Transcript Monetary policy, asset prices and financial stability

JEM027
Monetary Economics
Monetary policy, asset prices
and financial stability
Tomáš Holub
[email protected]
December 14, 2015
Partly based on presentation of K. Šmídková from 2013
Institute of Economic Studies, Faculty of Social Sciences, Charles University in Prague
Monetary policy: the pre-crisis view
1 Central banks should take care of the economic welfare, together with other
institutions
2
Welfare is a too complex concept, so central bank is attributed one specific role
related to what it can really achieve in the long term: protect price stability by
using monetary policy
3 Price stability is too abstract concept, monetary policy is given explicit target
(2% CPI inflation)
4 Consensus about policy tool: short-term rates only
5 Consensus about forecasting tool: high-tech model such as DSGE (real
economy only, no financial sector) combined with expert judgment
JEM027 – Monetary Economics
1
Our basic model corresponds to this view
The loss function (strict IT)
The model
Inflation is the
only target
Forecasting model is consensual
Experts judge the
size of shocks
... where εt+1 and ηt+1 are white noise random shocks
Optimization problem
Interest rate the only tool
JEM027 – Monetary Economics
2
Monetary policy in this set-up
Optimal reaction function
Interest rates react only to variables in the model
Actual inflation in time t+2
Success is measured by deviation of inflation from the
target in a given time horizon
JEM027 – Monetary Economics
3
Inflation was stabilized: all was sunny
Bernanke (2004) in “Reflections on Monetary Policy 25 Years after October 1979”
“
The low-inflation era of the past two decades has seen not only significant
improvements in economic growth but also a marked reduction in economic
volatility, a phenomenon that has been dubbed "the Great Moderation" is
due to good monetary policy.
Greenspan (2005) in “Greenspan Era – Lessons for the Future”
“
US economy has prospered notably, inflation remains low, resilience to
shocks and flexibility enhanced.
Gaspar, Kashyap (2006) in “ECB colloquium held in honour of Otmar Issing”
“
”
”
The stability-oriented monetary policy clearly worked, the HICP inflation during
the first 7 years of the single monetary policy was very close, but not below 2%.
JEM027 – Monetary Economics
”
4
Criticism when IT was being introduced
The loss function (flexible IT)
Under flexible IT, some weight is put on output stability.
▪ When inflation targeting was being introduced, it was typically
criticised for leading to too restrictive policy (”inflation freaks“).
▪ Central bank put a lot of effort into explaining that in reality they
were doing flexible inflation targeting, contributing to stability of
output.
▪ Empirical analyses suggest that they were successful in achieving
this before the crisis.
JEM027 – Monetary Economics
5
The impact of the crisis: the prosperity is gone
UK GDP growth
US economic growth
Percent
Percentage change in GDP, annualised
Source: US Department of Commerce
JEM027 – Monetary Economics
6
IT and the crisis
▪ Even before the crisis, some people were saying that the IT alone is not
▪
▪
enough and that central banks should also take care of financial stability
(e.g. W. White, BIS)
The crisis has challenged the prevailing policy paradigm, including the IT
strategy.
The IT framework is now criticised for leading to too loose monetary
policy, that contributed to the emergence of bubbles!
10
9
Euro area - actual rate
Euro area - Taylor rule
8
7
6
5
4
3
2
1
0
I/99
I/00
I/01
I/02
I/03
I/04
I/05
I/06
I/07
I/08
JEM027 – Monetary Economics
7
Deviations from Taylor Rule and Housing Booms
Source: Ahrend, et al. (2008)
JEM027 – Monetary Economics
8
Some post-crisis quotes
▪ Greenspan (October 2008): he admitted that he was in a
state of shocked disbelief, because the whole intellectual
edifice had collapsed
▪ Krugman (2009): last year everything came apart
– There is no convergence of vision, as claimed by Lucas in
2003 (and also Blanchard in 2008, see next slide)
– The problem of depression-prevention has not been solved,
as claimed by Bernanke in 2004
JEM027 – Monetary Economics
9
About models: prior and after crisis
Blanchard (2008)
“
The state of macro is good... The battles of yesterday
were over...There is a broad convergence of vision.
”
Krugman (2009)
“
Comments from Chicago economists are the product of a
Dark Age of macroeconomics ... Hard-won knowledge
(...how to prevent depression) has been forgotten.
JEM027 – Monetary Economics
”
10
Monetary policy: the post-crisis debate
1 Central banks should take care of the economic welfare, together with other
institutions
2
Welfare is a too complex concept, so central bank is attributed one specific role
related to what it can really achieve in the long term : protect price stability by
using monetary policy
3 Price stability is too abstract concept, monetary policy is given explicit target
(2% CPI inflation)
4 Consensus about policy tool: short-term rates only
5 Consensus about forecasting tool: high-tech model such as DSGE (real
economy only, no financial sector) combined with expert judgment
JEM027 – Monetary Economics
11
Should we really put the blame on IT?
