Upload
others
View
3
Download
0
Embed Size (px)
Citation preview
Re-Thinking Monetary Policy Objectives Why Mess with Success?
Ken Kuttner
Williams College, PIIE and NBER
The Future Direction of Monetary Policy Frameworks and Strategies in Emerging Market Economies
Bank Negara Malaysia May 21, 2014
The sea change in monetary policy
• Worldwide, the emphasis on price stability has increased over the past 20 years.
– Narrowing of policy mandates, adoption of IT.
– True in the U.S. with no explicit framework change.
– Abandonment of exchange rate pegs.
• Has the trend gone too far? Is inflation targeting enough?
5/15/14 2
This paper
• Surveys theoretical arguments for and against expanding the set of policy targets.
• Examines the experiences of six Asian economies: Are they de facto already pursuing goals other than inflation?
5/15/14 3
The six Asian economies’ policy frameworks
• Indonesia: IT since 2005
• Korea: IT since 1998
• Malaysia: not clear, IMF classifies as “other”
• Philippines: IT since 2002
• Singapore: crawling peg/band since 1981
• Thailand: IT since 2000
5/15/14 4
1987-1999 2000-2013
Mean
Standard deviation Mean
Standard deviation
Indonesia 11.3% 15.3 7.0% 3.6
Korea 5.5 3.3 2.9 1.6
Malaysia 3.2 2.0 2.2 2.7
Philippines 8.7 4.4 4.0 3.5
Singapore 1.8 1.5 2.1 2.7
Thailand 4.6 2.9 2.5 3.0
Price stability – a success story
• The mean and variance of inflation fell relative to the pre-Asia crisis period in 5 of the 6 countries.
• Singapore is the exception – unchanged policy framework.
5/15/14 5
Theoretical considerations: What should central banks do?
• Theoretically, what belongs in the objective function?
• Can additional tradeoffs be managed?
• How to hold central banks accountable?
Re-thinking policy objectives
• Should policy pay attention to output?
• Does something else belong in the objective function?
5/15/14 7
The problem of goal proliferation
Must we accept parenthood for every economic development in the country? That is a hard thing for us to do. We would have a large family of children. Every time one of them misbehaved, we might have to spank them all. – Benjamin Strong, NY Fed President, 1914–1928
5/15/14 8
A financial stability goal?
• The Bernanke-Gertler view: asset price booms are equivalent to demand shocks.
• The central bank should react to booms only to the extent that they affect the output and inflation goals.
• BG model simulations: no gain from reacting directly to asset prices.
5/15/14 9
Financial instability: not just spending?
• Bordo and Jeanne: asset price collapse damages the economy’s supply side.
• Dupor: non-fundamental asset price fluctuations misdirect investment.
• Woodford: financial disruptions interfere with intermediation between savers and borrowers, causing inefficient consumption allocation.
5/15/14 10
Problems with a financial stability target
• Interest rates affect the whole economy, not just financial stability (the “blunt tool” problem).
• How much output or inflation must be sacrificed for certain increase in stability?
• What is a good empirical proxy for financial stability?
• How much of an impact does interest rate policy really have on financial stability?
• Accountability is almost impossible. 5/15/14 11
Why crisis prevention is hard to verify
• Consider a policy intended to increase the mean time between crises from 5 to 10 years.
• Assume crises follow a poisson process:
– There is a 39% chance of having a crisis within 5 years even if the policy is successful.
– There is a 13% chance of having no crisis for at least 10 years even if the policy is unsuccessful.
