Uncertain inflation outlook and monetary policy normalization in
the heterogeneous euro areaZsolt Darvas
Bruegel and Corvinus University of Budapest
Macro Economy Research Conference: “Monetary Policy Normalization Ten Years after the Great Recession”
Nomura Foundation
24 October 2018, Tokyo
The ECB’s core inflation forecast has proved to be overly optimistic. Would it work this time?
2
Note:
• Actual inflation is the thick
red line (moving 12 months
average rate of change)
• The solid lines show the
inflation forecast made in
each quarter
0
0.5
1
1.5
2
2.5
200
2Q
1
200
3Q
1
200
4Q
1
20
05
Q1
200
6Q
1
200
7Q
1
200
8Q
1
200
9Q
1
201
0Q
1
201
1Q
1
201
2Q
1
201
3Q
1
201
4Q
1
201
5Q
1
201
6Q
1
201
7Q
1
201
8Q
1
201
9Q
1
202
0Q
1
Note: ECB forecasts are available for the annual average
inflation. That’s why I use the 12 month average rate of
change for the actual data, which, in each December,
equals annual average inflation. In the chart the end
observation (in Q4 of various years) of each forecast curve
corresponds to the annual average inflation forecast
numbers published by the ECB. I have linearly interpolated
this Q4 annual average forecast data and the actual inflation
rate in the quarter of the date of the forecast.
Motivation
• Low/negative ECB interest rates and large expansion of balance sheet reflect non-standard monetary policies
• Inflation, the main objective of the ECB, reached 2% in summer 2018, but partly due to oil prices, while core inflation remains stable at 1%
• Inflation forecasts suggest an acceleration of core inflation in the coming years, yet the track record of inflation forecast is bad
• At the same time, the euro area remains rather heterogeneous with diverse inflation, growth and house price developments
• How to sequence monetary policy normalisation in the euro area? What lessons to draw from the monetary policy normalisation experiences of Sweden, US & UK?
3
Global comparison: central bank balance sheet (% GDP)
4
• Sizeable differences in
pre-crisis balance sheet
sizes
• Even more so after
2008
• Switzerland: main
foreign currency
purchase
• Others: purchase of
various securities,
including government
bonds
0
20
40
60
80
100
120
140
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
20
16
2017
2018
Bank of England
Bank of Japan
European Central Bank
Federal Reserve
Sveriges Riksbank
Swiss National Bank
Global comparison: banks’ reserves at the central bank relative to banking assets (%)
5
• There is no reserve
requirement in the UK
and Sweden, so excess
reserves cannot be
calculated
• We therefore show ratio
of total banks’ reserves
to bank assets
➢ Euro area has in fact
one of the lowest values
0
5
10
15
20
25
30
35
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
20
12
201
3
201
4
201
5
201
6
20
17
201
8
Bank of England
Bank of Japan
European Central Bank
Federal Reserve
Sveriges Riksbank
Swiss National Bank
Global comparison: 1-week interbank interest rates (%), 2 Jan 2000 – 17 Oct 2018
6
• Federal Reserve: sizeable
tightening since late 2015
• Bank of England: some
tightening since late 2017
• Other 4: low interest rates
remain
Note: I show the 1-week interbank rates and not
central banks’ interest rates, because for some
central banks the importance of certain rates
changed (e.g. for the ECB, the MRO rate was the
main determinant of short-term market rates
before 2008, since then the deposit rate is the
main determinant).
-2
-1
0
1
2
3
4
5
6
7
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
GBP LIBOR 1W
JPY LIBOR 1W
EUR LIBOR 1W
USD LIBOR 1W
SEK STIBOR 1W
CHF LIBOR 1W
Monetary policy normalisation questions
• Balance sheet: shrink or not? And to what level?
• Interest rates: when to raise and up to what ‘new normal’?
7
Lessons from monetary policy exit mistakes of Sweden, the US and the UK
8
Sweden: premature monetary policy exit followed by massive tightening
• After Lehman Brothers collapse: main rate cut form 4.75% in October 2008 to 0.25% in July 2009
• July 2010 - July 2011: rate increase to 2% for financial stability reasons
• This premature monetary policy exit led to high costs in terms of excessively low inflation, overly high unemployment and a higher real debt burden for households
• After July 2011: rate cut to -0.5% & quantitative easing
9
Swedish interest rates, 2 January 2008 –12 October 2018
10
-1
0
1
2
3
4
5
Ja
n 2
00
8
Ja
n 2
00
9
Ja
n 2
01
0
Ja
n 2
01
1
Ja
n 2
01
2
Ja
n 2
01
3
Ja
n 2
01
4
Ja
n 2
01
5
Ja
n 2
01
6
Ja
n 2
01
7
Ja
n 2
01
8
Repo rate
3-month treasury billrate
10-year governmentbond yield
QE
sta
rts
QE
en
ds
purc
ha
ses
incre
ased
purc
ha
ses
reduce
d
purc
ha
ses
reduce
d
purc
ha
ses
reduce
d
Note: The repo rate has been the
Riksbank's policy rate since 1994. The repo
rate is the rate of interest at which banks
can borrow or deposit funds at the
Riksbank for a period of seven days.
