AEA 2018 Finance in Macro Models

AEA Continuing Education
Program
DSGE Models and the

Role of Finance
Thomas Philippon,
New York University
January 7-9, 2018

Introduction Contracts Intermediation Dynamics Greece Regulation FinTech References
Finance in Macro Models
Thomas Philippon
NYU, NBER, CEPR
January 2018, AEA Continuing Education, Philadelphia

Does finance matter?
• Not much?
“Where enterprise leads finance follows."
Joan Robinson (1952)
• A lot?
“..that financial markets contribute to economic
growth is a proposition too obvious for serious
discussion.”
Merton Miller (1988)
Some (Contradictory) Views about Finance
• Financial Regulation
- Deregulation was responsible for the crisis; regulations should
be tightened
- Finance is over-regulated and credit supply is impaired
• Finance and Growth
- Financial development is important for growth
- The last useful financial innovation was the ATM
• Finance and Inequality
- Lack of access to finance creates poverty traps
- Finance increases income inequality
• Fin-tech and Bitcoins
Financial Contracts
Fundamentals of Financial
Contracting
Three Financial Frictions
• Debt overhang, Myers (1977)
• Monitoring & Moral hazard
• Screening & Adverse Selection

Basic Setup
Risk neutral investors, two periods t = 0, 1, limited liability

• Legacy assets & liabilities
- liquidity m, assets a with p ≡ Pr a = AH , (long term) debt D
! "
• New Investment
- Invest k to generate additional v = V with prob. q, otherwise
v =0
• Total income at t = 1
y = a+v
• Borrowing b to close funding gap
b = k −m
Debt Overhang
Total Senior Debt Equity
p AH D AH-D
1-p AL AL 0
Key Friction: Difficult to Renegotiate Senior Debt

• Free riding among dispersed creditors
• Risk of runs
How New Projects Are Evaluated
Total Senior Debt Junior Debt Equity
p AH+ vH D Rk AH-D+ vH-Rk

-k
1-p AL+ vL AL+ vL 0 0
New lenders break even Shareholder Value: vH-Rk

R=1/p 1. Risk shifting: vL irrelevant
2. Underinvestment: R>1
Underinvestment in Safe Projects
p AH+ v D k/p AH-D+ v-k/p

-k
1-p AL+ v AL+ v 0 0
• Myers (1977)
• Philippon-Schnabl (2009): (p,v) unknown to government
• Philippon (2010): bailouts in open economy
Philippon-Schnabl (2009)
v
Invest
Debt
Overhang
0 p
1
Risk Shifting
p AH +v D k/p AH-D+ v-k/p

-k
1-p AL -v AL -v 0 0
• NPV = (2p-1)v-k can be negative, e.g. when p<0.5

• But if v>k/p , shareholders like the project
Reluctance to Sell Assets
Total Senior Debt Equity
p (1-z)AH+m D (1-z)AH-D+m
-m
1-p (1-z)AL+m (1-z)AL+m 0
• Buyers: m=zE[A] Papers

• Equity: AH-D -z(1-p)(AH-AL) • Philippon-Schnabl (2009)
• Landier-Ueda (2009)
• Diamond-Rajan (2010)
Moral Hazard: Endogenous q

• Ignore existing assets and liabilities (a = D = 0): q is q H , or
q L with private benefit ψ . Incentive constraint with y e
payment to entrepreneur
ψ
y e > ymin
e
≡
qH − qL
• Therefore pledgeable income is q H (V − ymin

e )
# $
• Financing constraint is q H V − q H ψ−q L > k − m and
investment condition is
%
ψ
&
m > m̂ ≡ k − q H V −
qH − qL
- Existing assets can be used as collateral to increase pledgeable

income
Adverse Selection
• Quality of existing assets privately known: bad types a = Ab ,
good types a = Ag
- Good types safe, bad types risky
k
Ab < D < Ag −
q
• Issuance game & security design

- Myers and Majluf (1984), Nachman and Noe (1994)
• Pooling rate given perceived quality z (fraction of good types)
1
r (z) =
q + z (1 − q)
• Good types refuses to invest if
rk > qV
Time 1: Credit Market
1
q
Rate r
Limit Rate For

qV Strong Types
k
Pooling Rate
1
0 zI 1
Perceived Quality z
Time 1: Credit Market Equilibrium
1 Equilibrium Rate
q
Rate r
Limit Rate For

qV Strong Types
k
1
0 zI 1
Perceived Quality z
Taking Stock
• Debt overhang: legacy debt can limit/distort investment

• Moral Hazard & Adverse Selection: inside liquidity is
important
- Role for intermediation: monitoring (lower private benefit ψ )
and screening (acquire information about types)
• Collateralized borrowing: pledging existing assets to fund new
investment
- financial sophistication determines what can be collateralized
(e.g., patents in the US)
- adverse selection can limit pledgeability of existing assets
• How much are agents willing to pay for intermediation?
Intermediation
Measuring Financial Intermediation

Philippon (2015)
Price and Quantity of Financial Intermediation
• Price
u = r +ψ
- User cost of borrowers, expected return of savers

- Unit cost of intermediation: ψ
• Quantity
- Household finance bc
- Corporate finance bk , ek
- Liquidity services m
Traditional Banking
Modern Finance
Income Share of Finance Industry

.08
.06
.04
.02
1860 1880 1900 1920 1940 1960 1980 2000

year
VA fire, Hist. WN fire, Hist.

