princeton initiative

8/13/2019 Princeton Initiative

1/73

Global Liquidity

Hyun Song Shin

September 2011

Princeton Initiative in Macro, Money and Finance


2/73

Three Themes

Banking sector as driver of global financial conditions

Global banks (esp. European global banks) as transmission channel ofglobal liquidity conditions

US Dollar as currency underpinning global banking system

1


3/73

Mapping Global Liquidity

Consequences of global liquidity for United States

(Global flow of funds perspective)

Consequences of global liquidity for Europe(very briefly...)

Consequences of global liquidity for emerging/developing economies

2


4/73

Corporate Finance of Banking

A L

Assets

Equity

Debt

3


5/73

A L

Assets

Equity

Debt

A L

Assets

Equity

Debt

4


6/73

A L

Assets

Equity

Debt

A L

Assets

Equity

Debt

5


7/73

Asset growth

Leverage

growth

Slope = 1

0 A

B

6


8/73

Asset growth

Leverage

growth

Slope = 1Constant equity

growth of g

g

0

Constant equity

line

7


9/73

1998-3

1998-4

2007-3

2007-4

-.2

-.1

0

.1

.2

TotalAssetGrowth

-.2 -.1 0 .1 .2Leverage Growth

Lehman Brothers

1998-3

1998-4

2007-3

2007-4

-.2

-.1

0

.1

.2

TotalAssetGrowth


Merrill Lynch

1998-3

1998-4

2007-3

2007-4

-.2

-.1

0

.1

.2

TotalAssetGrowth


Morgan Stanley

1998-3

1998-42007-3

2007-4

-.1

0

.1

.2

To

talAssetGrowth


Bear Sterns2007-3

2007-4

-.05

0

.05

.1

To

talAssetGrowth

-.15 -.1 -.05 0 .05 .1Leverage Growth

Goldman Sachs

1998-3

1998-4

-.3

-.2

-.1

0

.1

To

talAssetGrowth


Citigroup Markets 98-04

Total Assets and Leverage

8


10/73

1998-4

2007-3

2007-4

2008-1

-.2

-.1

0

.1

.2

TotalAssets(logchange)

-.2 -.1 0 .1 .2Leverage (log change)

Asset weighted, 1992Q3-2008Q1, Source: SEC

Leverage and Total Assets Growth

9


11/73

What Drives the Leverage Cycle?

Value-at-Risk (VaR) is approximate worst case loss - smallest non-negative such that

Prob ( 0 ) for small 0

Value-at-Risk Rule (Basel). Maintain equityto limit failure prob to

= =

is Unit VaR (Value-at-Risk per dollar of assets). Leveragesatisfies

=1

10


12/73

Empirical implication:

ln = ln

so that

ln ln 1= (ln ln 1) (*)

Scatter chart of leverage changes against unit VaR changes should haveslope

1.

Evidence?

11


13/73

2007-Q3

2007-Q4 2008-Q1

2008-Q2

2008-Q3

2008-Q4

2009-Q1

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

-1 -0.8 -0.6 -0.4 -0.2 0 0.2 0.4

Leverage Growth

UnitVaR

Grow

th

Figure 1: Five (then four, three, then two) Wall Street banks, Adrian andShin (2011) Procyclical Leverage and Value-at-Risk

12


14/73

-5

0

5

10

15

20

Dec-01

Jun-02

Dec-02

Jun-03

Dec-03

Jun-04

Dec-04

Jun-05

Dec-05

Jun-06

Dec-06

Jun-07

Dec-07

Jun-08

Dec-08

Jun-09

Dec-09

Jun-10

Dec-10

Pre-Cri

sisStandardDeviations

Unit VaR

VaR/E

Leverage

Figure 2: Deleveraging keeps VaR in check: Adrian and Shin (2011)