The Riskbank’s repo rate: actual (thick red line) and Riksbank forecasts
11
-1
0
1
2
3
4
5
6
200
6Q
1
2007Q
1
200
8Q
1
200
9Q
1
201
0Q
1
201
1Q
1
201
2Q
1
201
3Q
1
201
4Q
1
201
5Q
1
201
6Q
1
201
7Q
1
201
8Q
1
201
9Q
1
202
0Q
1
202
1Q
1
• Apart from the short period
around 2010, the Riksbank
interest rate guidance
turned out to be grossly
inadequate
• Noteworthy that the 10-
year government bond
yield has not increased
despite the Riksbank’s
interest rate increase
forecast and the ending of
QE
Federal Reserve: ‘taper tantrum’ unnecessarily pushed-up the 10-year yield
• While QE3 was ongoing in the US, in early 2013 unemployment rate fell to 7.5%, nearing the 6.5% threshold which was announced earlier as the rate when the FED will start increase interest rates
• FOMC started to discuss “tapering” of QE in early May 2013: the 10-year yield increased from 1.7% to 3% in a few months –leading to a far larger tightening in financing conditions than the FED had intended
• Later, the 10-year yield has fallen back even below 1.7%, despite the actual tapering and ending QE and the first increase in federal funds rate in December 2015
12
US interest rates, 2 January 2000 – 15 October 2018
13
0
1
2
3
4
5
6
7
20
00
-01-0
3
20
00
-12-2
5
20
01
-12-1
7
20
02
-12-0
9
20
03
-12-0
1
20
04
-11-2
2
20
05
-11-1
4
20
06
-11-0
6
20
07
-10-2
9
20
08
-10-2
0
20
09
-10-1
2
20
10
-10-0
4
20
11
-09-2
6
20
12
-09-1
7
20
13
-09-0
9
20
14
-09-0
1
20
15
-08-2
4
20
16
-08-1
5
20
17
-08-0
7
20
18
-07-3
0
Federalfundseffectiverate
10-yeargovernment bondyield
QE
1
QE
1 e
nds
QE
3
QE
2
QE
3 e
nds
QE
2 e
nds
tap
er
tantr
um
Bank of England: some improper statements confusing markets
• July 2013: BoE releases a statement (an unusual move in the absence of a policy change) clarifying current policy and questioning whether the expected future rates were in line with economic developments
• Aug 2013: BoE introduces forward guidance, linking increase in interest rate to unemployment falling below 7%
• Feb 2014: BoE updates forward guidance, unlinking it from unemployment following the decrease of the unemployment rate below 7%
• June 2014: Mark Carney suggests that the interest rates could reach 2.5% in early 2017
14
UK interest rates, 2 Jan 2006 – 12 Oct 2018
15
0
1
2
3
4
5
6
Jan
-20
06
Jan
-20
07
Jan
-20
08
Jan
-20
09
Jan
-20
10
Jan
-20
11
Jan
-20
12
Jan
-20
13
Jan
-20
14
Jan
-20
15
Jan
-20
16
Jan
-20
17
Jan
-20
18
Bank rate 10-year government bond yield
Bre
xit
refe
ren
du
m
QE
1
QE
3
QE
2
QE
4
QE
1 e
nds
QE
3 e
nd
s
QE
2 e
nd
s
Co
nfu
sin
g
sta
tem
ents
QE
4ends
The Bank Rate is the single
most important interest rate in
the UK: the interest rate the
Bank of England pays to
commercial banks that hold
money with the Bank of
England.
Key conclusions from the QE exit experiences of the three countries
• Learn from the mistakes
• Clear forward guidance is crucial
• Premature exit has to be avoided
• Long-term interest rates have not increased when asset purchases have been stopped; not even after the first few rate increases (US, UK)
16
New normal in monetary policy?
