ï¿½ï¿½ @ 5 7 5 = B 0 F 8 O PowerPoint · 2020. 4. 28. · Title: ï¿½ï¿½ @ 5 7 5 = B 0 F...

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


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).


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))


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.


• 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.


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


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


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


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


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



loss function, credit-to-GDP and credit weight 1, baseline volume of shock (left), 2


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



loss function, credit growth and credit weight 0, baseline volume of shock (left), 2


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



loss function, credit growth and credit weight 1, baseline volume of shock (left), 2


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

ï¿½ï¿½ @ 5 7 5 = B 0 F 8 O PowerPoint · 2020. 4. 28. · Title: ï¿½ï¿½ @ 5 7 5 = B 0 F...

Documents

Transcript of ï¿½ï¿½ @ 5 7 5 = B 0 F 8 O PowerPoint · 2020. 4. 28. · Title: ï¿½ï¿½ @ 5 7 5 = B 0 F...