Procyclical Leverage and Value-at-Risk 13


15/73

-5

0

5

10

15

20

Dec-01

Jun-02

Dec-02

Jun-03

Dec-03

Jun-04

Dec-04

Jun-05

Dec-05

Jun-06

Dec-06

Jun-07

Dec-07

Jun-08

Dec-08

Jun-09

Dec-09

Jun-10

Dec-10

Pre-CrisisStandardDeviat

ions

Unit VaR

Implied

Vol

CDSSpread

Figure 3: Unit VaR tracks VIX with accounting lag: Adrian and Shin (2011)

Procyclical Leverage and Value-at-Risk 14


16/73

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

Trillionpounds

Equity

Other Liabilities

Total MMFfunding

CustomerDeposits

Figure 4: Total Liabilities of Barclays (1992 - 2007) (Source: Bankscope)

15


17/73

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

Trillionpounds

Total Assets

Risk-WeightedAssets

Figure 5: Barclays, risk-weighted assets and total assets (Source:

Bankscope) 16


18/73

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

18.0%

1992

1994

1996

1998

2000

2002

2004

2006

2008

201

0

TotalCapitalRatio

Tier 1Ratio

Equity/Total

Assets

Figure 6: Barclays, capital ratios (Source: Bankscope)

17


19/73

0.0

0.5

1.0

1.5

2.0

2.5

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

201

0

TrillionEuros

Equity

Other Liabilities

Total CustomerDeposits

Figure 7: BNP Paribas total liabilities (Source: Bankscope)

18


20/73


21/73

0.0%

2.0%

4.0%

6.0%

8.0%

10.0%

12.0%

14.0%

16.0%

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

201

0

TotalCapitalRatio

Tier 1 Ratio

Equity/TotalAssets

Figure 9: BNP Paribas capital ratios (Source: Bankscope)

20


22/73


23/73


24/73

Turning Credit Risk Model on Its Head

Turn credit risk model on its head and think of it as credit supply model

Fix . Determine credit supply

=

1

1+

1+

() (0 1)

is ratio ofnotional assetstonotional debt

[ is normalized leverage measure, with (0 1)]

23


25/73

BIS Banking Statistics

BISlocational banking statistics

Classification based onresidence Branches/subsidiaries of global banks classified under host country Consistent with balance of payments and national income statistics Cross-border claims

BISconsolidated banking statistics

Classification based onnationality of parent

Foreign claims = cross-border claims + local claims International claims = cross-border claims + local claims in foreign

currency

24


26/73


27/73

2008Q1

2.0

3.0

4.0

5.0

6.0

7.0

8.0

9.0

10.0

11.0

1

999Q1

2

000Q2

2

001

Q3

2

002

Q4

2

004

Q1

2

005Q2

2

006Q3

2

007

Q4

2

009Q1

2

010Q2

TrillionDollars

US charteredcommercialbanks' totalfinancial

assets

US dollarassets ofbanks outsideUS

Figure 11: US dollar cross-border foreign currency claims and US commercial

bank total assets (Source: Flow of Funds, Federal Reserve and BISlocational banking statistics, Table 5A)

26


28/73

Figure 12: US Dollar-denominated assets and liabilities of euro area banks(Source: ECB Financial Stability Review, June 2011, p. 102)

27


29/73

USHouseholds

USBorrowers

USBankingSector

EuropeanGlobalBanks

border

Wholesale

funding market

Shadow banking

system

Figure 13: European global banks add intermediation capacity for

connecting US savers and borrowers 28


30/73


31/73

0

50

100

150

200

250

300

350

400

450

Dec-07

Feb-0

8

Apr-0

8

Jun-0

8

Aug

-08

Oct-08

Dec-0

8

Feb-0

9

Apr-0

9

Jun-0

9

Aug

-09

Oct-09

Dec-0

9

Feb-1

0

Billion

Dollars

Japan

Canada

Ireland

Switzerland

Netherlands

France

Germany

UK

USA

Figure 15: Claims outstanding on Federal Reserve Term Auction Facility

(TAF) on US and non-US banks (Source: Federal Reserve disclosures onTAF)

