View
6
Download
0
Category
Preview:
Citation preview
A CASE FOR LEANING
AGAINST THE WIND IN A
COMMODITY-EXPORTING
ECONOMY
Irina Kozlovtceva
Alexey Ponomarenko
Andrey Sinyakov
Stas Tatarintsev
Research and Forecasting Department
Moscow, April 2020
The views expressed in this presentation are solely
those of the authors and do not necessarily reflect the
official position of the Bank of Russia.
2 Motivation
Commodity price changes are important source of relative price (real exchange
rate) changes and macroeconomic volatility in commodity-exporting economies
(CEEs)
• see Medina and Soto (2007) for Chile, Bergholt et al. (2017) - Norway,
Kreptsev and Seleznev (2017) - Russia
75
80
85
90
95
100
105
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
commodity exporters
non-comodity exporters
Figure 1. Median real effective exchange rate indices for commodity and non-commodity exporters
3 Motivation
1. Inflation targeting (IT) central banks may face a trade-off when a commodity
price shock hits.
Bergholt (2014, 2017), Charnavoki (2010), Allegret et al. (2015), Hamann et al.
(2016) for commodity exporters:
Exchange rate changes and exchange rate pass-through to inflation
vs.
Effect of the shock on economic activity
Kormilitsina (2011), Bodenstein et al. (2012), Plante (2014) for net commodity
importers:
Pass-through of imported commodities prices to inflation
vs.
Effect of the shock on economic activity
The first issue: How do IT central banks react to commodity shocks in
practice? or Following Bernanke, et al. (1997): How does monetary policy
systematically react to commodity shocks?
4 Motivation
2. Commodity price changes may be a source of credit booms and busts in CEEs
• Theory: Gonzales et al. (2016), Bejarano et al. (2016), Carvalho, F. A. et al
(2017), Gourinchas (2018)
• Empirics: IMF (2015), Shousha (2016), Gonzales et al. (2016)
The second issue: What is contribution of (systematic) monetary policy
reaction under IT to commodity prices in observed dynamics of real credit
after a commodity shock?
5 Motivation
3. Relative price changes (under commodity shocks) may misguide the central
banks regarding inflation and financial stability trade-off - Borio et al. (2014,
2016, 2018), BIS annual economic report (2018)
Leaning against the wind (LAW), Svensson (2013), is a way to solve the trade-off,
when a central bank lacks effective macroprudential policy.
See discussion in Andrian and Liang (2016), Svensson (2017a), Svensson
(2017b), Agenor and Pereira da Silva (2019). Ajello et al. (2019)
The third issue: from the normative point of view, can monetary policy
under IT in a CEE better stabilize inflation and output if it takes into account
its own effects on volatility of real credit (that feeds back into inflation and
output volatility)?
or Can LAW alone lead to better macroeconomic outcome?
6 Contribution
The first issue: How do IT central banks react to commodity shocks in
practice? or Following Bernanke, et al. (1999): How does monetary policy
systematically react to commodity shocks?
- Review DSGE models of IT-CEEs central banks, estimated over period of IT
- Estimate panel SVARs and Local projection models (LPM) for IT countries
The second issue: What is contribution of (systematic) monetary policy
reaction under IT to commodity prices in observed dynamics of real credit
after a commodity shock?
- Apply the Sims and Zha (1995, 2006) approach and (as a robustness check)
that by Bernanke et al. (1997)
The third issue: Can LAW alone lead to better macroeconomic outcome?
- compare inflation targeting and LAW in an estimated DSGE model for Russia
from Kreptsev and Seleznev (2017). It is a small open economy model with a
commodity-exporting sector (oil). It incorporates a friction similar to Bernanke, et
al. (1999) with a banking sector as in Gerali et al. (2010) and a country risk-
premium elastic to oil-price changes as in Gonzales et al. (2016).
7 The first issue: findings
• We estimate panel SVARs and Local projection models (LPM) for IT countries
grouped as in Table 1. to check impulse responses of policy rates and real
credit to a positive commodity price shock. Datasamples: IT period.
