rose
to 3.7% in the fourth quarter of 2002 from 3.4% a year ago. Upward
pressure from labour cost developments is likely to militate against a
more pronounced decline in inflation, owing to weaker economic growth.
Finally, higher productivity increases in some countries may be
associated with persistently higher inflation there (the so-called Balassa-
Samuelson effect).
-1
-2
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
61
AN ASSESMENT OF ECB POLICY
60 6
Expected inflation (lhs)
50 5
Actual inflation, % yoy (rhs)
40 4
30 3
20 2
10 1
0 0
1991 1993 1995 1997 1999 2001 2003
21
In fact, ‘perceived inflation’ in the wake of the euro cash introduction is still
high.
62
ADJUSTING TO LEANER TIMES
18 M1 growth, % yoy
16
M3 growth, % yoy
14
12
10
0
1981 1984 1987 1990 1993 1996 1999 2002
22
In a recent speech, Marian Bell, member of the Bank of England Monetary
Policy Committee, illustrated this point: “In the five years to April 1933, the
stock of both the M1 and M2 aggregates fell around 30% (in the US). In Japan,
there was a sharp slowdown in the annual growth of broad money (M2 plus
CDs) from double-digit rates of growth in 1990 to around zero in 1992” (see
Bell, 2002).
63
AN ASSESMENT OF ECB POLICY
3.0 Forecast
2.5 Output Gap
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
1990 1992 1994 1996 1998 2000 2002 2004
64
ADJUSTING TO LEANER TIMES
and drag the rest of Euroland with it. How seriously do we need to
consider these risks in Euroland?
i) Asset prices
Consider the development of stock prices first. Since their all time high
reached in March 2000, Euroland stock prices (measured by the DJ
EURO STOXX index) fell by about 50%. Thus, the performance of the
Euroland stock market was even worse than that of the US market, where
prices fell by around 40%. However, it seems that the decline in stock
prices has had a smaller impact on consumption and investment in
Euroland compared to the US.
According to a recent ECB study (see ECB Monthly Bulletin, September
2002), the marginal propensity to consume is around 1 cent per euro of
equity wealth compared to 3-7 cents per dollar in the US. Thus, a
permanent decline in stock prices by 50% would imply a long-term
decline in consumption from its counterfactual level by only about 0.3%.
There is an even weaker relationship between stock price developments
and gross fixed capital formation in Euroland. As the ECB study shows,
gross fixed investment rose slowly relative to GDP between 1995 and
2001 while stock prices first surged and then plunged. Hence, the ECB
concludes that “…the direct impact of stock market developments on
economic activity…can be expected to be rather limited, though
discernible” (ECB Monthly Bulletin, September 2002, p. 30).
Given widespread home ownership, house price developments may have
a stronger influence on consumption than stock price changes. A sharp
drop in house prices may not only weaken consumption by making
consumers feel less wealthy, but also boost mortgage defaults if it
compresses homeowners’ cash flow (e.g. through a decline in rents
received from let properties) or wipes out their equity capital. 23
According to the ECB, following a period of stability between 1993 and
1997, house prices began to rise in the Euroland average as of 1998.
House price inflation peaked at a little more than 7% in 2000, easing back
to about 6% in 2002 (see ECB Monthly Bulletin , October 2002, p. 26).
With consumer price inflation running at around 2%, this translates into a
real house price increase of 4%, which is hardly worrisome. Nevertheless,
the average rate of house price inflation reflects considerable differences
among Euroland countries. Prices have increased more strongly in some
23
For the relationship between house prices and economic activity, see Girouard
and Blöndal (2001).
65
AN ASSESMENT OF ECB POLICY
2.0 1.2
1.1
2.5 1.0
trough at April 2003
0.9
3.0 March 85/
June 01 0.8
3.5 0.7
t-6 t-5 t-4 t-3 t-2 t-1 t t+1 t+2 t+3 t+4
years before trough - years after trough
67
AN ASSESMENT OF ECB POLICY
signs of credit rationing. A fragile German banking sector raises the risk
for the entire Euroland banking system.
10
4
Credit to the private sector, % yoy
2
1992 1994 1996 1998 2000 2002
iv) Germany
After robust growth in 2000, the German economy slowed sharply in
2001 and recorded marginally positive growth in 2002. The outlook for
2003 is not much better. At the same time, inflation has come down to
less than 1% while credit growth has slowed sharply. It is widely
believed that real interest rates are too high for Germany and the Stability
and Growth Pact imposes severe constraints on fiscal policy as an
instrument for supporting demand. With the government having
difficulties implementing much needed structural reforms, concerns are
growing that Germany is locked into economic stagnation, which could
turn into outright deflation.