5/15/14 12
Debt growth tempers Riksbank’s stimulus
A monetary policy that to a greater degree must prioritise the short-term development of inflation of course places a greater responsibility on other policy areas for managing the risks associated with the high and growing indebtedness of households. A factor of importance here is that the Riksbank's need to prioritise short-term stimulus measures to a greater extent will in itself lead to an increase in household debt. – Minutes of Riksbank monetary policy meeting, February 14, 2014
5/15/14 13
Some think this is a bad idea
The repo rate is something of a blunt instrument for managing debt levels, however. According to the Riksbank’s analysis, substantial changes in the repo rate would be needed to have any appreciable effect on the debt-to-income ratio. The NIER does not therefore believe there is currently any reason for monetary policy to be influenced by household debt. – “The Swedish Economy,” National Institute of Economic Research, March 2014
5/15/14 14
Other tools exist for financial stability
5/15/14 15
Indonesia
<-
loose
nin
g /
tig
hte
nin
g -
>
interest rate, %
2006 2007 2008 2009 2010 2011 2012
-2
-1
0
1
0
4
8
12
Reserve Req
Interest rate
Korea
<-
loo
sen
ing
/ t
igh
ten
ing
- >
interest rate, %
2006 2007 2008 2009 2010 2011 2012
-2
-1
0
1
0
2
4
6
Reserve Req
LTV limit
DSTI limit
Interest rate
Malaysia
<-
loo
sen
ing
/ t
ighte
nin
g -
>
interest rate, %
2006 2007 2008 2009 2010 2011 2012
-2
-1
0
1
0.0
1.0
2.0
3.0
Reserve Req
LTV limit
Risk wt
Tax
Interest rate
Philippines
<-
loose
nin
g /
tig
hte
nin
g -
>
interest rate, %
2006 2007 2008 2009 2010 2011 2012
-2
-1
0
1
0
2
4
6
8
Reserve Req
Interest rate
Singapore
<-
loo
sen
ing
/ t
igh
ten
ing
- >
interest rate, %
2006 2007 2008 2009 2010 2011 2012
0
1
2
3
0.0
1.0
2.0
3.0Reserve Req
LTV limit
Exposure
Tax
Interest rate
Thailand
<-
loo
sen
ing
/ t
ighte
nin
g -
>
interest rate, %
2006 2007 2008 2009 2010 2011 2012
-2
-1
0
1
0
2
4
Reserve Req
LTV limit
Risk wt
Interest rate
• Reserve requirements, LTV limits, etc., are widely used in Asia.
Source: Kuttner & Shim (2014)
What about an exchange rate goal?
• The exchange rate is something of an afterthought in the canonical New Keynesian model.*
– Taylor: exchange rate changes = cost shocks, respond only if they affect inflation.
– Galí & Monacelli: target domestic inflation, ignore the exchange rate.
*The term “open economy” appears only once in Woodford’s Interest and Prices, on page 200. The exchange rate is never mentioned.
5/15/14 16
Rationalizing a weight on the exchange rate
• This requires direct costs of exchange rate fluctuations, not just pass-through effects.
• Possible sources:
– Liability dollarization (“original sin”)
– Trade promotion
– Devereux & Engel: Local currency pricing
5/15/14 17
Revealed preference: What have central banks been doing?
• Realized output and inflation volatility.
• Policy response to external conditions.
• Mini case study of 2007-08 commodity price shock.
Multiple objectives and macro volatility
Inflation volatility
Outp
ut
vo
lati
lity
Efficient policy frontier
Adding anotherobjective
• Adding objectives requires sacrificing output and/or inflation stabilization.
5/15/14 19
Hawkishness and macro volatility
Inflation volatility
Outp
ut
vo
lati
lity
Increasedinflation
emphasis
Efficient policy frontier
("Taylor curve")
• Increased emphasis on inflation:
– Price stability comes at the expense of higher output volatility.
5/15/14 20
Asia’s “pretty good moderation”
• Excluding crises, output volatility has fallen (except in Singapore).
Indonesia
1987 1990 1993 1996 1999 2002 2005 2008 2011
0.0
5.0
10.0
15.0
20.0
Korea
1987 1990 1993 1996 1999 2002 2005 2008 2011
0.0
5.0
10.0
15.0
20.0
Malaysia
1987 1990 1993 1996 1999 2002 2005 2008 2011
0
4
8
12
16
Philippines
1987 1990 1993 1996 1999 2002 2005 2008 2011
0.0
2.5
5.0
7.5
10.0
12.5
Singapore
1987 1990 1993 1996 1999 2002 2005 2008 2011
2.5
5.0
7.5
10.0
12.5
15.0
17.5
Thailand
1987 1990 1993 1996 1999 2002 2005 2008 2011
0
4
8
12
16
5/15/14 21
Explaining output and inflation stability
Inflation volatility
Outp
ut
vo
lati
lity
Reduction in in shocks'
Increasedinflation
emphasis
emphasisoutput
Increased
variance
• Smaller shocks and/or better policy shift policy frontier inward.