30


32/73

0

50

100

150

200

250

Dec-07

Feb-0

8

Apr-0

8

Jun-0

8

Aug

-08

Oct-08

Dec-0

8

Feb-0

9

Apr-0

9

Jun-0

9

Aug

-09

Oct-09

Dec-0

9

Feb-1

0

Billion

Dollars

Japan

Canada

Ireland

Switzerland

Netherlands

France

Germany

UK

Figure 16: Claims outstanding on Federal Reserve Term Auction Facility

(TAF) on non-US banks (Source: Federal Reserve disclosures on TAF) 31


33/73


34/73

31-Dec-08

30-Jun-08

-100

0

100

200

300

400

500

600

700

800

900

Mar-8

5

Se

p-8

6

Mar-8

8

Se

p-8

9

Mar-91

Se

p-92

Mar-94

Se

p-9

5

Mar-97

Se

p-9

8

Mar-0

0

Se

p-01

Mar-0

3

Se

p-04

Mar-0

6

Se

p-07

Mar-0

9

Se

p-1

0

Billion

Dollars

Interoffice Assets of ForeignBanks in US

Net Interoffice Assets ofForeign Banks in US

Figure 18: Interoffice assets of foreign banks in the United States (Source:Federal Reserve, series on Assets and Liabilities of U.S. Branches andAgencies of Foreign Banks)

33


35/73


36/73


37/73

Fund CDs and timedeposits Commercialpaper Corporatenotes Repos Total Net assets,$ billionsFidelity Cash Reserves 91 / 73 28 / 27 54 / 34 70 / 70 63 / 51 128JPMorgan Prime Money Market 98 / 94 35 / 31 57 / 39 73 / 73 67 / 62 120Vanguard Prime Money Market 94 / 69 39 / 25 0 / 0 68 / 68 33 / 24 106BlackRock Liquidity Temp 95 / 91 4 / 4 37 / 17 13 / 13 51 / 47 68Reserve Primary 98 / 88 24 / 18 54 / 51 18 / 18 43 / 37 65Schwab Value Advantage 91 / 64 24 / 19 58 / 48 67 / 67 54 / 40 61GS FS Prime Obligations 0 / 0 0 / 0 0 / 0 2/2 0 / 0 56Dreyfus Inst Cash Advantage 85 / 71 32 / 25 33 / 24 0 / 0 62 / 51 49Fidelity Inst Money Market 100 / 91 44 / 44 51 / 36 45 / 45 61 / 54 47Morgan Stanley Inst Liq Prime 4 / 4 19 / 19 0 / 0 91 / 91 37 / 37 34Dreyfus Cash Management 92 / 75 46 / 30 31 / 31 0 / 0 70 / 56 33AIM STIT Liquid Assets 95 / 69 25 / 20 27 / 16 84 / 84 57 / 45 32Barclays Inst Money Market 67 / 57 10/6 30 / 21 21 / 21 24 / 19 31

Merrill Lynch Premier Inst Portfolio 92 / 80 32 / 25 46 / 36 45 / 45 60 / 51 26Fidelity Inst Money Market: Prime 100 / 90 33 / 33 51 / 34 15 / 15 56 / 47 21Total 92 / 78 26 / 22 47 / 33 51 / 51 50 / 42 878Share of asset class in assets 34 26 13 11 100

Figure 21: US prime money funds assets in non-US/European bankobligations (% each asset class) mid-2008 (Source: Baba, McCauley andRamaswamy, BIS Quarterly Review 2009)

