tCHPgg
Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium
Member
Libraries
http://www.archive.org/details/fiscalpolicyfinaOOcaba
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
FISCAL POLICY AND FINANCIAL DEPTH
Ricardo
J.
Caballero
Arvind Krishnamurthy
Working Paper 04-22
May 3, 2004
Room
E52-251
50 Memorial Drive
Cambridge,
This
MA 021 42
paper can be downloaded without charge from the
Network Paper Collection at
Social Science Research
http://ssrn.com/abstract=548384
MASSACHUSETTS INSTITUTE
OF TECHNOLOGY
DEC
7 20(M
LIBRARIES
Fiscal Policy
Ricardo
J.
and Financial Depth
Arvind Krishnamurthy*
Caballero
This draft:
May
2004
03,
Abstract
Most economists and observers place the
recent Argentine
crisis.
lack of fiscal discipline at the core of the
This begs the question of how countries
(pre-Maastricht) could run large
fiscal deficits
like
and accumulate debts
Belgium or
Italy
beyond those of
far
Argentina, without experiencing crises nearly as dramatic as that of Argentina?
Why
that Argentina cannot act like Belgium or Italy and pursue expansionary
fiscal
is it
policy during downturns?
their financial depth,
We
argue that advanced and emerging economies
and show that lack of
way that can overturn standard Keynesian
financial
fiscal
depth constrains
policy prescriptions.
empirical support for this viewpoint. Crowding out
is
fiscal
We
when the crowding out
JEL Codes:
coefficient exceeds
policy in a
also provide
systematically larger in emerging
markets than in developed economies. More importantly, this difference
ing crises,
differ in
is
extreme dur-
one in emerging market economies.
E44, E62, F34, F41
Keywords: Sudden
stops, financial depth, fiscal deficits, liquidity crises, populism,
crowding out.
•Respectively:
MIT and NBER;
krishnamurthy@northwestern.edu.
We
Northwestern
University.
are grateful to Fernando Broner,
E-mails:
Guy
caball@mit.edu,
Debelle,
a-
Jaewoo Lee, and
Alessandro Pavan for their comments and to Francisco Gallego for excellent research assistance. Caballero
thanks the
NSF
for financial support.
Introduction
1
Most economists and observers place the
gentine
crisis.
lack of fiscal discipline at the core of the recent Ar-
This begs the question of how countries
could run large
and accumulate debts
fiscal deficits
Belgium or
like
far
beyond those of Argentina, with-
Why
out experiencing crises nearly as dramatic as that of Argentina?
cannot act
We
like
Belgium or
Italy
Italy (pre-Maastricht)
and pursue expansionary
is it
that Argentina
policy during downturns?
fiscal
provide an answer to these questions based on the observation that advanced and
emerging market economies
depth constrains
differ in their financial depth.
policy in a
fiscal
We
show that
lack of financial
way that can overturn standard Keynesian
fiscal
policy
prescriptions.
By
financial
depth we mean the supply of funds available to the government and private
sector of an emerging market. Investing in an emerging market requires far
than investing in an advanced one.
exchange rate
tion.
risk,
Segmentation
investors
who have
depth of a country
As
it
requires knowledge of political risk,
and the degree and form of corporate,
and government corrup-
judicial
a prevalent feature of emerging markets.
is
We
refer to the small set of
the investment expertise on these markets as specialists.
is
The
financial
limited by the liquidity controlled by these specialists.
in our previous
model an external
For example,
more expertise
work
crisis as
(see Caballero
and Krishnamurthy 2001, 2002, 2003, 2004), we
an event in which financial depth
country faces a quantity financial-constraint on
its
crowds out private investment; therefore loose
fiscal
The crowding-out problem
is
We
of the country's assets.
amplified
illustrate
if
is
borrowing.
policy
expansionary
limited.
1
In this context, the
Any government expenditure
may
in fact
fiscal policy
be contractionary.
worsens the quality
two channels of amplification.
First, as
the rising
share of public debt to private assets reduces the aggregate liquidity of the country's assets,
specialists increase their required liquidity
Second,
if
premium and
this further reduces financial depth.
the lack of fiscal discipline sparks investors' fears regarding the
fiscal responsibility
of the government, specialists endogenously lower their valuation of the country's assets
financial depth also
We
1
reduced further.
provide empirical support for our crowding-out hypothesis by examining the
ential response of
2
is
See Broner et
emerging and advanced economies to
al (2003) for
and
2
fiscal shocks.
We
first
differ-
extend the
extensive evidence of limited supply of funds during external crises.
This second, signaling channel has a resemblance with some of the explanations in the literature on
"expansionary
fiscal
contractions" sparked by Giavazzi and
(1990) and Drazen (1990).
Hemming
et al (2002).
See
More
also, e.g.,
generally, the tight connection
environment of tight capital flows
is
behavior of the Brazilean yield curve.
