International Monetary FundJanuary 18, 2018

Bailing Out the People? When Private Debt Becomes Public

Samba Mbaye, Marialuz Moreno Badia, and Kyungla Chae

Fiscal Affairs Department

The views expressed in this presentation are those of the authors and do not necessarily represent the views of the IMF, its Executive Board, or IMF management.

1

“While the focus has been on the Dow Jones and

Wall Street, we are addressing the real pain felt by Mr. and Mrs. Jones on Main street. They are why we must pass this

legislation today.”

Nancy Pelosi, Statement in support of the Emergency Economic

Stabilization Act(October 3, 2008)

Marialuz Moreno Badia International Monetary Fund

Apparent “debt migration” after the GFC

2

Advanced Economies: Gross Debt(Percent of GDP)

Sources: Eurostat; Global Debt Database; and authors’ calculations.

Spain Gross Debt(Percent of GDP)

Results CausalityData & MethodologyMotivation Conclusions

40

50

60

70

80

90

100

160

180

200

220

240

260

280

300

320

340

360

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

Total debt

Private debt

General government debt (right scale)

30

70

110

150

190

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

Private debtGeneral government debtGeneral government debt minus capital injections and call on guarantees by financial institution

49.5%

64.9%

Marialuz Moreno Badia International Monetary Fund

The Million Dollar Questions…

3

Through What Channels?

When Does Overall Indebtedness Increase?

Does Private Deleveraging Always Trigger a Rise in Public Debt?

Results CausalityData & MethodologyMotivation Conclusions

Marialuz Moreno Badia International Monetary Fund

4

Global Debt DatabaseBuilding on the Fall 2016 Fiscal Monitor

543 episodes

153 countries

1950-2016

AEs, EMEs, LICs

Crisis & Non-crisis

Incidence of Private Deleveraging, 1950–2016(Number of Episodes)

Unparalleled Coverage of Private Deleveraging

Results CausalityData & MethodologyMotivation Conclusions

Number of episodesSwitzerland 9Myanmar 8Argentina 8Canada 6

Japan 4United States 3Germany 3

Marialuz Moreno Badia International Monetary Fund

Private Deleveraging: Are We Measuring the Right Thing?

5

Private Debt and Leverage Flows in the United States(Annual change, percent of GDP)

Sources: Global Debt Database; and authors’ calculations.

-13

-8

-3

2

7

1951

1956

1961

1966

1971

1976

1981

1986

1991

1996

2001

2006

2011

2016

Macro-factorsLeverage flowsPrivate debt ratio

Lev. flows

Macro

Private debt

+Lev.

flowsMacro

Private debt

+

Case 1 Case 2

Results CausalityData & MethodologyMotivation Conclusions

Marialuz Moreno Badia International Monetary Fund

A Descriptive Framework to Tell the Story of Deleveraging

6

Private Debt and Leverage Flows in the United States(Annual change, percent of GDP)

Sources: Global Debt Database; and authors’ calculations.

-13

-8

-3

2

7

1986

1991

1996

2001

2006

2011

2016

Macro-factors

Leverage flows

Private debt ratio

Normal times

T+1

T+2

T+3T+4

T-1

T-2

T-3T-4

T

T+1

T+2T+3

T+4

T-1

T-2

T-3T-4

T

Gourinchas and Obstfeld (2012)= + +

Results CausalityData & MethodologyMotivation Conclusions

Marialuz Moreno Badia International Monetary Fund

Private Deleveraging since World War II

7Sources: Global Debt Database; and authors’ calculations.Note: AEs = advanced economies; EMEs = emerging market economies; LIDCs = low-income developing countries.