Clearly, in a low-inflation environment, any further drop in oil prices or
rise in the euro could induce a temporary decline in the German price
level. But would a sinking price level in Germany signal harmful
deflation in that country or raise deflation risks for Euroland? First, a
decline in the German price level in an environment of positive inflation
rates at the Euroland level would not necessarily signal the beginning of a
malign deflation spiral. With German prices declining, competitiveness
69
AN ASSESMENT OF ECB POLICY
24
European Central Bank (2002), Report on financial structures, ECB Frankfurt.
70
ADJUSTING TO LEANER TIMES
sensitive areas, as seems to have happened in the export sector. Higher costs
caused by exploding non-wage labour costs will simply raise prices in the
domestic sector. This leads to a loss in competitiveness of the formal domestic
sector against the informal sector, the so-called shadow economy. With demand
shifted from the formal to the informal sector, growth of official GDP and the tax
base declines, while officially recorded unemployment increases.
There are indeed signs of rapid growth in Germany’s shadow economy.
According to Friedrich Schneider, an economics professor at the University of
Linz and an expert in estimating the shadow economy, unrecorded German
nominal GDP grew at annual average rate of 5.5% in 1995-2003 compared with
2.25% for recorded GDP (see Schneider, 2003). If we add Professor Schneider’s
estimates of the shadow economy to official GDP and employment, total real
GDP and employment grew at annual average rates of 1.7% and 1.1%,
respectively, compared with 1.3% and 0.5% for the official counterparts.
If Germany’s key economic problem is not a traditional lack of competitiveness
that can be cured by (‘internal’) real exchange rate devaluation, it may take
longer for structural reforms to show the expected pay-off. For instance, a
reduction in unit labour costs as a result of a cut in payroll taxes that were made
feasible by social security reform, may not lead to a quick and meaningful boost
in export demand. Rather, lower labour costs may be needed to induce a shift of
resources from the shadow to the official economy.
Another implication of our analysis is that the appreciation of the euro may not
have the widely expected devastating effect on German exports. Given their past
experience, German companies may be more capable of dealing with a
deterioration of their price competitiveness than some of their Euroland
competitors. Hence, countries that in the past relied more heavily on improved
price competitiveness to boost growth and employment, may feel more affected
by euro appreciation. If this argument holds, the gap in growth between
Germany and France, which benefited from larger gains in cost competitiveness
in the past, could narrow in the future.
72
ADJUSTING TO LEANER TIMES
25
The econometrically estimated Taylor equation should not, of course, be used
as a benchmark to assess the ECB’s performance. However, our calibrated
Taylor equation is based on economically plausible, rather than econometrically
estimated, parameter values. Hence, this equation has a more normative
character and may serve as a benchmark to analyse ECB action.
73
AN ASSESMENT OF ECB POLICY
pre-emptive action if such risks appear on the horizon. This view has
been reinforced by the ECB’s recent clarification of their goal of price
stability (see the following section). Moreover, monetary policy still has
considerable room to move rates lower. And even when the level of
nominal rates approaches zero, a determined central bank may ease
monetary policy further by pumping liquidity into the economy.
74
ADJUSTING TO LEANER TIMES
ECB has made no major mistakes during the first four years of
operations.
Secondly, the critics have argued that the upper limit of ‘less than 2%’ is
too low. However, there were numerous exogenous shocks to inflation
during the first four years of EMU (oil, food, euro cash introduction,
exchange rate changes) that could all be absorbed without much
monetary policy stress. Despite the numerous shocks, inflation has
averaged close to 2% since the beginning of EMU, and there have been
no signs that this average will move higher or lower in coming years.
Much has also been made of the results of the Balassa-Samuelson (B-S)
effect in the context of EMU enlargement. However, in a recent study, we
put this effect at 1% to 2.5% for the accession countries (Gros et al.,
2002b). Taking the average of 1.75% (which is also the estimated B-S
effect for Poland) and assuming the share of EMU accession countries at
6% of EMU GDP would imply an increase in EMU inflation by 0.1% due
to the B-S effect. That hardly justifies raising the ECB’s inflation target.
Thirdly, the critics have claimed that the first pillar (analysis of monetary
and credit aggregates) would be redundant and lead to confusion. We do
not share the view that it is redundant. Apart from being able to signal a
long-term inflation risk, a close analysis of credit and monetary
developments helps to keep an eye on asset price inflation (see below).