Pagano
(1990),
and discussed by Blanchard
Alesina and Perotti (1995), Giavazzi and Pagano (1996), and
between
fiscal
policy and cost of credit in an
consistent with the evidence in Favero and Giavazzi (2003)
on the
(IADB
results in
economies than
1997) and show that
We
advanced economies.
in
expansion on private investment.
policy
fiscal
is
indeed more pro-cyclical in emerging
then turn to estimating the
effect of a fiscal
We show that this coefficient is more negative in emerging
economies than in advanced economies. However, our main results are from a "differencein-difference" regression.
in tranquil times
much
is
We
show that the
difference
between the response
in crises
and
bigger (more negative) in emerging economies than in advanced
ones.
we make the
In section 2
basic connection between crowding out
and
financial
depth
Sections 3 and 4 are the core of the paper and
during sudden stops of capital inflows.
present two models of extreme crowding out. In the
liquidity of a country's assets as the share of
first
one we we model the decline in the
government debt
in total assets rises, while in
the second one we highlight the negative signaling of a government that refuses to adjust
accounts.
its fiscal
The simple dynamic model
value. Section 5 present empirical evidence,
and section
We
consider a government that has a stock of debt,
private sector has a total of / projects that
marginal product of
I
of independent
6 concludes.
r.
it
D, which
it
The
needs to refinance.
needs to fund. Each of these projects has a
In sum, the financing need of the government and the private sector
+ D.
The
international interest rate
is
r*
During normal times, there are
.
lenders (or collectively they have sufficient funds), so that
investment are financed at the interest rate of
Our
central assumption
strained.
At
there are
many
by
is
Crowding out
2
is
of illiquidity in Section 3
q,
that during
this point, the country
where q
is
specialists,
is
an external
crisis this supply of
funds
is
con-
each with some limited funds to lend. The specialists are indexed
is
than
q.
whose opportunity cost
is less
for funds
>
r*). In
aggregate, the supply
the mass of funds available for lending by
with the supply of funds yields,
I
+ D = F{q).
Since the private sector has marginal product of
borrow funds. Thus
government debt and private
has limited financial depth. Formally, we assume that
given as F(q), where F(-)
Equating the demand
many
r*.
their effective opportunity cost of lending (q
schedule of funds
specialists
is
all
sufficiently
r, it is
willing to
pay up to r in order to
in equilibrium,
I
+D=
F(r).
(1)
^From
during a
this relation,
crisis
we
see an immediate constraint that the loss of financial depth
places on fiscal policy: If the government increases
crowding out of private investment. Figure
1
illustrates this scenario.
we measure gross returns and on the horizontal the amount
external
crisis,
the interest rate jumps to the
which pins down the
the government,
maximum
means one
less
D, there
maximum
On
is
a one-for-one
the vertical axis
of external loans.
During an
return the private sector can offer,
(and actual) loans the country receives.
Any new
loan to
loan to the private sector.
F
D
Figure
In the next sections
potentially
more
In this section
of
Crowding out
we develop dynamic
illiquidity
we show
government debt
falls.
:
model and
illustrate
that
in the
if
and crowding out
specialists
economy
have any liquidity preference, then as the share
increases, the liquidity of all of the country's assets
Specialists require a larger liquidity
supply of funds. The reason behind this
is
versions of this simple
drastic forms of crowding out.
Aggregate
3
1
premium
in response,
effect is that
ultimately linked to the productivity of
its
and further reduce their
the aggregate liquidity of a country
private assets.
Government
assets
may be
backed by domestic transfers but they do not themselves generate aggregate returns. Thus,
as crowding out increases
country's assets
assets decrease, the liquidity of the
falls.
Our dynamic model
to which
and returns from private
is
a fairly straightforward extension of the previous static model,
we add a government
in order to derive the
dynamics of government debt.
Time
continuous.
is
Our
with either the reversal of the
any time interval
default. In
Xdt.
The
attack event
is
crisis,
the reversal of the
dt,
crisis
We
equal to F(q).
is
precipitate a government
occurs with exogenous probability
endogenously determined by the behavior of specialists (see below).
period financial depth
crisis
may
or a final attack which
For now, we denote the attack probability in the next dt as
During the
and finishes
analysis starts with the onset of an exogenous crisis
is
and the supply of funds to the country
limited
introduce some liquidity concerns
among
<
We
the specialists.
The long-term
that there are "long-term" and "short-term" specialists.
opportunity cost of funding of q
fitdt.
>
Short-term specialists have cost of funding of q
In addition to higher
q.
We model this by
specialists also face the possibihty of liquidity shocks.
have
specialists
F=
with corresponding total mass of funds of
q,
assume
F(q).
short-term
q's,
assuming that with
flow probability of 5dt a sunspot occurs in the next interval of time. If the sunspot occurs,
may
then the short-term specialists
Loans from
short term
projects
(i.e.
is r,
specialists to
exit the
As
The government
effect of its actions
next dt and cease lending.
before, since the marginal product of the private sector's
the interest rate on these loans
—
for the
both the government and private sector are assumed to be
instantaneous).
excess return of (r
market
also
is
q)dt to an investor of type
r.
If
q.
If
loans are repaid, they yield a flow
there
default, the return
is
values an expansionary fiscal policy, while
We make
on private sector investment.
is
— 1.
it
completely ignores the
this
extreme assumption in
order to capture the short horizon of the type of governments that concern us here.
flow benefit
sets g t
equal
=
g
is gtdt.
>
0,
So, the
government either continues an expansionary
or balances the
budget by setting gt
t
.
growth
t
As mentioned, the
in public debt
t
.
is
over
its
debt,
crisis
may end
+gt.
t
in a final attack in
and has to reconcile
its
debt.
which the government
We assume that
But not
all
governments perceive the same
type, 6, that parameterizes
how concerned
cost.