Results CausalityData & MethodologyMotivation Conclusions

100

105

110

115

120

125

130

135

0

10

20

30

40

50

60

-5 -4 -3 -2 -1 0 1 2 3 4Start of deleveraging

3.4 years

Private debt drop from peak to through: 11.8%

DeleveragingEpisode

5-yr prior to deleveraging

GDPPrivate debt

World: Private Deleveraging Episodes

Number of episodes: 543Probability of deleveraging: 28%

Private Deleveraging Episodes by Income Group

AEs EMEs LIDCs

Length (years) 3.9 3.3 3.5

Size (drop from peak to through, % of GDP) -18.6 -15.2 -6.2

Peak private debt (average, % of GDP) 132.7 52.6 20.6

Pre-delev. change in debt (average of 5-yr cum. change, % of GDP) 16.9 8.8 3.1

Pre-delev. growth (average of 5-yr, percent) 3.0 4.0 3.3

Growth during delev. episodes (average, percent) 3.5 4.1 3.9

Number of episodes 114 181 248

Probability of deleveraging (share of deleveraging obs.) 25% 27% 30%

AEs EMEs LIDCs

Length (years) 3.9 3.3 3.5

Size (drop from peak to through, % of GDP) -18.6 -15.2 -6.2

Peak private debt (average, % of GDP) 132.7 52.6 20.6

Pre-delev. change in debt (average of 5-yr cum. change, % of GDP) 16.9 8.8 3.1

Pre-delev. growth (average of 5-yr, percent) 3.0 4.0 3.3

Growth during delev. episodes (average, percent) 3.5 4.1 3.9

Number of episodes 114 181 248

Probability of deleveraging (share of deleveraging obs.) 25% 27% 30%

Marialuz Moreno Badia International Monetary Fund

The Typical Deleveraging Episode

8Sources: Global Debt Database; and authors’ calculations.Note: The dotted line denotes the 95 percent confidence interval for each conditional mean.

-7

-2

3

-4 -3 -2 -1 0 1 2 3 4 5-6

-4

-2

0

2

4

-4 -3 -2 -1 0 1 2 3 4-2%

-1%

0%

1%

2%

-4 -3 -2 -1 0 1 2 3 4 5

70

90

110

-5 -4 -3 -2 -1 0 1 2 3 4

1. Private Debt 2. Growth 3. Public Debt 4. Total Debt (stock)

Private debt to GDPLeverage flowsMacro-related changes

Results CausalityData & MethodologyMotivation Conclusions

Years (t) Years (t) Years (t) Years (t)

Marialuz Moreno Badia International Monetary Fund

The Private-Public Debt Tango: A Global Phenomenon

9Sources: Global Debt Database; and authors’ calculations.

3. Public Debt2. Real Growth 4. Total Debt (Stock)

Results CausalityData & MethodologyMotivation Conclusions

-9

-7

-5

-3

-1

1

3

-4 -3 -2 -1 0 1 2 3 4 5 6 7-3

-2

-1

0

1

-4 -3 -2 -1 0 1 2 3 4 5-2

-1

0

1

2

3

4

5

-4 -3 -2 -1 0 1 2 3 4 50

20

40

60

80

100

120

-5 -4 -3 -2 -1 0 1 2 3 4

1. Private Debt

Emerging market economiesAdvanced economies Low-income developing countries

Years (t) Years (t) Years (t) Years (t)

Marialuz Moreno Badia International Monetary Fund

Bailout or Growth Story?

10

Advanced Economies: Sources of Change in Public Debt during Private Deleveraging

Sources: Global Debt Database; and authors’ calculations.

-4

-2

1

3

5

t-4 t-3 t-2 t-1 t t+1 t+2 t+3 t+4 t+5

CAPB (positive=deficit)

Macro-flows

Stock-flow adjustment

Public debt to GDP

Results CausalityData & MethodologyMotivation Conclusions

Marialuz Moreno Badia International Monetary Fund

It is NOT just a Crisis Story

11Sources: Global Debt Database; and authors’ calculations.

1. Private Debt

-8

-6

-4

-2

0

2

4

-4 -3 -2 -1 0 1 2 3 4 5 6 7

2. Public Debt 3. Total Debt 4. Real Growth

-2

-1

0

1

2

3

4

5

-4 -3 -2 -1 0 1 2 3 4 5 6 70

20

40

60

80

100

120

140

-4 -3 -2 -1 0 1 2 3 4 5 6 7-4

-3

-2

-1

0

1

-4 -3 -2 -1 0 1 2 3 4 5 6

Results CausalityData & MethodologyMotivation Conclusions

Years (t) Years (t) Years (t) Years (t)

Without crisis With crisis

Marialuz Moreno Badia International Monetary Fund

The “ Unsuccessful” Deleveraging

12Sources: Global Debt Database; and authors’ calculations.Note: The dotted line denotes the 95 percent confidence interval for each conditional mean.

Results CausalityData & MethodologyMotivation Conclusions

1. Private Leverage Flows 2. Real Growth 3. Public Debt 4. Total Debt

-9

-7

-5

-3

-1

1

3

-4 -3 -2 -1 0 1 2 3 4 5 6 7-6

-5

-4

-3

-2

-1

0

1

2

-4 -3 -2 -1 0 1 2 3 4-2

0

2

4

6

-4 -3 -2 -1 0 1 2 3 475

80

85

90

95

100

105

110

115

120

-5 -4 -3 -2 -1 0 1 2 3 4

High-growth Growthless

Years (t) Years (t) Years (t) Years (t)

Marialuz Moreno Badia International Monetary Fund

Fiscal Policy to the Rescue?