Historical experience in Japan, and more recent experience in the US and
the UK, show that a too-narrow focus on consumer price inflation in a
one-to-two-year time horizon, may expose central banks to the risk of
missing the emergence of a liquidity-driven asset price bubble. Such a
bubble may not have a tangible effect on consumer prices during its
expansion phase, but could cause consumer price deflation when it bursts.
Hence, to avoid financial instability and deflation, a separate analysis of
money and credit developments appears warranted.
Moreover, we do not think the monetary pillar continues to lead to
confusion. Markets and analysts have gained a sufficiently good
understanding of how the ECB looks at the monetary pillar, so that the
release of an M3 growth figure above consensus expectations currently
fails to trigger any market reaction. Should monetary developments turn
into a risk for price stability in the medium-term, the ECB could easily
signal this to the markets and change market reaction to the M3 releases.
A few clarifications
Given that criticism of the strategy is not well founded, it did not come as
a surprise that the review brought mostly presentational changes, such as
76
ADJUSTING TO LEANER TIMES
a reordering of the two pillars and the foregoing of the fixed annual
reassessment of the M3 reference value. There was, however, one
important clarification. Apart from reconfirming the definition of price
stability as “a year-on-year increase in the HICP for the euro area of
below 2%”, the Council also agreed “that in the pursuit of price stability
it will aim to maintain inflation rates close to 2% over the medium term”.
Some commentators have interpreted this as a ‘relaxation of the inflation
goal’. We disagree. The ECB’s clarification essentially establishes the
way it pursued its goal of price stability over the last four years as a
benchmark for its future conduct. Since 1999, the monthly year-on-year
inflation rate averaged 2.0%, close to the ECB’s clarified definition of
price stability (see Figure 4.18). During that period, the ECB tended to
embark on monetary tightening when inflation trended upward, and it
eased when inflation trended downward. The momentum of monetary
policy tightening rose the more that inflation exceeded the 2% mark, and
easing became more aggressive the further that inflation fell below 2%
(in a deliberate attempt to minimise the risks of deflation). In effect, the
ECB behaved as if it wanted to see inflation of less than 2%, but not
much less. This pragmatic approach to the pursuit of price stability now
appears to have been formalised.
The reordering of the two pillars – with the economic analysis now
coming first and the monetary analysis second – and the omission of an
annual reassessment of the reference value of M3 growth, are further
reconciliations of theory with practice. During most of its existence, the
ECB has conducted monetary policy on the basis of its economic
analyses and forecasts, and used the monetary analysis only as a
crosscheck for the longer-term inflation outlook. This approach has now
become the new benchmark. It is a little surprising that the Council could
not agree to add a reference to the monetary pillar as a gauge for asset
price inflation. ECB Chief Economist Ottmar Issing had played with this
idea in a number of speeches delivered in the months before the
conclusion of the strategy review, but apparently could not convince his
colleagues of such a reference. In any case, omission of an explicit
reference to the monitoring of asset price developments through money
and credit developments will not preclude future use of the monetary
pillar in the strategy for this purpose.
77
AN ASSESMENT OF ECB POLICY
3.5 5.0
3.0 4.5
2.5
4.0
2.0
3.5
1.5
3.0
1.0 HICP, % yoy (lhs)
78
ADJUSTING TO LEANER TIMES
26
In this respect, the ECB has already stated its intention to improve statistical
coverage of financial indicators, such as the launching of a loan officer survey
and the compilation of new indicators of indebtedness.
27
Ireland is an interesting case. The Central Bank of Ireland notes: “Between
1996 and 2000, new house prices rose by 92% and existing house prices by
126%. Nominal year-on-year growth rates for prices of existing houses peaked at
over 35% in 1998. It should be noted that overall liquidity in the Irish economy
(measured by either money supply growth rates or private credit sector credit)
was highly correlated with property price increases” (see G10 Contact Group…,
2002).
79
AN ASSESMENT OF ECB POLICY
12 16
M3 14
10
Private loans 12
Asset prices 10
8
(rhs)
8
6 6
4
4
2
0
2
-2
0 -4
1992 1994 1996 1998 2000 2002
80
ADJUSTING TO LEANER TIMES
20 Money growth 16
Period averages: M growth
18 14
Inflation
16 Period averages: inflation
12
14
12 10
10 8
8 m 6
6
4
4
2 2
0 0
1972 1976 1980 1984 1988 1992 1996 2000
14
M3, % yoy
12
HICP, % yoy
10
0
1981 1984 1987 1990 1993 1996 1999 2002
Our conclusion from the above discussion is that, with monetary and
especially credit developments closely linked to financial stability, a
separate monitoring of these variables puts the central bank in a better
position to achieve financial and hence long-term price stability.