We
is
unable to
the government cares about
the growth in the stock of debt because the cost of dealing with the attack
D
that
to,
D = rD
roll
= —rD The
fiscal policy
The
is
increasing in
assume that a government has a
with the debt. In particular, a government
it is
of type 8 views the cost of debt as,
(1
3
Note
essential.
also that
What
is
we have assumed
- B)C(D)
full
is
that there
illiquidity after the
except of a measure zero event (when the final
See below.
crisis
0,
C" >
0.
than partial default or restructuring. This
default, rather
important to our model
This holds as long as the degree of
C>
is
a discrete expected
run
is
substantial,
takes place at the
first
loss
which
when
is
is
not
default takes place.
the case
instant in which
in
our model
it is
feasible).
Our
preferred interpretation of 6
and expropriate
as a "populist"
expropriation
then
is
[itOdt.
so that the analogue of (1)
<
6
<
We
may
react
assume that a government of type
In any interval of time dt, the probability of
1.
Therefore, the supply of funds from specialists
F(r -
is
8),
/j, t
is,
It
The model has
that in the attack event, the government
financial assets.
all
8 expropriates with probability
is
+D =
t
a sovereign principle built
regardless of whether these were issued
F{r -
fi t 0t).
(2)
as default/expropriation occurs on
in,
all
debts,
by the country's private sector or the government.
Thus, foreign investors value corporate and government debt equally.
We
bility
next describe the liquidity concerns of specialists and link this to the attack proba-
fj. t
of Sdt.
We
.
As noted
earlier,
The sunspot may
serve to coordinate an attack.
assume that the sunspot
show that any short-term
for the next dt
reason
order
a sunspot occurs in the next interval of time with flow probability
is
dt.
is
only observed by the short-term specialists.
specialist that
and not renew
— 1, which
is
the other hand,
if
his financing
on any government or corporate
is
benefit.
assume that q > r
—
no problem
in
if
is
of
is
is
any
events.
We
specialists,
oiF—F assets.
unwinding the corporate position, since these are short-term loans
not repay these loans without
question
is
4
a sunspot occurs specialists will unwind a position
on completed projects. The problem
will
which
to withdraw.
do not observe the sunspot of the short-term
the do not cease lending based on this sunspot.
is
is
on the other hand, continue lending regardless of
for all levels of Dt. Also, since they
There
q)dt,
The
This ensures that they always earn a surplus on specialist lending
59.
Then, in aggregate,
—
Thus, as long as there
chance of default, the optimal strategy for the short-term specialist
specialists,
ir
loans.
the attack occurs and the government defaults, the cost
an order of magnitude larger than the
The long-term
easy to
has seen a sunspot will withdraw from the country
that the benefit of continuing to invest in the country
On
It is
may
first
arise
with the government loans. The government
securing financing from other specialists.
whether the specialists that did not observe the liquidity shock have
Thus the
sufficient
resources to finance the government bonds of the exiting specialists. These resources will
be
insufficient
if,
A > F,
There
is still
their financing.
term
small.
specialists.
the possibility of a "run" equilibrium in which
We
are implicitly ruling this out.
In this case,
we
(3)
all
of the long-term specialists do not renew
One can imagine a sunspot
also coordinating the short-
are analyzing a situation where the probability of this sunspot
is
very
which case the government
in
6
is
not able to refinance
debt and defaults with probability
its
5
even
satisfied, so that
a sunspot occurs, there
if
government's debt. In this case,
and
in distress
F
D
Thus, as
F(r) to
t
0.
D
> F,
t
a sunspot always results
from
falls
In other words as the government crosses through the threshold, the
become
illiquid,
and crowding out
deficits lower
is
more than
More
one-for-one.
generally,
the liquidity of a country's assets, crowding out will be
severe.
when
D
is large,
t
the marginal lender
Whenever the
Note also that
is,
liability.
is
a short-term specialist
This
is
who
is
fi-
the source of the instability.
(1983), there
is
a "run" equilibrium, which we resolve with a
latter takes place,
it
precipitates an attack.
Diamond and Dybvig
sunspot.
That
not
sufficient liquidity to refinance all of the
For large levels of
nancing an implicitly long-term government
in
is
debt level a government can take on without risking a final attack.
government
Intuitively,
As
this inequality
S.
58).
country's assets
more
=
is
F,
passes through this threshold the supply of funds from specialists
F(r —
as long as
=
maximum
a
is
/it
\i t
<
For small levels of Dt
There are two cases to consider.
illiquidity
has a more severe effect on worse governments
(i.e.
higher
f?'s).
as government debt passes through the threshold, a worse government experiences
stronger crowding-out.
The model can be
solved recursively because of the assumption that the government
not concerned about private investment, and because the interest rate
the
The government problem can be
crisis.
investor's
If
the
constant during
solved without regard for what investors think
of its actions, since the residual claimant of these actions
The
is
is
the domestic private sector.
problem can be solved next, taking the actions of the government as given.
crisis
ends before the attack occurs, the debt
repaid with taxes (which the
is
government does not internalize), and the government's perception of the benefits of a
expansion vanishes as
5
well.
may be
able to change the promised interest rate on
the government to increase the interest rate would smooth our results somewhat but
anything qualitatively.
interest rate
it
There would be a region of
The main
pays on
its
fiscal
6
In a richer model, the government
discussed above.
difference
is
bonds above
crisis
and
default, linked to the
it
its
debt. Allowing
would not change
same considerations
that prior to the default event, the government would raise the
r,
thereby delaying the
crisis.
This would be
realistic as
highly
indebted governments (relative to the size of the specialists pool and private sector's assets) would see
frequent interest rate spikes.
rise
6
The
point, however,
is
that there
is still
a
Dt beyond which there
is
no
more
interest
that can prevent the crisis from taking place.