WN fin, NIPA VA fin, NIPA
Outstanding Credit / GDP

1
.5
0
1920 1930 1940 1950 1960 1970 1980 1990 2000 2010
year
Farm, Hist Farm, FoF

Household, Hist Household, FoF
Nonfarm Business, Hist Nonfarm Business, FoF
Neoclassical Growth with Intermediation
• Households’ consumption and liquidity services: u (ct , mt ), life

cycle borrowing/lending
• Costs of intermediation services
- ψm for liquid assets, ψc for households credit, ψk for firm
credit
• Non financial businesses
- user cost
1−α
kα = ,
r + δ + ψk
• Financial intermediation income
ytf = ψc bc,t + ψm mt + ψk kt ,
Aggregation
• Assumption: Constant Relative Costs: ψi ,t = µi ψt , with

the normalization µc = 1.
• Quantity of assets
qt ≡ bc,t + µm mt + µk kt ,
• Estimate the µ ′ s with micro data on rates, returns and

issuance costs
Intermediated Assets & Services

2
1.5
1
.5
0
1880 1900 1920 1940 1960 1980 2000

year
Business Credit & Equity Household Credit

Liquidity Services M&As
Financial Intermediation (in the U.S.)

.08
4
Intermediated Assets/GDP
.06
3
Share of GDP
.04
2
.02
1
1880 1900 1920 1940 1960 1980 2000 2020
year...
Share of GDP Intermediated Assets/GDP
Source: Philippon (AER, 2015)

Unit Cost in the U.S.

Raw Unit Costs
.03
.025
.02
.015
.01
.005
0
1880 1900 1920 1940 1960 1980 2000 2020

time
2012 Data New Data
Source: Philippon (AER, 2015)

Unit Costs, Global Comparison
Source: Bazot (2013)

Constant Returns to Scale

Evidence of Constant Returns to Scale

3.5
3
2.5
2
1.5
1
1950 1960 1970

year
Unit Cost Real Assets Real GDP PC
Notes: Series normalized to one in 1950

Quality Adjustments with Heterogenous Borrowers

Corporate Finance
• Heterogenous inside equity (retained earnings) x. Monitoring

µ to prevent cash flow diversion.
- IC constraint, etc.
• Aggregate monitoring
µ̄t = µh + (1 + r )(xh − xl ) st ,
kl
where s ≡ kl +kh
• Income from corporate finance intermediation
f
yk,t = ϕt kt + ζt µ̄t .
where ϕ is an asset management fee

Household Finance
• Heterogenous labor endowment η ∼ F (.)

• Credit: marginal cost ϕ , fixed cost κ w
- Participation for η > η̂
- Household debt market
b̄c 1
' ## $ $
= λ − (1 − ϕ )−1 η − κ dF (η )
w 2+r η >η̂
- Income
f
yc,t = ϕ b̄c,t + κt w (1 − F (η̂t )) .
Calibration
Rate Deprec. Growth Labor Sh. CRRA

r = 0.05 δ = 0.1 γ = 0.02 α = 0.7 ρ =1
High cash Low cash Asset Mgt Fee

xh = 0.62 xl = 0.1 ϕ = 0.01
Calibration
Moments
B Cost share HH D Frac Top 20 Liquid

µ̄k
b̄k /y ϕ +ζ b̄k
s b̄c /y 1 − F (η̂ ) ... m/y y f /y
data 0.806 0.0205 0.20 0.73 0.84 0.468 0.71 0.0585
model 0.811 0.0208 0.199 0.73 0.84 0.468 0.71 0.0580
Implied Parameters
Monit Firms Slope Ineq. Fix Liq D Liq S

r kh
ζ= 3.3 k∗ = 0.622 λ = 2.07 H = 0.875 κ = 0.023 ν = 0.0181 ψm = 0.019
Calibrated Model
A. CORP. FINANCE INCOME SHARE D. PARTICIPATION
0.03 0.9
0.028 0.8
0.026
0.7
0.024
0.022 0.6
0.02 0.5
0.1 0.2 0.3 0.4 1 1.5 2 2.5
B. CORPORATE DEBT / GDP E. HOUSEHOLD DEBT / GDP

1 0.75
0.7
0.9
0.65
0.8 0.6
0.55
0.7
0.1 0.2 0.3 0.4 1 1.5 2 2.5
C. CORPORATE ADJUSTMENT F. HH ADJUST. FACTOR

1.3
1.2 1
1.1
1 0.95
0.9
0.9
0.1 0.2 0.3 0.4 1 1.5 2 2.5
Share of Low Cash Firms Fixed Cost Index
Quality-Adjusted Quantity of Intermediation

5
4
3
2
1
1880 1900 1920 1940 1960 1980 2000

year
Assets Firm Adj. Firm & HH. Adj.

Adjusted Unit Cost

.03
.025
.02
.015
.01
.005
0
1880 1900 1920 1940 1960 1980 2000

year
Raw Firm Adj. Firm & HH Adj.