• Vector autoregression model (VAR) is:
Yt = 𝛼 + Φ1Yt−1 + ⋯ + ΦpYt−p + εt
Yt = (GlobalGDPt, ComdtyPriceIndext, GDPt, Inflationt, Policy ratet, Creditt)
• Local projection model (LPM) as in Jorda (2005)
Commodity exporters Non-commodity exporters
(importers)
Emerging
market
economies
Brazil, Chile, Colombia,
Indonesia, Mexico, Peru,
Philippines, South Africa
Armenia, Georgia, Guatemala,
Poland, Romania, Serbia, Turkey
Developed
market
economies
Australia, Canada, New
Zealand, Norway Czech Republic, Israel, Sweden,
United Kingdom
8 The first issue: findings
Figures 2-3. IRF in VAR model for credit and policy rate for CE EME (Exporters) and NCE DE (Importers)
• IT CE EMEs tend to reduce policy rates after a positive commodity shock
(Chart on the left), while commodity importers and CE AEs raise policy rates.
Possible explanations of the differences in IRFs for CE EMEs vs CE AEs:
anchored inflation expectations, prudent fiscal policy in AEs, cyclicality of
capital flows, elasticity of domestic output gap to commodity prices (PC
correlation).
9 The first issue: findings
Table 2. Statistically significant IRFs in VAR/LPM model for credit and policy rate
• Possible explanations of the differences in IRFs for CE EMEs vs CE AEs:
- anchored inflation expectations,
- prudent fiscal policy in AEs (fiscal rule),
- cyclicality of capital flow, size of ERPT
- elasticity of domestic output gap to commodity prices (PC correlation).
10 The second issue: findings
• Calculations using the Sims and Zha (1995) approach show that the
endogenous nominal rates reduction under inflation targeting in commodity-
exporting EMEs accounts for 20% of real credit increase on average after a
positive oil price shock
11 The third issue: Comparing IT LAW in a DSGE model
Figure 4. The block scheme of the model with the banking sector (from Kreptsev, Seleznev (2017))
12 The third issue: Comparing IT LAW in a DSGE model
Changes in the basic DSGE model:
1. Loss function has the following form (as in Verona et.al., 2017):
𝐿 = 𝑣𝑎𝑟 𝜋 + 𝑣𝑎𝑟 𝑌 + 𝛼𝑐𝑟𝑣𝑎𝑟(𝐶𝑟)
Where: 𝑣𝑎𝑟 𝜋 – variance of inflation, 𝑣𝑎𝑟 𝑌 – variance of GDP, 𝑣𝑎𝑟(𝐶𝑟) – variance
of credit to GDP ratio, 𝛼𝑐𝑟 – weight of credit variance on the loss function.
2. We add a credit variable to the policy rule.
𝑅𝑡
𝑅∗ =𝑅𝑡−1
𝑅∗
𝜙𝑅 𝜋𝑡
𝜋∗
1−𝜙𝑅 𝜙𝜋 𝐶𝑟𝑡
𝐶𝑟𝑡−1
𝜙𝑐𝑟
𝑒𝑒𝑡𝑅
Where: 𝑅𝑡 – policy interest rate at 𝑡, 𝜋𝑡 – inflation rate at 𝑡, 𝐶𝑟𝑡 – credit cycle variable
(credit growth or change in credit-to-GDP), 𝑒𝑡𝑅 – monetary policy shock, 𝜙𝑅, 𝜙𝜋, 𝜙𝑐𝑟
– interest rate inertia, inflation weight and credit cycle weight coefficients
respectively. Variables with an asterisk represent the steady state levels.
3. We estimate the loss function value in case of different oil price shock volatilities.
13 The third issue: Comparing IT LAW in a DSGE model
• We take simulations of the DSGE model for different sizes of oil-price shock
(s.e. from baseline model, multiplied by 1, 2, 5, 10, 20, 50 and 100) and
different weight of the credit cycle variable in MP rule (from 0 to 1 in
increments of 0.1)
• In each case we calculate loss function values for different shocks and
weights without using Leaning Against the Wind principle in MP rule and with
LAW
• We calculate the relative loss function as the ratio of losses with the LAW to
the losses without it for the same shock size and the same weight.
