International Credit Supply Shocks A. Cesa-Bianchi1 1

A. Ferrero2

A. Rebucci3

Bank of England and Center for Macroeconomics 2 Oxford University Johns Hopkins Carey Business School and NBER

3

18th Jacques Polak Annual Research Conference IMF - November 2-3, 2017

*The views expressed in this paper do not necessarily reflect the position of the Bank of England.

Introduction

1

Fact 1: Capital inflows are typically associated with expansions and asset price surges Cross-border Credit

House Price

-2

-1

0

+1

+2

+3

0 -5 -10 -3

-2

+1

+2

+3

-3

-2

-2

-1

0

+1

+2

+3

Current Account / GDP Pctg. Point

0

0

0 -2 -4

-5 -2

-1

0

+1

+2

+3

-3

-2

GDP

-1

0

+1

+2

+3

6 4 2 0 -1

0

+1

+2

+3

Pctg. Point

4 2 0 -3

-2

-1

0

+1

-2

-1

0

+1

+2

+3

Real Short-term Int. Rate

6

-2

-3

Consumption Percent

Percent

0

5

2

-3

-1

10 0 -10 -20 -30

Real Exch. Rate (USD) Percent

Percent

Real Eff. Exch. Rate

-3

Percent

Percent

Percent

0 -10 -3

Equity Price

5

10

+2

+3

5

0 -3

-2

-1

0

+1

+2

+3

All Economies

NOTE. Each panel plots the median across all boom-bust episodes, using a 6-year window that goes from three year before the peak to three years after the peak. In each panel, time 0 marks the peak of the boom-bust cycle in cross-border bank claim growth (i.e., the last period of a boom in which cross-border bank claims displays a positive growth rate), which is also depicted with a vertical line. All variables are expressed in percentage changes, with the exception of the short-term interest rate and the current account over GDP which are expressed in percentage points.

Introduction

2

Fact 2: Some countries seem more sensitive than others to the volatility of capital inflows Cross-border Credit

House Price

-2

-1

0

+1

+2

+3

0 -5 -10 -3

-2

+1

+2

+3

-3

-2

-2

-1

0

+1

+2

+3

Current Account / GDP Pctg. Point

0

0

0 -2 -4

-5 -2

-1

0

+1

+2

+3

-3

-2

GDP

-1

0

+1

+2

+3

6 4 2 0 -1

0

+1

+2

+3

All Economies

Pctg. Point

4 2 0 -3

-2

-1

0

+1

Advanced Economies

-2

-1

0

+1

+2

+3

Real Short-term Int. Rate

6

-2

-3

Consumption Percent

Percent

0

5

2

-3

-1

10 0 -10 -20 -30

Real Exch. Rate (USD) Percent

Percent

Real Eff. Exch. Rate

-3

Percent

Percent

Percent

0 -10 -3

Equity Price

5

10

+2

+3

5

0 -3

-2

-1

0

+1

+2

+3

Emerging Economies

NOTE. Each panel plots the median across all boom-bust episodes, using a 6-year window that goes from three year before the peak to three years after the peak. In each panel, time 0 marks the peak of the boom-bust cycle in cross-border bank claim growth (i.e., the last period of a boom in which cross-border bank claims displays a positive growth rate), which is also depicted with a vertical line. All variables are expressed in percentage changes, with the exception of the short-term interest rate and the current account over GDP which are expressed in percentage points.

Introduction

3

100

80

60

40

20

0

Introduction

100

80

60

ARG AUT CHN DNK FIN DEU GRC ITA JPN LUX MYS MLT PHL SGP CHE BGR NZL NOR BRA EST IDN PRT CAN ISR SWE AUS BEL CZE FRA ISL IRL LVA LTU MEX POL RUS ZAF ESP THA UKR USA

120 IND GBR

NLD

50

CHE COL HKG DEU JPN AUT FRA

60

COL HKG HUN KOR SVN URY

0 SGP AUS GBR ISL USA

90

80

70

ZAF NZL FIN DNK MEX ARG AUS CHL NLD PER GBR ISR IDN IND BRA SWE THA LUX IRL BEL PRT ITA GRC SVN ESP CZE ISL PHL MLT POL SRB RUS URY LVA EST NOR TWN BGR CHN SVK SGP HRV HUN LTU

NZL DNK CHE NLD

80

HRV

20

CHE DNK NZL SWE AUS CAN GBR NOR MLT CZE ZAF TWN SGP HKG POL CHN MYS KOR THA HUN ISR MEX UKR BRA IND IDN RUS SVK PHL ISL SRB LVA HRV CHL COL SVN EST PER ARG LTU MAR BGR URY

40

RUS SVN ARG BGR IDN PER UKR BRA POL CZE IND SVK LTU PHL MAR HRV HUN MEX CHN COL EST LVA ITA CHL GRC THA KOR ISR ZAF FRA TWN AUT BEL MLT MYS LUX FIN JPN ESP HKG CAN IRL DEU PRT NOR SWE

60

USA DEU ITA AUT ESP BEL FRA PRT FIN NLD GRC LUX IRL JPN

Fact 3: Countries differ in important dimensions, and the EMs vs. AEs divide may not be whole story Mortgage Debt / GDP 100

Home Ownership

Share of foreign currency debt 40

Max Loan to Value (LTV)

40

NOTE. Each bar corresponds to a country. The lighter (yellow) bars are classified as emerging markets and the darker (blue) bars as advanced economies. See the data appendix for variable definitions and data sources.

