^S^CE%^^
-si*
\
Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium IVIember Libraries
http://www.archive.org/details/currentaccountsiOOkraa
©
4.-
HB31
.M415
working paper
department
of economics
i
CURRENT ACCOUNTS IN DEBTOR AND CREDITOR COUNTRIES
Aart Kraay
Jaume Ventura
No.
97-12
July, 1997
massachusetts
institute of
technology
50 memorial drive
Cambridge, mass. 02139
~
L
9
t** C
WORKING PAPER
DEPARTMENT
OF ECONOMICS
CURRENT ACCOUNTS IN DEBTOR AND CREDITOR COUNTRIES
Aart Kraay
Jaume Ventura
No.
97-12
July, 1997
MASSACHUSEHS
INSTITUTE OF
TECHNOLOGY
50 MEMORIAL DRIVE
CAMBRIDGE, MASS. 021329
siOL 3
1
1997
Current Accounts in Debtor and
Creditor Countries
Aart Kraay
The World Bank
and
Jaume Ventura
M.I.T.
July
1997
in international economics: What
response
transitory
income
shock such as a temporary
Is the current account
to a
Improvement In the terms of trade, a transfer from abroad or unusually high
production? To answer this question, we construct a world equilibrium model in which
productivity varies across countries and international borrowing and lending takes
place to exploit good investment opportunities. Despite Its conventional Ingredients,
the model generates the novel prediction that favourable Income shocks lead to
current account deficits In debtor countries and current account surpluses In creditor
countries. Evidence from thirteen OECD countries broadly supports this prediction of
Abstract This paper reexamines a classic question
:
the theory.
We are grateful to
Rudi Dornbusch for discussing these ideas with us. We also thank Daren
Acemoglu, Jakob Svensson and participants in seminars at Harvard, Princeton, MIT,
Rochester and The World Bank for their useful comments. Further comments are welcome.
Please contact the authors at akraay@worldbank.org (Kraay) orjaume@mit.edu (Ventura).
The views expressed herein are the authors', and do not necessarily reflect those of the World
Bank.
Introduction
This paper reexamines a classic question
in
international economics:
What
is
the current account response to a transitory income shock such as a temporary
improvement
in
the terms of trade, a transfer from abroad or unusually high
To answer
production?
this question,
productivity varies across countries
we
and
construct a world equilibrium model
international borrowing
place to exploit good investment opportunities. Despite
its
in
which
and lending takes
conventional ingredients,
the model generates the novel prediction that favourable income shocks lead to
current account deficits
in
debtor countries and current account surpluses
countries. Evidence from thirteen
OECD countries broadly supports this
creditor
in
prediction of
the theory.
A
how
simple thought experiment reveals
natural our result
is
as a benchmark
case. Consider a country that receives a favourable transitory income shock.
Suppose
further that this country
saves
this
shock and has two investment choices,
domestic capital and foreign loans. To the extent that the shock does not affect the
expected
is
profitability of future
investments at
home and
abroad, a reasonable guess
that investors allocate the marginal unit of wealth (the
assets
in
the
same
proportions as the average unit of wealth. Since by definition the
share of a debtor country's wealth invested
increase
in
income shock) among
wealth (savings) results
in
in
domestic capital exceeds one, an
a greater increase
in
domestic capital
(investment), leading to a deficit on the current account (savings
Conversely,
in
as a portion
of this wealth increase is invested
creditor countries the increase in wealth
account surplus
in
creditor countries.
minus investment).
exceeds investment
at
home,
abroad. This produces a current
The sharp
assumptions.
result that
First,
comes
out of this simple example follows from three
the income shock
is
saved. Second, investing
not an option for the country. Third, the marginal unit of wealth
assets as the average one
is.
We
maintain the
paper without (excessive) apologies. The
allocated
is
of
we
feel the jury is
still
consumption-smoothing as a savings motive
we focus on
here.^
if
and only
if
equivalently,
the share of domestic capital
if
and only
if
foreign debt
in
among
failures of the
out regarding the relative importance
at the
The second assumption can be
other assumptions, a favourable income shock
is
a basic tenet of
consumption-smoothing models of savings. Despite some empirical
simplest of these models,
foreign capital
two assumptions throughout the
first
assumption
first
is
in
business cycle frequency that
easily removed.^
If
we keep
leads to a current account
still
the
deficit
the country's wealth exceeds one or,
exceeds the stock
of
outward foreign
investment. Otherwise a favourable income shock leads to a current account surplus.
The
bulk of the theoretical effort of this paper
is
devoted
to
assessing the
merit of the third assumption underlying our simple example, namely, that the
marginal unit of wealth (savings)
wealth.
To do
so,
we
uncertainty
invested
in
the
same
proportions as the stock of
construct a simple world equilibrium model
varies across countries
good investment
is
and
international borrowing
opportunities. In the model,
and random changes
in
we
technology.
which productivity
and lending takes place
distinguish
In
in
to exploit
between production
each date, some countries have
"good" production functions that exhibit high average productivity, while other
countries have "bad" production functions that exhibit low average productivity.
normal times, production functions do not change but output
is
uncertain.
We
In
use the
The importance
of consumption-smoothing depends on the frequency of the data one is
obviously important for the analysis of quarterly data (most people spend more
than they earn over Christmas and other holidays, and somewhat less than they earn in other
times), and almost as surely is a bad theory for understanding savings rates over a quarter of a
century. See Deaton (1992) for a survey of evidence on intertemporal models of savings.
^
analyzing.
^
It
Moreover,
economies
is
this
that
assumption is consistent with the strong home equity preference in OECD
has been documented by French and Poterba (1991) and Tesar and Werner
(1992). Lewis (1995) surveys alternative explanations for this
phenomenon.
term output shock to refer to production surprises. These shocks do not affect the
probability distribution of future productivity and,
income or wealth
effects
as a
in their
economic environment
level of productivity,
that
change
These events have
production functions to "good" ones, or vice versa.
on the average
they have only transitory
on investors. Occasionally, countries perform economic
reforms or experience changes
effects
result,
and we
label
them
their "bad"
persistent
productivity shocks.
Since productivity shocks change the probability distribution of future productivity,
they both have income or wealth effects on investors, and also affect their investment
strategies.
In
our basic model,
we assume
aversion and have no labour income.
that investors exhibit constant relative risk
As a
result, the
shares
wealth invested
of
domestic capital and foreign loans depend only on asset characteristics,
returns
and
volatilities.
Since these are not affected by output shocks,
the marginal unit of wealth (the output shock)
is
in
in
we
find that
find that positive output
these countries, and
creditor countries. This distinction
we
expected
invested as the average one
Since countries with high productivity are debtors,
lead to current account deficits
i.e.
does not apply
to current
in
is.
shocks
account surpluses
to productivity shocks.
We find
instead that favourable productivity shocks always lead to current account deficits, as
investors react to the increase
in
the expected return to domestic capital by
increasing their holdings of domestic capital and reducing their holdings of foreign
loans.
The usefulness
of this
benchmark model
is
that
it
highlights the set of
assumptions that underlie our example: shocks have only transitory income
investors exhibit constant relative risk aversion,
We then
proceed
to relax
and there
these assumptions.
is
First,
effects,
no labour income.
we
find that
if
relative risk
aversion decreases with wealth, positive output shocks raise wealth and induce
investors to take riskier investment positions.
As a
invested
share
if
in risky
labour income
domestic capital exceeds
is
its
result, the
in
share of the shock
wealth. Second,
we show that,
less risky than capital income, positive output shocks raise the
ratio of financial to
human
wealth and hence expose the investor to greater
induces investors to take safer investment positions
the share of the shock invested
wealth.
We obtain a simple
in
domestic capital
rule to
in their financial
falls
short of
its
risk.
This
wealth, and so
share
in financial
determine when a positive output shock leads to a
current account deficit: the country's debt has to exceed a certain threshold that
depends on how
attitudes towards risk vary with wealth
and the size
income. This threshold can be either positive or negative, and
constant relative
risk
Our research
account.^
The
is
of labour
zero
in
the case of
aversion and no labour income.
naturally relates to existing intertemporal
early generation of intertemporal models,
models
of the current
such as Sachs (1981,1982)
Obstfeld (1982), Dornbusch (1983) and Svensson and Razin (1983), were designed
to
study the effects of terms of trade shocks and to develop rigorous theoretical
foundations for the Harberger-Laursen-Metzler
effect.
We
share with these models
the notion that countries save transitory income shocks so as to
smooth consumption
over time. However, since these models abstract from capital accumulation, income
shocks can only be invested
transitory
in
income shocks lead
Simply allowing
result of this paper,
foreign loans.
to current
for capital
As a
result they predict that positive
account surpluses
accumulation
is
countries.
not sufficient to obtain the main
however. Subsequent contributions by Sachs (1981), Persson
and Svensson (1985) and Matsuyama (1987) extended the
models
in all
to include capital
models were designed
to
early intertemporal
accumulation by investors with perfect foresight. These
analyze the current account response to persistent shocks
to the profitability of investment.''
that the return to investment
is
Since the assumption of perfect foresight implies
certain, arbitrage requires that the marginal product of
capital equal the world interest rate. This condition,
combined with the assumption
diminishing returns at the country level, uniquely determines the domestic stock of
"
See Obstfeld and Rogoff (1 995)
These shocks correspond to our
for
a survey of these models.
productivity shocks.
of
capital independently of the country's wealth.
raise wealth but
in
do not
Hence, transitory income shocks which
affect the marginal product of capital are again only invested
foreign loans, leading to current account surpluses
One can understand
important implications for
Once investment
is
in all
countries.
our contribution as recognizing that investment
how
risk
has
the current account reacts to transitory income shocks.
modelled as a
equates the return on investment
risky activity, the appropriate arbitrage condition
to the world interest rate plus
a risk premium. Since
the latter increases with the share of wealth held as risky domestic capital, transitory
income shocks which do not
must
in
part
In particular,
capital
be invested
we
in
domestic capital
find that the
do
affect the profitability of investment, but
for the arbitrage condition to
share of the income shock that
exceeds the income shock
itself in
raise wealth,
is
invested
debtor countries, but not
in
be
in
satisfied.
domestic
creditor
countries.^
The paper
is
organized as follows: Section
1
develops the basic model.