36


38/73

ABCP Sponsor Location and Funding Currency ($ million)Currency /Sponsor

Location

U.S. dollars Euro Yen Other Total

Belgium 30,473 4,729 0 0 35,202Denmark 1,796 0 0 0 1,796

France 51,237 23,670 228 557 75,692Germany 139,068 62,885 0 2,566 204,519

Italy 1,365 0 0 0 1,365

Japan 18,107 0 22,713 0 40,820Netherlands 56,790 65,859 0 3,116 125,765

Sweden 1,719 0 0 0 1,719

Switzerland 13,082 0 0 0 13,082nited Kingdom 92,842 62,298 0 3,209 158,349

United States 302,054 0 0 2,996 305,050

Total 714,871 219,441 22,941 12,444 969,697

Figure 22: ABCP sponsor location and funding currency January 1, 2007(Source: Acharya and Schnabel, IMF Economic Review 2009, data fromMoodys)

37


39/73

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

1980Q1

1982

Q1

1984

Q1

1986Q1

1988Q1

1990Q1

1992

Q1

1994

Q1

1996Q1

1998Q1

2000Q1

2002

Q1

2004

Q1

2006Q1

2008Q1

201

0Q1

Trillion

Dollars

Municipal securities

Agency and GSE

Treasury

Other Assets

Open market paper

Time and savingsdeposits

Repos

Figure 23: US Money market mutual fund assets (Source: Federal Reserve,

Flow of Funds) 38


40/73

USHouseholds

USBorrowers

USBankingSector

EuropeanGlobal

Banks

border

Wholesale

funding market

Shadow banking

system

Subprime safeliquid claims

Figure 24: Expanding lending capacity of European banks draw USD fundingto finance increased USD lending through shadow banking system

39


41/73

USD Billion Total BondHoldings Treasury Agency Corporate CorporateMBSTotal 6,642 2,194 1,413 3,035 594

Advanced 3,508 963 508 2,037 350Offshore 762 85 111 566 204

Emerging/developing 2,373 1,147 794 432 40

China 894 477 387 29 9Japan 976 622 231 123 17

Cayman Islands 461 29 56 376 157United Kingdom 500 48 28 424 90

Luxembourg 469 56 42 371 39

Belgium 372 15 33 323 19Ireland 261 16 30 215 33

Switzerland 155 40 18 97 20

Germany 166 46 15 105 33Netherlands 136 17 24 96 32

France 90 17 11 62 31

Figure 25: Foreign holding of US bonds in mid 2007 (Source: US Treasuryand Milesi-Ferretti (2009), G20 Mumbai Volume)

40


42/73

Gross Positions versus Net Positions

Largegross positionscreated by European banks impact on USfinancialconditions.

But net positions (current account imbalances) are small since assetsand liabilities net out.

Eurzone has near-balanced current account

UK has current account deficit Borio and Disyatat (BIS working paper, 2011)

Focusing on Global Savings Glut (net positions) misses the Global

Banking Glut (gross positions)

41


43/73

Why did European banks expand so much?Two candidate explanations:

Basel II and EU Capital Adequacy Directive (CAD) allowed Europeanbanks to expand assets without incurring rising risk-weighted assets

Advent of Euro opened up cross-border banking market within theeurozone

[Shin (2011) Mapping Global Liquidity, IMF Mundell-Fleming Lecture]

42


44/73

Implications for Current Conjuncture in Europe

Europe has atwin crisis, combiningbanking crisiswithsovereign debtcrisis

Emerging economy crises of 1990s were twin crises, combiningbanking crisiswithcurrency crisis

Deleveraging by European banks will impact not only eurozone, but also

US shadow banking system Capital flows to emerging economies (see below) Emerging Europe, especially

43


45/73

Landscape of Global Banking

Borrowers

in A

Borrowers

in B

Borrowers

inC

Banks

in A

Banks

in B

Banks

in C

Global

BanksWholesale

Funding

Market

44


46/73

Borrowers

in A

Borrowers

in B

Borrowers

inC

Banks

in A

Banks

in B

Banks

in C

Global

BanksWholesale

Funding

Market

45


47/73

100

0

50

100

150

200

250

300

350

400

450

500

Mar.1

999

Mar.2

000

Mar.2

001

Mar.2

002

Mar.2

003

Mar.2

004

Mar.2

005

Mar.2

006

Mar.2

007

Mar.2

008

Mar.2

009

Mar.2

01

0

Ireland

Spain

Turkey

Australia

South Korea

Chile

Brazil

Egypt

South Africa

Figure 26: External claims (loans and deposits) of BIS reporting banks oncounterparties listed on right (Source: BIS locational banking statistics