• No change in preferences: proportional volatility reduction.
• Greater inflation emphasis: relatively larger impact on inflation volatility.
5/15/14 22
60 1 2 3 4 5
9
0
1
2
3
4
5
6
7
8
Inflation volatility
Ou
tpu
t v
ola
tili
ty
TH
ID
KR
MY
SG
PH
No distraction or inflation obsession
• Proportional reduction in output and inflation volatility.
• No evidence of single-minded inflation targeting.
• Consistent with narrower set of policy objectives.
• Singapore is the exception: larger shocks? Different policy objectives?
5/15/14 23
Policy responses to external conditions
• Regress change in policy interest rate on:
– Lagged change in and level of policy rate – Current and lagged Fed funds rate – Current and lagged Asia risk spread* – Quantitative Easing (QE) proxy – Fed’s discount window and TAF lending
• Similar specifications for exchange rate change and FX intervention (change in reserves)
• Reduced form. All regressors are exogenous.
• Limitation: no controls for domestic conditions.
5/15/14 24 *BofA Merrill Lynch Asia Emerging Markets Corporate Plus Sub-Index Option-Adjusted Spread
Reaction to the US federal funds rate
Domestic policy rate Δ Exchange rate FX reserves Short run Long run FF change Lag FF level
Indonesia
Korea 0.31
Malaysia 0.14 + +
Philippines 0.51 0.33 +
Singapore 1.00 0.53 – +
Thailand + +
• In some cases domestic policy rates track the Fed funds rate.
• Funds rate often allowed to affect the exchange rate.
• Malaysia and Thailand use FX intervention.
• Does not necessarily imply concern with exchange rate stability. 5/15/14 25
Plots corroborate the regressions Indonesia
2002 2004 2006 2008 2010 2012
0
1
2
3
4
5
6
5.0
7.5
10.0
12.5
15.0
17.5US Fed funds
Policy rate
Korea
2002 2004 2006 2008 2010 2012
0
1
2
3
4
5
6
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5US Fed funds
Policy rate
Malaysia
2002 2004 2006 2008 2010 2012
0
1
2
3
4
5
6
1.8
2.2
2.6
3.0
3.4US Fed funds
Policy rate
Philippines
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
0
1
2
3
4
5
6
2
3
4
5
6
7
8
9US Fed funds
Policy rate
Singapore
2002 2004 2006 2008 2010 2012
0
1
2
3
4
5
6
0.0
1.0
2.0
3.0
4.0US Fed funds
Policy rate
Thailand
2002 2004 2006 2008 2010 2012
0
1
2
3
4
5
6
1.0
2.0
3.0
4.0
5.0US Fed funds
Policy rate
5/15/14 26
Reaction to U.S. Quantitative Easing
Domestic policy rate
Δ Exchange rate
FX reserves
Indonesia + +
Korea +
Malaysia +
Philippines
Singapore +
Thailand + +
• QE led to exchange rate appreciation.
• Orders of magnitude: $100 billion → 1–3% Δ exchange rate.
• A largely laissez-faire response to QE – even in Singapore.
• Unexpected reaction in TH and ID: domestic conditions? 5/15/14 27
Reaction to Asia risk spread
Domestic policy rate
Δ Exchange rate
FX reserves
Indonesia – –
Korea – –
Malaysia – –
Philippines + –
Singapore – –
Thailand + – –
• Thailand & Philippines: slight tightening (1 pp Δ spread → 1–2 bp).
• Significant depreciation: 1 pp Δ spread → 2–7% Δ exch rate.
• In some cases FX sales: 1 pp Δ spread → $3 billion Δ reserves.