Section 2 presents the main result of the paper. Section 3 explores the robustness of
this result.
Section 4 presents empirical evidence for thirteen
OECD
countries.
Section 5 concludes.
economy in which
accumulation and investment risk. The latter arises from a stochastic
depreciation rate. This model is used to show that cross-country differences in the rate of time
preference could explain the Feldstein-Horioka finding that savings and investment are highly
correlated in a cross-section of countries. Interestingly, he finds a U-shaped relationship
Zeira (1987) provides an overlapping-generations model of a small open
there
is
capital
between the steady-state level of debt of a country and its rate of time preference. He does not
however explore the effects of transitory income shocks, as we do here.
A Model
1.
of International
The world
equilibrium
international borrowing
Borrowing and Lending
model presented here
and lending
is
average
some
average
assume
stock markets.
enough
This
is
r,
all
which
We
own
their capital stocks
assume
is
determined
enough
in real
production functions.
In
We
that international loans
to
in
of equity in
markets
is
high
the domestic stock market.
generate the strong home-equity
economies.^
We draw a distinction
the state of technology.
world equilibrium.
that the cost of operating in foreign stock
an extreme, yet very popular device
preference observed
in
and are financed by sales
domestic investors and firms trade
that only
that
countries are allowed to borrow and
that the penalties for default are large
are riskless. Firms
the rate of time preference.^
have "bad" production functions
productivity. Investors in
lend from each other at an interest rate
We
in
countries have "good" production functions that exhibit high
productivity, while other countries
exhibit low
that
results from differences in investment
opportunities across countries rather than differences
At each date,
based on the view
between production uncertainty and random changes
in
normal times, countries have time-invariant but stochastic
use the term output shocks to refer to production surprises
which occur during these normal times. Since these shocks do not
probability distribution of future productivity, they
have only
affect the
transitory
income or
wealth effects on investors. Occasionally, countries perform economic reforms or
experience other changes
on
their
average
in their
economic environment
level of productivity.
and model them as random changes
®
See
See
in
label these events
have persistent effects
as productivity shocks
the production function. Since productivity
and cLarida (1990) for world equilibrium models in which borrowing and
motivated by cross-country variation in rates of time preference.
Obstfeld (1994) for a discussion of the effects of financial integration in a model similar
Buiter (1981)
lending
^
We
that
to ours.
is
shocks change the probability
income or wealth
effects
A number of
j=1,2,...
.
on investors, and also
is
used
for
many
atomistic countries, indexed by
we
we assume
that there exists a single
consumption and investment. This device permits us
intertemporal trade
Finally,
infinitely
This allows us to rule out large-country effects and concentrate on the pure
effects of international linkages. Also,
which
affect their investment strategies.
simplifications serve to highlight the bare essentials of our
We consider a world with
arguments.
have
distribution of future productivity, they both
restrict
and eliminate the complications
that arise from
our analysis to the steady state of the model
average growth and the
in
to
focus on
commodity
shocks as opposed
to global shocks.
trade.
which both world
interest rate are constant. This allows us to focus
effects of country-specific
good
on the
While these
assumptions simplify the analysis considerably, we are convinced that removing them
would not
affect the thrust of our
arguments.
Firms and Technology
Production
is
random. Let
qj
and
kj
be the cumulative production and the
stock of capital of the representative firm of country
technology of
this country.
representative firm
Conditional on
is
E[d9j ]=dt
and E[d9jd9m]=0
depreciation)
if
that, conditional
in
ttj
as the state
of
the production function of the
+ akj-dej
a positive constant and the
which states
Also, define
is:
dqj =7ij-kj-dt
where a
ttj,
j.
country
j
is
GjS
(1)
are Wiener processes with E[dej]=0 and
j>m. Equation
on the state
(1) is
simply a linear production function
of technology, the flow of output (net of
a normal random variable with instantaneous mean
;
and variance-covariance matrix defined by E[dqi^]=a^kf
E[dqj]=7rjkjClt
E[dqjdqni]=0
not
j^vn.
if
change the
Realizations of
dt
and
d0jS are output shocks. Since these
tine
probability distribution of future productivity, they
shocks do
have only income or
wealth effects on the owners of the firm, but do not affect their investment strategies.
Average
productivity varies across countries
of the countries are in a high-productivity regime,
and over
=n
n-^
,
=n, withn < k The dynamics
low-productivity regime, n^
.
each date,
time. At
while the other half are
of
follow
Kj
half
in
a
a Poisson-
directed process:
with probability
fO
dTii
=
\n-K
where
g(7i;i)
=
•{_
\n
'
—n
if
TCj
.,
'
If
7ii
=
7t
=
7t
E[dKjd9j]=0 and E[dTtjd7rm]=0
rare events
(i.e.
;
if
(j),
dt
(j)dt
n and n are
Equation
j;^m.^
positive constants with
(2) states that
they occur with probability that goes to zero
time) that are uncorrelated across countries
(1).
(j)
(2)
with probability
\g{n-.)
'
-
1
i
Since the probability of a change
in
changes
in
the
is
in
regime are
limit of
and with the output shocks
regime
n - 5 < a^
in
continuous
Equation
small, productivity levels are
persistent. Since high(low)-productivity countries expect productivity eventually to
decline (increase), productivity levels also exhibit mean-reversion. Realizations of the
dTTjS
are productivity shocks. Since these shocks change the probability distribution of
future productivity, they
and they also
have both income or wealth
effects
on the owners
of the firm,
affect their investment strategies.
Since there are
infinitely
productivity countries
many
countries,
change regime.
regime, half of the countries
will
It
each period a
follows that
always be
in
if
fraction
(t)dt
of the high(low)-
half of the countries
each regime.
are
initially in
each
There are many
technology.
The
identical firms in
representative firm
is
each country with free access
divided into
kj
to existing
shares which have a (constant)
value of one and deliver an instantaneous dividend equal to the flow of production
per unit of capital.
We assume that the
distributions of the current
shocks
d9j
realizations of past
and
dTij
are
shocks and the
known by
probability
investors before
production starts. However, the realizations of the contemporaneous shocks d9j and
dnij
are only
investors
known
after production is
must commit
their
completed and output
resources before production
is
observed. Since
starts, their
investments are
subject to uncertainty related to the contemporaneous realizations of the shocks.
Since investors can freely trade equity after production
investments are not subject to uncertainty related
It
=
—
-
dt
+ d9j Therefore the expected
.
the representative firm are
tij
and
completed,
their
to future realizations of the
follows that the return process perceived by investors
doOj
is
return
and
is
Tijdt+adcoj,
volatility of
shocks.
where
holding a share of
o, respectively.^
Consumption and Investment Strategies
Each country contains many
utility
identical
consumer/investors with a logarithmic
function:
EjlnCj-e-P'dt
(3)
Despite the fact that productivity shocks affect the return to investment, they do not contribute
mean and variance of the return process since both they occur infrequently (with
probability of order dt) and they are small (with magnitude of order dt).
to the
where q
is
the consumption of the representative
be the wealth
respectively.
constraint
of this
We
consumer and
assume
consumer
the share of wealth that
that aj(0)>0 for
all
j.
in
is
country
held
in
Let
j.
Hj
and
Xj
equity,
Then, the consumer's budget
is:
daj
= [((jtj - r)
+ r)
Xj
•
- Cj
aj
dt
+ aXj-aj-dcOi
(4)
This budget constraint illustrates the standard risk-return trade-off behind investment
decisions.
If 7i:j>r,
expected return
by aaj.
In
increases
to wealth
Appendix
1
in
by
the share of wealth allocated to equity raise the
(nj-r)aj
we show
,
,
at the cost of raising the volatility of this return
that the solution to the
consumer's problem
is:
Cj=p-aj
-r
Tt|
Equation
(5)
consumption
(5) states that
of asset characteristics
i.e.
substitution effects of
changes
consumers. Equation
(6)
depends only on asset
aj.
This
is
and
r, tij
in
shows
is
a fixed fraction of wealth and
c. This is the
is
independent
well-known result that income and
asset characteristics cancel for logarithmic
that the share of wealth allocated to
characteristics,
i.e.
r, tcj
and
nothing but the simple investment rule
introduction.
10
a,
and not on the
we used
in
each asset
level of wealth,
the example
in
the
World Equilibrium
To
find the
world interest
^(aj -kj) =
international loans,
n=
lim
—
•
j^~ J
^
—
'-^j
are constant.
and the investment
rule in
Equation
(6) to obtain:
In
.
Appendix
1
we show
that there exists
a
lim-.Tai
j^-J
in
market-clearing condition for
(7)
Kj
'-'
steady-state
0,
tlie
= n-a^
T
where
we use
rate,
.^,
which both the world growth rate and the world average productivity
In
what
follows,
we assume
that the world
economy
is
already
in this
steady state.
Using the world interest rate
Equation
(6)
we
71:
ki
= 1+
.
and
its
Equation
(7)
and the investment
rule in
find that the world distribution of capital stocks is given by:
''
Equation
in
-7t^
'
•^i
o'
J
(8)
(8) states that the capital
stock of a country
is
increasing
productivity. Interestingly, this world of stochastic linear
in
both
its
wealth
economies does not
generate the usual corner solution of a world of deterministic linear economies
which
all
the capital
productivity. In the
is
located
presence
in
the country or countries that have the highest
of investment risk, these
are ruled out by investors as excessively
risky. In fact,
extreme investment strategies
since
we have assumed
productivity differences are not too large relative to the investment risk,
S-u < a^
,
all
in
countries hold positive capital stocks
11
in
equilibrium.
i.e.
that
Although world average growth
To see
cross-section of growth rates.
is
constant, this world
this,
economy
exhibits
a
rich
substitute Equations (5)-(7) into (4), to find
the stochastic process for wealth:
da.