Table 7A) 46


48/73

100

0

50

100

150

200

250

300

350

Mar.1

999

Mar.2

000

Mar.2

001

Mar.2

002

Mar.2

003

Mar.2

004

Mar.2

005

Mar.2

006

Mar.2

007

Mar.2

008

Mar.2

009

Mar.2

01

0

Australia

South Korea

Indonesia

Malaysia

Thailand

Figure 27: External claims (loans and deposits) of BIS reporting banks oncounterparties listed on right (Source: BIS locational banking statistics

Table 7A) 47


49/73

100

0

100

200

300

400

500

600

700

800

Mar.1

999

Mar.2

000

Mar.2

001

Mar.2

002

Mar.2

003

Mar.2

004

Mar.2

005

Mar.2

006

Mar.2

007

Mar.2

008

Mar.2

009

Mar.2

01

0

Slovakia

Poland

Ireland

Spain

Turkey

Figure 28: External claims (loans and deposits) of BIS reporting banks oncounterparties listed on right (Source: BIS locational banking statisticsTable 7A) 48


50/73

100

0

500

1,000

1,500

2,000

2,500

3,000

Mar.1

99

9

Mar.2

00

0

Mar.2

00

1

Mar.2

00

2

Mar.2

00

3

Mar.2

00

4

Mar.2

00

5

Mar.2

00

6

Mar.2

00

7

Mar.2

00

8

Mar.2

00

9

Mar.2

01

0

Latvia

Lithuania

Estonia

Iceland

Figure 29: External claims (loans and deposits) of BIS reporting banks oncounterparties listed on right (Source: BIS locational banking statisticsTable 7A) 49


51/73

0

100

200

300

400

500

600

700

800

Greece Ireland Spain Australia South Korea

Billion

Dollars

Other

Japan

United States

United Kingdom

Netherlands

France

Germany

Figure 30: International claims of BIS reporting banks on counterpartiesin countries listed on right (Dec 2010) (Source: BIS consolidated banking

statistics Table 9D)

50


52/73

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Greece Ireland Spain Australia SouthKorea

Other

Japan

United States

United Kingdom

Netherlands

France

Germany

Figure 31: International claims (by percent) of BIS reporting bankson counterparties in countries listed on right (Dec 2010) (Source: BIS

consolidated banking statistics Table 9D)

51


53/73

Claims of European Banks on Counterparties in Korea

0

50

100

150

200

250

2005-Q2

2005-Q4

2006-Q2

2006-Q4

2007-Q2

2007-Q4

2008-Q2

2008-Q4

2009-Q2

2009-Q4

201

0-Q2

201

0-Q4

BillionDollars

Other European BISReporting countries

Switzerland

United Kingdom

France

Germany

Figure 32: International claims of European BIS-reporting banks oncounterparties in Korea (Source: BIS consolidated banking statistics Table9D)

52


54/73

Credit Supply

Notation for balance sheet of bank

C

E

f1r1L

Bank

53


55/73

Credit Supply as Flip Side of Credit Risk Model

Vasicek (2002) model, backbone of Basel capital requirements.1

Borrower repays the loan when 0, where

= 1 () +

+p

1

()c.d.f. of standard normal, and{} independent standard normals

Pr ( 0) = Pr

+p

1 1 ()

= 1 ()= 1http://www.moodyskmv.com/conf04/pdf/papers/dist loan port val.pdf

54


56/73

Bank diversifi

es away idiosyncractic risk

Conditional on , defaults are independent.