4

This paper

I

Traditionally, push-pull factor analysis of capital flows and their impact

I

This paper focuses on one particular push shock ⇒ A change in the leverage constraint of global banks that shifts the international credit supply

Introduction

5

This paper

I

Traditionally, push-pull factor analysis of capital flows and their impact

I

This paper focuses on one particular push shock ⇒ A change in the leverage constraint of global banks that shifts the international credit supply

I

Questions (1) What are the mechanisms through which capital inflows lead to macroeconomic booms? (2) What are the characteristics that account for the differences in sensitivity across countries?

Introduction

5

This paper: What we do & What we find I

What we do • Theory: Open economy model with international financial intermediation • Empirics: Heterogeneous panel VAR model for more than 50 countries

Introduction

6

This paper: What we do & What we find I

What we do • Theory: Open economy model with international financial intermediation • Empirics: Heterogeneous panel VAR model for more than 50 countries

I

Three main takeaways 1. Leverage shock is expansionary both in the model and in the data 2. In the average economy, the shock has sizable impact and explains a significant fraction of macroeconomic and asset price variance 3. In the cross-section, a stronger transmission is associated with higher max LTV ratios and shares of FX liabilities

Introduction

6

This paper: What we do & What we find I

What we do • Theory: Open economy model with international financial intermediation • Empirics: Heterogeneous panel VAR model for more than 50 countries

I

Three main takeaways 1. Leverage shock is expansionary both in the model and in the data 2. In the average economy, the shock has sizable impact and explains a significant fraction of macroeconomic and asset price variance 3. In the cross-section, a stronger transmission is associated with higher max LTV ratios and shares of FX liabilities

I

Important implication LTV ratios and shares of FX liabilities, which can be influenced by policy, are linked to final outcomes

Introduction

6

Selected related literature

I

Global financial cycle Rey (2013, 2016); Passari and Rey (2015); Bruno and Shin (2015a,b); Miranda-Agrippino and Rey (2015); Dedola, Rivolta, and Stracca (2015); Forbes, Reinhart, and Wieladek (2016); Cerutti, Claessens, Rose (2017); Aoki, Benigno, and Kiyotaki (2016); Boz and Mendoza (2014); Cetorelli and Goldberg (2011, 2012)

I

House prices and capital flows in the United States Aizenman and Jinjarak (2009); Gete (2009); Bernanke (2010); Justiniano, Primiceri and Tambalotti (2014); Favilukis, Ludvigson and Van Nieuwerburgh (2017); Ferrero (2015)

I

Sensitivity of consumption to asset price and credit shocks Jappelli and Pagano (1989); Almeida, Campello, and Liu (2006); Calza, Monacelli, and Stracca (2014); Berger, Guerrieri, Lorenzoni, and Vavra (2016); Mian, Sufi, and Verner (2016)

Introduction

7

The Model

Model

8

Overview of the model I

Model

Two-period, two-country, two-good, endowment economy with no uncertainty

9

Overview of the model I

Two-period, two-country, two-good, endowment economy with no uncertainty

I

Impatient Home household (i ∈ [0, n]) • Borrows in domestic currency (b) and foreign currency (f ) to consume and

purchase housing services (h1 )

• Subject to collateral constraint: b + s1 f ≤ θ qh1

Model

9

Overview of the model I

Two-period, two-country, two-good, endowment economy with no uncertainty

I

Impatient Home household (i ∈ [0, n]) • Borrows in domestic currency (b) and foreign currency (f ) to consume and

purchase housing services (h1 )

• Subject to collateral constraint: b + s1 f ≤ θ qh1 I

Patient Foreign household (i ∈ (n, 1]) • Saves via deposits (d) and equity (e) in a global bank • Subject to adjustment cost on equity position

Model

9

Overview of the model I

Two-period, two-country, two-good, endowment economy with no uncertainty

I

Impatient Home household (i ∈ [0, n]) • Borrows in domestic currency (b) and foreign currency (f ) to consume and

purchase housing services (h1 )

• Subject to collateral constraint: b + s1 f ≤ θ qh1 I

Patient Foreign household (i ∈ (n, 1]) • Saves via deposits (d) and equity (e) in a global bank • Subject to adjustment cost on equity position

I

Global bank • Channels funds from lenders to borrowers • Subject to capital requirement: e ≥ χ (b/s1 + f )

Model

9

Overview of the model I

Two-period, two-country, two-good, endowment economy with no uncertainty

I

Impatient Home household (i ∈ [0, n]) • Borrows in domestic currency (b) and foreign currency (f ) to consume and

purchase housing services (h1 )

• Subject to collateral constraint: b + s1 f ≤ θ qh1 I

Patient Foreign household (i ∈ (n, 1]) • Saves via deposits (d) and equity (e) in a global bank • Subject to adjustment cost on equity position