Dynamics
Macro-Finance Dynamics
Asset Pricing, Cochrane (2017)
• Equity premium 6%
- Just as likely to be 1% (in good times) or 11% (in bad times)
• Pricing kernel
% &− γ
Ct+1
Λt+1 = β Ξt+1
Ct
- Aggregate consumption growth not volatile enough, γ = 25 is

just silly.
• Ξt+1 needed. Cochrane (2017) reviews 10 models, including:
Habits, LR risk with EZ preferences, disasters, ambiguity
aversion, behavioral biases
Epstein-Zin Recursive Utility

• An important class of recursive utility function uses a CES
aggregator (ε ) where risk aversion is γ
1
% # ( )$ 1− ε & 1−ε
1−ε 1−γ 1−γ
Ut = (1 − β ) C + β Et Ut+1
• This leads to a pricing kernel

⎛ ⎞ε −γ
% &− ε
Ct+1 Ut+1
Λt+1 = β
⎜ ⎟
Ct
⎝# ( )$ 1−1 γ ⎠
1−γ
Et Ut+1
- Typical calibration of long run risk model à la Bansal and

Yaron (2002) has ε close to 1 and γ > 1
- Household worried not only about bad news to consumption
but also to future utility
- See also Cochrane’s discussion of disaster risk
Friction-less Benchmark: Q-Theory
• Firm value 0 1
∞
Vt = Et ∑ Λt,t+j Πt+j
j=0
• Capital accumulation
Kt+1 = (1 − δt ) Kt + It
• Dividends & Adjustment Costs

% &2
ϕk It
Πt = Rk,t Kt − Pk,t It − Pk,t Kt − δt
2 Kt
Friction-less Benchmark: Q-Theory

• Recursive
Vt (Kt ) = max Πt + Et [Λt+1 Vt+1 (Kt+1 )]

It
Vt It
• Define Vt ≡ Kt and net investment xt ≡ Kt −δ:
ϕk
Vt = max Rk,t −Pk,t (xt + δt )− Pk,t x 2 +(1 + xt ) Et [Λt+1 Vt+1 ]
x 2
• FOC Pk,t (1 + ϕk xt ) = Et [Λt+1 Vt+1 ] can be written as
1 # k $
xt = Qt − 1
ϕk
where
Et [Λt+1 Vt+1 ] Et [Λt+1 Vt+1 ]
Qtk ≡ =
Ptk Ptk Kt+1
Tobin’s q and Investment
.13
5
.12
4
.11
3
I/K
Q
.1
2
.09
1
.08
1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
I/K Usual Q
Q-Theory: Bonds vs Stocks, Philippon (2009)

Et [Λt+1 Vt+1
e
] Et [Λt+1 Vt+1
b
]
• Let et = k and bt = k for equity and bonds
Pt Kt+1 Pt Kt+1
qt = et + bt
• Suppose you observe bt : can you re-construct qt ? Leland

(1994) and Leland (1998): debt with coupon c and average
maturity 1/φ . One unit of principal at t repays
t +1 t +2 ... τ ...
c +φ (1 − φ ) (c + φ ) ... (1 − φ )τ −t−1 (c + φ ) ...
• Risk “neutral” debt pricing, with book leverage l , in state ω :
1
E π min (c + φ ) l + (1 − φ ) b ω ′ ; V ω ′ |ω
2 3 ! " ! "4 5
b (ω ) =
1 + r (ω )
Bond q and Investment, Philippon (2009)

Bond q and Investment, Philippon (2009)
.13
1.8
1.7
.12
1.6
.11
1.5
Bond Q
I/K
1.4
.1
1.3
.09
1.2
.08
1.1
1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
I/K Bond Q
Models with Financial Frictions

• Financial Accelerator
- Financial frictions can amplify and propagate macroeconomic
shocks
• First generation: Backward loop, Bernanke and Gertler (1989);
Bernanke et al. (1999)
- Based on simple moral hazard model: dynamics of cash on
hand
- Bad shock: mt ↘=⇒ Kt +1 ↘=⇒ mt +1 ↘
• Second generation: Forward loop, Kiyotaki and Moore (1997)
- Collateralized borrowing by “specialists”
Bt ≤ θt Qt Kt +1
- Second best use of capital and fire sales: if constraint binds, K

needs to be operated by non-specialists, and Q goes down
- Bad shock: Qt ↘=⇒ Kt +1 ↘=⇒ Qt +1 ↘
- But Q is forward looking Qt ↘⇐= Et [Qt +1 ] ↘
Intermediary-Driven Crisis
• The 2008-2009 crisis centered around financial intermediaries

• New models focused on financial intermediaries, such as He
and Krishnamurthy (2012) and Gertler and Kiyotaki (2013).
- Banker with survival probability σ maximizes
Vt = Et [Λt +1 ((1 − σ )nt +1 + σ Vt +1 )]
subject to budget constraint
Qt kt = dt + nt
and incentive/collateral constraint
θ Qt kt ≤ Vt
• Promised rate on deposits is Rt so
nt = (Zt + Qt ) kt−1 − Rt dt−1

Discussion
• What we get from Gertler and Kiyotaki (2013)