Keep fi

xed but diversify: increase number of borrowers, reduce face valueof individual loans

In the limit, realized value of assets is function of only

() (1 + ) Pr ( 0|)= (1 + ) Pr

+

p1 1 () |

= (1 + ) 1()1

55


57/73

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

3

6

9

12

15

Dens

ityover

assetre

alizations

rho=0.03

rho=0.07

rho=0.15

Asset realization densities for three values of [= 01, (1 +) = 1]

1


58/73

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 10

4

8

12

16

20

Densityover

asset

realizations

epsilon=0.05

epsilon=0.10

epsilon=0.15

Asset realization densities for three values of [= 02, (1 +) = 1]

2


59/73

c.d.f. of

() = Pr ( )= Pr

1 ()=

1 ()

=

1

1 () +

p1 1

(1 + )

Common risk factordetermines shape of the density, with largerimplyingfatter tail.

Value-at-Risk (VaR) rule: keep enough equity to limit insolvencyprobability to 0

56


60/73

Private credit determined from

Pr ( (1 + ) ) =

1()+

11

(1+)(1+)

!=

Notional liabilitiesNotional assets

=(1 + ) (1 + )

= 1 () 1 ()

1

(1)

where

()

1()1()1

57


61/73

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.00.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Figure 33: Plot of notional debt to assets ratio (). This chartplots as a function of with = 0001. Dark line is when = 001.

Light line is when = 0005.

58


62/73

Supply of CreditCredit supplyand demand for funding is obtained from (1) and balancesheet identity=+

=

1 1+1+ =

1+1+

1 1

Aggregation holds due to proportionality

Leverage = 11 1+1+

Risk premium is well-defined

Risk premium = (1 ) (1 + ) 1

59


63/73

Risk premium is decreasing as assets expand by sliding down the creditdemand curve. Lending standards are eroded in this sense.

0 f

E

111

1/

Credit

Supply

11

f

rC

r

Supply of credit

60


64/73

Double-decker model of Global Liquidity

C

RE

M

GE

i1f1r1

L

L

Regional Bank Global Bank

Figure 34: Bruno and Shin (2011) Capital Flows, Cross-Border Banking

and Global Liquidity

61


65/73


66/73

Global, Regional and Idiosyncratic Risk Factors

1 () +

+

p1

=p

+p

1 Regional bank defaults when

1

((1 + ) ) =

11

() +p

1

1

()

Or when 0

1 ()p1 1 ()=

p +

p (1 ) 1 ()

p1 1 ()

63

f f f


67/73

Asset realization is deterministic function of global risk factor

() = (1 + ) Pr ( 0|)

= (1 + ) Pr 1()+

11()

(1) q 1= (1 + )

q 1

1()+

11()

(1)

Quantiles follow from the c.d.f. of ().

() = Pr ( () )= Pr

1 ()

=

1 ()

64


68/73

N i l li bili i (1 )


69/73

Notional liabilities

Notional assets = (1 + )

(1 + )

=

1()1()

11()

(1)

( )

Cross-border loan supply

=

1

1+

1+

66


70/73

0 L

1/

11

i

fLS

fLD

f

11 r

i

EG

111

Figure 36: Equilibrium cross-border lending

67

Capital Flows and Domestic Credit


71/73

Capital Flows and Domestic Credit

Market clearing for

1+1+ 1 1=

1 1+1+

Private credit

= + 1 1+1+

Total privatecredit =

Aggregate bank capital (regional + global)

1 spread regionalleverage

globalleverage

68

Risk premium in recipient economy


72/73

Risk premium in recipient economy

(1 ) (1 + ) 1

Equilibrium stock of cross-border lending

=+

1+1+

1 1+1+

Total cross-border lending

=Global and weighted regional bank capital

1 spread regionalleverage

globalleverage

69

Comparative Statics


73/73

Comparative Statics

Banking sector capital flows:

increase with (bank ROE)

increase with bank leverage (fall with VIX)

increase in change in bank leverage (fall with VIX)

fall with interaction between ROE and VIX

(Explored empirically in Bruno and Shin (2011))

70

princeton initiative

Documents