7ti
-71
+ n-c^ -p
a+-
TZi-n
dt
+ a+-
dCO;
(9)
a.
The growth
rate consists of the return to the country's wealth
to wealth ratio.
and
is
The
first
term
in
Equation
(9) is the
minus the consumption
average or expected growth
rate,
larger in high-productivity countries, since these countries obtain a higher
average return on
component
in
their wealth.
the growth rate,
The second term
and
is
more
in
Equation
(9) is the
unexpected
volatile in high-productivity
countries, since these countries hold a larger fraction of their wealth
12
in
risky capital.
^
2.
Determinants of the Current Account
The model developed above describes a world
equilibrium
in
which high-
have
productivity countries borrow from low-productivity countries since the former
access
investment opportunities than the
to better
productivity countries borrow
this, let
fj
is
limited only
be the net foreign assets
by
of country
j,
latter.
The amount
bear
their willingness to
i.e. fj=aj-kj
,
that highrisks.
To see
and use Equation
(8) to
find that:
Since n < n <
jt
high-productivity countries are debtors, fj<0, while low-productivity
,
countries are creditors,
risk,
fj>0.
Equation (10) shows
the volume of borrowing and lending
is
that, for
a given
level of
investment
larger the larger are the cross-country
productivity differentials. Also note that, for a given productivity differential, the
volume
of
borrowing
country can
is
move from
larger the lower
is
the investment
risk. Finally,
lender to borrower (borrower to lender)
if
observe that a
and only
if it
experiences a positive (negative) productivity shock.^°
Next
First,
we
we examine
the behavior of the curent account
derive the stochastic process for the current account and
salient features.
determine
how
Second, we provide an
intuition
to the
^°
sudden and large current account
^^
in
main economic forces
deficit that turns
debtor should be seen as a positive development. This notion
held beliefs
comment on
its
that
we emphasize
the current account between debtor and creditor countries.^
in
world, a
as
world equilibrium.
the current account responds to shocks. Throughout,
the differences
in this
in this
is
a country from creditor to
clearly at
odds with widely-
policy circles.
The reader might ask why emphasize
productivity distinction.
the debtor/creditor distinction instead of the high/low
There are two reasons. From a theoretical viewpoint, one could defend
13
Current Account Patterns
Since the current account
Ito's
lemma
to (10)
is tfie
and use Equation
ctiange
net foreign assets,
in
i.e. dfj,
(9) to find tfie following stochastic
we
process
apply
for
foreign assets:
/
df;
=
Kj
a+
-71
4-71-a^
-
-p f-dt+|a+^
'-
fdcOi+f
^
(11)
J
Equation (11) states that net foreign assets follow a mixed jump-diffusion process
and provides a complete characterization
of their
dynamics as a function
of the
du:
forcing processes dOj
and
parameters of the model
Consider
account
in
first
drtj
a, p,
(remember
(j),
that dcO:
n and n
the prediction of the
dt
+ dB:), andallthe
.
model
for the
average or expected current
debtor and creditor countries. Taking expectations of
E[d,,]
=
TCi
+
-
-Tt
(1 1)
yields:
<t)g(Ttj)
+ n-<f -p
fdt +
•f-dt
Kj
-n
(12)
choice by pointing out that the debtor/creditor distinction is more robust, in our model, only
cross-country differences in average productivity determine whether a country is a debtor or a
creditor. If we assume, for instance, that high average productivity is associated with high
this
production, it is then possible that the high-productivity technology might be
unappealing enough to turn high-productivity countries into creditors (see Devereaux and Saito
(1 997)). In this case, all the results presented in this paper would still hold true for the
volatility in
debtor/creditor distinction, but not for the high/low productivity distinction. From an empirical
viewpoint and anticipating that the predictions of the model will be confronted with the data,
one could defend the use of the debtor/creditor distinction based on the observation that
existing measures of net foreign asset positions of countries are of
those of their average productivity.
14
much
better quality than
'
Equation (12) separates the expected current account into two pieces, whicli capture
and expected changes
the effects of expected savings
The
term reflects consumption
first
"tilting"
by agents.
in
If
asset returns, respectively.
the expected return to
wealth exceeds (does not exceed) the rate of time preference,
'
-71
Ttj
a+—
to
^
+
their
"tilt"
Ti
-a,2 > (<)
consumption
same
assets
in
means
that, in
deficits,
the
p
,
consumers
profile.
will find
These savings
or dissavings are allocated across
proportions as the existing stock of wealth. Ceteris paribus, this
growing economies, debtors countries on average run current account
economies with negative growth. The second term
effects of
mean
above
long-run average.
reversion
The opposite
It
expected
is
(increase) their holdings of capital
means
is
Equation (12) captures the
in
to decline, while the
in
converse
is
on average experience current account
The balance
growing economies, and depends on
in
true for
and instead lend (borrow from) abroad. Ceteris
that debtor countries
ambiguous
is
is
productivity induce investors to reduce
surpluses, while creditor countries on average experience deficits.
the two effects
true in
productivity. Since productivity in debtor countries
in
creditor countries. Reductions (increases)
paribus, this
optimal to save (dissave) so as
it
while creditors have surpluses on average.
its
i.e.
all
of
the
parameters of the model. The faster the economy grows and the smaller the meanreverting
component
of productivity
run current account deficits
and
is,
the
more
likely
it
shocks
To see
this,
note that
it
that
on average debtors
creditors run current account surpluses.
Both output and productivity shocks contribute
account.
is
to the
follows from Equation (11)
variance of the current
and the
definitions of the
that:
Var
[dfj
=
\2
r
J
V^i-^7
2
rV^\\
a+^
(l)fdt
^
15
(13)
In
normal times
with probability close to one)
(i.e.
cl7i:j=0
and
fluctuations in the current
account are driven by the output shocks. These shocks have small effects
dt'"^)
but occur with high probability (of order
the current account
dates
captured by the
first
order
Their contribution to the variance of
some
term of Equation (13). At
with probability close to zero) dnj^tO
(i.e.
account
is
1).
(of
and the behavior
infrequent
of the current
dominated by productivity shocks. Although these shocks occur with small
is
probability (of order dt), they
do have large
effects
on the current account
(of
order 1)
since they induce a reallocation of investors' portfolios. Their contribution to the
variance of the current account
is
captured by the second term of Equation (13).
The Current Account Response
We
are
now ready
that the
response
country
is
n-
a+—
to
to
Shocks
examine perhaps the most novel
of the current
finding of this paper,
account to an output shock depends on whether a
a debtor or a creditor. This result follows directly from Equation
(11).
Since
-K
>
,
a positive output shock,
i.e.
d9j>0, leads to a current account deficit
in
o
debtor countries and a current account surplus
intuition for this result,
we
in
To develop an
creditor countries.
focus on the savings-investment balance.
The permanent-income consumers who populate our world economy save
in
order to smooth their consumption over time. Since the output shock represents a
transitory increase
of
in
whether a country
models
income,
is
it
is
saved
a debtor or a
(recall
creditor,
Equation
and
is
a
(9)).
This
is
true regardless
typical feature of intertemporal
of the current acount.
Having decided
to
save the output shock, investors must then decide how
allocate these additional savings
between domestic equity and foreign loans.
16
to
We
depart from previous intertemporal models of the current account
how we model
in
Since the investor's desired holdings of equity are equal
this decision.
country's stock of capital
in
equilibrium, Equation (6)
to the
can be interpreted
in
terms
of
a
familiar arbitrage condition:
=r + cT^--!-
7t
(14)
Equation (14) states that expected rate of return
interest rate,
r,
return to equity
and the
domestic capital
premium
Kj,
must equal the world
plus the appropriate risk or equity premium, o^(k/ai).
logarithmic investors, this risk
risk
to equity,
premium
is is
nothing but the covariance between the
return to the investors' wealth.
investors' wealth, the larger
in
world of
In this
is
this
The
larger
is
the share of
covariance and the larger
that investors require to hold the marginal unit of equity.
The
is
the
additional
savings that result from the output shock allow investors to increase their holdings of
risky
domestic equity without increasing the
keep the share
of equity in their portfolios constant.
(the output shock) is invested in the
definition the
risk of their portfolios,
same
provided that they
Thus, the marginal unit of wealth
proportions as the average one. Since by
share of a debtor country's wealth devoted to domestic capital exceeds
one, an increase
in
wealth (savings) results
in
a greater increase
in
domestic capital
(investment), leading to a deficit on the current account. Conversely,
countries the increase
wealth increase
is
in
in
creditor
wealth exceeds investment at home, as a portion of
invested abroad. This produces a current account surplus
this
in
creditor countries.
This discussion emphasizes the importance of allowing for investment
predicting the current account response to
form of uncertainty
investment
is
is
important, existing
a riskless
activity,
an output shock. To the extent
models
of the current
risk in
that this
account that assume
or else abstract entirely from capital accumulation.
17
provide a misleading description of
how
the current account responds to output
shocks.^^
We now turn to the
o+—
since
of the current
account
>
,
the second term of Equation
o
productivity shock,
account
deficit in
This effect
i.e. d7ij>0,
(1 1
shocks.
)
shows
that
a positive
generates an income effect that leads to a current
debtor countries and a current account surplus
formally equivalent to that of an output shock
is
to productivity
-n
n-
First,
response
in
creditor countries.
and requires no
further
discussion.
Second, a productivity shock has a rate-of-return
changes the
effect
probability distribution of future productivity.^^
on investment since
When
it
a creditor (debtor)
country receives a positive (negative) productivity shock, the expected return to
equity increases
their portfolio in
reflected
in
of this investment
the current account are
(of
they would
^^
The
result,
response
is
generates an investment
a current account
boom
(bust).
deficit (surplus)
and
the third term of Equation (11). Since rate-of-return effects of
shocks consist
productivity
shocks
This induces investors to hold a larger (smaller) fraction of
domestic equity and, as
The counterpart
is
(falls).
order
dt).^"
in
larger (of order 1) than the
large spikes
premium observed
feature of real economies.
should be larger
much
income effects
of the
on
same
Since productivity shocks are infrequent but have large effects,
show up as
large equity
of reallocations in the stocks of assets, their effects
A
in
in
a time series of the current account.
is an important
premium, o^+Hj-n,
our knowledge, this result is new and has
the data suggests that investment risk
prediction of this
model
debtor countries. To the best of
been tested yet.