I

Global bank • Channels funds from lenders to borrowers • Subject to capital requirement: e ≥ χ (b/s1 + f )

Capital ratio Model

9

Equilibrium: Graphical analysis I

Model

Equilibrium can be represented in the ‘quantity of credit – interest rate’ space

10

Equilibrium: Graphical analysis I

Equilibrium can be represented in the ‘quantity of credit – interest rate’ space 5.5

Non-binding region Binding region

5

B

4.5

4

3.5 Demand Supply (Binding)

3 0.566

Model

0.567

0.568

0.569

0.57

0.571

0.572

10

International credit supply shock I

Experiment Reduction of equity requirement for global banks (χ ↓) • A push shock from Home country’s perspective ↑

Model

11

International credit supply shock I

Experiment Reduction of equity requirement for global banks (χ ↓) • Credit flows into Home country, lending rate falls

(a) Lending Rate

6

B

0.9598

B' 4

3 0.566

B

s1

R

5

(b) Exch. Rate

0.96

0.9596

0.567

0.568

0.569

0.57

0.571

0.572

0.9594 0.566

(c) House Price

0.854

0.567

0.568

0.569

0.57

0.853

0.85 0.566

Model

B

c1

q

1.95

B 0.567

0.568

0.569

0.57

0.571

0.572

B'

1.945

B'

0.851

0.571

(d) Consumption

1.955

0.852

B'

0.572

1.94 0.566

0.567

0.568

0.569

0.57

0.571

0.572

12

International credit supply shock I

Experiment Reduction of equity requirement for global banks (χ ↓) • Real exchange rate appreciates

(a) Lending Rate

6

B

0.9598

B' 4

3 0.566

B

s1

R

5

(b) Exch. Rate

0.96

0.9596

0.567

0.568

0.569

0.57

0.571

0.572

0.9594 0.566

(c) House Price

0.854

0.567

0.568

0.569

0.57

0.853

0.85 0.566

Model

B

c1

q

1.95

B 0.567

0.568

0.569

0.57

0.571

0.572

B'

1.945

B'

0.851

0.571

(d) Consumption

1.955

0.852

B'

0.572

1.94 0.566

0.567

0.568

0.569

0.57

0.571

0.572

13

International credit supply shock I

Experiment Reduction of equity requirement for global banks (χ ↓) • House prices increase (if binding borrowing constraint)

(a) Lending Rate

6

B

0.9598

B' 4

3 0.566

B

s1

R

5

(b) Exch. Rate

0.96

0.9596

0.567

0.568

0.569

0.57

0.571

0.572

0.9594 0.566

(c) House Price

0.854

0.567

0.568

0.569

0.57

0.853

0.85 0.566

Model

B

c1

q

1.95

B 0.567

0.568

0.569

0.57

0.571

0.572

B'

1.945

B'

0.851

0.571

(d) Consumption

1.955

0.852

B'

0.572

1.94 0.566

0.567

0.568

0.569

0.57

0.571

0.572

14

International credit supply shock I

Experiment Reduction of equity requirement for global banks (χ ↓) • Consumption increases

(a) Lending Rate

6

B

0.9598

B' 4

3 0.566

B

s1

R

5

(b) Exch. Rate

0.96

0.9596

0.567

0.568

0.569

0.57

0.571

0.572

0.9594 0.566

(c) House Price

0.854

0.567

0.568

0.569

0.57

0.853

0.85 0.566

Model

B

c1

q

1.95

B 0.567

0.568

0.569

0.57

0.571

0.572

B'

1.945

B'

0.851

0.571

(d) Consumption

1.955

0.852

B'

0.572

1.94 0.566

0.567

0.568

0.569

0.57

0.571

0.572

15

Empirics

Model

16

A Heterogeneous Panel VAR model I

Objective Identify an international credit supply shock in the data (1) Transmission and relative importance for the average economy (2) Differential impact across countries

Empirics

17

A Heterogeneous Panel VAR model I

Objective Identify an international credit supply shock in the data (1) Transmission and relative importance for the average economy (2) Differential impact across countries

I

Reduced-form VAR for country i is

Xit = ai + bi t + ci t2 + F1i Xi,t−1 + uit , where

Xit =

Empirics



LEVt

KFit

Cit

HPit

RERit

CAit /Yit



17

A Heterogeneous Panel VAR model I

Objective Identify an international credit supply shock in the data (1) Transmission and relative importance for the average economy (2) Differential impact across countries

I

Reduced-form VAR for country i is

Xit = ai + bi t + ci t2 + F1i Xi,t−1 + uit , where

Xit = I



LEVt

KFit

Cit

HPit

RERit

CAit /Yit



LEVt : Leverage of US Broker-Dealer sector (Flow of Funds) • Empirical proxy for global banks’ leverage • Common to all countries

Empirics

17

A Heterogeneous Panel VAR model I

Objective Identify an international credit supply shock in the data (1) Transmission and relative importance for the average economy (2) Differential impact across countries

I

Reduced-form VAR for country i is

Xit = ai + bi t + ci t2 + F1i Xi,t−1 + uit , where

Xit = I



LEVt

KFit

Cit

HPit

RERit

CAit /Yit



KFit : Cross-border claims of BIS reporting banks on country i • All instruments, to financial and non-financial sectors