▪ Deviation from Taylor rule means that the policy was too loose compared
to the inflation targeting standard (not that IT is too loose in itself)
▪ The subprime mortgage bubble that led to the global financial crisis
originated in the US, which was not pursuing inflation targeting, but
following its ”just-do-it strategy“ with the dual objective (and focus put on
core rather than headline inflation)
▪ In Europe, the problem of too loose policy related to countries that had
surrendered their monetary policy either due to euro adoption, or due to
a currency board regime; it was not a problem of IT countries;
▪ IT performed quite well empirically during the crisis, relative to other MP
regimes.
 The debate should not be about replacing inflation with another
known MP regime, but about adding the MP-FS nexus.
JEM027 – Monetary Economics
12
Post-crisis issues
▪ How to link monetary policy, financial stability and regulation/supervision?
▪ MP response to asset prices: Greenspan‘s ”cleaning“ view challenged;
should monetary policy ”lean against the wind“?
▪ Inclusion of the asset prices into the targeted index:
e.g. imputed rents?
▪ Inclusion of financial imperfections into macro models
(for sure, but how?)
▪ Number of objectives vs. number of tools: active counter-cyclical use
of regulatory powers as well (i.e. macro-prudential tools)?
JEM027 – Monetary Economics
13
Monetary policy – financial stability nexus (i)
Is it possible / optimal to have a Chinese wall in between?
?
Price stability
Financial stability
Financial developments
Interest rates
Macro-prudential tools
JEM027 – Monetary Economics
14
Two (separate) targets and models?
The loss function (pure IT)
The loss function (FS targeting)
What we need to know
The model
The model
Optimization problem for year t
Optimization problem for year t
JEM027 – Monetary Economics
15
Leaning against the wind?
Shall monetary policy sometimes be more restrictive than needed to for
achievement of the inflation target in order to reduce the risk of bubbles?
Price stability
Financial stability
Financial developments
Interest rates
Macro-prudential tools
JEM027 – Monetary Economics
16
Leaning against the wind – arguments in favour
▪ Bubbles have high (first-order) social costs.
▪ Once bubbles burst, it affects adversely achievement of MP objectives –
trying to avoid them is in line with these objectives, you just need to take
a longer-term view.
▪ Pre-emptive reaction is less harsh than a postponed one.
▪ Reacting only when bubble bursts leads to asymmetry (the „Greenspan
put“ resulting in excessive risk-taking).
▪ If macro-prudential tools are not developed, using monetary policy may be
a second-best option.
▪ Pre-emptive reaction to bubbles and imbalances advocated by various
studies (e.g. Cecchetti et al, 2000)
JEM027 – Monetary Economics
17
Leaning against the wind – arguments against
▪ Bubbles are hard to detect, there are many false alarms.
▪ Responding to these false alarms imposes unnecessary costs on the
economy.
▪ Having two objectives with one instrument only runs against the Tinbergen
principle: trying to chase two rabbits, you may lose both of them, and
undermine the credibility of your policy framework.
▪ Interest rates are too blunt a tool to prick the financial bubbles.
▪ The first best is to develop macro-prudential tools.
JEM027 – Monetary Economics
18
Leaning against the wind – Swedish experiment
▪ Sweden was leaning against the wind in 2011-2013.
▪ Lars Svensson, Riskbank‘s Board member, has criticised this heavily.
▪ In December 2013, Riksbanks started to move away from this policy:
– Riskbank‘s December 2013 policy decision: ”Deputy Governor Per Jansson
began by saying that this was probably the most difficult repo-rate decision he
had been involved in during his time as a member of the Riksbank's Executive
Board.“
– Centralbanking.com (8 January 2014): Riksbank board grappling with ‘genuine
policy dilemma: ”The Riksbank cut its repo rate by 25 basis points to 0.75% last
month in a bid to bring "unexpectedly low" inflation back towards its 2%
target…The move marked a departure from the bank's previous policy of
using interest rates to curb Sweden's large and growing household debt
stock (known as ‘leaning against the wind'), which currently stands at about
170% of average disposable income.“
JEM027 – Monetary Economics
19
Leaning against the wind – Swedish experiment
JEM027 – Monetary Economics
20
Leaning against the wind – Norway
Interest rate trajectory
according to different monetary policy criteria
Source: Norges Bank
7
6
Historical interest rate
Actual interest rate
Criterion 1 (March 2012 forecast)
Criteria 1&2 (March 2012 forecast)
Criteria 1,2&3 (March 2012 forecast)
5
4
3
2
1
0
3/08
9/08
3/09
9/09
3/10
9/10
3/11
9/11
3/12
9/12
3/13
9/13
3/14
9/14
3/15
9/15
▪ Criterion 1 = strict IT; Criterion 1&2 = flexible IT; Criterion 1,2&3 = flexible
▪
IT with leaning.
It seems that the Norges Bank was leaning in 2012-2013, but the fall in oil
prices and deteriorating macro situation have enforced rate cuts.
JEM027 – Monetary Economics
21
Debate about the targeted index
▪ Inflation target is based on CPI that includes prices of consumed
goods
▪ Missing from the index are: prices of commodities (gold, silver, etc.)
and assets (equities, houses, etc.)