As Equation (5) shows, income and
is
that this equity
not
^^
substitution effects of changes in the expected return to
our world of logarithmic consumers. In models with more general preferences
these rate-of-return effects of productivity shocks could be associated with consumption
booms or busts, depending on the balance of their income and substitution effects.
^*
And, for that matter, much larger than the income effects of output shocks (of order dt'^).
equity cancel
in
18
3.
Investment Strategies
The theory developed above
output shocks
this result
is
different in debtor
predicts that the current account response to
and
creditor countries. Instrumental in deriving
were our assumptions regarding how investors trade
These assumptions ensured
foreign loans coincide.
is
and
return.
and average propensities
that the marginal
However, there
risk
a long and distinguished literature that
analyzes how optimal investment strategies depend on attitudes towards
size
and stochastic properties
and changes
returns
in
A
risk,
the
income and the correlation between asset
of labour
the investment opportunity set
investor's environment.^^
to invest in
general finding of
and other aspects
this literature is that
of the
one should not
expect that marginal and average propensities to invest coincide.
The purpose
holds
in
show
of this section is to
that a modified version of our result
a generalized model that allows attitudes towards
wealth and introduces riskless labour income. ^^
here, marginal
find
and average propensities
a simple rule which determines
account
deficit:
attitudes towards risk vary with wealth,
risk
the generalized
positive output
and
(2)
shocks lead
that
However, we
to
a current
depends on
(1)
how
the size of labour income. This
is
zero
aversion and no labour income.
result that favourable output
model presented
shock leads
exceed a threshold
threshold can be either positive or negative, and
constant relative
vary with the level of
to invest in foreign loans differ.
when a
the country's debt has to
In
risk to
to current
in
the
benchmark case
We therefore
account
of
have the modified
deficits in sufficiently
indebted countries. Otherwise, they lead to current account surpluses.
See Merton (1995) for an overview of this research, and Bodie, Merton and Samuelson
(1992) for an example with risky labour income.
do not explore the implications for our argument that arise from the possibility that asset
We
be correlated with changes in the investors' environment. These correlations give rise
a hedging component in asset demands that greatly depends on the specifics of the model.
returns
to
19
Two Extensions
To
allow attitudes toward risk to vary with the level of wealth,
following Stone-Geary
utility
we adopt
the
function:
oo
Ejln(Cj+pj)e~P'dt
where the
PjS
(15)
are constants, possibly different across countries.
—
The
coefficient of
C:
relative risk aversion,
i.e.
,
For a given level of consumption or wealth,
important for our purposes,
relative risk aversion
as
if
risk
uses labour
to
aversion
is
time, as follows.
decreasing
in Pj.
their level of
consumption increases.
produce the single good.
^^
that there
is
^^
an additional technology
Normalizing the labour force of each
country to one, the flow of output produced using the second technology
A,j
•
dt
.
Labour
productivity,
?ij
,
is
assumed
to
be constant although
across countries. Workers are paid a wage equal to the value of
product,
i.e.
^jdt.
More
Pj<0 (Pi>0), investors exhibit decreasing (increasing)
To introduce labour income, we assume
that
and over
varies across countries
The existence
of labour
it
their
income complicates only
is
given by
might vary
marginal
slightly the
consumer's budget constraint:
cbj =[((7tj -r)-Xj
+r)aj +^j -Cj]dt + Xj
^^
a^
adcOj
(16)
As is well-known, consumers with Stone-Geary preferences might choose negative
consumption. We ignore this in what follows.
The assumption of an aggregate linear technology between labour and capital is much less
restrictive that it might seem at first glance. It arises naturally in models where some form of
factor-price-equalization theorem holds. One could, for example, use the model in Ventura
(1997) to endogenously generate a linear technology. We do not do so here to save notation.
20
To ensure
that
that 3,(0) >
The
countries hold positive capital stocks
all
representative
follows the
this
dynamics
in
consumer
and the
Xi
(2). In
depend on both
income,
X,j=0,
country] maximizes (15) subject to
equilibrium) belief that
in
Appendix
1
,
we show
J
that the solution to
illustrate
o
how
and investment
optimal consumption
and the presence
of labour
exhibit constant relative risk aversion, |3j=0,
Equations (17) and (18) reduce
model (Equations
equity cancel. Equation (18)
decreases with wealth
consumers
constant and
-r
(5)
to the
and
(6)).
if
shows
and only
if
As
To
income. Note
and there
is
first
no labour
before, consumption
in
that the share of wealth
X.j+Pj>0.
rules
consumption and investment
wealth and income and substitution effects of changes
Pj>0,
is
J
Tti
rules of the previous
r
is:
attitudes towards risk
consumers
in
(17)
'
=
Equations (17) and (18)
if
we assume
a+-^
r-a^
if
(correct
Equation
y
in
residing
generalized consumer's problem
Ci=p-
that
equilibrium,
n-G^
the budget constraint (16)
Ttj
in
linear
the expected return to
devoted
to equity
interpret this condition, note
exhibit increasing relative risk aversion,
is
and so choose
first
that
to allocate
a smaller share of wealth to risky domestic capital as their wealth increases. Second,
note that
in
the presence of riskless labour income,
wealth raise the
ratio of financial to
investor to greater
risk.
In
human
A.j
>0, increases
in financial
wealth, and, ceteris paribus, expose the
response, agents adopt less aggressive investment
21
strategies,
and the share
Finally, holding
i.e.
of financial wealth
devoted
to risky
domestic capital
constant the level of wealth, investors with low relative
high values of
high values of \,
risk
falls.
aversion,
and/or a relatively large stream of riskless labour income,
pj,
devote a larger share of
will
i.e.
wealth to risky domestic capital.
their
World Equilibrium
To compute
we impose once
the world equilibrium interest rate,
J
1
market-clearing condition for international loans, lim
investment rule (18) to find that Equation
1
-
•
^^aj -
valid.
still
kj
=0
Appendix
1
and use the
shows
that,
if
"^
--^^.j +Pj =
~*°°
lim
^
(7) is
again the
,
there exists a steady-state
which both the world growth rate
in
j=i
and world average
productivity are constants. In
restriction regarding the
the world
economy
The world
is in
what
follows,
we assume
cross-country distribution of parameters
is
that this
satisfied
and
that
the steady state.
distribution of capital stocks
is
now
given by a straightforward
generalization of Equation (8):
ki
=
(
^i+^A
(
1
+
T^\-A
'
(19)
-
<J
As
before, the capital stock of a country
wealth,
and moreover, the world
the restrictions that 3,(0) >
71
increasing
in
both
distribution of capital stocks
—
-a
—
is
J
j-
and ti-ti < a^
.
22
,
jointly
Its
is
productivity
and
non-degenerate since
ensure that
all
countries
hold positive capital stocl<s
have a high tolerance
have large domestic
We find,
world
economy
in
for risk
whose
residents
and/or high labour productivity,
high
and/or
exhibits
Equations (17)-(18) and
a
that while the world
rich
average growth
cross-section of growth rates.
(7) into
+ K-a^-p
a+-
i.e.
Pj
X,j,
capital stocks.
once more,
da.
equilibrium.^^ In addition, countries
the budget constraint
1
•dt
To see
is
constant, the
this,
Equation (16)
substitute
to obtain:
L +a
-Tt
Kj
^i+Pi
+
in
rate
+ o+
1
a(7r-(/)
+
dco,
a(7r-o^)
(20)
As
before, high productivity countries
grow
faster
and experience more
volatile
growth than low productivity countries. Perhaps the most interesting difference
between
this
(/\.j+Pj<0),
growth rate and the special case
Equation
both the growth rate of a country and the
(increase) with the level of wealth. This
described n Equations (18).
share of
in
their
If
is
(9) is that,
volatility of
if
^j+Pj>0
the growth rate decline
a consequence of the investment strategy
X,j+Pj>0 (>.j+Pj<0), investors invest
a smaller
(larger)
wealth to risky domestic capital as their wealth increases, lowering
(raising) both the
expected return on
their
wealth and the
volatility of that return.
Determinants of the Current Account
We are now ready to examine the
more general model. The world
behaviour of the current account
distribution of loans
is
in this
given by the following
generalization of Equation (10):
19
The
first
parameter
restriction
ensures the
first
23
parentheses
in
Equation (19)
is
positive.
:
^2
)
As
before, the
first
advantage
term
in
in
which, conditional on the level
borrow from low-productivity countries
home. This
of better investment opportunities at
Equation (21).
In addition,
characteristics of labour
is
order
in
reflected
in
the
the international pattern of borrowing and
lending reflects cross-country differences
who have a
a
[
model describes a world equilibrium
of wealth, high-productivity countries
to take
(21)
n-a
attitudes towards risk
in
and the
income across countries. Countries populated by investors
high (low) tolerance for risk and/or a large (small) stream of riskless
labour income ?ij+Pj>0 (^j+(3j<0) are, ceteris paribus, more
We find the
current account by applying
Ito's
likely to
lemma
to
be debtors.
Equation (21) and
using Equations (17) and (20)
\2
f
Ttj
a+—
dfjV
+ 7c - a - p
•dt
K-O
/
-7t^ (
+ a+-
X-.
+P
\
fi+-
7t-a^
J
f
•dcoj
+
J
V
^i+pi
V n-c
(22)
The
current account again follows a mixed jump-diffusion process, with dynamics that
can be completely characterized
all
in
terms of the underlying shocks,
the parameters of the model, a, p,
iim-T'^i +p: =0,
(j),
k ,n
,
X,j
and
Pj.
the results of the previous section
Since
d9j
and
drtj,
and
we have assumed
may be
that
thought of as
describing the determinants of the current account for an average or "typical" country
where
>.j+Pj=0.