13Sources: Global Debt Database; and authors’ calculations.

1. Change in Public Debt 2. Real Growth 3. Total Debt (Stock)

-3

-1

1

3

5

7

9

11

-4 -3 -2 -1 0 1 2 3 4-6

-5

-4

-3

-2

-1

0

1

2

-4 -3 -2 -1 0 1 2 3 4

Years (t)

75

85

95

105

115

125

135

-5 -4 -3 -2 -1 0 1 2 3 4Years (t)

Results CausalityData & MethodologyMotivation Conclusions

Years (t)

Strong government Limited government

Marialuz Moreno Badia International Monetary Fund

Step 1: Estimate the probability (p) of receiving the treatment Step 2: Weigh each country by 1/p

Establishing Causality: a “Doubly Robust” Approach

14

Results CausalityData & MethodologyMotivation Conclusions

Recast private deleveraging into

a dichotomous “treatment”

Estimate the causal impact of private deleveraging using IPWRA

Inverse Propensity Weighting (IPW)

Regression Adjustment (RA)

Step1: Fit a linear model of growth and public debtStep 2: Use predictions from the latter as counterfactuals

Marialuz Moreno Badia International Monetary Fund

Our Specification

15

Results CausalityData & MethodologyMotivation Conclusions

Private deleveraging Growth Public debt

Lagged growthLagged GDP per capitaCrisis in the last 5 yearsLagged private and public debt3-yr cum. growth private and public debtLagged private interest rates

Lagged growthLagged GDP per capitaLagged institution quality (RofL)Lagged inflationLagged opennessLagged private and public debt

Lagged growth and GDP per capitaCrisis in the last 5 yearsLagged private and public debt Lagged public interest ratesLagged aged dependency ratioLagged institution qualityLagged fractionalization

Marialuz Moreno Badia International Monetary Fund

Causality Running from Private Deleveraging

16Sources: Global Debt Database; and authors’ calculations.

1. GDP (percent) 2. Public Debt (percent of GDP)

Results CausalityData & MethodologyMotivation Conclusions

-4

-3

-2

-1

0

1

t t+1 t+2 t+3 t+4 t+50

1

2

3

4

5

t t+1 t+2 t+3 t+4 t+5

Marialuz Moreno Badia International Monetary Fund

Conclusion

17

Implications for Debt Sustainability

…even Without a Crisis

You May End-up Paying your Neighbor’s Mortgage

Results CausalityData & MethodologyMotivation Conclusions

18

Marialuz Moreno Badia International Monetary Fund

The 2010 Growth Accident

19Sources: Global Debt Database; and authors’ calculations.Note: The dotted line denotes the 95 percent confidence interval for each conditional mean.

Results CausalityData & MethodologyMotivation Conclusions

-9

-7

-5

-3

-1

1

3

5

-4 -3 -2 -1 0 1 2 3 4 5 6 7

Leverage flowsChange in private debt ratio

-4

-2

0

2

4

6

-4 -3 -2 -1 0 1 2 3 4 5 6 775

85

95

105

115

125

-5 -4 -3 -2 -1 0 1 2 3 4 5 6 7-4

-3

-2

-1

0

1

2

-4 -3 -2 -1 0 1 2 3 4 5 6 7

1. Private Debt 2. Public Debt 3. Total Debt 4. Real Growth

Years (t) Years (t) Years (t) Years (t)

Marialuz Moreno Badia International Monetary Fund

Overlap and specification tests

20Sources: Global Debt Database; and authors’ calculations.

Results CausalityData & MethodologyMotivation Conclusions

02

46

8de

nsity

0 .2 .4 .6 .8 1Propensity score, del_fl~s=1

del_fl~s=0 del_fl~s=1

Fixed-effects IPWRA

GDP Growth (-2,-1) -0.013*** -0.003(0.004) (0.003)

Growth public debt (-2,-1) -1.959 1.865(2.971) (1.205)

GDP Growth (-3,-1) 0.010*** -0.003(0.004) (0.003)

Growth public debt (-3,-1) -1.761 1.257(3.123) (1.354)

GDP Growth (-4,-1) 0.017*** 0.006(0.008) (0.007)

Growth public debt (-4,-1) -3.235 0.959(3.644) (1.551)

GDP Growth (-5,-1) 0.029*** 0.013(0.012) (0.012)

Growth public debt (-5,-1) -7.318* -0.644(4.300) (1.688)

Distribution of Estimated Propensity ScoresEffects of Private Deleveraging on Past Growth and Public Debt