A11 that
we
is
require
is
that the government values g more during the
crisis
than during normal times.
The government's Bellman- Jacobi equation during the
0=
external crisis
is:
max AV'(Dt)(rDt +gt )-(\ + n(Dt ))V(Dt )+gt .-»{Dt )(l-9)C(Dt )}.
(4)
gt€{-rD t ,g}
Given the convexity of C(-),
stopping rule:
easy to see that the solution to this problem
is
it
J
—rD
|
otherwise.
t
The government begins with debt Dq.
assume that
D <
D"{9) V0.
Consider a 9 such that D*(9)
= — rDt
gt
D
.| (|)
also
F, which means that
D* (9) >
In this case, substituting
_aitfflli,
A
=
used the fact that
fj,
=
between setting gt equal to g and setting
condition gives us the
<
boundary condition,
-t-
where we have
t
than F.
strictly larger
is
into equation (4) gives the
nc
we
In order to keep the problem interesting,
7
Clearly no government will stop spending as long as
F.
a
A < D*{0),
if
j
is
5.
it
D =
At
t
(5)
o
D*(9) the government
is
indifferent
equal to —rDt. Manipulating this indifference
smooth pasting condition:
V'(D*(9))
=
-1.
This gives us an equation that we can solve for D*.
(6)
The
solution applies as long as
it is
greater than F.
Suppose that C(D)
= D1
>
with 7
D*(9)
then combining conditions
1,
= max
j \
The
first
term
in
squared brackets
spending sufficiently early that
A
one-for-one.
This holds as long as -
I
when
is
is
It
may
.
A
^nzgr
(7)
good type of government stops
never runs the risk of an attack, and crowding out
)
F
so that
it
is
its
only
creates
out.
F(r — 66). That
< F(r —
crowded
8
68).
is
no government ever
Without
fully
crowds out the private sector.
this assumption, the interest rate rises
above r
which unnecessarily complicates our analysis.
be surprising that g does not appear in the above first order condition, and hence in the expression
the private sector
D*
9.
(6) yields:
'
worse type of government continues spending beyond
We also assume that D" (0) <
8
-9) J
increasing in
is
more than one-for-one crowding
the
for
it
6(1
and
(5)
This
is
due to the
fully
linearity
out,
assumption
in
the objective function.
The
only role of g in our model
to control the speed at which the government accumulates debt along the path.
marginal flow-utility of government expenditure.
It
does not affect the
4
and crowding out
Fiscal fears
We now illustrate a second dynamic channel whereby crowding out is more than one-for-one.
Investors often worry that an emerging-market government
The government
in charge
around when the
that
may
it
may be
come
bills
may be
fiscally irresponsible.
too willing to run up expenditures, expecting not to be
due. Thus, another cost of fiscal expansions during a crisis
spark investor fear that the government
reduces financial depth as the
number
is fiscally
irresponsible.
is
This further
of specialists willing to lend to the country
falls.
Although there are some interactions between the informational problem we highlight
here and the liquidity mechanism in the previous section, our point
off
the liquidity mechanism. Thus,
we
simplify the liquidity story
constant (alternatively,
we
of the specialists are of the
the type 9
is
we make two modifications
is
best
are looking only in the region where
type, indexed by
not publicly known.
D
as in Section
q,
The unconditional
turning
of our previous model. First,
and assume that the attack parameter,
same
made by
>
2.
D).
fi,
is
exogenous and
We assume that
Second,
we assume
distribution of the latter
is
9
~
all
that
U[0,
9}.
Investors infer the type of the government from the history of government actions since
the beginning of the
crisis
and
its initial level
=
9t
The expected return on lending
E[9\{g s } s=0
an interval dt
in
-
(r
and the corresponding supply
of funds faced
/j,9 t
analysis
The
analysis of the problem
through
section.
is
,D
is
\.
is,
)dt,
solution
(8)
fj,9t
a stopping
is
it
r*.
I
C(D)
g
—rD
if
t
= D1
D
t
<D*(9),
otherwise,
with 7
>
1,
In particular, since
does not try to signal
its
type
identical to that of the previous
rule:
f
which, for the parametric case
>
very similar to the previous case.
Thus the government problem
is
is:
&).
not concerned with private investment,
its actions.
The
t
conducted in the region where r —
Our
the government
...
by the country
F(r is
of debt,
has:
Since D*'(8)
>
0,
the more populist a government
is,
the slower
its fiscal
tightening.
Investors understand this and update their priors with respect to the government's type
based on the path of government's expenditures.
government
is
= g,
If gt
worse than that which would have stopped at D*
have that
g> max (o,l- \ +^7
Conversely,
of
all
the fiscal deficit
is
know
=D
t
.
that the type of
Inverting (9),
we
X
r
eliminated, investors learn that the value of 9
is
the best
those that were possible before adjustment took place.
The
crisis
if
investors
solid line in Figure 2 illustrates the
path of expected default,
goes on and the government does not adjust
its fiscal deficit.
/j,9,
as the external
The dashed
line,
on the
other hand, shows the path of expected default for the best 9 possible, given the level of
Dt.
When
line to the
a government adjusts,
shifts
it
the market perception from a point on the solid
corresponding point on the dashed
updating stops, 9 t
>
=
9
line.
At
this
time there
is full
revelation
and
W >t.
tie
t*6.