If this ratio is smaller than one, we can conclude, that monetary policy benefits
from considering the credit cycle.
14 The third issue: Comparing IT LAW in a DSGE model
Figure 5. Relative loss function in case of the credit-to-GDP ratio in monetary policy
for equal weights of output, inflation and credits in loss function
0.7
0.75
0.8
0.85
0.9
0.95
1
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
Rela
tive losses
𝜙𝑐𝑟
Relative losses, credit to GDP, 𝛼 𝑐𝑟=1
1 2 5 10 20 50 100
15 The third issue: Comparing IT LAW in a DSGE model
Figure 6. Relative loss function in case of the credit-to-GDP ratio in monetary policy
for equal weights of output, inflation and credits in loss function
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
Rela
tive losses
𝜙𝑐𝑟
Relative losses, credit growth, 𝛼𝑐𝑟=1
1 2 5 10 20 50 100
16 The third issue: Comparing IT LAW in a DSGE model
Figure 7-10. IRF of interest rate (left column) and real credit (right column) for
𝜙𝑐𝑟 = 0.1 (top row) and 𝜙𝑐𝑟 = 0.5 (bottom row)
-0.002
-0.0015
-0.001
-0.0005
0
0.0005
0.001
0.0015
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
R baseline R C2GDP R CG
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Cr baseline Cr C2GDP Cr CG
-0.003
-0.002
-0.001
0
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
R_baseline R C2GDP R CG
0
0.002
0.004
0.006
0.008
0.01
0.012
0.014
0.016
0.018
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Cr baseline Cr C2GDP Cr CG
17 Conclusion
• IRF analysis in panel VARs and LPMs finds that not all commodity-exporting
countries included in the estimation reduce interest rates in response to
higher oil prices: only emerging market economies do. Real credit tends to
grow after a positive shock in commodity-exporting EMEs, but tends to
decline in all other country groups.
• In the DSGE model we show that LAW outperforms IT when commodity
price shocks become a relatively important source of volatility, thus
supporting our empirical findings.
• Even when the financial stability risks associated with the volatility of credit
developments are negligible, a moderate leaning against the wind policy is
still preferable
18 References
Adrian, Tobias, and Nellie Liang (2016), “Monetary Policy, Financial Conditions, and
Financial Stability,” Staff Report No. 690, Revised December 2016, Federal Reserve Bank
of New York
Agénor, Pierre-Richard and Luiz A Pereira da Silva, Integrated inflation targeting - Another
perspective from the developing world, BIS, 2019
Ajello, A., Laubach, T., López-Salido, D., & Nakata, T. (2019). Financial Stability and
Optimal Interest Rate Policy. International Journal of Central Banking.
Allegret, Jean Pierre, and Mohamed Tahar Benkhodja. "External shocks and monetary
policy in an oil exporting economy (Algeria)." Journal of Policy Modeling 37.4 (2015): 652-
667.
Bank of International Settlements (BIS), Moving forward with Macroprudential frameworks,
Chapter IV, Annual economic report, 2018
Bejarano, J., Hamann, F., Mendoza, E. G., & Rodríguez, D. Commodity Price Beliefs,
Financial Frictions and Business Cycles, 2016
Bergholt, Drago, Vegard H. Larsen, and Martin Seneca (2017): Business Cycles in an Oil
Economy. Journal of International Money and Finance, in press
19 References
Bergholt, Drago. Monetary Policy in Oil Exporting Economies. No. 5/2014. Centre for
Applied Macro-and Petroleum economics (CAMP), BI Norwegian Business School, 2014.
Bernanke, B. S., Gertler, M., Watson (1997). Systematic monetary policy and the effects of
oil price shocks. Brookings papers on economic activity, 1997(1), 91-157.
Bodenstein, Martin, Luca Guerrieri, and Lutz Kilian. "Monetary policy responses to oil price
fluctuations." IMF Economic Review 60.4 (2012): 470-504.
Borio, C. E., Disyatat, P., Juselius, M., & Rungcharoenkitkul, P. (2018). Monetary policy in
the grip of a pincer movement.