81
AN ASSESMENT OF ECB POLICY
28
Taking account of cyclical effects on credit losses is known as dynamic
provisioning. Although it could make an important contribution to financial
stability in Euroland countries, dynamic provisioning is presently practiced only
82
ADJUSTING TO LEANER TIMES
in Spain. (The role of micro policies for financial stability is discussed in G10
Contact Group…, 2002).
29
In recognition of the shortcomings of pure inflation targeting, Bernanke and
Gertler (1999) have advocated a more flexible approach to inflation targeting,
providing a unified framework for achieving both general macroeconomic and
financial stability. Thus, it is conceivable that a revised ECB strategy and a more
flexible inflation-targeting strategy converge to the same model.
83
AN ASSESMENT OF ECB POLICY
84
CHAPTER 5
THE REVIVAL OF THE EURO
s discussed in earlier parts of this report, the external value of the
85
THE REVIVAL OF THE EURO
existence. But it is not only the change of direction in the exchange rate
that seems to justify a reconsideration of the issue; two other factors have
emerged to make the subject more topical.
One is the re-emerging divergence over the role of macroeconomic
policies between the monetary and political authorities in the US on the
one hand and in the euro area on the other hand. The other factor is the
limited attention given to the issue by the European Convention, which
delivered its (in some other respects quite radical) draft Constitutional
Treaty to the European Council in June 2003.
We shall deal with these two factors in turn, leading up to the argument
that there is indeed a good case for reinforcing the political element in the
governance of the euro area and, in particular, in an external role. But
first it is necessary to recall the main provisions of the present system.
5.1 How has the present framework for decision-making and external
representation worked so far?
The Maastricht Treaty seems to offer a clear answer to the question of
who is responsible for the external value of the euro. The formulation of
the primary objective of monetary policy as the maintenance of price
stability leaves little room for any exchange rate target. The treaty went to
considerable lengths to minimise the potential threats to the pursuit of the
primary objective arising from public sector deficits and their financing,
as well as from exchange rate considerations. But a certain tension
persisted in the latter area between this ambition and the traditionally
central role of the political authorities in shaping exchange rate policy.
The Treaty of Maastricht determined in Art. 109 (to be slightly modified
in the proposed Art. III 223 of the Draft Constitutional Treaty, see Box
5.1) a role for the political authorities to participate in the management of
the external value of the euro in two distinct circumstances.
First, the Council of Ministers is empowered to enter into “formal
agreements on an exchange-rate system” for the euro vis-à-vis non-EU
currencies. But two preconditions will greatly complicate any such
decision: it has to be reached unanimously and there has to be prior
consultations with the Eurosystem to assure consistency with the
objective of price stability. This remains a very unlikely scenario.
Secondly, in the absence of any formal agreements, the Council of
Ministers may, by qualified majority “formulate general orientations for
exchange-rate policy” – again after assuring that such orientations are
“without prejudice to the primary objective of price stability”.
86
ADJUSTING TO LEANER TIMES
Box 5.1 Rules for determining exchange rate policy for the euro area
1. By way of derogation from [Article III-222 (ex 33)], the Council, acting
unanimously on a recommendation from the European Central Bank or from
the Commission, following consultation with the European Central Bank
with a view to reaching a consensus compatible with the objective of price
stability and after consultation with the European Parliament in accordance
with the procedure laid down in paragraph 3 for the arrangements there
referred to, may conclude formal agreements on a system of exchange rates
for the euro in relation to non-Union currencies. The Council may, acting by
a qualified majority on a recommendation from the European Central Bank
or the Commission and after consulting the European Central Bank in an
endeavour to reach a consensus consistent with the objective of price
stability, adopt, adjust or abandon the central rates of the euro within the
exchange-rate system. The President of the Council shall inform the
European Parliament of the adoption, adjustment or abandonment of the
central rates of the euro.
2. In the absence of an exchange-rate system in relation to one or more third-
country currencies as referred to in paragraph 1, the Council, acting either on
a recommendation from the Commission and after consulting the European
Central Bank or on a recommendation from the European Central Bank, may
formulate general orientations for exchange-rate policy in relation to these
currencies. These general orientations shall be without prejudice to the
primary objective of the European System of Central Banks, to maintain
price stability.