- value function is linear in net worth Vt = νt nt , so aggregation
is straightforward
- key choice for intermediary is leverage Qnt tkt
- recessions driven by liquidity and intermediation risk
• Notes
- What is missing from Fig 1 in GG?
- What is the link between Ξ, time varying risk aversion, and the
Gilchrist-Zakrajsek EBP (excess bond premium)?
A Prototypical Crisis Model
• The basic elements of a modern financial crisis model are:

- Output depends on asset price
Y = Y (Q) ≤ Y ∗
- Asset price is forward looking
E [Z ′ + Q ′ ]
Q=
Rk
- Risk spread increases with intermediary distress
s = Rk − r
- Monetary policy is constrained
r ≥0
Household Finance
• Most macro finance of the 1990’s models had representative

agent. Action was on firm side
• Household balance sheet were important in the crisis. Two
classes of models
- Two agents with different β Eggertsson and Krugman (2012)
- Bewley-style models Kaplan and Violante (2011)
Greece
Greece: A Macro-Financial Tragedy

Three Financial Frictions
• In 2007, Greek GDP per capita was around 22,600 euros and
the unemployment rate was 8.4%.
• In 2014, Greek GDP per capita was around 17,000 euros and
the unemployment rate was 26.6%
• What happened?
Benchmarking: the Comparison Group
• Sudden Stops
- Combination of capital flow reversal & large drop in domestic
output
- Extend Calvo et al (2006), Korinek & Mendoza (2013)
- 49 sudden stops
• Sovereign Defaults
- from Gourinchas & Obstfeld (2012) based on literature
- default on domestic or external debt
- 65 default episodes
• Lending booms/busts
- defined as in Gourinchas et al (2001)
- deviation of credit/output from trend
- 114 boom/busts
The Incidence of Crises
Sudden Stop Defaults Credit Booms ‘Trifecta’ #
AE 13 Greece 18 Greece 22
EM 36 64 96 9 57
Total 49 65 114 10 79
Benchmarking 1: GDP Relative to All Sudden Stops

Collapse of 25%
Benchmarking 1: Aggregate Domestic Investment/Output

Collapse of 50%
Benchmarking 2: Other Crises

Sovereign Default? Credit Bust?... Trifecta

Benchmarking 3: Compared to EM Floaters & Peggers

Benchmarking 4: Endogenous Peg?

Empirical Lessons
1. Greek crisis significantly more severe persistent and backloaded

than typical sudden stop
2. Greek crisis significantly more severe persistent and backloaded
than ‘Trifecta’ episodes
3. Greek crisis more severe than for peggers (even Estonia or
Latvia)
4. Collapse in aggregate investment unprecedented in its
persistence and magnitude
Model
• Small Open Economy in a currency union (r , π F exogenous)

• Standard NK DSGE à la Galì (2011) with financial frictions
- Government (B g , T , G , r g )
- Banks (V , r d )
- Households (B h , C , r h )
- Firms (I , K , r k )
• Various shocks
ζt# = ρ # ζt−1
#
+ σ # εt#
Government
• Budget constraint
g
Btg Bt−1
g + τt Y t = G t + T t +
Rt ΠHt
• Fiscal rule (spending and social transfers)
gt = Fl gt−1 − Fn nt − Fr rtg − Fb b gt + ζtspend
• Tax rate
τt = τ̄ + ζttax
• Government funding cost
rtg = rt + dtg
Bg # g ( ) $
dtg = d¯g h
bt − Et [yt+1 ] − Et πt+1 + ζtdg
Y
Households
6 ! "1−γ ! "1+φ 7 0 1 εh
∞ ε −1 ε −1
Cit Ni 1 i hε 1 i hε εh −1
i
U = E0 ∑ βit − t ; Cit ≡ (1 − ϖ ) εh
CH,th +ϖ εh
CF ,t h
t=0 1−γ 1+ϕ
• Borrowers, mass χ , Bth ≤ B̄th ,
PH,t Bth !
Pt Cbt = (1 − τt ) Wt Ntb + − 1 − dtp PH,t−1 Bt−1
h
+ PH,t Ttb
"
Rt h
dtp = −d¯y yt + d¯b bth + ζtdef

h
b̄th = ψbh b̄t−1 − ξ bh rtd + ζtbh
• Savers, β > βb (mass 1 − χ ),
Pt Cst = (1 − τt ) Wt Nts + R̃t PH,t−1 St−1 − PH,t St + PH,t Tts

Non-Financial Firms
• Capital-producing firms:
- Q theory
• Goods-producing firms:
- Convert capital and labor into goods.
- Cobb-Douglas with constant TFP.
- Financing friction: pay part of wage bill in advance.
Intra-period loan with funding cost r k .
Price and Wage Rigidity
• Wage-calvo process yields a Phillips curve for wages
πtw = β Et πt+1
w
− λw (wt − γ ct − ϕ nt ) + ζtw
• Price-calvo process yields a Phillips curve for domestic prices
πth = β Et πt+1
h
+ λp mct + ζtπ h ,
where mct is log real marginal cost in terms of domestic goods.

• ζtw : wage markup shock, ζtπ h : domestic price markup shock
Banks
• Domestic deposits and foreign loans

• Lend to households, firms and government
• Subject to capital requirement
Btk Bth
% &
Vt ≥ κ +
Rtk Rth
where Vt is franchise value.