Borio, Claudio. "Revisiting three intellectual pillars of monetary policy." Cato J. 36 (2016):
213.
Borio, Claudio. "The financial cycle and macroeconomics: What have we learnt?." Journal
of Banking & Finance 45 (2014): 182-198
Carvalho, Fabia A., and Marcos R. Castro. "Macroprudential policy transmission and
interaction with fiscal and monetary policy in an emerging economy: a DSGE model for
Brazil." Macroeconomics and Finance in Emerging Market Economies 10.3 (2017): 215-
259.
20 References
Charnavoki, Valery. "International risk sharing and optimal monetary policy in small
commodity-exporting economy." New Economic School, Job Market Paper (2010).
Gerali, A., Neri, S., Sessa, L., & Signoretti, F. M. (2010). Credit and Banking in a DSGE
Model of the Euro Area. Journal of Money, Credit and Banking, 42, 107-141.
González, Andrés, Franz Hamann, and Diego Rodríguez. "Macroprudential policies in a
commodity exporting economy." CBRT, the BIS or the IMF. (2016): 69.
Gourinchas, Pierre-Olivier. "Monetary Policy Transmission in Emerging Markets: An
Application to Chile." Central Banking, Analysis, and Economic Policies Book Series 25
(2018): 279-324
Hamann, Franz & Bejarano, Jesús & Rodríguez, Diego & Restrepo-Echavarria, Paulina,
2016. "Monetary Policy in an Oil-Exporting Economy," Review, Federal Reserve Bank of
St. Louis, vol. 98(3), pages 239-261.
International Monetary Fund. 2015. World Economic Outlook: Adjusting to Lower
Commodity Prices. Washington (October).
Kormilitsina, A. (2011). Oil price shocks and the optimality of monetary policy. Review of
Economic Dynamics 14(1), 199 – 223
21 References
Kreptsev, D., Seleznev S., DSGE model of the Russian economy with a banking sector,
Bank of Russia Working Paper Series, 2017
Plante, Michael. "How should monetary policy respond to changes in the relative price of
oil? Considering supply and demand shocks." Journal of Economic Dynamics and Control
44 (2014): 1-19.
Shousha, S. (2016). Macroeconomic effects of commodity booms and busts: The role of
financial frictions. Unpublished Manuscript.
Sims, Christopher A., and Tao Zha. "Does monetary policy generate recessions?"
Macroeconomic Dynamics 10.2 (2006): 231-272.
Svensson, L.E., 2013. ‘Leaning Against the Wind’ Leads to a Higher (Not Lower)
Household Debt-to-GDP Ratio. Working Paper
Svensson, Lars EO. "Cost-benefit analysis of leaning against the wind." Journal of
Monetary Economics 90 (2017a): 193-213.
Svensson, Lars EO. "How Robust Is the Result That the Cost of ‘Leaning against the
Wind’Exceeds the Benefit?." (2017b).
APPENDIX
23 VAR Model
Variables: GGDP – Global GDP; ComPI – Commodity Price Index for non-
exporting inflation targeting countries, and prices of Oil, Metals, Copper
and Precious metals for exporting countries, depending on the
commodity they export; GDP; Inflation (derived from the consumer price
index); Nominal policy rate of a central bank; Credit. Credit and GDP
variables are seasonally adjusted. All series except interest rates are in
log differences. In addition, data standardisation is carried out.
Vector autoregression model (VAR) is:
Yt = 𝛼 + Φ1Yt−1 + ⋯ + ΦpYt−p + εt
Where 𝑡 = 1, … , 𝑇 denotes time, Yt is a 𝑞 × 1 vector of variables,
Following Kilian and Lewis (2011) we choose a conventional ordering in
a Cholesky decomposition: from external variables (global GDP and
commodity prices) to domestic variables:
Yt = (GGDPt, ComPIt, GDPt, Inflationt, Policy ratet, Creditt)
εt~ 𝑁(0, Ω)
24 VAR Model
Vector autoregression model (VAR) is:
Yt = 𝛼 + Φ1Yt−1 + ⋯ + ΦpYt−p + εt
Where 𝑡 = 1, … , 𝑇 denotes time, Yt is a 𝑞 × 1 vector of variables,
Following Kilian and Lewis (2011) we choose a conventional ordering in
a Cholesky decomposition: from external variables (global GDP and
commodity prices) to domestic variables:
Yt = (GGDPt, ComPIt, GDPt, Inflationt, Policy ratet, Creditt)
εt~ 𝑁(0, Ω)
25 Local Projection Model
Local projection model (LPM), following Jorda (2005), is:
s = [0 ∶ h]
In s = 0 evaluates:
yt+0 = 𝛼 + B11yt−1 + ⋯ + Bp
1yt−p + C0Xt + ut+00
IRF0 = C0d𝑗
Where d𝑗 represents the ‘structural shock’ to the 𝑗𝑡ℎ element in yt.