Comparing Equations
(22)
and
24
(11),
we see
that they are identical
drt:
K;
-n
—
—^
provided that
we
generalize
a straightfonward manner provided that
in
replace
used. Accordingly,
we
fj
restrict
insights obtained from the
to output
shocks.
Since o +
K- -n
—
>
account
with
+
i
Thus,
.
n-a
ail
of the results in Section
this
modified definition of debt
is
ourselves here to a brief discussion of the additional
more general mode! regarding the response
,
2
Equation (21
)
shows
of the current
that a positive output shock,
i.e.
o
d9j>0, leads to a current account deficit
countries where
in
fj
—
^i+Pi
+—
^<
and a
n-a
surplus
in
countries
where
fj
—^
+—
>
.
This result
is
again best understood
in
n-a
terms of the savings-investment balance. As
savings behaviour
consumption
in
arbitrage condition
in
a'-^-7
aj
the
premium
is
is in
we
results differ from
are allocated across
obtain the following generalization of the
^—r
'-^
(23)
(K-a^)-aj+^j+Pj
The
the product of two terms.
equity returns
domestic equity,
Where our
how these savings
on equity and the return on an investor's
volatility of
of the previous section,
Equation (14).
K =r +
return
model
the face of transitory income shocks.
assets. Rearranging Equation (18),
risk
the
very standard and reflects the desire of agents to smooth their
is
the existing current account literature
The
in
o^-(k/aj).
and the
larger
The second term
25
is
is
first is
portfolio,
the covariance between the
which
is
greater the larger
the share of wealth invested
is
in
the coefficient of relative risk aversion
(7i-a^)-aj
—r
of the investor's value function,
section ?.j+PpO and this coefficient
depends on both
was equal
income. Most important
for
our results
is
invested
in
falls (rises)
shock
now
that raises investors' wealth.
exactly the
if
relative
that this term
wealth provided that Xj+Pj<0 (^j+Pj>0). Suppose
positive output
same
If
and Equation
(23)
more general
setting,
it
importance of riskless labour
is
decreasing (increasing)
in
that a country experiences a
the marginal unit of wealth
proportions as the average
>lj+Pj<0 (>.j+Pj>0),
the model of the previous
to one. In this
and the
attitudes towards risk
In
.
the overall risk
unit,
no longer holds. Hence,
arbitrage condition to be satisfied, the marginal unit of wealth invested
domestic capital must exceed (be less than) the average
is
premium
for the
in risky
unit.
This result qualifies the relationship between debt and the response of the
current account to output shocks.
(debtor)
If
^j+pj<0 (/Vj+Pj>0), a country
and yet experience a current account
favourable income shock.
If
may be
deficit (surplus) in
relative risk aversion
a creditor
response
to
a
decreases with wealth, positive
output shocks that raise wealth induce investors to take riskier investment positions.
As a
result, the
exceeds
some
its
share of the marginal
share
in
unit of
wealth invested
average wealth. Depending on the magnitude
creditor countries might run current
account
less risky than capital income, positive output
human wealth and hence expose
to take safer
^°
The
in
deficits.
domestic capital
in their financial
falls
short of
its
risk.
wealth,
share
of this effect,
is
ratio of financial to
This induces investors
and so the share
in financial
coefficient of relative risk aversion of the value function tells us
values different
domestic capital
Since labour income
shocks raise the
the investor to greater
investment positions
shock invested
in risky
of the
wealth. Hence,
how the consumer
as apposed to the coefficient of relative risk aversion of the
utility function, which tells us how the consumer values different lotteries over consumption.
The latter depends only on preferences, while the former depends on both preferences and
other aspects of the consumers' environment.
lotteries in wealth,
26
some
debtor countries might run current account surpluses
in
response
to
a
favourable output shock.
In
account
summary, we
to
threshold
find
a simple rule to determine the response of the current
a favourable output shock.
-fj
>
—
'
—^
,
If
the level of debt exceeds the following
then favourable output shocks lead to a current account
Tt-a
deficit.
Otherwise, they lead to a current account surplus. This threshold can be
positive or negative,
Finally,
and
in
the special case of the previous sections
note that using Equation (21)
we can
rewrite this condition as
high-output shocks lead to current account deficits
current account surpluses
Once
again,
remember
in
equal to zero.
7tj>7i.
That
high-productivity countries
low-productivity countries.^^
that this
differences across countries
in
is
is
a consequence
in volatilities.
See
of
our assumption that there are no
footnote 11.
27
is,
and
4.
Empirical Evidence
In this section,
we
present
some
preliminary empirical evidence that broadly
supports the theory developed above. This evidence
as suggestive that
of the theory, but rather
behaviour of the current account
and
in
we
not intended as a formal test
is
are capturing
the real world.
We
some aspects
of the
begin by assuming that
X-^+^i
are unobservable. Hence, the threshold level of debt above which output
TCj
shocks lead
to current
account
cannot be observed. Under
deficits
this
assumption,
the content of our theory can be understood as a probabilistic statement that the
higher
is
the level of debt of the country, the
lead to current account deficits.^^
We take
likely
per capita
favourable output shocks
GNP
growth as an imperfect
shocks emphasized by the theory. This measure
measure
of the
does not
distinguish
between output and
output shocks are present
in
the data,
the current account with per capita
levels of debt. Accordingly,
of
more
we
study
imperfect since
it
productivity shocks. Yet to the extent that
we would
GNP
is
growth
how
expect
is
to find that the correlation of
smaller
in
countries with higher
this correlation varies with the level of
debt
a country.
Data
For our empirical work,
we
require appropriate
measures
of debt
and the
We construct a measure of debt using data on the international
investment positions (MPs) of OECD economies as reported in the International
current account.
Monetary Fund's Balance
of estimates of stocks of
^^
The
of
Payments
Statistics
Yearbook. The
IIP is
a compilation
assets corresponding to the various flow transactions
unconditional probability of an output shock leading to a current account deficit
Pr(^+Pi>-rfj)=Pr(7ij>7i)=1/2.
However, conditional on the
distribution of Xj+Pj, this probability
is Pr(;\.j+Pj>-rfjlfj)
28
level of
which
is
in
the
is
debt of the country and for any
increasing
in
fj.
capital
account of the balance
of
payments, valued
at nnarket prices.
We
measure
debt as minus one times the net holdings of public and private bonds, and other long-
and short-term
ratio to
used
capital of the resident official
GNP. However, as noted
in
and
non-official sectors,
measure
the introduction, this
share of wealth held as claims on domestic
to infer the
take into account the fact that the countries
in
of
expressed as a
debt cannot be
capital, since
it
does not
our sample can hold their wealth
three forms: debt, domestic capital, and capital located abroad. Accordingly,
and holdings
subtract outward foreign direct investment
of foreign equity
in
we
by domestic
residents from debt to arrive at an "adjusted debt" measure. Figure A1 plots the time
series for debt
and adjusted debt
for the thirteen
able to construct these variables. ^^ Table
sample
13
of
OECD
measures. The
located abroad.
it
economies
for
which
we
are
presents an overview of the data for the
is
possible to construct adjusted debt
column reports the net external debt
first
GNP,
fraction of
countries for which
1
OECD
of country
j,
expressed as a
while the second column reports the holdings of claims on capital
The
third
column reports the difference between the
first
two
columns, our "adjusted debt" measure.
We
measure the current account as the change
in
the international
investment position of a country, expressed as a fraction of
measured
at
market prices, the change
in
in
measure
of the current account,
the stock of foreign assets. Figure A2, which plots the
conventional measure of the current account and the change
the countries
change
The
in
are
the IIP reflects both the within-period
transactions which comprise the conventional flow
as well as revaluations
GNP. Since MPs
in
the IIP for each of
our sample, reveals that the contribution of revaluation effects to the
in
the IIP
is
final
sample
of countries
complete description
substantial
of the
is
data
in
most countries.
determined by the limited data
provided in Appendix 2.
is
29
availability for
these series.
A
,
Current Account Cyclicality
We
are
now ready
to
Debtor and Creditor Countries
in
examine how the
cyclicality of the current
with the level of debt of a country. Table 2 presents a
first
column reports the average
level of adjusted
account varies
The
look at the evidence.
first
debt over the period 1971-93, while
the second column reports the time-series correlation of the current account surplus
GNP)
(expressed as a share of
each
debt
of the countries in
is
while
account
In six
is
GNP
which plots the
is
(Sweden
is
account (on the
There
is
the simple correlation between the two variables
Although highly suggestive, the results
in
is
if
time series correlations
of the current
is
their
level of
in
in
Figure
First,
adjusted debt.
we
We then
in
Figure
1
1
should be interpreted with
To
the extent that country-
poses no
particular
productivity are important in the data, the simple
may obscure
1
pool
highlighted
-0.54.
Table
variations over time
account within countries. To address
following strategy.
by
changes
the only exception),
vertical axis) against the level
specific levels of productivity are constant over time, this
However,
is
a clear negative relationship, and
caution as they pool information within countries.
difficulties.
period, for
negative, the current account
the only exception). This pattern
cyclicality of the current
same
out of seven countries where adjusted
countercyclical
of adjusted debt (on the horizontal axis).
some
growth over the
out of six countries where adjusted debt
procyclical (Japan
is
our sample.
positive, the current
in five
with per capita
all
this
concern,
in
the cyclicality
we adopt
the
country-year pairs of observations and rank them
divide the
sample
in
adjusted debt. Then, for the two subsamples,
two
at
a particular threshold
we compute
the cyclicality of
the current account for the two subgroups and test whether they are significantly
different.^^
^'*
We
remove country means from
all
equality of current account cyclicality
variables before computing the correlations.
in
the two groups as follows:
First,
we
We test for
regress the current
account on per capita income growth in the two subsamples. Under the assumption that the
two point estimators are independent and asymptotically normal, we can construct the usual
Wald statistic for the null hypotheses that the slope coefficients are equal in the two
30
Figures 2(a)-2(d) present the results of this robustness check for various
subsamples
each
of the data, in
current account
figure, the bold (solid) line plots the cyclicality of the
debtor (creditor) countries for the corresponding level of adjusted
in
debt at which the sample
is
divided
in
The dashed
two, indicated on the x-axis.
line
reports the p-value for a test of the null hypothesis that the cyclicality of the current
account
is
equal
the two groups. Figure 2(a) uses data for
in
entire period from
1971
to
countries over the
all
1993, and reveals that current accounts are clearly
procyclical in creditor countries
and countercyclical
values of the threshold,
range
of
level.