• No capital requirement for sovereign exposure
• Bank funding costs
2 p 5
rtd = rt + ζtrd + ξ d LEt dt+1
Summary of Funding Costs

• Key equations
- Funding cost: banks –> households & firms
rtk = rtd
- Banks: sudden stop and capital loss
rtd rt + ζtrd + ξ d LEt dtp+1

2 5
=
dtp = −d¯y yt + d¯b bt−1 + ζ def
t
- Government
rtg = rt + dtg
Bg # g ( ) $
dtg = d¯g bt − Et [yt +1 ] − Et πth+1 + ζtdg
Y
- Households
rth = rtd + Et dtp+1
2 5
Doom Loops
No direct doom loop, but indirect GE feedback loops:

• Sovereign risk shock ζtdg :
- Government funding costs increase → Government raises taxes
and reduces expenditure → Output declines → Expected costs
of default on private-sector loans increase → Funding costs for
private sector increase and investment drops.
• Sudden stop ζtrd :
- Funding costs for private sector increase → Output and
investment drop → Fiscal revenues drop → Expected costs of
default on sovereign loans increase → Government funding
costs increase.
Impulse Response: Sovereign Risk Shock

Sov Risk
Output Consumption Investment Employment

0.05 0.1 0.1 0.05
Saver
0.05 Borrower 0.05
0 0 0 0
-0.05 -0.05
-0.05 -0.1 -0.1 -0.05

5 10 15 5 10 15 5 10 15 5 10 15
-3 Funding Cost Sovereign Yield NPL/Total Loans Inflation

×10
2 0.05 0.2 0.01
rk
1 zrd 0.1 0.005
0 0 0 0
-1 -0.1 -0.005
-2 -0.05 -0.2 -0.01

5 10 15 5 10 15 5 10 15 5 10 15
Govt Finance Govt Debt Household Debt Current Account/GDP

0.05 0.1 0.01 0.01
Rev.
Spend. 0.05 0.005 0.005
0 0 0 0
-0.05 -0.005 -0.005
-0.05 -0.1 -0.01 -0.01

5 10 15 5 10 15 5 10 15 5 10 15
Years Years Years Years
Impulse Response: Sudden Stop

Sudden Stop
Output Consumption Investment Employment

0.1 0.1 0.5 0.1
Saver
0.05 0.05 Borrower 0.05
0 0 0 0
-0.05 -0.05 -0.05
-0.1 -0.1 -0.5 -0.1

5 10 15 5 10 15 5 10 15 5 10 15
Funding Cost Sovereign Yield NPL/Total Loans Inflation

0.02 0.01 0.5 0.01
rk
0.01 zrd 0.005
0 0.005 0 0
-0.01 -0.005
-0.02 0 -0.5 -0.01

5 10 15 5 10 15 5 10 15 5 10 15
Govt Finance Govt Debt Household Debt Current Account/GDP

0.1 0.04 0 0.02
Rev.
0.05 Spend. 0.03 0.01
0 0.02 -0.05 0
-0.05 0.01 -0.01
-0.1 0 -0.1 -0.02

5 10 15 5 10 15 5 10 15 5 10 15
Years Years Years Years
Bayesian Estimation of the Model

• Standard techniques (Herbst & Schorfheide (2015))
• Period: 1999 to 2015
• Calibrate steady state parameters
• Estimate dynamic parameters
Observable Description Shock Shock Description

Gt + Tt Government spending ζtspend Govt. spending shock
τt Yt Government revenues ζttax Tax rate shock
Rtg Greek government spread over EZ average ζtdg Sovereign risk shock
Rtk SME spread over EZ average ζtr Funding cost shock
dtp ζ def
! "
exp Non-performing loans/total loans, def = npl Private default shock
Πt Greece CPI - EZ CPI ζ πh PPI cost push shock
Bth Household debt ζtbh Household credit shock
Πw
t Greek Wage Inflation - EZ Wage Inflation ζw Wage inflation shock
Table: Observables and Shocks

Calibrated Parameters-I
Parameter Description Value
β Discount Factor 0.97
α Capital Share 1/3
εh Elasticity between H and F 1
εf Elasticity between exports 1
ϕ Inverse labor supply elasticity 1
γ Risk Aversion 1
ϑ Price Stickiness 0.5
ε Elasticity of Substitution Goods 6
ϑw Wage Stickiness 0.5
εw Elasticity of Substitution Labor 6
εr Elasticity of R to NFA 0.0001
ϕk Adjustment Cost 1
δ Depreciation 0.07
FC Fixed cost of production, 10% of Y 0.0955

Calibrated Parameters-II
ϖ Openness (Martin and Philippon (2014)) 0.3
χ Fraction of Impatient (Martin and Philippon (2014)) 0.65
∆ Annual lending spread of 2% 1.02

B̄ h Household debt to GDP of 50% 0.5
Y
Bg Government debt to GDP of 120% 1.2
Y
G Government consumption to GDP of 20% 0.2
Y
T Public social expenditure to GDP of 20% 0.2
Y
d¯h Steady state default rate for Households 5.4%
d̄ k Steady state default rate for Corporates 5.4%