Figure
Expected Default
2:
Again, this environment exhibits a more extreme form of crowding out than that in
the static model of Section
2.
During the external
crisis,
the country faces an aggregate
financial constraint:
It
Taking
beliefs
+ D = F{r-fi9
t
about the type of government as given,
investment one-for-one (as in Section
does not leave beliefs unchanged.
2).
(10)
t ).
fiscal
expenditure crowds out private
But an expansionary
fiscal
policy during the crisis
This negative updating further reduces the supply of
funds to the country, and private investment
falls
more than one-for-one with the
expenditure.
fiscal
The
other face of this perverse relation between
financial resources
and there
a
is
is
jump
fiscal
made
in the resources
by
available
The reason
deficit early.
is
policy
and the
availability of
Adjustment leads 6 to
the great benefit of adjustment.
specialists.
9
fall
sharply
Note, however, that
same point as would cutting the
cutting the deficit late does not take the economy to the
is
rise in
that along the path, investors have learned that the government
more populist than a government
that reacts early. That
which include the perceived quality of
its
is,
the country's "fundamentals,"
government, are no longer the same. 10
Empirical Evidence
5
This section begins with some facts on the cyclical behavior of public
with and without financial depth.
crowding-out
is
It
deficits in countries
concludes with tests supporting the hypotheses that
larger in emerging economies than in
advanced ones and, most importantly,
that this difference rises significantly during crises.
Cyclically of Deficits
5.1
Let us contrast the behavior of
advanced economies
fiscal variables in
vis a vis
emerging
Beginning with an example, we contrast the experience of Italy during the
economies.
1980s with that of Argentina and Brazil in the late 1990s. Each of these country-episodes
known
for a high fiscal deficit within its respective
is
comparison group and the centrality of
the deficit in public debate about macroeconomic outcomes. Panel (a) in Figure 3 presents
the evolution of public debt and overall
fiscal deficit as
the 1980s. Debt
axis while the deficit
Panels (b) and
1990s.
It is
is
(c)
reported on the
deficits) are significantly larger for Italy
maximum
in Argentina. Public debt in Italy
recent example of this scenario
seeing Lula's
10
We
sector,
fiscal
the sharp
is
fall in
austerity plan, realized that he was
isolate
less
large,
for
right axis.
was above 15%,
may
find that a
government
is
was below
when
and
in
4%
Brazil.
investors, after
populist than feared.
our main points.
is
it
as large as in Argentina
One
of these assumptions
not concerned about signaling since the cost of a bad signal
but not late when the bad reputation
(i.e.
Argentina and Brazil
Brazil's sovereign spreads
which does not concern the government during the
government, then one
is
is
than
deficit in Italy
was more than twice
have made extreme assumptions to
government
measured on the
repeat this figure for Argentina and Brazil, respectively, during the late
the relevant periods. While the
A
is
GDP for Italy during
apparent from this figure that both the level and change of public debt
roughly public
9
left
a percentage of
crisis.
If
is
paid in
full
is
that the
by the private
we reintroduce some concern by the
willing to stop spending early
when the
signaling gain
already too hard to undo. See Angeletos et al (2003) for
recent developments on policy signaling models.
10
It is also interesting to
this section, that Brazil
point out, although this
made
is
not the main point that concerns us in
a significant effort to reduce
its deficits
while Argentina did
not.
(a) Italy
19B0
1982
1381
1983
1984
(b)
1995
198S
1986
1987
19
Argentina
2000
1999
1995
X
-Public Debt
1988
Overall Deficit -
-
-
Primary Deficit
|
Figure 3 Debt and Deficits
:
The
cyclical behavior of these deficits
Italy the deficit
between the
component
for
when looking only
component
of
and 1990s
for
not in Argentina and Brazil.
in the 1960s,
when
and the
cyclical
at the expenditure side.
GDP
using the Hodrick-Prescott
available.
is
The
developed economies and for the 1990s
11
for
11
correlation
of
GDP
is
This difference
-0.38 for Italy,
filter.
In
correlation between the
and 0.83 and
For these introductory numbers
Later on in our regressions
similar conclusions.
The
component
Argentina and Brazil, respectively.
government expenditures and
n The cyclical components are computed
we use data beginning
it is
of the public deficit
and 0.02 and 0.28
also apparent
cyclical
countercyclical, while
cyclical
-0.53 for Italy,
is
is
also very different across these economies.
is
we use data
for the
1980s
emerging markets. These shorter samples yield
0.51, for Argentina,
and
Brazil, respectively.
They can be
These patterns extend beyond these few economies.
ferences between emerging
and advanced economies. The top
of Table
above evidence while the bottom report the medians of similar
advanced economies.
12
While the
try/episodes in our example,
is
generalized to
1
reproduces the
statistics for
differences are not as dramatic as for the
emerging and
extreme coun-
apparent that the use of countercyclical
it is still
fiscal
a reality for advanced economies but not for most emerging market economies.
Public Deficit,
GDP
Government Expenditures,
Argentina
1.96%
83.03%
Brazil
28.37%
50.83%
Italy
-52.69%
-37.94%
Emerging (median)
-4.41%
45.60%
Advanced (median)
-47.09%
9.08%
Table
We
1:
dif-
policy
13
GDP
Procyclicality of Fiscal Policy
argued with our models that an important candidate for explaining the differences
between both groups of countries
is
financial depth. In
emerging markets, limited funding
constrains the use of fiscal policy during crises.
Measuring financial depth as the ratio of credit to the private sector over GDP, Argentina
and Brazil have
70% during
More
on indices
GDP).