17
Secular decline in global real interest rates
18
Trends in short-term real interest rates, 1870-2016
Source: Del Negro, Marco, Domenico Giannone, Marc P. Giannoni and Andrea
Tambalotti (2018) ‘Global Trends in Interest Rates’, NBER Working Paper No. 25039
Explanations of the secular
decline in global real rates by
Del Negro et al (2018):
• Increase in the premium that
international investors are
willing to pay to hold safe and
liquid assets (scarcity of safe
assets in the context of a
global saving glut)
• Lower economic growth
Central bank balance sheet
• No benchmark for ‘normal’ balance
• Balance sheet depends on the way monetary policy is conducted, on the exchange rate regime, past monetary policy actions, central bank tasks, profit distribution
• Arguments in favour of larger balance sheet: • (1) Lower equilibrium interest rate → zero lower bound will likely be
reached more frequently → unconventional monetary policy would be use more regularly;
• Larger balance sheet could (2) improve monetary transmission, (3) provide safe assets, (4) reduce banks’ incentives for excessive maturity transformation
• Arguments against larger balance sheet: • It exposes the central bank to financial risk and undue political influence
19
Monetary policy exit when the inflation outlook is uncertain
20
Recall: The ECB’s core inflation forecast has proved to be overly optimistic. Would it work this time?
21
Note:
• Actual inflation is the thick
red line (moving 12 months
average rate of change)
• The solid lines show the
inflation forecast made in
each quarter
0
0.5
1
1.5
2
2.5
200
2Q
1
200
3Q
1
200
4Q
1
20
05
Q1
200
6Q
1
200
7Q
1
200
8Q
1
200
9Q
1
201
0Q
1
201
1Q
1
201
2Q
1
201
3Q
1
201
4Q
1
201
5Q
1
201
6Q
1
201
7Q
1
201
8Q
1
201
9Q
1
202
0Q
1
Note: ECB forecasts are available for the annual average
inflation. That’s why I use the 12 month average rate of
change for the actual data, which, in each December,
equals annual average inflation. In the chart the end
observation (in Q4 of various years) of each forecast curve
corresponds to the annual average inflation forecast
numbers published by the ECB. I have linearly interpolated
this Q4 annual average forecast data and the actual inflation
rate in the quarter of the date of the forecast.
Ray of hope: wage growth seems to accelerate
22
• Following repeated failures in
predicting wage growth, more
recent data suggest a pick-up in
wage growth
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
20
02Q
1
20
03Q
1
20
04Q
1
20
05Q
1
20
06Q
1
20
07Q
1
20
08Q
1
20
09Q
1
20
10Q
1
20
11Q
1
20
12Q
1
20
13Q
1
20
14Q
1
20
15Q
1
20
16Q
1
20
17Q
1
20
18Q
1
20
19Q
1
20
20Q
1
Compensation of employees in the
euro area: actual data (thick red line)
and ECB forecasts
But: labour force participation continues to expand – good news for the people, bad news for inflation
23
Labour force participation rate (age 15-64, % of population)
• While Americans were
fleeing the labour
market in 2000-2015,
there has been a
steady increase in
euro area labour force
participation65
70
75
80
199
7Q
1
199
9Q
1
200
1Q
1
200
3Q
1
200
5Q
1
200
7Q
1
2009Q
1
201
1Q
1
201
3Q
1
201
5Q
1
201
7Q
1
Euro area (19countries)
Japan
United Kingdom
United States
But 2: Underemployment is higher than unemployment, moderating wage growth
24
0
2
4
6
8
10
12
14
2001
2003
2005
2007
2009
2011
2013
2015
2017
Germany
Underemployment rate
Unemployment rate
02468
1012141618
2001
2003
2005
2007
2009
2011
2013
2015
2017
France
Underemployment rate
Unemployment rate
0
2
4
6
8
10
12
14
16
2001
2003
2005
20
07
2009
20
11
2013
20
15
2017
Italy
Underemployment rate
Unemployment rate
0
5
10
15
20
25
30
35
2001
2003
2005
2007
2009
2011
2013
2015
2017
Spain
Underemployment rate
Unemployment rate
Sources: Underemployment: Bell, David N.F. and David G. Blanchflower (2018) ‘Underemployment in the US and Europe’, NBER
Working Paper No. 24927; Unemployment: Eurostat; 2018 data refer to the average of January – August 2018.
ECB deposit facility interest rate, main asset purchase announcement dates and 10-year government bond yields of four countries
25
-1012345678
2000
20
01
2002
2003
2004
2005
2006
20
07
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
France 10-year Germany 10-year
Italy 10-year Spain 10-year
ECB deposit facility rate
"Whatever it takes" speech
Mo
nth
lyp
urc
ha
se
€
80
bn
Mo
nth
lyp
urc
ha
se
€
60
bn
Mo
nth
lyp
urc
ha
se
€
60
bn
Mo
nth
lyp
urc
ha
se
€
30
bn
Mo
nth
lyp
urc
ha
se
€
15
bn
,
the
n z
ero
Note: for asset purchases the announcement
dates are indicated; the actual changes to
purchased volumes took effect typically about 2
months later.