3. By way of derogation from [Article III-222 (ex 33)], where agreements on
matters relating to the monetary or exchange-rate system are to be the subject
of negotiations between the Union and one or more states or international
organisations, the Council shall, acting on a recommendation from the
Commission and after consulting the European Central Bank, decide the
arrangements for the negotiation and for the conclusion of the agreements.
These arrangements shall ensure that the Union expresses a single position.
The Commission shall be fully associated with the negotiations.
4. Without prejudice to Union competence and agreements as regards economic
and monetary union, Member States may negotiate in international bodies
and conclude international agreements.
Source: Draft Treaty establishing a Constitution for Europe (CONV802/03), Article III (International
Agreements) revised text of 12 June.
87
THE REVIVAL OF THE EURO
the political authorities have a more limited role than their counterparts in
the United States (or Japan).
The pre-eminence of the monetary authorities in Euroland is reinforced
by the fact that official international reserves in the latter two countries
are owned by the respective Treasuries, which are accountable to their
political authorities, whereas in the euro area ownership of reserves has
been transferred to the Eurosystem (i.e. the ECB). In the transatlantic
relationship there is a certain asymmetry between the respective
assignments of responsibility, but that has not prevented a joint
understanding between the dominant actors that foreign exchange
interventions should be used very sparingly while domestic
considerations are allowed to dominate in the design of monetary policy.
Developments over the early years of the euro have not made further
clarification of the assignment of responsibilities quite as urgent an issue
as was expected at the start. During the initial two years of euro
depreciation, there was only one publicly known controversy – associated
with the suggestion by the then German Finance Minister Oskar
Lafontaine, to seek a target zone arrangement for the euro. Although
France was not totally unsympathetic to the idea, no effort was made
during 1999 nor most of 2000 to formulate ‘general orientations’ for the
euro exchange rate. To have done so would have been difficult; while the
Eurosystem became increasingly concerned about the inflationary impact
of the continuing slide in the euro, several participating governments
were not unhappy with the consequent improvements in competitiveness,
at a time when there were still perceptions of slack in their economies,
even though these perceptions now seem to have been unfounded.
In retrospect, the output gap in the euro area seems to have been
eliminated in the course of 2000, as we have argued in the analysis of
productivity developments, hence weakening the case for any major
stimulus from external demand through a weak euro. But as the euro
weakened further in the spring and summer of 2000, there was a
convergence of views. The concern over inflation in the Eurosystem
coincided in the end with a sense of political embarrassment among
governments, over the performance of the euro in the currency markets,
creating support for interventions to stem the slide. Such interventions
were undertaken jointly with other G7 governments in September – even
though it was fully recognised that the normal case for intervention,
based on a build-up of short-term positions against a currency, was not
observable. The main cause of depreciation was a major and rather steady
outflow of long-term funds from the euro area, primarily towards the
88
ADJUSTING TO LEANER TIMES
5.2 The increasing importance of the external side for the euro area
The appreciating euro and policy assignments in the euro area
The euro has gradually moved back near its starting level vis-à-vis the US
dollar, i.e. €1=$1.17. This in itself should not be a source of concern,
since most of the available research suggests that this is fairly close to,
though probably still below, the longer-term equilibrium for the most
important bilateral exchange rate. But as argued previously, in effective
rate terms, the euro is back to its strength of 1995. This is due to the close
correlation of some other important third-country currencies with
movements in the dollar/euro rate, notably the pound sterling, the yen and
several East Asian currencies; some of the latter are even pegging to the
US dollar. Still it has to be noted that although the pace of appreciation
has been swift at times since the spring of 2002, it has been fairly smooth
and hence not providing any justification for action on the grounds of
correcting disorderly market conditions. Appreciation has, as argued
above, helped to reduce inflation. The euro area still has a small surplus
89
THE REVIVAL OF THE EURO
90
ADJUSTING TO LEANER TIMES
91
THE REVIVAL OF THE EURO
92
ADJUSTING TO LEANER TIMES
initially sought by the administration, will have its major effect in 2003-
04, through the advancement of earlier announced tax cuts and will
clearly have the effect of bringing demand growth temporarily back
above the growth of potential output. Despite the fiscal expansion, not
only short-term, but also long-term interest rates remain at historically
very low levels, facilitated by the low inflation expectations that have
been encouraged through the loud concerns expressed about the risks of
deflation and the vigorous way in which the Federal Reserve has
announced its intention to deal with that risk. Were this risk to be
recognised as exaggerated, interest rates would no doubt move up quite
strongly. Combined with the relatively high levels of debt, particularly in
the household sector, the prospect of the continuing strength of consumer
demand looks precarious.