LaPorta,
nificant
the 1980s
14
We
the period of large public
of financial
report
deficits).
That
is,
OLS and IV
GDP
and liquid
fiscal
pro-
liabilities
(using legal origins as instruments, along the lines of
effect of financial
more
two measures of
development (private credit over
et al., 1998) results. Virtually all
and negative
variables.
(i.e.
in the late 1990s, while in Italy the ratio exceeds
generally, Table 2 presents cross-country regressions of
cyclicality
over
25% and 30%
ratios of
combinations
tell
the same story: there
is
development on the degree of procyclicality of
financially developed economies experience
a sigfiscal
more countercyclical
fiscal policy.
These
results are economically significant.
For instance, a representative country in
the top quartile of the distribution of private credit has a correlation between the cyclical
12
The sample
in this exercise
the 1960-2002 period.
The
corresponds to 88 emerging and 22 advanced economies with information in
classification of
emerging and advanced economies follows that of the IMF.
13
This was one of the central messages in
14
The
IADB
(1997).
source of our measures of financial development
is
the Financial Structure Database of the World
Bank. Private credit includes credit by commercial banks and other financial institutions. Liquid
liabilities
include currency and deposits (time and interest-bearing) in banks and other financial intermediaries. (See
Beck
at
al.
(1999) for a detailed description of the original sources.)
12
components of
GDP
and
quartile of the distribution
Dependent
while that of a country located in the bottom
deficit of -0.39;
is
-0.04.
15
GDP
Expenditures and
variable: Correlation of:
Public Deficit and
GDP
Private credit
OLS
IV
Number
-0.093
-0.098
(0.007)
(0.027)
-0.290
-0.374
(0.048)
(0.004)
90
90
of countries
Liquid
OLS
IV
Number
Table
2:
of countries
Procyclicality of Fiscal Variables
liabilities
-0.157
-0.100
(0.003)
(0.073)
-0.505
-0.439
(0.019)
(0.008)
85
85
and Financial Development. Robust standard
errors are reported in parentheses.
Evidence of State Dependent Crowding-Out
5.2
We now
turn to assessing
how
crowding-out of private investment varies across advanced
and emerging market economies, especially during severe contractions and
we
D
and C,
erwise.
15
X
is
=
A/it-i
+ aD + PDitCu + jX + 6X C
it
indicator function that takes a value of one
this,
The median country
there
GDP,
is
a
public deficit over
"crisis"
and zero oth'
in
We also have conducted
credit
if
(11)
it .
it
16 17
a group of controls, including a constant, and the relative price of capital.
the top quartile
is
South Africa with a ratio of private credit to
while the median country in the bottom quartile
16
it
respectively, are (private or total) investment over
GDP, and an
is
Nepal where private credit to
GDP
is
GDP
of
is
crisis indicator.
The
results are unaffected
partly due to standard problems for interest rates to appear significant
investment equations. As well as due to the fact that in practice crowding out takes place through
channels which are only partially captured by domestic interest rates and bank loans.
17
Aside from our specific
50%,
only about 10%.
robustness checks including the domestic real interest rate and domestic private
growth as well as interactions of these variables with the
by these additions. Probably, this
in
For
estimate:
Iit
/,
crises.
tests, this specification
is
justified in
13
more
detail by Serven (2003).
many
All specifications include fixed effects
hand
side
is
and the lagged dependent variable on the
right
instrumented using the second lag of the dependent variable.
Data and samples
5.2.1
We obtain the data from multiple sources. Total investment and the relative price of capital
are from Heston et al. (2002). We construct private investment by removing government
investment from total investment. We obtain the former from the Government Finance
Statistics of the
Growth
IMF
(GFS). The latter
of private credit
and
is
also the source for the public deficit information.
were obtained from the World Bank's World
real interest rates
Development Indicators.
Our panels
are unbalanced, with the sample restricted to countries that have a
of five observations.
We
split
the sample into two groups:
economies and another including 13 emerging economies.
system to allocate countries to each of these groups.
in that classification:
Ireland, Italy, Netherlands,
UK, and USA. For
less
New
one including 18 advanced
We
use the IMF's classification
We include all the
Australia, Belgium, Canada,
minimum
advanced economies
Denmark, Finland, France, Greece,
Zealand, Norway, Portugal, Spain, Sweden, Switzerland,
developed economies we restrict the sample to those countries that
Moreover, we exclude the
are sufficiently developed so as to have access to capital flows.
transition economies because they experienced shocks
during the 1990s. These two criteria plus the
reduce that sample of emerging markets
to:
and reforms of a very
minimum
different
nature
of five observations for each variable
Argentina, Chile, Colombia, Egypt, Indonesia,
South Africa, Thailand, and Venezuela.
India, Mexico, Malaysia, Peru, Philippines,
We
study periods when international capital flows are relevant for each of the groups: the 1980s
and 1990s
for the
advanced economies, and the 1990s
emerging market economies.
for the
A key variable for us is the indicator of crisis. For this, we use three indicators built from
the current account to GDP ratio, GDP growth, and country risk. The latter is measured
as 100
for
minus the Euromoney country
advanced economies and
take
first
risk rating,
which
is
available for the 1980s
emerging economies.
for the 1990s for
differences of each of these variables. Crises are periods
Note that the popular
EMBI/EMBI+
constructed by
JPMorgan
is
emerging economies and at most from 1994. The Euromoney country
papers, for instance Haque, et
Euromoney index
is
40%, including
rise in
political risk,
all
and 1990s
crises indicators
these are located in
countries, of current
only available for a subsample of
risk rating has
been used
in other
(1996) use this indicator to study determinants of country risk.
built using polls of economists
with an increase meaning a
of
al.,
Our
when
the highest (lowest) quartile of the distribution of changes, across
ls
18
and
creditworthiness and
economic
risk,
is
political analysts.