Empirics

18

A Heterogeneous Panel VAR model I

Objective Identify an international credit supply shock in the data (1) Transmission and relative importance for the average economy (2) Differential impact across countries

I

Reduced-form VAR for country i is

Xit = ai + bi t + ci t2 + F1i Xi,t−1 + uit , where

Xit =



LEVt

KFit

Cit

HPit

RERit

CAit /Yit



I

All variables are in real terms (except LEVt and CAit /Yit ) and in log-levels (except CAit /Yit )

I

Mean group estimator [Pesaran and Smith (1995); Pesaran (2006)] over 1985:Q1-2012:Q4 sample period

Empirics

19

Identification of international credit supply ‘push’ shock in the data I

Empirics

In the model shocks to leverage shift global supply of cross-border bank credit

20

Identification of international credit supply ‘push’ shock in the data I

In the model shocks to leverage shift global supply of cross-border bank credit

I

In the data LEVt arguably is exogenous to conditions in individual country

i

• Unlikely to be driven by country-specific ‘pull’ factors • Drop US from sample

Empirics

20

Identification of international credit supply ‘push’ shock in the data I

In the model shocks to leverage shift global supply of cross-border bank credit

I

In the data LEVt arguably is exogenous to conditions in individual country

i

• Unlikely to be driven by country-specific ‘pull’ factors • Drop US from sample

I

Empirics

Implementation with country-by-country Cholesky factorization with LEVt ordered first Shocks

20

Identification of international credit supply ‘push’ shock in the data I

In the model shocks to leverage shift global supply of cross-border bank credit

I

In the data LEVt arguably is exogenous to conditions in individual country

i

• Unlikely to be driven by country-specific ‘pull’ factors • Drop US from sample

I

Implementation with country-by-country Cholesky factorization with LEVt ordered first Shocks

I

Robustness • Control for globally synchronized pull shocks • Drop ‘not so small’ open economies

Empirics

20

Transmission consistent with model and stylized facts on boom-bust episodes in cross-border credit Leverage 6

Percent

Cross-border Credit

Consumption 0.4

2

0.3

1.5 4 1

0.2

2

0.5

0.1

0

0

0

5

10 15 20 25 30 35 40

5

House Price

10 15 20 25 30 35 40

5

Real Exch. Rate

10 15 20 25 30 35 40

Current Account

0

1

Percent

0 -0.2 0.5

-0.1

-0.4 -0.6

0 5

10 15 20 25 30 35 40

Quarters

Empirics

-0.2

-0.8 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

21

The shock explains a substantial fraction of the forecast error variance of domestic variables Leverage

100

30

Cross-border Credit

Consumption

20

Percent

15 20 50

10 10 5

0

0 5

10 15 20 25 30 35 40

House Price

25

Percent

20

10 15 20 25 30 35 40

Real Exch. Rate

20

5

10 15 20 25 30 35 40

Current Account

15

15 10

15 10 10

5 5

5 0

0 5

10 15 20 25 30 35 40

Quarters

Empirics

0 5

0 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

22

Understanding cross-country heterogeneity

I

Error bands for IRFs and FEVDs are relatively wide ⇒ Significant differences across countries

I

Does heterogeneity follow specific patterns?

Empirics

23

Understanding cross-country heterogeneity

I

Error bands for IRFs and FEVDs are relatively wide ⇒ Significant differences across countries

I

Does heterogeneity follow specific patterns?

I

Conjecture Transmission affected by country characteristics. Focus on two characteristics that have a clear counterpart in the model • Maximum LTV limit (θ i ) • Share foreign currency liabilities (ψi )

I

Empirics

Compare theoretical predictions with data

23

Loan-to-Value ratios I

Prediction 1 A larger LTV ratio (higher θ ) implies a higher sensitivity of Ci , HPi , and RERi to shocks to χ • If constraint binds, higher θ leads to higher house price response, and hence

larger collateral effect and amplification • Higher θ leads to higher credit and demand, and hence larger real exchange

rate response

Empirics

24

Loan-to-Value ratios I

Prediction 1 A larger LTV ratio (higher θ ) implies a higher sensitivity of Ci , HPi , and RERi to shocks to χ • If constraint binds, higher θ leads to higher house price response, and hence

larger collateral effect and amplification • Higher θ leads to higher credit and demand, and hence larger real exchange

rate response

House Price

Consumption

IRs (Max)

1 0.5 0

IRs (Max)

4

1.5

IRs (Max)

Real Exch. Rate 0

2 0

40

60

80

(Home Ownership) x (max LTV)

-2 -3

Correlation: 0.4 t-Statistic: 2.9

Correlation: 0.3 t-Statistic: 2.0

-0.5

-1

40

60

80

(Home Ownership) x (max LTV)

Correlation: -0.2 t-Statistic: -1.3

40

60

80

(Home Ownership) x (max LTV)

NOTE: LTV is maximum LTV weighted by homeownership rate.

