Download Reserve Accumulation, Growth and Financial Crises

Reserve Accumulation, Growth
and Financial Crises
Gianluca Benigno
Luca Fornaro
IGC Workshop on Fiscal and Monetary Policy
LSE, November 2012
1
Research questions
I
What explains the spectacular accumulation of foreign exchange
reserves in developing countries?
I
Why do we observe a positive relationship between growth and
current account surpluses?
2
Reserve accumulation in developing countries
Reserves (% of GDP)
60
50
Developing countries
East Asia
40
30
20
10
0
3
GDP growth and current account (1980-2010)
4
GDP growth and reserve accumulation (1980-2010)
5
Empirical evidence
I
Empirical regularities first emphasized by Gourinchas and Jeanne
(2011) and by Alfaro, Kalemli-Ozcan and Volosovych (2011)
I
These facts are hard to reconcile with the neoclassical growth model
I
In the neoclassical growth model:
I
Faster growth is associated with higher capital inflows
I
The competitive equilibrium is efficient, hence no role for public
intervention in capital flows
6
Our contribution
I
We develop a theory of public intervention in capital flows
I
Key elements:
I
I
Knowledge externalities in the tradable sector
I
International borrowing constraint
The combination of these two elements provides an incentive for the
government to accumulate reserves in order to stimulate growth
7
Our contribution (cont’d)
I
Accumulation of reserves is associated with exchange rate
undervaluation and faster growth
I
Financial frictions create imperfect substitutability between private
and public capital flows
I
The possibility of using reserves during crises amplifies the positive
relationship between reserve accumulation and growth
I
The welfare gains from an appropriate reserve policy are substantial
(in the order of a 1 percent permanent increase in consumption in
our baseline calibration)
8
Related literature
I
Theories of reserve accumulation: Durdu et al. (2010), Jeanne
and Ranciere (2011), Dooley et al. (2003), Aizenman and Lee
(2007), Rodrik (2009), Korinek and Serven (2010)
I
Related empirical evidence: Gourinchas and Jeanne (2011),
Alfaro, Kalemli-Ozcan and Volosovych (2011), Rodrik (2008), Cerra
and Saxena (2008)
9
Plan of the talk
I
Model
I
Explanation of the mechanisms
I
Reserve management in an economy opening to capital flows
I
Welfare
10
Model
I
Small open economy
I
Two sectors: tradable and non-tradable
I
Households, firms, foreign investors, government
11
Households
I
Expected lifetime utility
"
E0
I
∞
X
C 1−γ
βt t
1−γ
t=0
#
Consumption aggregator
Ct = CtT
ω
CtN
1−ω
I
Supply inelastically one unit of labor during each period
I
Budget constraint
N
CtT + PtN CtN = Wt + ΠT
t + Πt
12
Real exchange rate and non-tradable sector
I
Real exchange rate
PtN =
I
1 − ω CtT
ω CtN
Firms in the non-tradable sector maximize
N
ΠN
LN
t = Pt
t
αN
− Wt LN
t
13
Firms: tradable sector
I
Produce using labor LT
t , imported inputs Mt and knowledge Xt
YtT = Xt LT
t
I
α T
Mt1−αT
Dividends
T
T
M
ΠT
t = Yt − Wt Lt − P Mt − Bt+1 + RBt − Tt
I
Firms maximize
E0
"∞
X
#
β
t
λ t ΠT
t
t=0
14
Working capital
I
Working capital requirement: a fraction φ of the imported inputs has
to be paid before production takes place
φP M M
| {z }t
work. cap. requirement
I
=
DtG +
|{z}
gov. loans
DtP
|{z}
loans from foreign investors
We assume a zero interest rate on intraperiod loans
15
Borrowing constraint
I
To prevent defaults foreign investors impose the borrowing limit
−RB
| {z }t
+
bonds maturing in period t
I
DtP
|{z}
intratemporal loan at time t
≤
κt
|{z}
Xt
credit shock
Binding borrowing constraint interferes with:
I
Consumption smoothing
I
Import of intermediate goods
16
Knowledge accumulation
I
Knowledge evolves according to
Xt+1 = ψXt + Mtξ Xt1−ξ
I
This is meant to capture spillovers of foreign knowledge through the
imports of intermediate goods
I
Externality: since knowledge is non-excludable firms do not
internalize the impact of their actions on the future stock of
knowledge
17
Discussion of growth process
I
Cross-country knowledge spillovers: Klenow and Rodriguez-Clare
(2005)
I
Transmission of knowledge trough trade: Coe, Helpman and
Hoffmaister (1997), Amiti and Konings (2007), Blalock and Gertler
(2004), Park, Yang, Shi and Jiang (2010)
I
Tradable sector as engine of productivity convergence: Rodrik
(2012)
I
Knowledge externalities: Romer (1990), Grossman and Helpman
(1991), Aghion and Howitt (1992)
18
Government
I
Collects taxes to finance reserve accumulation
I
Uses reserves to provide working capital loans to firms (efficiency
loss as in Gertler and Karadi (2009))
FXt+1 = R FX FXt + Tt − DtG