The index
goes from
The
to 100,
a weighted average of analytical indicators (weight
and economic performance),
credit indicators (weight of
20%,
payment record and rescheduling), and market indicators (weight of 40%, access to bond markets, selldown
on short-term paper, and access to discount available
for forfeiting).
14
account and country risk
(GDP
growth). Table 3 summarizes the fraction of observations
identified as crises for each set of countries.
Definition of Crises
Period
Growth
CA
Country Risk
Emerging economies
1990s
7.3
7.3
9.3
Advanced economies
1980s
5.1
7.0
8.8
Advanced economies
1990s
2.9
2.1
2.0
Table
Main
5.2.2
Fraction of crises-observations, by countries and periods
3:
results
Tables 4 and 5 presents our main results from estimating equation (11).
reports results for private investment while the latter does
it
The former
for total investment.
half of each table contains the results for emerging market economies, while the
reports the results for the advanced economies.
across
•
most of the
Crowding out
is
6
We
its
Dn
is
much
larger in
rows).
our hypothesis, while in advanced economies the extent of crowd-
and
emerging markets crowding out
crises times, in
during crises (sum of coefficient in
most cases crowding out during
The long run
coefficient
for
similar across tranquil
rises significantly
• In fact, in
bottom half
conclusions are quite clear and robust
present in advanced and emerging economies but
is
Most importantly
ing out
The top
specifications:
the latter group (coefficient in the
•
The
table
crises
Du
and DaCit).
exceeds one even in the short run.
estimates, which simply divide the short run results by one minus the
on Iu-i, typically exceed two - a very extreme form of crowding
out.
Conclusion
have shown how limited financial depth during
crises constrains fiscal policy
use as a countercylical policy instrument. In fact, using
Emerging markets
factors.
set the
crises invariably
it
in this fashion
and
may
backfire.
stem from a combination of bad luck and financial
Argentina was no exception to these factors.
However, one of the factors that
Argentine experience apart was the poor response of the authorities to the
phases of the
crisis.
limits
Argentina was too late in adjusting
15
its fiscal
accounts.
initial
Along with
Emerging Countries
0.475
0.535
0.439
(0.000)
(0.000)
(0.000)
-0.739
-0.662
-0.793
(0.000)
(0.000)
(0.000)
-0.664
-0.159
-0.681
(0.043)
(0.502)
(0.023)
-2.009
-3.403
-0.291
(0.043)
(0.000)
(0.780)
Obs. /Countries
106/13
106/13
106/13
Time Period
1990s
1990s
1990s
Crisis indicator
Growth
CA
Country Risk
Tranquil
-1.408
-1.424
-1.414
Crisis
-2.672
-1.766
-2.627
0.482
0.488
0.472
(0.000)
(0.000)
(0.000)
-0.178
-0.170
-0.229
(0.000)
(0.000)
(0.000)
0.101
0.177
0.057
(0.346)
(0.029)
(0.223)
-1.357
-2.155
-0.332
(0.064)
(0.001)
(0.373)
Obs. /Countries
297/18
297/18
297/18
Time Period
1980-1990s
1980-1990s
1980-1990s
Crisis indicator
Growth
CA
Country Risk
Tranquil
-0.344
-0.332
-0.434
Crisis
-0.149
0.014
-0.326
Iit-1
B
it
DitCu
Cit
Long-Run Crowding-Out
Advanced Countries
ht-i
D
it
DuCit
C%t
Long-Run Crowding-Out
Table
4:
Private Investment. P-values are presented in parentheses. Covariates include the
(log of)relative price of capital
and interactions
16
of this variable with the crisis indicator.
Emerging Countries
0.504
0.537
0.455
(0.000)
(0.000)
(0.000)
-0.746
-0.728
-0.800
(0.000)
(0.000)
(0.000)
-0.482
-0.066
-0.624
(0.099)
(0.779)
(0.019)
-2.384
-2.583
-0.444
(0.013)
(0.006)
(0.660)
Obs. /Countries
112/13
112/13
112/13
Time Period
1990s
1990s
1990s
Crisis indicator
Growth
CA
Country Risk
Tranquil
-1.504
-1.572
-1.468
Crisis
-2.476
-1.715
-2.613
0.436
0.448
0.450
(0.000)
(0.000)
(0.000)
-0.215
-0.211
-0.255
(0.000)
(0.000)
(0.000)
0.083
0.164
0.063
(0.405)
(0.019)
(0.127)
-1.105
-2.094
-0.437
(0.106)
(0.000)
(0.178)
Obs./Countries
309/18
309/18
309/18
Time Period
1980-1990s
1980-1990s
1980-1990s
Crisis indicator
Growth
CA
Country Risk
Tranquil
-0.381
-0.382
-0.464
Crisis
-0.234
-0.085
-0.349
ht-l
D
it
DuCit
Cu
Long-Run Crowding-Out
Advanced Countries
Iit-1
D
it
DuCit
Cu
Long-Run Crowding-Out
Table
5:
Total Investment. P-values are presented in parentheses. Covariates include the
(log of)relative price of capital
and interactions of
17
this variable
with the
crisis indicator.
the political environment, this poor response worsened the quality of Argentina's assets by
reducing aggregate liquidity and reigniting fears of populism.