Bk Corporate debt to GDP of 50% 0.5
Y
ψsk Working Capital Constraint 1
τ Tax rate, budget balance in SS 0.436
L Leverage scaling 1
Calibrated Parameters-III
Fb Elasticity of govt. spending to public debt 0.05
Fn Elasticity of govt. spending to employment 0.025
Fr Elasticity of govt. spending to the int. rate 0.5
Fl Persistence of govt. spending 0.75

Data Inputs
Govt. Revenue Govt. Spending Govt. Yield Private Funding Cost

0.3 0.03
0.1 0.15
0.2
0.05 0.02
0.1 0.1
0
0.01
-0.05 0 0.05
-0.1 0
-0.1 0
2000 2005 2010 2015 2000 2005 2010 2015 2000 2005 2010 2015 2000 2005 2010 2015
NPL/Total Loans (Obs.) dlog GDP Deflator (Obs.) Household Debt Wage Inflation (Obs.)
0.02 0.02
1 1
0
0.5 0
0.5 -0.02
0
-0.02 -0.04
-0.5 0 -0.06
-0.04
2000 2005 2010 2015 2000 2005 2010 2015 2000 2005 2010 2015 2000 2005 2010 2015
Fit of the Model

Output Investment
0
0.1
-0.2
0
-0.4
-0.1 -0.6
2000 2005 2010 2015 2000 2005 2010 2015
PPI Inflation Current Account/GDP

0.02
0 0
-0.02
-0.05
Data
-0.04
Model
-0.1
-0.06
2000 2005 2010 2015 2000 2005 2010 2015
Wage Inflation
0.02
0
-0.02
-0.04 Data
-0.06 Model
2000 2005 2010 2015

Decomposition of Output and Investment
GDP
0.2
0.15
Investment
0.1 Spending 0.6
Tax 0.4
0.05
Credit Demand
0
Sudden Stop 0.2
Priv. Def. 0
-0.05 Sov. Risk

-0.2
Markup
-0.1 Wage Markup -0.4
-0.15 -0.6
-0.8
-0.2
-1
2000 2005 2010 2015

2000 2005 2010 2015
Years
Years
Decomposition of Sovereign Debt

Govt. Debt
0.8
0.6
Spending
Tax
Credit Demand
0.4
Sudden Stop
Priv. Def.
Sov. Risk
0.2
Markup
Wage Markup
0
-0.2
2000 2005 2010 2015

Decomposition of Private Default

Private Default
0.5
Spending
Tax
Credit Demand
0
Sudden Stop
Priv. Def.
Sov. Risk
-0.5
Markup
Wage Markup
-1
2000 2005 2010 2015

Decomposition of Government Spending

Govt. Spending
0.4
0.3
Spending
0.2 Tax
Credit Demand
Sudden Stop
0.1
Priv. Def.
Sov. Risk
0 Markup
Wage Markup
-0.1
-0.2
2000 2005 2010 2015

Key Lessons
• Fiscal trajectory prior to 2009 unsustainable. Stimulates

output initially, but depresses it later on.
• First phase of the crisis (2009-2013)
- Sovereign risk
- Sudden stop
• Second phase of the crisis (2013-..)
- Non-performing loans
- Price markups.
4 Counterfactual Exercises
1. Low leverage (EME leverage)

2. Banking union
3. Fiscal discipline
4. Price flexibility (Latvia)
Counterfactual I: EME Leverage
Greece Typical EME Min Max

Credit / GDP 1.01 0.46 0.025 1.46
Sovereign Debt / GDP 1.38 0.343 0.063 0.68
Current Account /GDP -0.083 -0.039 -0.10 +0.17
Table: Leverage and Imbalances Before Sudden Stop
Notes: Average from t-6 to t-2 where t is sudden stop.
Counterfactual I: EME Leverage

GDP Investment
0.15
0
0.1
0.05 -0.2
0
-0.05 -0.4
-0.1 Data
-0.6 CFact
-0.15 Model
2000 2005 2010 2015 2000 2005 2010 2015
Current Account/GDP Govt. Spending

0.3
0.04
0.02 0.2
0
-0.02 0.1
-0.04
0
-0.06
-0.08 -0.1
2000 2005 2010 2015 2000 2005 2010 2015

Counterfactual II: Banking Union

GDP Investment
0.15 0.2
0.1
0
0.05
0 -0.2
-0.05 -0.4
-0.1 Data
-0.6 CFact
-0.15 Model
2000 2005 2010 2015 2000 2005 2010 2015

0.3
0.04
0.02
0.2
0
-0.02 0.1
-0.04
0
-0.06
-0.08
-0.1
2000 2005 2010 2015 2000 2005 2010 2015

Counterfactual III: No Discretionary Spending

GDP Investment
0.15 0
0.1
0.05 -0.2
0
-0.4
-0.05
-0.1 Data
-0.6 CFact
-0.15 Model
2000 2005 2010 2015 2000 2005 2010 2015

0.3
0.04
0.02 0.2
0
-0.02 0.1
-0.04
0
-0.06
-0.08 -0.1
2000 2005 2010 2015 2000 2005 2010 2015