Empirics

24

Share of foreign currency debt I

Prediction 2 A larger share of foreign currency debt (higher ψ) may imply a higher sensitivity of Ci , HPi , and RERi to shocks to χ • Higher ψ implies larger collateral and endowment valuation effects (↑), and

larger debt valuation effect (↓) • Depending on which effect dominates, higher ψ can lead to both

higher/lower amplification

Empirics

25

Share of foreign currency debt I

Prediction 2 A larger share of foreign currency debt (higher ψ) may imply a higher sensitivity of Ci , HPi , and RERi to shocks to χ • Higher ψ implies larger collateral and endowment valuation effects (↑), and

larger debt valuation effect (↓) • Depending on which effect dominates, higher ψ can lead to both

higher/lower amplification

House Price

Consumption

IRs (Max)

1 0.5 0 Correlation: 0.5 t-Statistic: 4.8

-0.5 0.2

0.4

0.6

0.8

Share of foreign currency debt

IRs (Max)

4

1.5

IRs (Max)

Real Exch. Rate 0

2 0

-1 -2 -3

Correlation: 0.6 t-Statistic: 5.1

-2 0.2

0.4

0.6

0.8

Share of foreign currency debt

Correlation: -0.4 t-Statistic: -3.7

0.2

0.4

0.6

0.8

Share of foreign currency debt

NOTE: Share of foreign currency liabilities computed using BIS banking data.

Empirics

25

Robustness checks

I

Control for synchronized pull shocks

Go

• Augment vector of endogenous variables with world GDP

I

Drop ‘not so small’ economies that can affect global credit supply

Go

• Japan, Switzerland, UK, and Germany

I

Exclude lagged country variables from the leverage equation

I

Scatter plots vs. VARs on sub-groups

I

VAR vs. Local Projections

Empirics

Go

Go

Go

26

Conclusions I

Theory • Expansionary push shock triggered by changes in leverage of global banks

I

Empirics • Identified shock to US broker-dealers’ leverage explains a significant share of

domestic variance • Transmission consistent with model (both time series and cross-section)

Conclusions

27

Conclusions I

Theory • Expansionary push shock triggered by changes in leverage of global banks

I

Empirics • Identified shock to US broker-dealers’ leverage explains a significant share of

domestic variance • Transmission consistent with model (both time series and cross-section)

I

Policy implications • Max LTV ratios and shares of FX liabilities associated with sensitivity to shock • Macro-pru can try to influence them and hence affect final outcome

I

Next on agenda: Quantitative model and policy analysis

Conclusions

27

Appendix: Event Study

Appendix

28

Event study: Methodology

I

Event study follows Mendoza and Terrones (2008)

I

Boom (Bust) = At least 3 consecutive years of ∆ ln KFit > 0 (< 0)

I

134 boom, 81 bust, and 50 boom-bust episodes

I

Observe economy’s behavior around boom-bust cycles’ peak

Appendix

29

Event Study: Summary Statistics Mean Across Episodes Boom

Number Duration Max Min Amplitude

Bust

Boom-bust

ALL

AE

EM

ALL

AE

EM

ALL

AE

EM

2.4 7.3 32.6 5.0 131.6

2.5 8.8 28.5 3.7 130.1

2.3 6.1 35.9 5.9 132.8

1.4 4.4 -4.2 -20.4 -53.2

1.1 3.7 -4.6 -17.5 -36.9

1.6 4.8 -4.1 -21.9 -61.3

0.9 12.7 36.3 -21.8 103.5

0.8 13.4 29.5 -19.2 115.7

0.9 12.4 40.5 -23.5 96.0

Median Across Episodes Boom

Number Duration Max Min Amplitude

Bust

Boom-bust

ALL

AE

EM

ALL

AE

EM

ALL

AE

EM

2.0 6.0 28.5 3.0 105.5

2.0 8.0 26.0 2.0 121.0

2.0 5.0 31.0 4.0 84.0

1.0 4.0 -3.0 -18.0 -42.0

1.0 3.0 -3.0 -15.0 -30.0

2.0 4.0 -3.0 -19.0 -51.5

1.0 12.0 29.0 -19.0 80.5

1.0 13.0 27.0 -18.0 106.0

1.0 12.0 31.0 -20.0 39.0

NOTE. Number is number of episodes; Duration is length of episodes in years; Max and Min are maximum and minimum growth rate of cross-border credit during episode, respectively; Amplitude is cumulative sum of growth rate of cross-border credit over episode.

Appendix

30

Appendix: Data

Appendix

31

Data

I

Global variable • Global banks’ leverage: US Broker-Dealers’ leverage (LEVt )

I

Country-specific variables • International credit: cross-border claims of BIS reporting banks (KFit ) • Macro variables: private consumption (Cit ) and current account to GDP

(CAit /Yit )

• Asset prices: house prices (HPit ) and real exchange rate vis-a-vis the US

dollar (RERit )

I

Sample: 57 countries between 1977 and 2012 (unbalanced) Data sources

Appendix

32

International credit claims I

Cross-border total claims (all instruments, to financial and non-financial sectors) of BIS reporting banks on country i

KFit =

N

∑

j=1(j6 =i)

KFij,t

I

Important role of banks in international financial intermediation in the run up to the global financial crisis