▪ This omission would not matter if both – consumer and asset –
prices developed similarly
▪ However, asset prices are more volatile and this volatility can make
the whole economy volatile
▪ Since low CPI inflation does not guarantee stable asset prices,
question follows: is it enough to target CPI inflation?
JEM027 – Monetary Economics
22
The view that targeted index must change
▪ Targeting CPI not enough, extend the index (Filardo, 2000 explored
how much it would help)
▪ The crisis partially caused by neglecting asset price booms,
specially on housing market (USA, UK, Spain)
▪ With extended new index, monetary policy would have been
tighter, ceteris paribus
▪ Monetary policy strategy as a such would remain unchanged
▪ Volatility would be smaller, so would economic costs
▪ ESCB discussed incorporating housing prices into HICP via owner
occupied housing
JEM027 – Monetary Economics
23
The new index and the model
The loss function (pure IT)
Change index
The model
... where εt+1 and ηt+1 are white noise random shocks
Optimization problem
JEM027 – Monetary Economics
24
Question: which is which?
▪ Assume I want to combine
two indices into one to get
the new target index for
Spain
Which is which
(CPI, housing, new
index)?
What will new index
imply for policy
rates?
JEM027 – Monetary Economics
25
Answer: new index
▪ The new index is
black
▪ Housing more
volatile
▪ New index will
lead to more
volatile rates
JEM027 – Monetary Economics
26
Modelling financial imperfections (seminar topic)
▪ External finance premium (fin. accelerator; credit channel)
– The price of credit depends on net worth
– Bernanke, Gertler (1989); Bernanke, Gertler, Gilchrist (1999);
Christiano, et al. (2003); Beneš, Kumhof (2011)
▪ Collateral constraints (balance sheet channel)
– Volume limits on borrowing
– Hart, Moore (1994); Kiyotaki, Moore (1997); Kocherlakota (2000);
Iacoviello (2005); Iacoviello, Neri (2010)
▪ Models with banks
– Bank lending channel as well as balance sheet channel; existence of
multiple interest rates
– Goodfriend, McCallum (2007); Cúrdia, Woodford (2009); Christiano,
Trabandt, Walentin (2007); Gerali et al. (2009; 2010)
JEM027 – Monetary Economics
27
Macro-prudential policy: How to define the target?
▪ How exactly should the explicit target for macro-prudential policy
be defined is not clear yet
▪ Hans Gersbach 2009: equity capital requirements (minimum level)
▪ Claudio Borio and Mathias Drehmann 2009: operational framework
difficult, take the measurement challenge seriously, find barometer
of distress (EWI)
▪ Gabriele Galati and Richhild Moessner 2010 and 2011: set limit to
systemic risk
▪ IMF 2011: set limit to a range of systemic risk indicators
JEM027 – Monetary Economics
28
Two mandates under one roof ?
▪ Carmine Di Noia et al (1999): Inflation rate higher where two
mandates in one institution (credibility loss), based on 12
countries with 2-2 and 12 countries with 2 in 1 institutional
set-ups
▪ C.A.E. Goodhart (2000): Arguments for 2 in 1 alternative
(transmission of information, payment system) as well as
against it (the balance of power important, conflicts of
interest)
▪ 2013: BoE, CNB, ECB... many central banks have two
mandates now
JEM027 – Monetary Economics
29
Macroprudential tools: examples
▪ The idea that interest rates alone cannot prevent various
asset bubbles and their costly consequences leads also to
the proposal that we need a set of macroprudential tools
▪ Examples of macro-prudential tools: counter-cyclical capital
buffer, discretionary capital ratios to different classes of
assets, extensive stress testing of portfolios to force macro
risk internalization, credit growth ceilings, LTVs, removal of
tax policies that encourage excessive risk taking (mortgage
subsidies)
JEM027 – Monetary Economics
30
Actual use of macro-prudential tools (i)
JEM027 – Monetary Economics
31
Actual use of macro-prudential tools (ii)
JEM027 – Monetary Economics
32
Summary: to great moderation and beyond
1980s
▪ Inflation, lost credibility
1900s
▪ Emergence of inflation targeting, consensus on
explicit target, policy tool and forecasting model
2000s
▪ Great Moderation: Romer re-writes textbooks
(How LM Curve Disappeared), new generation
of forecasting models (DSGE without financial
variables)
2010s
▪ Financial crisis challenged nearly everything;
financial stability is becoming a new objective,
macro-prudential tools are being developed…
JEM027 – Monetary Economics
33
Summary: the post-crisis situation
1
Economic welfare: still desirable, but cannot be linked
only to price stability and output gap stabilisation
2
Central banks: many were attributed with another
role (financial stability)
3
Explicit target: CPI may be modified to better include
housing prices
4
Instruments: macro-prudential tools are being
developed and employed in practice
5
Forecasting tools: macro models being extended to
include a financial part (so far as simulation rather than
forecasting tools); better EWS needed, also work in
progress
JEM027 – Monetary Economics
34