Counterfactual V: Low Price Stickiness

GDP Investment
0.15 0
0.1
0.05 -0.2
0
-0.4
-0.05
-0.1 Data
-0.6 CFact
-0.15 Model
2000 2005 2010 2015 2000 2005 2010 2015

0.3
0.05
0.2
0 0.1
0
-0.05
-0.1
2000 2005 2010 2015 2000 2005 2010 2015

Greece, Conclusion: What Would Have Helped?
• What we can say

- Exposure Y+10%, I+15%
- Banking union Y+10%, I+30%
- Sound fiscal Y+15%, I+20%
- More flexible prices Y+15%, I+20%
• Open issues
- Uncertainty (political, EZ risk)?
- Early sovereign default?
- Devaluation?
Regulation
Systemic Risk
Systemic Risk (with Acharya, Pedersen, and Richardson)
• Need to avoid two pitfalls

- Too narrow: only leveraged, short-term wholesale-funded
banks are systemic
- Too broad: “Everything is systemic”
• Not much consensus beyond banking
- Are insurance companies systemic?
- Asset managers?
• Regulatory approach
- Size + Interconnectedness + Substitutability
• Our initial approach (2010)
- Systemic risk when aggregate capital shortfall
- Tax/regulate contribution to shortfall –> SES
Basic Model of Systemic Risk
• J financial firms, i = 1, ..J, two dates t = 0, 1, normalize r f = 0

• Time 0
ai = wi ,0 + bi
• Time 1, random gross return q
wi = qi ai − di
• Debt priced fairly
bi = E [min (qi ai , di )] .
Basic Model of Systemic Risk
• Firm i ’s problem
Ui = max E [max (qi ai − di ; 0) − wi ,0 − τi ]

di ,q̃i
• Regulator cares about aggregate externality:

J
max ∑ Ui + E
{τi }i i =1
s.t.
E ≡ −eE [1W −κ A (κ A − W )] .
Reduced Form Externality

Optimal Pigouvian tax
Proposition
In the simple model with exogenous externality, the efficient
outcome is obtained by a tax
τi = e × Pr (W < κ A) × MESi + τ0
where the systemic expected shortfall is defined by
MESi ≡ E [κ ai − wi | W < κ A]
• Look for MES at https://vlab.stern.nyu.edu

BAC MES
MES & SCAP

Measuring Leverage Not Easy

Liquidity & Maturity
Source: Hanson, Shleifer, Stein, Vishny, 2014

Bank Capital Regulations
• ? identify four constraints on bank equity

• $ equity per $1 of asset
- Risk based capital (RBC): kRBC × wi , capital requirement *
risk weight
- Supplementary leverage ratio (SLR): kSLR
- Post stress RBC: kRBC ,STRESS × wi + NLRi , RBC + net loss
rate
- Post stress SLR: kSLR,STRESS × wi + NLRi
- GS more likely to be bound by SLR, while WF bound by RBC
Required Capital Ratios

Distance from Requirements
?
Estimated Capital Charges
• Compute capital charge for bank j under j’s most binding

constraint
?
Banks vs Non-Banks
• Agreement that banks (and shadow banks) can create

systemic risk
• Much more controversial for non-banks
- MetLife sued FSOC January 2015 over their decision to
designate MetLife as a SIFI
• Insurance Companies play important role in financial system
- 30%–35% of bonds issued by European financial firms
- Fire sales can happen (Dick-Nielsen-Feldhutter-Lando,
Ellul-Jotikasthira-Lunblad, Manconi-Massa-Yasuda)
- Market activities: McDonald and Paulson (2015)
FinTech
Are Financial Innovations Useful?

What did NOT happen in Finance
Retail Trade: Relative Price

1.5
1.4
1.3
1.2
1.1
1
1960 1975 1990 2005

year
Relative Wage & Eduction in Finance
Source: Philippon & Reshef (QJE, 2012)

Relative Wage: Financiers vs Engineers

Employment Shares & Relative Wages

Pay & Regulation

Regulation: Where We Are & Where We Want To Be

Current Strategy Has Run Its Course
Will not work: entrenched interests, coordination costs, intractable

design problem
My Proposal: Strategy 2
My Proposal: Strategy 2
This Will Not Happen Automatically

This Will Not Happen Automatically

Finance and Climate Change
• New risks
• Insurance
- impact on insurance liabilities from climate- and
weather-related events, such as floods and storms that damage
property or disrupt trade
• Asset management
- Revaluation risks from adjustment towards a lower-carbon
economy
- Stranded assets : Fossil fuels companies: mainly valued on
their reserves. But reserves exceed the budget of the planet:
2,795 GtCO2, but budget : 1,437 GtCO2
Low Carbon Indices
• How would you go about providing financial incentives for

reducing carbon footprint
• Option 1: exclude poluters from portfolio and/or bet on
replacements
- very costly and complicated
• How do you buil an investment strategy when you think there
is a risk in the long run but you have no idea when risk might
materialize?
• Option 2: start from existing indexes and reduce footprint
- start from MSCI index & match with carbon emission data
- choose porfolio that achieves X% reduction of Emission
Intensity and Z% reduction in exposure to stranded assets
- optimize and re-balance regularly : low tracking error
Low Carbon Indices: Theory