• (Contradicts Reinhart & Rogoff’s “Fear of Floating” model.) 5/15/14 28
Plots again corroborate regressions
Indonesia
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
1
3
5
7
9
5.0
7.5
10.0
12.5
15.0
17.5Asia spread
Policy rate
Korea
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
1
3
5
7
9
2.0
2.5
3.0
3.5
4.0
4.5
5.0
5.5Asia spread
Policy rate
Malaysia
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
1
3
5
7
9
1.8
2.2
2.6
3.0
3.4Asia spread
Policy rate
Philippines
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
1
3
5
7
9
2
3
4
5
6
7
8
9Asia spread
Policy rate
Singapore
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
1
3
5
7
9
0.0
1.0
2.0
3.0
4.0Asia spread
Policy rate
Thailand
2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
1
3
5
7
9
1.0
2.0
3.0
4.0
5.0Asia spread
Policy rate
5/15/14 29
A mini case study of 2007-08
• Needed: a “natural experiment.”
– Ideally, one involving conflicts between goals.
– The absence of a “divine coincidence.”
• Hard to find one among the six Asian economies.
• A less-than-ideal experiment:
– Commodity prices spiked in 2007-08, causing inflation to surge.
– How did policy respond?
– Experiment cut short by financial crisis, alas.
5/15/14 30
Central banks ignored inflation rise
Indonesia
2006 2007 2008
0
2
4
6
8
10
12
14
Korea
2006 2007 2008
0
1
2
3
4
5
6
Malaysia
2006 2007 2008
0
1
2
3
4
5
Philippines
2006 2007 2008
0
2
4
6
8
Singapore
2006 2007 2008
0
2
4
6
8
Thailand
2006 2007 2008
0
2
4
6
8
Policy rate
CPI inflationTarget
Average inflation rate
Average inflation
Core CPI
5/15/14 32
Explanations for the policy non-response?
• Commodity price shock thought to be temporary.
– Thailand: targeting core inflation
– But why the rate cuts in some countries?
• Weak level of economic activity?
– No, real GDP was close to or above trend.
• External factors, i.e. the exchange rate?
– Apparently, a reaction to appreciation in the Philippines, Thailand and Singapore.
– To be expected for Singapore, of course.
5/15/14 33
Do exchange rates explain policy rates?
• Apparently yes in Philippines, Singapore and Thailand.
Indonesia
Rup
iah
/US
D percen
t
2006 2007 2008
8950
9000
9050
9100
9150
9200
9250
9300
8
9
10
11
12
13
Korea
Won
/US
D percen
t
2006 2007 2008
920
940
960
980
1000
1020
4.0
4.2
4.4
4.6
4.8
5.0
5.2
Malaysia
Rin
ggit
/US
D
percen
t
2006 2007 2008
3.2
3.3
3.4
3.5
3.6
3.7
3.8
3.25
3.30
3.35
3.40
3.45
3.50
3.55
Philippines
Pes
o/U
SD p
ercent
2006 2007 2008
40.0
42.5
45.0
47.5
50.0
52.5
5.0
5.5
6.0
6.5
7.0
7.5
8.0
Singapore
SG
D/U
SD p
ercent
2006 2007 2008
1.35
1.40
1.45
1.50
1.55
1.60
1.65
0.5
1.0
1.5
2.0
2.5
3.0
3.5
Thailand
Bah
t/U
SD p
ercent
2006 2007 2008
32
34
36
38
40
3.2
3.6
4.0
4.4
4.8
Policy rate
Exchange rate
5/15/14 35
Bottom line: should there be other objectives?
• Justifying objectives beyond output and inflation requires extending standard macro theory.
• Targeting financial stability would present major practical problems.
• Exchange rate stability may already be a de facto goal. (Old news.)
• Entwined with the search for additional tools:
– Non-interest rate (macroprudential) policies,
– Sterilized foreign exchange intervention.
5/15/14 36
The proof of the pudding is in the eating
• Asian inflation targeters: flexible IT + some degree of exchange rate management (not news).
– They do not overreact to supply-driven inflation.
– FX intervention often used instead of interest rate.
• “Flexible/Eclectic IT” seems to have worked well.
– Inflation is lower and more stable.
– Output is less volatile.
• Why mess with success?
5/15/14 37