The contrasting approaches to macroeconomic policies imply a major
potential for conflicts across the Atlantic. The US authorities clearly find
the euro area unresponsive to the need to restore growth outside the
United States, while the euro area authorities are increasingly concerned
about the perceived lack of fiscal responsibility in the United States. In
the absence of other corrective mechanisms, the adjustment falls squarely
on the US current account and depreciation of the US dollar. The drain of
demand from the deteriorating external position, as the result of
continuing faster growth in the United States, overwhelms the effect of
improving US competitiveness through a weaker dollar.
Thirdly, the US current account deficit has become larger than any
realistic prospect of financing it smoothly through an inflow of longer-
term funds. The US current account is now widely expected to reach
about 5% of GDP during the next year. Even with parallel growth in the
United States and its trading partners, the US current account has shown a
historical tendency to deteriorate, and growth will continue to be faster
this year and next. A return to a more neutral stance of monetary policy in
the light of the fuller use of resources would add further to the size of the
US current deficit and bring it unmistakably into unsustainable territory,
which even major depreciation could not correct. This would no doubt be
more painful for the euro area economy than for the United States itself,
although it could be argued that it has been brought about primarily by
US domestic policies. For this reason, the role of reinforced external
representation at the political level should not principally be to attribute
blame for past events and policies, but to discuss with the US policy-
makers and others in global macroeconomic fora how the unsustainable
position could unfold in a more constructive way.
93
THE REVIVAL OF THE EURO
94
ADJUSTING TO LEANER TIMES
95
THE REVIVAL OF THE EURO
30
The wording of this article is still under discussion. See modifications in
CONV836/03, 27 June 2003.
96
ADJUSTING TO LEANER TIMES
and a half years. This solution would also make sense in light of the
formal recognition of the Eurogroup in the constitution (albeit as an
informal forum) and the fact that for some time to come, the countries of
the eurozone might constitute a minority within the EU.
97
REFERENCES
Ahearne, A., J. Gagnon, J. Haltmaier and S. Kamin (2002), Preventing
Deflation: Lessons from Japan’s Experience in the 1990s,
International Finance Discussion Paper No. 729, Board of Governors
of the Federal Reserve System, June, pp. 38-39.
Begg, D., F. Canova, P. De Grauwe, A. Fatás and Ph. Lane (2002),
Surviving the Slowdown , Monitoring the European Central Bank No.
4, Centre for Economic Policy Research (CEPR), London, April.
Bell, Marian (2002), ‘Six Months on the MPC: A Reflection on Monetary
Policy”, excerpt from a speech delivered to the confederation of
British Industry, Crawley, 9 December.
Bentolila, S. and J.F. Jimeno (2003), “Spanish Unemployment: The End
of the Wild Ride?”, CESIfo Working Paper No. 940, CESIfo,
Munich.
Bernanke, B. and M. Gertler (1999), “Monetary Policy and Asset Price
Volatility”, Economic Review, Federal Reserve Bank of Kansas City,
84 (4).
Berthold, N. and R. Fehn (2003), “Unemployment in Germany: Reasons
and Remedies”, CESifo, Working Paper No. 871, CESIfo, Munich.
Bonaglia, F., E. la Ferrara and M. Marcelli (2000), “Public capital and
economic performance, Evidence from Italy”, draft, OEDC, Paris.
Borio, C. and Ph. Lowe (2002), Asset prices, financial and monetary
stability: Exploring the nexus, BIS Working Paper No. 114, Bank of
International Settlements, Basel, July.
Clarida, R., J. Gali and M. Gertler (1998), “Monetary Policy Rules in
Practice: Some International Evidence”, European Economic
Review, 42, pp. 1033-1067.
European Central Bank (2002), Report on Financial Structures, ECB,
Frankfurt.
EEAG (European Economic Advisory Group) (2003), Report on the
European Economy 2003 by Lars Calmfors et al., CES/ifo, Munich.
European Central Bank (2002), Labour Market Mismatches in Euro Area
Countries, ECB, Frankfurt am Main.
European Commission (1997), Quest II: A Multi-Country Business Cycle
and Growth Model, Directorate General for Economic and Financial
Affairs, Economic Papers (International), No. 123:1-50, October.
98
ADJUSTING TO LEANER TIMES
100