I
θ
1−θ
Reserves cannot be negative and pay a return lower than the world
interest rate
19
Market clearing
I
Tradable good
CtT = YtT − P M Mt − Bt+1 + RBt − FXt+1 + R FX FXt − DtG
I
θ
1−θ
Non-tradable good
CtN = YtN
I
Labor
N
LT
t + Lt = 1
20
Intervention - tranquil times
I
I
When firms are not financially constrained an increase in reserves
leads to a higher use of imported inputs and faster growth
I
Increase in the stock of reserves
I
Decrease in consumption of tradables
I
Real exchange rate depreciation
I
Wages decrease and firms in tradable sector employ more labor
I
Use of imported inputs increases
I
Faster accumulation of knowledge
Focus on reserve accumulation rules of the form
FXt+1 − R FX FXt = χYtT
21
Intervention - tranquil times (FXt+1 − R FX FXt = χYtT )
GDP growth
Real exchange rate
0.045
Trade balance/GDP
0.08
0
0.06
0.04
−0.01
0.04
−0.02
0.035
0.02
−0.03
0
0.03
0
0.05
χ
0.1
0.15
−0.04
0
0.05
χ
0.1
0.15
Private net foreign assets/GDP Consumption of tradables
−0.16
0
0.05
χ
0.1
0.15
Aggregate consumption
0.162
0.415
0.16
−0.162
−0.164
0.158
0.41
0.156
0.405
0.154
−0.166
0.4
0.152
−0.168
−0.17
0
0.395
0.15
0.148
0.05
χ
0.1
0.15
0.39
0
0.05
χ
0.1
0.15
0
0.05
χ
0.1
0.15
22
Intervention - crises
I
When firms are financially constrained
Mt =
I
Xt κt + RBt + DtG
φP M
Government can increase the use of imported inputs by using foreign
exchange reserves to finance working capital
I
We assume that the government uses at most a fraction χWK of its
stock of reserves to finance working capital
23
Intervention - crises (cont’d)
GDP
Credit shock
Imported inputs
1
1.8
1.8
0.8
1.6
1.6
1.4
1.4
1.2
1.2
0.6
0.4
1
1
0.8
0.8
0.2
0
5
10
15
5
Time
10
15
5
Time
Real exchange rate
Private foreign debt
1.8
1.8
1.6
1.6
1.6
1.4
1.4
1.4
1.2
1.2
1.2
1
1
1
0.8
0.8
0.8
10
Time
15
5
10
15
Time
with intervention
15
Foreign exchange reserves
1.8
5
10
Time
5
10
15
Time
w/o intervention
24
Policy intervention and financial liberalization
I
To illustrate the properties of the model we look at the impact of
policy on an economy that it is opening to capital flows (i.e.
B0 = FX0 = 0)
I
1. We look at the effect on growth and capital flows by comparing
an economy without intervention to one with the optimal policy rule
(χ = 0.09, χWK = 1)
I
2. We compute the welfare gains from policy intervention
I
We assume two possible realizations for the credit shock kH > kL
25
Calibration
Table 1: Parameters
Parameter
Symbol
Value
Risk aversion
γ
2
Interest rate on private borrowing
R
1.04
Discount factor
β
1/R
Labor share in output in tradable sector
αT
0.65
Labor share in output in non-tradable sector
αN
0.65
Share of tradable goods in consumption
ω
0.341
Price of imported inputs
PM
1
0.1
Borrowing limit
κL
Probability of bad credit shock
1 − ρH
0.1
Probability of exiting bad credit shock
1 − ρL
0.5
Working capital coefficient
φ
0.33
Elasticity of TFP w.r.t. imported inputs
ξ
0.15
Constant in knowledge accumulation process
ψ
0.34
Interest rate on reserves
R FX
1
Efficiency of government intervention during crises
θ
0.5
26
Reserve management, growth and capital flows
Private NFA/GDP
Reserves/GDP
−0.05
−0.1
0.3
−0.15
0.2
0.1
−0.2
0
2 4 6 8 10 12 14
Years since liberalization
Knowledge growth
0
0
Current account/GDP
0.04
0.02
0
−0.02
−0.04
2 4 6 8 10 12 14
Years since liberalization
0
Probability binding constraint
2 4 6 8 10 12 14
Years since liberalization
GDP
0.2
0.4
0.04
0.035
0.1
0.2
0.03
0.025
0
2 4 6 8 10 12 14
Years since liberalization
Consumption of tradables
0
0
0
0
2 4 6 8 10 12 14
Years since liberalization
Consumption of nontradables
2 4 6 8 10 12 14
Years since liberalization
Real exchange rate
0
0.4
0.3
0.2
−0.05
0.2
0.1
0
0
2 4 6 8 10 12 14
Years since liberalization
−0.1
0
2 4 6 8 10 12 14
Years since liberalization
No intervention
0
0
2 4 6 8 10 12 14
Years since liberalization
Optimal policy
27
Welfare
Consumption equivalent in percent
1.5
1
χWK = 0
0.5
χWK = 0.25
χWK = 0.5
0
WK
χ
= 0.75
χWK = 1
−0.5
−1
0
0.05
χ
0.1
0.15
28
Social planner
I
The social planner does not accumulate reserves
I
The first best can be replicated by subsidizing the purchase of
intermediate inputs
I
Subsidies to exporters can conflict with trade agreements
I
Reserve accumulation can be used to circumvent the restrictions
imposed by trade agreements
29
Conclusions
I
We provide a novel framework able to reproduce the positive
correlation between reserve accumulation, current account surplus
and growth observed in the data
I
Future research:
I
Interaction between reserve management and capital controls
I
Global imbalances and reserve accumulation
30