In s = 1: h evaluates:
yt+s = 𝛼 + B1s+1yt−1 + ⋯ + Bp
s+1yt−p + ut+ss
We obtain
IRF1 = B11C0d𝑗
⋮ IRFs = B1
sC0d𝑗
where
yt+s = (GGDPt+s, ComPIt+s, GDPt+s, Inflationt+s, Policy ratet+s, Creditt+s)
Xt = ComPIt
and 𝑡 = 1, … , 𝑇 denotes time, yt+s is a 𝑞 × 1 vector of variables. Xt is a variable of
commodity price index.
26 Comparing IT LAW in a DSGE model
Figure 11-13. Comparison of relative loss functions for different weights of credits in
loss function, credit growth, baseline volume of shock (left), 2 times higher (right),
50 times higher (bottom)
0.9
1.1
1.3
1.5
1.7
1.9
2.1
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
𝜙𝑐𝑟
1 0,5 0,25 0
0.8
1
1.2
1.4
1.6
1.8
2
2.2
2.4
2.6
2.8
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
𝜙𝑐𝑟
1 0,5 0,25 0
0.5
1.5
2.5
3.5
4.5
5.5
6.5
7.5
8.5
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
𝜙𝑐𝑟
1 0,5 0,25 0
27 Comparing IT LAW in a DSGE model
Figure 14-16. Comparison of relative loss functions for different weights of credits in
loss function, credit-to-GDP, baseline volume of shock (left), 2 times higher (right),
50 times higher (bottom)
0.8
0.82
0.84
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
𝜙𝑐𝑟
1 0,5 0,25 0
0.8
0.82
0.84
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
𝜙𝑐𝑟
1 0,5 0,25 0
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
0 0 . 1 0 . 2 0 . 3 0 . 4 0 . 5 0 . 6 0 . 7 0 . 8 0 . 9 1
𝜙𝑐𝑟
1 0,5 0,25 0
28 Comparing IT LAW in a DSGE model
Figure 17-19. Comparison of relative loss functions for different weights of output in
loss function, credit-to-GDP and credit weight 0, baseline volume of shock (left), 2
times higher (right), 50 times higher (bottom)
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
1.6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
1.6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
1.6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
29 Comparing IT LAW in a DSGE model
Figure 20-22. Comparison of relative loss functions for different weights of output in
loss function, credit-to-GDP and credit weight 1, baseline volume of shock (left), 2
times higher (right), 50 times higher (bottom)
0.8
0.82
0.84
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0.8
0.82
0.84
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0.75
0.8
0.85
0.9
0.95
1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
30 Comparing IT LAW in a DSGE model
Figure 23-25. Comparison of relative loss functions for different weights of output in
loss function, credit growth and credit weight 0, baseline volume of shock (left), 2
times higher (right), 50 times higher (bottom)
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0
1
2
3
4
5
6
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0
1
2
3
4
5
6
7
8
9
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
31 Comparing IT LAW in a DSGE model
Figure 26-28. Comparison of relative loss functions for different weights of output in
loss function, credit growth and credit weight 1, baseline volume of shock (left), 2
times higher (right), 50 times higher (bottom)
1
1.05
1.1
1.15
1.2
1.25
1.3
1.35
1.4
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0.8
0.85
0.9
0.95
1
1.05
1.1
1.15
1.2
1.25
1.3
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
1
1.05
1.1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
1 0.5 0
Recommended