Figures 2(b), (c) and (d) present the
debtors. Moreover, for a wide
in
this difference is significant at the
same
information for three
the data. Figures 2(b) and 2(c) restrict the sample to the 1971-81
subperiods respectively, and reveal that the difference
much more pronounced
following the
the latter period. However,
1973 and 1979
Figure 2(d), the pattern
In
in
oil
5-10 percent
in
if
subsamples
and 1982-93
current account cyclicality
we
is
done
in
re-emerges.
the theory developed above, output shocks are saved
creditor countries, while the differential current
in
account behavior
in
both debtor and
the two sets of
countries arises from the the differential investment response to output shocks.
debtor countries, a fraction greater than one of these savings
capital, while in creditor countries
fourth
columns
theory.
The
third
GNP
per capita
correlation
of
Table 2 provide
a
this fraction
some rough
^^
In all
subsamples.
A
The
countries, there
portion of
is
at
shocks
correlation
and
between
a strong positive
annual frequencies,
to
income are saved
evidence that the
the two groups of countries.
defined as net national savings, expressed as a fraction of
in
cyclicality of the
in
31
third
indicators of these two pieces of the
rejection of this null hypothesis constitutes
current account differs
is
some
In
allocated to domestic
smaller than one.
between savings and per capita Income growth
consistent with the view that at least
Savings
is
is
column reports the within-country time-series
growth and savings.
is
drop the two years
shocks from the 1971-82 subperiod, as
we emphasize
of
GNP.
order to smooth consumption over time.
relationship
final
column
between the
of
As
in
the theory, there
no obvious
is
debt and the cyclical behaviour of savings.^^ The
level of
Table 2 reports the within-country time series correlation between
savings and the current account. Consistent with the theory, there
is
a strong
negative relationship between the level of debt of a country and the correlation of
savings and the current account.
In six
out of seven debtor countries, savings and
the current account are negatively correlated, while
in five
out of six creditors, they
are positively correlated (Sweden and Japan again are exceptions).
Other Explanations for the Debtor/Creditor Distinction
A
notable feature of business cycles
be highly correlated across
countries.^''
in
OECD
economies
is
that they tend to
This observation suggests two possible
alternative explanations which might account for the difference in current account
cyclicality
between debtor and
expansions tend
series correlation
six-month
LIBOR
countercyclical
in
to
be associated with increases
between
is
creditor countries.
OECD average
0.55. Thus,
it
is
First,
in
OECD-wide economic
interest rates.
GNP
per capita
In fact,
the time-
growth and growth
in
the
possible that current accounts are
debtor countries only because domestic
booms
coincide with high
present and expected future debt service obligations.
Second,
to the extent that countries hold claims
account fluctuations also
much
reflect
on foreign
capital, current
domestic and foreign residents' decisions on how
capital to hold abroad. This too
can
potentially
account
for the countercyclicality
®
This of course does not rule out other explanations for the procyclicality of savings. For
if booms redistribute wealth from individuals with low savings propensities to
individuals with high savings propensities, savings would also be procyclical. However, as long
example,
as individuals allocate their wealth using investment rules such as the ones discussed
paper, our effects would still arise.
See Costello (1993) and Kraay and Ventura (1997)
32
in this
accounts
of current
in
debtor countries. To see wtiy, suppose
that,
consistent with the
notion that they are high-productivity countries, debtor countries tend to invest less
abroad than foreigners invest
marginal unit of wealth
in
among
them. Then, provided that
assets
in
the
same
all
investors allocate their
proportions as their average
wealth, favourable income shocks that are correlated across countries
current account deficits
in
will
produce
debtor countries simply because foreigners purchase more
claims on debtor country capital than residents of the debtor country purchase
our sample, the outward investment of debtor countries
abroad.
In
5.6%
GNP,
of
while inward investment
in
countries, the corresponding figures are
To adequately
these countries
was on average
was 9.8%.
In creditor
15.2% and 11.1%.
differentiate our theory
from these alternatives,
we
revisit
evidence presented above, conditioning on global shocks such as changes
interest rates
and
foreign income. This
is
done
in
Table 3 and Figure
3,
in
the
world
which report
the partial correlation between the current account surplus and domestic per capita
GNP
growth, controlling for the growth rate of the 6-month
rate of
OECD
per capita
GNP
excluding the country
in
LIBOR and
question.^®
the growth
Now
the cross-
country correlation between the level of adjusted debt and the cyclicality of the
current account remains negative,
sample
of dividing the
full
levels of debt.
The
we can now
the
same
in
and
is
equal to -0.55. Figure 4 reports the results
of countries into debtors
pattern which
emerges
is
and
creditors at various threshold
similar to that in Figure 2(a).
However,
only reject the null hypothesis that the cyclicality of the current account
is
debtor and creditor countries over a smaller range of threshold levels of
debt.
A
further concern might be that productivity shocks account for a larger share of fluctuations
per capita GNP growth in debtor countries than in creditors. Although there are no a priori
reasons to believe in this asymmetry, it is possible that our results are driven by it. We did
experiment with controlling for various proxies for productivity shocks, and obtained
in
substantially similar results to those reported here.
33
5.
Concluding Remarks
Using a model with quite conventional ingredients,
prediction that,
if
investors exhibit constant relative risk aversion
income, favourable output shocks lead
and surpluses
in
average one. Since by
is
to current
definition the
in
as a portion of
in
account surplus
in
wealth increase
is
creditor countries.
aversion varies with wealth and
in
a
to
unit of
proportions as the
deficit
in
a greater
on the current account.
exceeds investment
at
home,
invested abroad. This produces a current
We
have also shown
that,
if
investors' risk
the presence of labour income, there
determine when a positive output shock leads
rule to
same
wealth (savings) results
creditor countries the increase in wealth
this
debtor countries
share of a debtor country's wealth devoted to
domestic capital (investment), leading
Conversely,
and have no labour
deficits in
distributed across assets in the
domestic capital exceeds one, an increase
in
account
derived the novel
Under these assumptions, the marginal
creditor countries.
wealth (the income shock)
increase
we have
to
is
a current account
a simple
deficit:
the
country's debt has to exceed a threshold that can be either positive or negative, and
is
zero
in
the case of constant relative risk aversion and no labour income.
also provided
some
suggestive evidence from thirteen
OECD
countries that
We
have
is
consistent with this prediction.
To make progress one must be
been shy about doing
so.
Sovereign
features of real economies that
however
that our results
willing to
risk
we have
would survive
and
left
all
make assumptions, and we have
foreign investment are two important
unmodelled here.
is
the
investment
absence
is
of
most
adjustment costs to investment.
not "bang-bang" as
We feel
confident
but the most extreme versions of models
that incorporate these elements. In our opinion, the
paper
not
we have assumed
glaring omission of this
In real
economies
here. Abstracting from
adjustment costs greatly simplifies the analysis, while permitting us to isolate the
34
main economic forces
at wor[<.
The drawbacl< however
is
that the effects of
all
these
forces are collapsed into a single instant, effectively precluding any meaningful
discussion of the dynamic aspects of the current account responses to shocks.
are currently extending the theory to determine
to
shocks
affect the results
how
We
sluggish investment responses
presented here and, hopefully, derive
new
results
regarding the dynamic aspects of the current account responses to both output and
productivity shocks.
35
.
References
R.C. Merton and W.F. Samuelson (1992). "Labor Supply Flexibility and
Portfolio Choice in a Life Cycle Model." Journal of Economic Dynamics and
Control. Vol. 16, pp 427-449.
Bodie,
Z.,
Buiter,
W.
(1981). "Time Preference and International Lending and Borrowing in an
Overlapping-Generations Model". Journal of Political Economy. Vol. 89, pp.
769-97.
Clarida, R. H. (1990). "International Lending
Stationary Equilibrium". International
and Borrowing in a Stochastic,
Economic Review. Vol. 31, pp 543-558.
Costello, D. (1993). "A Cross-Country, Cross-Industry
Growth". Journal of
Political
Economy.
Comparison
of Productivity
Vol. 101, No. 2, pp. 207-222.
Deaton, A. (1992). Understanding Consumption. Oxford: Clarendon Press.
Devereaux, M.B. and M Saito (1997). "Growth and Risk-Sharing with Incomplete
International Asset Markets". Journal of International Economics. Vol. 42, pp
453-481
Dornbusch, R. (1983). "Real Interest Rates,
Home Goods, and
Optimal External
Borrowing". Journal of Political Economy. Vol 91, pp. 141-153.
French, K. and
J. M. Poterba (1991). "Investor Diversification and International
Equity Markets". American Economic Review. Vol. 31, pp. 222-226.
Kraay,
A and
J.
Ventura (1997). "Trade and Fluctuations". Manuscript, M.I.T. and
the World Bank.
Lewis, K. (1995). "Puzzles in International Financial Markets", in G. Grossman and
K. Rogoff, eds. Handbook of International Economics. Amsterdam: Elsevier.
Matsuyama,
K. (1987). "Current Account Dynamics in a Finite Horizon Model".
Journal of International Economics. Vol. 23, pp. 299-313.
Merton, R. C. (1992). Continuous-Time Finance. Cambridge: Basil Blackwell.
36
and the Terms of Trade: Is There are
Laursen-Metzler Effect?". Quarterly Journal of Economics. Vol. 97, pp. 251270.
Obstfeld, M. (1982). "Aggregate Spending
Obstfeld, M. (1994). "Risk-Taking, Global Diversification
Economic Review. Vol 84, pp. 1310-1329.
and Growth". American
and K. Rogoff (1995). "The Intertemporal Approach to the Current
Account' in G. Grossman and K. Rogoff, eds. Handbook of International
Economics. Amsterdam: Elsevier.