I

Three examples Argentina

300

Brazil

150

200

100

100

50

Ireland

100

50

0

0 1980

Appendix

1990

2000

2010

0 1980

1990

2000

2010

1980

1990

2000

2010

33

Leverage of US Broker-Dealers I

Leverage is defined as Assets/Equity of the US broker dealer sector from the Federal Reserve’s Flow of Funds

I

Empirical proxy for global banks’ leverage • [Bruno and Shin (2015); Rey (2013)]

30

Leverage

25 20 15 10 1985

Appendix

1989

1993

1997

2001

2005

2009

34

Data sources: Countries I

24 Advanced Economies: Australia, Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Italy, Japan, Luxembourg, Malta, Netherlands, New Zealand, Norway, Portugal, Spain, Sweden, Switzerland, UK, and US

I

33 Emerging Economies: Argentina, Brazil, Bulgaria, Chile, China, Colombia, Croatia, Czech Republic, Estonia, Hong Kong, Hungary, India, Indonesia, Israel, Korea, Latvia, Lithuania, Malaysia, Mexico, Morocco, Peru, Philippines, Poland, Russia, Serbia, Singapore, Slovakia, Slovenia, South Africa, Taiwan, Thailand, Ukraine, and Uruguay

I

Sample: 1970:Q1–2012:Q4 (subject to data availability)

Back

Appendix

35

Data sources: Quantities I

Cross-border banking flows. Foreign claims (all instruments, in all currencies, locational by residence) of all BIS reporting banks vis-` a-vis all sectors deflated by US consumer price inflation. Source: BIS.

I

GDP. Real index. Source: OECD, IMF IFS, Bloomberg.

I

Consumption. Real private final consumption index. Source: OECD, IMF, IFS, Bloomberg.

I

Current account to GDP ratio. Current account balance divided by nominal GDP. Source: OECD, IMF IFS, Bloomberg.

Back

Appendix

36

Data sources: Prices I

House prices. Nominal house prices deflated by consumer price inflation. Source: Cesa-Bianchi et al (2015, JMCB)

I

Short-term interest rates. Short-term nominal market rates. A real ex-post interest rate is obtained by subtracting consumer price inflation. Source: OECD, IMF, IFS, Bloomberg.

I

Consumer prices. Consumer price index. Source: OECD, IMF IFS, Bloomberg.

I

Equity prices. Equity price index deflated by consumer price inflation. Source: OECD, IMF IFS, Bloomberg.

I

Exchange rate vis-` a-vis US dollar. US dollars per unit of domestic currency. A real exchange rate is obtained with US and domestic consumer price inflation. Source: Datastream.

I

Real effective exchange rate. Index (such that a decline of the index is a depreciation). Source: IMF IFS, BIS, Bloomberg.

Back Appendix

37

Cross-border credit: Banks vs. non-Banks

US Dollars (Trillions)

30 25

Banks Non-banks Total

20 15 10 5 0 1977

1981

1985

1989

1993

1997

2001

2005

2009

2013

Back

Appendix

38

Appendix: Model

Appendix

39

Households I

Home country (starts with zero initial credit)

max

{c1 ,c2 ,h1 ,f }

u(c1 ) + βu(c2 ) + v(h1 )

with β ∈ (0, 1) and h0 given, subject to

c1 + qh1 − b − s1 f c2 where

ct ≡ I

= pH1 y + qh0 = pH2 y − Rb b − s2 Rf

α c1 − α cHt Ft α α (1 − α )1− α

Collateral constraint

b + s1 f ≤ θqh1

Appendix

40

Households I

Foreign country (1 > β∗ > β)

max u(c1∗ ) + β∗ u(c2∗ )

{c1∗ ,c2∗ ,d,e}

subject to

c1∗ + d + e + ψ(e) c2∗

∗ ∗ y = pF1 ∗ ∗ = pF2 y + Rd d + Re e + Π

with ψ0 , ψ00 > 0, and ∗

c∗ α c∗1− α c = ∗ α ∗ H F ∗ 1− α ∗ α (1 − α ) ∗

∗

Appendix

41

Global financial intermediaries I

Balance sheet Assets Loans (H currency): Loans (F currency):

I

Profits

Π = Rf +

Liabilities b/s1 f

Deposits (F currency):

d

Equity (F currency):

e

Rb b − Rd d − Re e − φ s2



b s1



where φ(·) is cost of swapping loans in Foreign currency (with φ0 , φ00 > 0) I

Leverage constraint (capital requirement)

e≥χ Appendix



b +f s1



42

Equilibrium: Analytical characterization I

Take limit for n → 0 ⇒ Home becomes small open economy

I

Abstract from intermediaries portfolio problem • Fix the ratio between domestic and foreign currency liabilities (η )

I

All households are risk-neutral and housing (land) is in fixed supply

I

Then, we can solve analytically • Terms of trade from goods market equilibrium (⇒ Real exchange rate) • Credit demand and credit supply

I

Represent the equilibrium in the {f , R} space

Appendix

43

Parameters

Parameter

Description

β β∗ κ λ θ η χ y = y∗

Country H discount factor Country F discount factor Normalized marginal utility of housing Degree of openness LTV ratio Share of foreign debt Capital requirement Endowments