The
recent experience of Brazil under President Lula reflects the other side of the coin.
Faced with deteriorating external financial conditions, and contrary to expectations, Brazil's
government endorsed tight
Markets were positively surprised that the
fiscal discipline.
government was not as populist as many feared.
The
reaction was a sharp reversal of
capital outflows.
Our model captures
of the country's assets
these events. Slow fiscal adjustment weakens investors' perception
through two channels:
it
lowers the perceived quality of the gov-
ernment; and
it
investments.
Conversely, early adjustment can result in a dramatic improvement in the
reduces the liquidity of the country's assets by crowding out productive
country's performance.
Our evidence points
clearly in the direction of a crowding-out
severe in emerging market economies than in advanced ones.
ference rises during periods of crises.
mechanism that
More importantly,
In emerging markets, crowding-out
is
is
more
this dif-
more than
one-for-one during crises, suggesting that fiscal expansions at those times are in fact very
contractionary.
plain
why
This, together with the direct impact of capital flow reversals,
fiscal policy is
much
less countercyclical in
advanced ones.
18
may
ex-
emerging market economies than in
References
[1]
Alesina, A. and R. Perotti, "Fiscal Expansion
tries,"
[2]
Economic
Policy,
A
and
European Forum, Vol.
Angeletos, G.M, A.Pavan, and C.Hellwig,
Fiscal
Adjustments in
OECD
Coun-
21, 1995, pp. 205-48.
"Coordination and Policy Traps,"
MIT
mimeo, 2003.
[3]
Beck, Thorsten, Asli Demirgug-Kunt, and Ross Levine, "A
New
Database on Financial
Development and Structure" mimeo, June 1999.
[4]
NBER
[5]
Macroeconomics Annual,
ed.
in
by O. Blanchard and S.Fischer, 1990.
Broner, Fernando, Guido Lorenzoni, and Sergio Schmukler,
kets
[6]
A Comment"
Blanchard, O.J., "Can Severe Fiscal Contractions Be Expansionary?:
"Why do Emerging Mar-
Borrow Short term?" mimeo, July 2003.
Caballero, Ricardo and Arvind Krishnamurthy. "International and Domestic Collateral
Constraints in a Model of Emerging Market Crises," Journal of Monetary Economics.
December 2001, 48(3), pp. 513-548.
.
"A Dual Liquidity Model
nomic Review. May 2002, 92(2),
,
for
Emerging Markets," American Eco-
33-37.
"Excessive Dollar Debt: Underinsurance and Domestic Financial Un-
derdevelopment," Journal of Finance. April 2003, 58(2), pp. 867-893.
,
"Smoothing Sudden Stops," forthcoming
in Journal of Economic
The-
ory. 2004.
[7]
Caprio, G. and D. Klingebiel, "Episodes of Systemic and Borderline Financial Crises,"
mimeo World Bank, January
[8]
Diamond,
DW.
2003.
and P.H. Dybvig, "Bank Runs, Deposit Insurance, and Liquidity,"
Journal of Political Economy 91, 1983, pp. 401-419.
[9]
Drazen, A., "Can Severe Fiscal Contractions Be Expansionary?:
NBER
[10]
Macroeconomics Annual,
in
by O. Blanchard and S.Fischer, 1990.
Favero, C. and F. Giavazzi, "Understanding the Yield Curve in Emerging Markets:
Lessons from Brazil,"
[11]
ed.
A Comment"
Giavazzi, F.
NBER
IGER mimeo, August
2003.
and M. Pagano, "Can Severe Fiscal Contractions Be Expansionary?"
Macroeconomics Annual,
ed.
by O. Blanchard and S.Fischer, 1990.
19
in
[12]
,
"Non-Keynesian Effects of Fiscal Policy Changes: International Evi-
dence and the Swedish Experience,"
[13]
Haque, N. and Kumar,
S.
NBER
Reprints 2082, 1996.
and Nelson, M. and Mathieson, D.,"The Economic Con-
tent of Indicators of Developing
Country Creditworthiness," Staff Papers International
Monetary Fund 43(4), 1996.
[14]
Hemming,
R., Kell, M.S.,
and Mahfouz,
Stimulating Economic Activity
[15]
Heston, A.,R.
Summers and
S.,
"The Effectiveness of Fiscal Policy
in
— A Review of the Literature," IMF WP # 208, 2002.
B. Aten, "Penn World Table Version 6.1," Center for
International Comparisons at the University of Pennsylvania, October 2002.
[16]
Hodrick, R. and Prescott, E., "Post- War U.S. Business Cycless:
gation," Journal of
[17]
Money, Credit and Banking
IADB, Latin America
after a
Economic and Social Progress
[18]
La Porta, R. and
F. Lopez-de-Silanes
nance," Journal of Political
[19]
Decade of Reforms,
in Latin
Economy
America
and A.
An
Empirical Investi-
29, 1997, 1-16.
Inter- American
Development Bank,
series, 1997.
Shleifer
and R. Vishny, "Law and
106(6), 1998, 1113-1155.
Serven, L. "Real-Exchange-Rate Uncertainty and Private Investment in
Review of Economics and
Fi-
Statistics 85(1), 2003, 212-218.
20
LDC," The
Date Due
Lib-26-67
3 9080 02618 0031