Sovereign Default: The Role of
Expectations
João Ayres
Gaston Navarro
Juan Pablo Nicolini
Pedro Teles
Discussion by Mark Aguiar
1 / 15
Sources of Multiplicity
1. Endogenous future payments
2. Price-taking Behavior
2 / 15
Sources of Multiplicity
1. Endogenous future payments
I
Not really how sovereigns issue debt
I
Typically coupon is fixed and price is endogenous
I
Some floating rate debt has been issued
2. Price-taking Behavior
2 / 15
Sources of Multiplicity
1. Endogenous future payments
I
Not really how sovereigns issue debt
I
Typically coupon is fixed and price is endogenous
I
Some floating rate debt has been issued
2. Price-taking Behavior
I
More important departure
2 / 15
Eaton-Gersovitz Timing
No Default
Offer
q(d 0 ; y ),
Auction d 0
Default
VD
c=yd+q(d’;y)d’
Next period
owe d 0
Inherited
State: d
y realized
3 / 15
Key Assumptions
I
Bonds auctioned after period’s default decision made
4 / 15
Key Assumptions
I
Bonds auctioned after period’s default decision made
I
A strong form of intra-period commitment
I
Relaxed by Cole and Kehoe to generate additional equilibria
4 / 15
Key Assumptions
I
I
Bonds auctioned after period’s default decision made
I
A strong form of intra-period commitment
I
Relaxed by Cole and Kehoe to generate additional equilibria
Government’s face a price schedule
q(d 0 ) =
1 − F (yAut + d 0 )
R∗
4 / 15
Key Assumptions
I
I
Bonds auctioned after period’s default decision made
I
A strong form of intra-period commitment
I
Relaxed by Cole and Kehoe to generate additional equilibria
Government’s face a price schedule
q(d 0 ) =
I
1 − F (yAut + d 0 )
R∗
Acts as a monopolist in its own debt: Internalizes that by
issuing more debt, q will fall.
4 / 15
Eaton-Gersovitz Equilibrium
qHd'L=
qHd'L
1 - F HyAut + d 'L
R*
0.8
0.6
0.4
0.2
0.8
1.0
1.2
1.4
1.6
d'
5 / 15
Eaton-Gersovitz Equilibrium
Government’s Best Response: d ∗
1 - F HyAut + d 'L
qHd'L=
R*
qHd'L
0.8
0.6
0.4
0.2
d*
d'
6 / 15
ANNT Alternative
I
Government faces a price (scalar) for any amount of debt
issued
I
Faced with a price, government chooses the optimal amount
to issue
I
Equilibria are the intersection of the government’s best
response to the price and the creditors’ best response to debt
issuance policy
7 / 15
ANNT Alternative
I
Government faces a price (scalar) for any amount of debt
issued
I
Faced with a price, government chooses the optimal amount
to issue
I
Equilibria are the intersection of the government’s best
response to the price and the creditors’ best response to debt
issuance policy
I
Equilibrium price is now:
I
1 − F (yAut + B(d))
R?
0
where B(d) replacing d is the government’s debt-issuance
policy function given current state d (and associated policy
qBR(d))
qBR(d) =
7 / 15
ANNT Alternative
I
Government faces a price (scalar) for any amount of debt
issued
I
Faced with a price, government chooses the optimal amount
to issue
I
Equilibria are the intersection of the government’s best
response to the price and the creditors’ best response to debt
issuance policy
I
Equilibrium price is now:
I
1 − F (yAut + B(d))
R?
0
where B(d) replacing d is the government’s debt-issuance
policy function given current state d (and associated policy
qBR(d))
I
There now may be multiple equilibria
qBR(d) =
7 / 15
ANNT Equilibrium
qBR
0.8
0.6
0.4
0.2
0.8
1.0
1.2
1.4
1.6
GBR
8 / 15
ANNT Equilibrium
qBR
0.8
0.6
0.4
0.2
EG
GBR
9 / 15
Are Government’s Price Takers?
10 / 15
Are Government’s Price Takers?
I
Not really
10 / 15
Are Government’s Price Takers?
I
Not really
I
Fiscal authorities recognize that prices depend on debt
issuances
10 / 15
Are Government’s Price Takers?
I
Not really
I
Fiscal authorities recognize that prices depend on debt
issuances
I
Typical auction mechanics:
1. Announce projected debt issuances at time of budgeting (may
be updated)
2. Announce debt auction calendar (may be updated)
3. Before auction, announce target amount (in consultation with
primary dealers)
4. Collect bids (prices and quantities)
5. Finalize auction amount and price (may differ from target)
10 / 15
Key Features
I
Within an auction, bids induce a downward sloping “demand
curve”
I
Marginal price<Average price
11 / 15
8 Filling ® 881 ® - 20<<, FillingStyle ® Directive@Opaci
Marginal-Average
PlotStyleSpread
® Thin<
D&
Average minus Marginal HLeft AxisL
10Y Spreads with Germany HShaded - Right AxisL
0.14
0.12
4
0.10
0.08
0.06
2
0.04
0.02
0
0.00
2000
2005
2010
12 / 15
Key Features
I
Within an auction, bids induce a downward sloping “supply
curve”
I
Marginal price<Average price
I
However...bids are entered based on forecasted debt issuance,
not realized issuances
13 / 15
auction_analysis.nb
23
Announced versus Realized Issuance
Announcements vs Total Issuances
10
8
6
4
2
0
2011
2012
What happened on Jan 17 of 2012?
From the news: http://www.reuters.com/article/2012/01/12/us-spain-bonds-idUSTRE80B0GQ20120112
14 / 15
Assessment
I
Government’s know that prices are tied to debt issuances
I
Do not explicitly face full price schedule each period as in EG
I
May renege on debt targets
15 / 15
Assessment
I
Government’s know that prices are tied to debt issuances
I
Do not explicitly face full price schedule each period as in EG
I
I
May renege on debt targets
But also not price takers
15 / 15
Assessment
I
Government’s know that prices are tied to debt issuances
I
Do not explicitly face full price schedule each period as in EG
I
May renege on debt targets
I
But also not price takers
I
Repeated game – perhaps MPE not the right concept
15 / 15