Obstfeld, M.
Persson, T. and L. Svensson (1985). "Current Account Dynamics and the Terms of
Trade: Harberger-Laursen-Metzler Two Generations Later". Journal of
Political Economy. Vol. 93, pp 43-65.
Debt and Liabilities of Industrial Countries". Reserve
Australia Research Discussion Paper No. 9405.
Rider, M. (1994). "External
Bank
of
(1981). "The Current Account and Macroeconomic Adjustment in the
1970s". Brookings Papers on Economic Activity. Vol. 1, pp. 201-68.
Sachs,
J.
Sachs,
J.
(1982). "The Current Account
Scandinavian Journal
of
in
the
Macroeconomic Adjustment Process".
Economics. Vol. 84, pp. 147-159.
L. and A. Razin (1983). "The Terms of Trade and the Current Account:
The Harberger-Laursen-Metzler Effect". Journal of Political Economy. Vol.
Svensson,
91, pp. 97-125.
Tesar, L.L. and I.M. Werner (1992). "Home Bias and the Globalization of Securities".
NBER Working Paper No. 4218.
Ventura,
J.
(1997). "Growth
and Interdependence". Quarterly Journal
of
Economics.
Vol. 112, pp. 57-84.
Zeira, J. (1987), "Risk
and Capital Accumulation
in
a Small open Economy".
Quarterly Journal of Economics. Vol. 102, pp.265-279.
37
Appendix
Solution Details
1:
Optimal Consumption and Portfolio Rules
Here
we
derive the solution to the consumer's problem
in
the model of
Section 3, and note that the solution to the corresponding problem
obtains as the special case where Xj=pj=O.The representative
country
j
maximizes (15) subject
r is
The Bellman equation
for this
p- V(aj,7ij)=
f
max
,
ln(Cj
problem
>
+Pj) +
Cj,Xj>|^
—
consumer
1
residing
budget constraint (16) and the (correct
to the
constant and
equilibrium) belief that
Section
in
tij
follows the
dynamics
in
Equation
in
in
(2).^^
is:
3V(a|,7i;|)
r,
V
^-[((Kj
-I
-r)-Xi +r)-aj +Xj -c^.J +
(A1)
^
The
""^''"^'^Ixf -af -a^ +^[v(a,;.^ +g(Kj))- V(a,K,)]
aaf
first
order-conditions of this problem are:
1
3V(aj,7ij)
Cj+pj
aaj
=
^^
If
Ai=pj=0, the solution to this
state with
a constant
problem
(A2)
is
correct
interest rate.
38
even
if
the world
economy
is
not
in
a steady-
.
It
can easily be
.
verified that the following value function solves the
Bellman
equation:
V(ai,Tr:)
where
g(7tj)
= p-^ln aj+-^
does not depend on
(A4)
Substituting the derivatives of the value function
aj.
and (A3)
into the first order conditions (A2)
turn specialize to Equations (5)
+ g(7ij)
and
yields Equations (17)
for the
(6)
case
and
(18).
These
in
of l^-^^-Q.
The World Steady-State
Here we show
1
that,
lim
if
-•
"
^^j
+Pj =
,
there exists a steady-state
in
j=i
which both the world average productivity and world growth rate are constants. Let H
= n and L be the
be the set of countries with
Ttj
Remember that each group
contains the
average wealth of countries
in
,
set of countries with
same number
the two groups
is
a =
-
lim
—
J^-J
•
ya
^
interest rate
Next
do so
in
.
i
Also, define s
as
Using
.
^
a+a
r
=
s
•
t!
+ (1 - s)
we show that there
four steps. First,
the condition
=
'
we
ds=0 requires
7t
lim
2
—
^aj
•
and
J6H
this notation,
we can
write the world
- o^
exists
a
distribution of
wealth s* at which ds=0.
note that, except for the cases
that
=n
of countries. Therefore, the
•'^"^
J
a=
Ttj
^^ =
a
^a
.
Second,
39
we
In
which a =
We
or a = 0,
note that Equation (19), the
J
''
assumption
that lim
.1—loo
J->-
—1 V^,
^^
+(3:
•
.
=0, and
thefact that a fraction
of countries
(t)dt
'
^
I
1=1
change regime each
da =
period, jointly imply that:
n-Tz
+ (1-S)
+ S-7C + (1-S)-7t-G^-p •a+{j)-(a-a)^-dt+
(A5)
+ a + (1-s)
0-S
da
lim
—
y
a,
dco
—\ +s-7t + (1-s)-7i-a-p
a + ({)-(a-a)>-dt+
a
(A6)
+ o+s
Third,
we
—\-
lim
— -"y 3;
-dco
note that, conditional on the ajS and given that the shocks
dcoj
are
independent across countries, a straightforward application of the Law of Large
Numbers shows
that lim
j^°»
ds=0
if
and only
J
2
-dco,
= lim—
J-*- J
jsH
y
a,
dco:
=0.
Fourth,
it
follows that
JeL
if:
s(1-s)
An
2
—
-y a,
(1-2S)
k-tC
+ 2(7i:-7t) + (t)(1-2s) =
(A7)
analysis of this equation reveals that there exist a solution s*e (1/2,1]. Hence,
taking this distribution of wealth as an
initial
and the world
interest rate are constant.
growth rate
also constant.
is
condition, the world
The reader can
40
easily
average productivity
check
that the world
Appendix
2:
Data Sources
This paper uses data on international investment positions (IIPs) reported
the International Monetary Fund's Balance of
Subject to
Edition).
availability, this
Payments
Statistics
Yearbook
source reports data on the stocks
in
(5th
of various
assets held abroad by residents of a country, and the corresponding stocks of
domestic assets held by non-residents. These stocks of assets are valued
prices,
and hence changes
which are not captured
In
three
in
in
these stocks
the usual flow
reflect
measure
of the IIP:
market
unrealized capital gains and losses
of the current account.
order to empirically implement our model,
components
at
we need
to distinguish
between
claims on foreign capital held by the residents of a
country (outward equity claims), claims on domestic capital held by non-residents of
that country (inward equity claims),
and net holdings
of the international
Outward equity claims are measured as outward foreign
residents' holdings of corporate equities abroad.
Finally, net
direct investment plus
Similarly,
measured as inward FDI plus non-residents' holdings
of
bond.
inward equity claims are
domestic corporate equities.
bond holdings are proxied by the non-reserves
residual of the
IIP,
which
includes net public and private bond holdings and other long- and short-term capital.
Our sample
of countries
concerns about data
We then
checked the
quality.
was determined
We
began
with a
both by data availability and
sample
of
20
OECD
economies.^"
overall IIP series for these countries against that reported in
Rider (1994), which presents independent estimates of IIPs based on extensive
research into national sources. For most countries, these two sources correspond
Greece, Ireland and Iceland were immediately dropped from the sample due to extremely
coverage. The Balance of Payments Statistics Yearbook reports balance of
payments data for an aggregate of Belgium and Luxembourg only. This reduces the original
sample of 24 OECD economies by four.
limited data
41
quite closely.
Zealand due
However, we were forced
to
to
drop Belgium/Luxembourg and
major discrepancies between the two sources. Next,
New
we excluded
Portugal and Switzerland since IIP data were available only for eight years for each
of these countries.
Finally,
we dropped Denmark, Norway, Turkey
since stock data on the subcomponents of the IIP
This resulted
in fairly
complete series on the
IIP
we
and
from the sample
required were not available.
its
components
between 1970 and 1993. The remaining missing values
in
of
payments
obtain a balanced panel of 24 annual observations for each country
drawn from the OECD's
^^
in this
13 countries
the sample were obtained
by cumulating the corresponding flow items from the balance
The remaining data used
for
^^
and
in
order to
series.
^^
paper (GNP and net national savings) are
national accounts.
Stock data on equity holdings not available
for
Norway and Turkey. Stock data on FDI
not
available Turkey.
54 out
of
a
total of
312 observations were obtained
1970s.
42
in this
manner,
all
of
them
in
the early
Table 1- Debt and Adjusted Debt
(percent of
GNP. average, 1970-1993)
Debt
Outward Investment
Adjusted Debt
(1)
(2)
(3)=(1)-(2)
Debtor Countries
Finland
24.3
3.2
21.1
Canada
26.0
13.3
12.7
Australia
16.6
6.6
10.0
Sweden
17.1
8.6
8.5
Austria
8.5
2.4
6.1
Italy
6.5
3.3
3.3
Spain
4.5
1.7
2.8
Creditor Countries
France
Japan
2.2
6.4
-4.2
-3.2
3.5
-6.7
United States
3.2
11.1
-7.9
Germany
United Kingdom
-3.5
5.8
-9.3
6.3
27.8
-21.5
Netheriands
-0.6
36.3
-36.9
Debtor Average
14.8
5.6
9.2
0.7
15.2
-14.4
Creditor Average
>c
CO
CO
>*—
c
3
o
o
<
r- CM CO ._ CO 1- 00
T-;
CO t^ '^ '^
in 05
CM
cp
O O
o o d o o o o
o c
c QJ
,o t
3
TO
o
m
i_ .c
k_
o 1
o
o
m To o
o To o o o
•«*•
-"S-
CD
1-^
q S
9 o
CD
Cvl
O
W O
1^
^O
o
re
^
o
5
o
o
re
CM •<*- OO
CD CM CM
CM T- TjCD CO CD
r^ '^ r^
^ O CO CM
d d d d d d d
in f^ in en CM t^
CD
CM
-^
C3 c6
^.
CM 1^
Tf CD
o o
Tl-
lO
in
d d
o
CM
d
re
O
0)
1-
o
Ql
r:
O
O
m CD CM
CD
d d d d d d
* O
o d
o
CD
CD U1 CM CO OO
LO IT) IT)
LO -^
CD
cb
r--
§•5.
c
re
M
o
Q
c
O o
o
o
< o
c Z
E O
re
o
o
>«
O
c
O
u
u
<
o
c
(^j
^
d ^
^ o
d d
re
O
01
a.
o
O
c
0)
O
I
-
CL
-°
Z
T3
O
So
CM
V
re
.