Value

0.9 0.99 0.85 0.79 0.92 0.43 0.1 1

I

Adjustment cost parameters pin down equity and loans risk premia

I

In turn, the level of risk premia will determine whether the equilibrium lies in the unconstrained/constrained region

Appendix

44

Appendix: Identification

Appendix

45

Estimated international credit supply shock I

Orthogonalized leverage innovations for each of the country-specific models (light solid lines) can differ slightly across countries • Lagged feedback from the rest of the system to leverage equation • Models are estimated over different sample periods (depending on data

availability)

20

Percent

10 0 -10 -20 -30 1985

1989

1993

1997

2001

2005

2009

NOTE. The light solid lines are the orthogonalized leverage innovations for each of the country-specific models. The dark solid line is the cross-country average of the country-specific leverage innovations. The dotted lines are the average of the one-standard deviation bands, equal to 7.5 percent per quarter.

Back Appendix

46

Brokers-Dealers’ leverage innovations and their underlying determinants I

Leverage is exogenous in our model, but in the data various factors can affect the leverage of US Broker-Dealers xt ∆FFRt

(1)

(2)

(3)

-2.477** [-2.364]

eMP

(5) -2.613** [-2.536]

-0.0497 [-0.650]

RLt − Rt

-0.900 [-1.642]

VIXt Obs. Adj. R2

(4)

111 0.049

91 0.005

111 0.024

-0.00182** [-2.057]

-0.00195** [-2.252]

111 0.037

111 0.091

NOTE. The Table reports a regression of the leverage innovations (average across countries) on their possible determinants:

eLEV = βxt . ∆FFRt is the first difference of the real (ex-post) federal fund rate; eMP is Romer and Romer (2004) monetary t

policy shock; RL t − Rt is the slope of the US yield curve; VIXt is the VIX index. The regressions also include a constant and world GDP (not reported).

Back Appendix

47

Appendix: Identification Robustness

Appendix

48

Identification robustness: controlling for globally synchronized ‘pull’ shocks I

Small open economy assumption rules out local factors can drive LEVt • No single country can affect leverage of global banks

I

But LEVt could be affected by globally synchronized factors

I

Synchronized shocks should affect world GDP • Augment vector of endogenous variables with world GDP

Xit =

I



Ytw

LEVt

KFit

Cit

HPit

RERit

CAit /Yit



Shock to leverage of US broker-dealers still identified with Cholesky Back

Appendix

49

IRFs to leverage shock (Identification robustness) Leverage

Cross-border Credit 1.5

4

1

2

0.5

0.2

Percent

6

Consumption

0.1

0

0 0 5

10 15 20 25 30 35 40

5

House Price

10 15 20 25 30 35 40

5

Quarters Current Account

Real Exch. Rate

0.8

0

0.6

-0.2

0.4

-0.4

10 15 20 25 30 35 40

0.1 0 -0.1

0.2

-0.6

0

-0.8 5

10 15 20 25 30 35 40

Quarters

-0.2 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back

Appendix

50

Variance decomposition (Identification robustness) Leverage

100

20

Cross-border Credit

Consumption

15

Percent

15 10

50

10

5

5 0

0 5

10 15 20 25 30 35 40

House Price

20

0 5

10 15 20 25 30 35 40

Real Exch. Rate

15

5

10 15 20 25 30 35 40

Quarters Current Account

15

15 10

10

5

5

10 5 0

0 5

10 15 20 25 30 35 40

Quarters

0 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back

Appendix

51

Identification robustness: World equity prices I

World GDP might not have enough forward looking component to capture globally synchronized pull shocks

I

Estimate a VAR with world equity prices (world MSCI index) instead of GDP

I

Results are robust qualitatively, but a bit weaker quantitatively • World equity prices incorporate information, like risk premia, also captured

by the leverage variable

Back

Appendix

52

Identification robustness: World equity prices I

Impulse responses Leverage

Cross-border Credit

Consumption

1.5

0.2

6

Percent

1 4 0.5 2

0.1 0

0 0

-0.1 5

10 15 20 25 30 35 40

House Price 0.6

5

10 15 20 25 30 35 40

Real Exch. Rate

0.2

5

10 15 20 25 30 35 40

Quarters Current Account

0.1

0 0

0.4

-0.2

0.2

-0.4

0

-0.1

-0.6 -0.2

-0.2 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

53

Identification robustness: World equity prices Forecast error variance decompositions Leverage

80

Cross-border Credit

Consumption 10

10

60

Percent

I

40 5

5

20 0

0 5

10 15 20 25 30 35 40

0 5

House Price 15

10 15 20 25 30 35 40

5

10 15 20 25 30 35 40

Quarters Current Account

Real Exch. Rate 10

10

5

5

10

5

0

0 5

10 15 20 25 30 35 40

Quarters

0 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

54

Appendix: VAR Robustness

Appendix

55

VAR robustness: Drop large countries (IRF) Leverage

Percent

6

Cross-border Credit

Consumption 0.4

2

0.3 4

0.2

1 2

0.1 0

0

0 5

10 15 20 25 30 35 40

5

House Price

10 15 20 25 30 35 40

5

Real Exch. Rate

10 15 20 25 30 35 40

Current Account

0

Percent

1

0

-0.2

-0.1

-0.4

0.5

-0.6 0

-0.2

-0.8 -0.3 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

56

VAR robustness: No feedback from SOE I

Block exogenous VAR: no feedback from endogenous variables in country i to Broker-Dealers’ leverage