Low Carbon Indices: Historical Performance

References I
Bansal, R. and A. Yaron (2002). Risks for the long run: A potential
resolution of asset pricing puzzles. forthcoming, Journal of
Finance.
Bernanke, B. and M. Gertler (1989). Agency costs, net worth and
business fluctuations. American Economic Review 79, 14–31.
Bernanke, B., M. Gertler, and S. Gilchrist (1999). The financial
accelerator in a quantitative business cycle framework. In J. B.
Taylor and M. Woodford (Eds.), Handbook of Macroeconomics,
Volume 1C. Elsevier Science, North Holland.
Cochrane, J. H. (2017). Macro-finance.
Eggertsson, G. and P. Krugman (2012). Debt, deleveraging, and
the liquidity trap:a fisher-minsky-koo approach. Quarterly Journal
of Economics (127(3)), 1469–1513.
Gertler, M. and N. Kiyotaki (2013). Banking, liquidity and bank
runs in an infinite horizon economy. Working Paper, NYU.
References II
He, Z. and A. Krishnamurthy (2012). A model of capital and

crises. Review of Economic Studies.
Kaplan, G. and G. L. Violante (2011, August). A Model of the
Consumption Response to Fiscal Stimulus Payments. NBER
Working Papers 17338, National Bureau of Economic Research,
Inc.
Kiyotaki, N. and J. Moore (1997, Apr). Credit cycles. Journal of
Political Economy 105 (2), 211–248.
Leland, H. E. (1994). Bond prices, yield spreads, and optimal
capital structure with default risk. Working Paper 240, IBER,
University of California, Berkeley.
Leland, H. E. (1998). Agency costs, risk management, and capital
structure. Journal of Finance 53, 1213–1243.
References III
Martin, P. and T. Philippon (2014). Inspecting the mechanism:

Leverage and the great recession in the eurozone. Working Paper
NYU.
Myers, S. C. (1977). Determinants of corporate borrowing. Journal
of Financial Economics 5, 147–175.
Myers, S. C. and N. S. Majluf (1984). Corporate financing and
investment decisions when firms have information that investors
do not have. Journal of Financial Economics 13, 187–221.
Nachman, D. and T. Noe (1994). Optimal design of securities
under asymmetric information. Review of Financial Studies 7,
1–44.
Philippon, T. (2009). The bond market’s q. The Quarterly Journal
of Economics 124 (3), 1011–1056.
References IV
Philippon, T. (2015). Has the us finance industry become less

efficient? on the theory and measurement of financial
intermediation. The American Economic Review 105 (4),
1408–38.

AEA 2018 Finance in Macro Models

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

AEA 2018 Finance in Macro Models

Uploaded by

Copyright:

Available Formats

AEA Continuing Education

DSGE Models and the

January 7-9, 2018

Finance in Macro Models

NYU, NBER, CEPR

January 2018, AEA Continuing Education, Philadelphia

Does finance matter?

Joan Robinson (1952)

Some (Contradictory) Views about Finance

Three Financial Frictions

• Debt overhang, Myers (1977)

• Monitoring & Moral hazard

• Screening & Adverse Selection

Risk neutral investors, two periods t = 0, 1, limited liability

Total Senior Debt Equity

Key Friction: Difficult to Renegotiate Senior Debt

Total Senior Debt Junior Debt Equity

p AH+ vH D Rk AH-D+ vH-Rk

New lenders break even Shareholder Value: vH-Rk

Total Senior Debt Junior Debt Equity

p AH+ v D k/p AH-D+ v-k/p

Total Senior Debt Junior Debt Equity

p AH +v D k/p AH-D+ v-k/p

• NPV = (2p-1)v-k can be negative, e.g. when p<0.5

Total Senior Debt Equity

• Buyers: m=zE[A] Papers

Moral Hazard: Endogenous q

• Therefore pledgeable income is q H (V − ymin

- Existing assets can be used as collateral to increase pledgeable

• Issuance game & security design

• Good types refuses to invest if

Time 1: Credit Market

Limit Rate For

Time 1: Credit Market Equilibrium

Limit Rate For

• Debt overhang: legacy debt can limit/distort investment

Measuring Financial Intermediation

Price and Quantity of Financial Intermediation

- User cost of borrowers, expected return of savers

Income Share of Finance Industry

1860 1880 1900 1920 1940 1960 1980 2000

VA fire, Hist. WN fire, Hist.

Outstanding Credit / GDP

Farm, Hist Farm, FoF

Neoclassical Growth with Intermediation

• Households’ consumption and liquidity services: u (ct , mt ), life

• Assumption: Constant Relative Costs: ψi ,t = µi ψt , with

• Estimate the µ ′ s with micro data on rates, returns and

Intermediated Assets & Services

1880 1900 1920 1940 1960 1980 2000

Business Credit & Equity Household Credit

Financial Intermediation (in the U.S.)

Share of GDP Intermediated Assets/GDP

Source: Philippon (AER, 2015)

Unit Cost in the U.S.

1880 1900 1920 1940 1960 1980 2000 2020

2012 Data New Data

Source: Philippon (AER, 2015)

Unit Costs, Global Comparison

Source: Bazot (2013)

Constant Returns to Scale

Evidence of Constant Returns to Scale

1950 1960 1970

Unit Cost Real Assets Real GDP PC