C^ $^
3
a)
^^
o
^
Q.
>-:
°
vn T- CO <»
00 CD CO CM
C3>
f-^
CO
ai CM CD
CO
CM
CJ)
Q
D
0)
I/I
u
0)
k.
o
0)
Q
E
o
c
3
O
c
3
o
O
0)
o
o
re
TO
™
.E
.b
li-
re
.2
c
re
a)
o
re
w a) 01
3 g
5 3
O < W <
^
CO
ii
're
(2.
W
i
^
(U
o
c
re
re
O
o) 2
5 «
>
0)
(U
Ll-i?
B
5 o D Z
<
u.
o
1-
O
p
<
c
o
m
.ti
s^
a u
o
o
Table 3- Current Account Cyclicality Controlling for Global Shocks
Adjusted Debt
(percent of
GNP)
Account
Growth
Partial Correlation of Current
with Per Capita
GNP
Partial Correlation of
Debtor Countries
Finland
21.1
-0.03
Canada
12.7
-0.20
-0.13
Australia
10.0
0.11
-0.02
Sweden
8.5
0.18
-0.02
Austria
6.1
0.02
-0.15
Italy
3.3
-0.33
-0.20
Spain
2.8
-0.31
-0.31
France
-4.2
-0.14
-0.12
Japan
-6.7
-0.13
0.08
United States
-7.9
0.06
0.24
Germany
-9.3
0,14
-0.03
United Kingdom
-21.5
0.12
0.09
Netherlands
-36.9
0.45
-0.03
Debtor Average
Creditor Average
9.2
-0.08
-0.13
-14.4
0.24
0.01
-0.55
-0.36
-0.05
Creditor Countries
Correlation w/ Adjusted Debt
Savings
with Current Account
tf)
T3
0)
o
DC
ro
(A
_o
O
>>
O
C
3
O
u
u
<
c
0)
^u.
3
o
o
3
O)
sjojipajo puB sjoiqaa
AjlIEDipAo JBMJ lS9i
^
'
en
00
o
o
'
^~
d
(p
d
in
d
u;
*
d
<'
^
lenbg
p|B/v\ JOJ
1
/:
s;
3n|BA-d
CO
rsj
T-
d
d
d
c
OLO
>
\
600
900
1
/-^Z
ZOO
900
900
/
C
3
O
u
u
900
900
WO
t'O'O
L'O'O
<
£00
c
200 1
0)
k.
1_
3
o
LOO
_
Creditors
O
O)
-w en
J
LOO
^
000
%
LOO- o
(U
LOO-
—
y
300-
CT)
~"^^^
O
in
£00- £
£00^^""^
•
MOwowo-
^^'^"•\~v^-^btors
—
3
'
(D
/^^V^X
900900900-
Q-_;
zoo-
/A
zoo-
\
60-0-
900-
OL'Oin
CM
CNJ
d
o
1
300- °
—
3^
^
lO
T—
d
T-
d
m
o
d
c3
{qvv\oJ9 dlMO BiidBQ jaj |Bay
'snidjng junooov iuajjno)jjo3
IT)
O
d
1-
d
'
u)
~
c3
j=
1-
»
sjo^ipojo puB sjo;qaa
A}!|BDn|oAo jeqj jsai
CD
o
d
IT)
<«
o
o
u|
p|B/\/\
n
o
|Bnbg
joj
sj
3n|BA-d
CM
o
Ol-O
—
oro
- 60'0
- 800
- 80-0
- 800
- ZOO
-
ZOO
- 900
-
C
3
O
u
u
900
<
V
c
V
w
3
O
1^
O
900
- 900
l_
n
soo a
soo T3
0)
t'OO
—
o
l-
^
3
TJ
£00 <
o
•"
300
'~
;-
000 >
1—
1-
20-0-
u
o
CO
•—
o
O
3
_.-
SO'O-
;-
wo-
- fO'OCO
- 900-
o
•4—'
- 900-
Q
- 900-
1 zoo.j
ZO'O•i
r-
800-
,v
— 600^ 6004-
CO
d
d
CN
og
o
o
(MVv\oj9
dN9
o
BiidEo jaj leay
'snidjns lunooov iuajjno)jjoo
d
n
o
2
n
JZ
- t-oo-
O)
0)
oro-
03
.
'
sjo;!pajo pue sjojqaQ
Ai!|eo!|dAo
en
00
d
d
}Em jsai
u;
lenbg
p|B/v\ jo^
CO
CD
d
d
sj
sn|BA-d
d
o
d
c\i
d
-
OL
E 10
Z 900
E soo
Z SCO
z
E
wo
wo
E t'oo
E eoo
E eoo
c
3
O
o
o
E 200
<
E
i-oo
E
mo
^->
(1>
c
'
E 000
0>
0)
3
o
*-<
E 000
Q
0)
tn
ro
^
>
E
i-o'o-
^>,=?
r—
Q_
E
i-oo-
**-
I LOO-
<u
30'0-
>
"-
o
=
re
o
o
t^
_
l>~.
o)
,-
.
.
-
-
-
-
,
"
E
.
1—
>.
»
,,_^
u
-
Ol
:
.
0)
l-
^
U
(D
i_
z eoo-
.
^
/-
3
cn
E
eo'o-
E
t'O'o-
E
E
wowo-
E soo-
z sooE so'o~ 900z
ZO'O-
E 800E 800"^
600in
CM
d
d
o
3
re
E 300- 2
o
E 300-
o
,'
O
<
o
'sn|djns lunoaov juajjno)jJoo
o
—
'
.
—
sjojipajQ puB sjojqaa u; jenbg
Aj!|B3!|dA3
O)
"
,
^
~
"
b
b
o
p|B/\A jo^
^
b
in
CD
r^
CO
b
•^
»Bm jsai
b
CO
b
CNI
•<—
b
b
o
/^
1
^^^^^^^
^^^'
^_,,-'^^
#
si
aniBA-d
^
#-»
1
01
'
-}-
.
t;^'
•
^
",
«
..^^
600
T
X-^
/
f"
V
-2
\
3
1-
^
)
k^
>
a>
i_
/
/
'
X\X
.
c
O
o
u
<
c
,
.
-
(A
•-
>
w^.
^^^^^
-*--^
1
CL
"
"
-
i
^ 200
'
c/)
1—
O
>
CD
<^•*yQ
^
(
1
k^
000 p
- £00-
- £007
1
V
j
/
'
t
WOWO-
-I-
»
,
_
''^
N,
-
-
.
_
X
^^^
"^--- ~\
s,^
^<
'•_
,'
-'-.'...
-
.1
b
m
r^
——-^
\
— —
o
V-
in
C3
b
(qvwojo
-
soo-
-L
-
\
!_::::=:-
BiidBQ jad |Bay
'snidjns junooov juajjno)jjoo
80-060-001-
1
o
Z0"0-
r\\
1—
"^
b
dN9
t'OO-
7 SO'O+
7 900-
^-•
1
'
OJ
-
-
>
r
*,
<
"S
Q
QJ
- 200-
(^
'''
y
^--^
E
-
-1-
?:
D)
000
3
7
T 300- 15
>
t
- 200- 2
o
+
- 200- -^
1
^
-
- ^oo- °
J
I
mo
mo- 1
- mo- <
—
I
-
-
(/)
75 .E
O 73
wo
7 £00
T
^ £00
\
<<y
/
'.
-
>^
'
^c
CD
o
O
- '
- t'OO
<
<^/:.,
f 900
- 900
—
-
WO
/
^
'•
900
m
o
b
'I—
b
'
r\
jC
o
CO
J3
Q
aQ)
=j
o
o
o
D
<
w
03
O2^6 CO
O EO
05
o Q ^
o
o
c
UJttf
o
^ O o
0)
13
(A
U
O
<
I
CO
<
ro
o
O
Q.
1_
D5
C
ro
o
O
"to
Q.
O
'^ _i
I-
11O
<
£=
ro
C
o
O
4->
o
8o
o
o
o
NT
o
o
CO
o
o
o
o
o
CNJ
"(5
_o
o
>«
O
*^
=)
LU
Q
c
3
O
u
u
<
c
o
o
pi
k.
3
O)
Q
-J
Z
o
o
SJOJJP8J3 puE sJo^qaQ u\ lenbg s;
AjMBDMoAo leqi jsai p|BAA Jo; 9n|BA-d
-
C3>
OO
r-
d
d
d
in
CD
d
\
CO
Ci
<r
o
d
—^_
CM
T-
d
d
o
n
1
1
f
;^
800
>c
'--.,_
"
-
-
»
^
£00
V>?
*
-
'n
eo'o
k
^^^^
,
1
01-
_
r
900
90'0
s
._/
900
:
SO'O
wo
i
t'O'O
o
O
^
voo
{
T-
I
i
r^-
.
/ -2
eoo
S CD
-CjQ
300
1-
3 O
O £:O
HO CO
T-
O
m
Co
= n
Value
T-
"
o
3-°
"^
Qi
<D
3
CD
>
J
ii
!
g
1
d
d
d
of
d
d
d d
Adjusted
" *
T-
'4-'
^
o
O
cn
c
e
Q.
\
CD
=
O
'
;'
\)
1
CM
,
:
i
SO'O-
£00- £
SO'O-
I
!
I
.11
WOwowosoo900900-
zoo-
10080'0-
600oroCM
d
If)
d
m
o
;uno3DV luajjno ^o uo!;b|8Jjoo
O
|Bj;jBd
i
MIT LIBRARIES
3 9080 01'439"2317
Si
Q
2
1
OJ
<
oc
s
< Q
<
3
D)
nl
1
1
Si
^(
\
\
V
tE
:8i
:sf
M
eJLi
t
£1
<
61
S Si
I
/
&>
i
i
8
f
<
w
s;
o
J
Q- in
a
§
%<
n
aj
>
"i:
re
C
^
a;
H
<
E
ai
u-
;Cmc
E °
,1,
-^
c >
CM
<
3
';Ui;.
IJ