ECB interest rate lift-off
• Key conclusion from other central banks:• Ending asset purchases might not increase interest rates (especially if maturing
asset holdings are reinvested)
• The key issues is the start date and the expected path of interest rate increases
• Premature monetary policy exit is dangerous
• My advise is to wait with interest rate increase until core inflation has actually reached a sufficiently high level. Since inflation has undershoot for long, prolonging rate increase to after a period of inflation overshooting would be desirable. In my view, the ECB’s current forward guidance is insufficient:
• “The Governing Council expects the key ECB interest rates to remain at their present levels at least through the summer of 2019, and in any case for as long as necessary to ensure the continued sustained convergence of inflation to levels that are below, but close to, 2% over the medium term.” (13 September 2018 ECB press release)
26
The euro area is heterogeneous, financial stability risks vary across countries
27
Core inflation: large heterogeneity in the euro area
28
Core inflation (moving 12 months average rate of change)
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
20
02
M0
1
20
04
M0
1
20
06
M0
1
20
08
M0
1
20
10
M0
1
20
12
M0
1
20
14
M0
1
20
16
M0
1
20
18
M0
1
Austria Belgium
France Germany
Netherlands
-3
-2
-1
0
1
2
3
4
5
20
02
M0
1
20
04
M0
1
20
06
M0
1
20
08
M0
1
20
10
M0
1
20
12
M0
1
20
14
M0
1
20
16
M0
1
20
18
M0
1
Greece Italy
Portugal Spain
-4
-2
0
2
4
6
8
10
20
02
M0
1
20
04
M0
1
20
06
M0
1
20
08
M0
1
2010M
01
20
12
M0
1
20
14
M0
1
20
16
M0
1
20
18
M0
1
Estonia Latvia
Lithuania Slovakia
Slovenia
-3
-2
-1
0
1
2
3
4
5
6
20
02
M0
1
20
04
M0
1
2006M
01
2008M
01
20
10
M0
1
20
12
M0
1
20
14
M0
1
20
16
M0
1
20
18
M0
1
Cyprus FinlandIreland LuxembourgMalta
Note: core inflation is defined as the overall index excluding energy, food, alcohol and tobacco
Credit growth: large heterogeneity in the euro area
29
60
70
80
90
100
110
120
130
140
150
160
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
Austria Belgium
France Germany
Netherlands BOOM
80
90
100
110
120
130
140
150
160
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
Greece Italy
Portugal Spain
BOOM
0
20
40
60
80
100
120
140
160
180
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
Estonia LatviaLithuania SlovakiaSlovenia BOOM
60
70
80
90
100
110
120
130
140
150
160
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
Cyprus Finland
Ireland Luxembourg
Malta BOOM
Note: BOOM refers to credit growth during historical episodes of housing booms.
Bank lending to households at constant prices (2013Q4=100)
House price developments: large heterogeneity in the euro area
30
80
90
100
110
120
130
140
200
9201
0201
1201
2201
32
01
4201
5201
6201
7201
8Greece Italy
Portugal Spain
BOOM
70
80
90
100
110
120
130
140
20
09
20
10
20
11
20
12
20
13
20
14
20
15
20
16
20
17
20
18
Estonia Latvia
Lithuania Slovakia
Slovenia BOOM
80
90
100
110
120
130
140
200
9
201
0
20
11
201
2
201
3
201
4
201
5
201
6
201
7
201
8
Austria Belgium
France Germany
Netherlands BOOM
80
90
100
110
120
130
140
150
200
9
201
0
201
1
201
2
201
3
201
4
201
5
201
6
201
7
201
8
Finland Ireland
Luxembourg BOOM
Note: BOOM refers to house price increases during historical episodes of housing booms.
Real house price index (2013Q4=100)
Should monetary policy aim to support financial stability?• Conceptually, monetary policy tools are ill-suited to fostering
financial stability goals
• Problem is even more severe in the heterogeneous euro area
• Macroprudential policy should play a major role
• Current macroprudential assessments: the overall risk to financial stability remains low, though there are certain vulnerabilities
• In some euro area countries certain vulnerabilities have already led to measures, like capital buffer for systemically important institutions, countercyclical capital buffers (CCyBs), and debt-to-income (DTI) ratio limits and loan-to-value (LTV) ratio limits
31
Five main take-aways
1. Premature monetary policy exits involves major risks: careful forward guidance is needed
2. In the new ‘normal’ central bank balance sheet policies will likely became part of the regular toolkit, especially if the natural rate of interest remains low
3. Stopping asset purchases would not increase long-term rates
4. The inflation outlook in the euro area is very uncertain: better to wait with interest rate increase till core inflation overshoots
5. The euro area is very heterogeneous: monetary policy cannot address financial stability concerns; instead, macroprudential policy should have major role
32