I

VAR for country i (abstracting from constant and time trend) is



I

LEVt xi,t





=

F11,i F21,i

0

F22,i



LEVt−1 xi,t−1





+

B11,i B21,i

0

B22,i



eLEV t exi,t



Identification: Cholesky decomposition as in the baseline

Back

Appendix

57

VAR robustness: No feedback from SOE (IRF) Leverage

Cross-border Credit

Consumption

8 3

0.4

Percent

6 0.3

2 4

0.2 1

2

0.1

0

0 5

10 15 20 25 30 35 40

House Price 1.5

0 5

10 15 20 25 30 35 40

Real Exch. Rate

0

Percent

10 15 20 25 30 35 40

Current Account 0

-0.2 1

5

-0.4

-0.1

-0.6 0.5

-0.2 -0.8

0

-1 5

10 15 20 25 30 35 40

Quarters

-0.3 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

58

VAR robustness: No feedback from SOE (FEVD) Leverage

100

60

Cross-border Credit

Consumption

40

Percent

30 40 50

20 20 10

0

0 5

10 15 20 25 30 35 40

House Price

50

0 5

10 15 20 25 30 35 40

Real Exch. Rate

30

5

10 15 20 25 30 35 40

Current Account

30

Percent

40 30 20

20

20

10

10

10 0

0 5

10 15 20 25 30 35 40

Quarters

0 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

59

VAR robustness: Local Projections Leverage

Cross-border Credit

Consumption

Percent

6 2

0.4

1

0.2

4 2 0 0

0 1

2

3

4

5

6

7

8

9 10

1

2

House Price

3

4

5

6

7

8

9 10

Real Exch. Rate

1

2

3

4

5

6

7

8

9 10

Current Account

0.1

0.5

Percent

0 1

0.5

0

-0.1

-0.5

-0.2 -0.3

-1 -0.4

0 1

2

3

4

5

6

Quarters

7

8

9 10

1

2

3

4

5

6

Quarters

7

8

9 10

1

2

3

4

5

6

7

8

9 10

Quarters

Back Appendix

60

VAR robustness: Local Projections (with REER) Leverage

Cross-border Credit

Consumption

3

Percent

6

0.4 2

4 2

1

0

0 1

2

3

4

5

6

7

8

9 10

0.2

0 1

2

House Price

3

4

5

6

7

8

9 10

1

2

Real Exch. Rate

3

4

5

6

7

8

9 10

Current Account 0.1

0.2

0

Percent

1 0

-0.1 -0.2

0.5

-0.2

0

-0.4

-0.3

1

2

3

4

5

6

Quarters

7

8

9 10

-0.4 1

2

3

4

5

6

Quarters

7

8

9 10

1

2

3

4

5

6

7

8

9 10

Quarters

Back Appendix

61

VAR estimated on ‘bins’: High and low share of foreign currency liabilities Leverage

Percent

6

Cross-border Credit

Consumption

2

0.4

1

0.2

4 2 0

0

0 5

10 15 20 25 30 35 40

5

House Price

10 15 20 25 30 35 40

5

Real Exch. Rate

10 15 20 25 30 35 40

Current Account

0 0

Percent

1

-0.2 -0.1

-0.4

0.5

-0.2

-0.6 0

-0.8 5

10 15 20 25 30 35 40

Quarters

-0.3 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

62

VAR estimated on ‘bins’: High and low maximum Loan-to-Value Leverage

Consumption

2

0.4

1.5

0.3

1

0.2

2

0.5

0.1

0

0

6

Percent

Cross-border Credit

4

5

10 15 20 25 30 35 40

0 5

House Price

10 15 20 25 30 35 40

5

Real Exch. Rate

10 15 20 25 30 35 40

Current Account

0 0

Percent

1 -0.2

-0.1

-0.4

0.5

-0.6

-0.2

-0.8

0

-0.3 5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

5

10 15 20 25 30 35 40

Quarters

Back Appendix

63

Appendix: Cross-Section

Appendix

64

Other characteristics I

Focus on share of foreign currency liabilities (1/(1 + η )) and the maximum LTV limit (θ ) as they have a clear counterpart in the model

I

But other characteristics might be relevant • Exchange rate flexibility • Controls on capital inflows • Mortgage credit over GDP

Max Loan to Value Foreign currency liability Exch. Rate flexibility Capital controls (inflows) Mortgage debt / GDP

Consumption

House Price

Exch. Rate

0.32 0.53 -0.40 0.23 -0.31

0.44 0.54 -0.41 0.32 -0.42

-0.21 -0.39 0.16 -0.28 0.25

NOTE. Correlation between the peak impulse response of selected variables (columns) and country characteristics (rows). See the appendix on data definition and sources.

Back Appendix

65