Ep Bab 15 Ingris-Halaman-3
Ep Bab 15 Ingris-Halaman-3
Ep Bab 15 Ingris-Halaman-3
Crises in an Open
Economy
E
conomic development takes place in the long term. Most of the processes dis-
cussed in the previous chapters, whether improving human welfare, increasing
saving, or shifting toward manufactured exports, take years and even decades
to bear significant results. If policy makers in developing countries gaze only
at the far horizon, however, they are unlikely ever to reach it. Much happens in the
short term, within a few months or a couple of years, to throw an economy off bal-
ance and make pursuing long-term strategies difficult and sometimes impossible.
Policy makers need to emulate a ship’s captain, who, always steering toward the port
of destination, nevertheless must deal decisively with any storms at sea.
Among the most dangerous and likely of these storms are changes in world
prices that throw the balance of payments into deficit, excessive spending that fuels
inflation or unsustainable debt, and droughts or other natural disasters that disrupt
production. Unless a government counteracts these economic shocks, they create
greater uncertainty and higher risk for private producers and investors, who take eva-
sive actions that reduce future investment, worsen the crisis, and cause development
efforts to flounder.
During the 1970s and 1980s, the late 1990s, and in Europe in 2010–11, as pointed
out in Chapters 12 and 13, many economies became unbalanced because of unstable
world market conditions and their own macroeconomic mismanagement. In Chap-
ters 5 and 11 through 13, we discussed the consequences of such macroeconomic
instability. Countries with overvalued exchange rates and rapid inflation were
unable to grow rapidly. Stabilization programs, many funded by the International
Monetary Fund (IMF), were intended to correct these macroeconomic imbalances.
545
5 4 6 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
1
In developing this and the next two sections, we acknowledge an intellectual debt to Shantayanan
Devarajan and Dani Rodrik, who wrote an excellent set of notes for their class on macroeconomics
for developing countries at Harvard’s John F. Kennedy School of Government in the late 1980s and to
Richard E. Caves, Jeffrey A. Frankel, and Ronald W. Jones, who develop the open economy model in Chap-
ter 19 of World Trade and Payments: An Introduction (Glenview, IL: Scott, Foresman, Little, Brown, 1990).
2
Among developing countries, China and India are large enough that they could become exceptions to
the small country rule, given continued growth in China and both greater growth and openness in India.
3
So called because it was developed by Australian economists, including W. E. G. Salter, “Internal Bal-
ance and External Balance: The Role of Price and Expenditure Effects,” Economic Record 35 (1959), 226–38;
Trevor W. Swan, “Economic Control in a Dependent Economy,” Economic Record 36 (March 1960), 51–66;
W. Max Corden, Inflation, Exchange Rates and the World Economy (Chicago: University of Chicago Press,
1977). Australia also is a small, open economy.
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 547
exportable goods and services. The Australian model lumps importables and export-
ables together as tradables and distinguishes these from all other goods and services,
called nontradables. (We use this specification again in Chapter 18’s discussion of
Dutch disease.)
Tradable goods and services are those whose prices within the country are deter-
mined by supply and demand on world markets. Under the small economy assump-
tion, these world market prices cannot be influenced by anything that happens within
the country and so are exogenous to the model (determined outside the model). The
domestic (local currency) price of a tradable good is given by Pt = ePt* where e is
the nominal exchange rate in local currency per dollar (pesos per dollar for Mexico
or rupees per dollar for Pakistan) and Pt* is the world price of the tradable in dollars.
Even if the supply of and demand for tradables change within an economy, the local
price will not change because domestic supply and demand have a negligible influ-
ence on the world price. Yet, changes in the nominal exchange rate change the domes-
tic price of tradables commodities. If a country devalues its nominal exchange rate, it
increases the amount of local currency required to purchase a dollar (e increases); if a
country revalues its nominal exchange rate it decreases the amount of local currency
required to purchase a dollar (e decreases). Devaluations thus increase the local cur-
rency price of tradables (all else equal), while revaluations tend to decrease the local
currency price of tradables. Because this model simplifies all tradables into one com-
posite good, the price of tradables Pt is best thought of as an index, a weighted average
of the prices of all tradables, much like a consumer price index.
Tradables include exportables, such as coffee in Kenya and Colombia, rice in
Thailand, beef in Argentina, cattle in West Africa, palm oil in Malaysia and Indone-
sia, copper in Peru and Zambia, oil in the Middle East, and textiles and electronics
in East Asia, and importables, such as rice in West Africa, oil in Brazil or Korea, and
intermediate chemicals and machinery in many developing countries.
Nontradables are goods and services, such as transportation, construction,
retail trade, and household services that are not easily or conventionally bought or
sold outside the country, usually because the costs of transporting them from one
country to another are prohibitive or local custom inhibits trade. Prices of nontrad-
ables, designated Pn, therefore, are determined by market forces within the economy;
any shift in supply or demand changes the price of nontradables. Nontradable prices
thus are endogenous to the model (determined within the model). The term Pn, like
Pt, is a composite or weighted average price incorporating all prices of nontradable
goods and services.
Production
frontier
Consumer
indifference
curve
Nontradables
1
N1
P = Pt /Pn
T1 Y1 = A 1
Tradables
the menu of possible combinations of outputs of the two kinds of goods, N and T. The
community indifference curves show consumer preferences between consumption
of tradables and nontradables.
Equilibrium is at point 1, the tangency of a consumer indifference curve and the
production possibilities frontier. At this point, the production of tradables, deter-
mined by the production frontier at point 1, is T1, equal to the demand for tradables,
determined by the indifference curve at 1, and similarly, for nontradables, supply
equals demand at N1. This is a defining characteristic of equilibrium in the Australian
model: At point 1, the markets for both goods are in balance. Put another way, there
is external balance (EB), because the supply of tradables equals demand, and inter-
nal balance (IB), because the supply of nontradables equals demand.
Point 1 simultaneously indicates the optimal (profit maximizing) combination of
tradables and nontradables for producers, the optimal (utility maximizing) combina-
tion of tradable and nontradables for consumers. For both producers and consum-
ers, this optimum occurs with respect to relative prices—in this case the relative price
of tradables to nontradables. This relative price is indicated by the slope of the price
line in Figure 15–1. This joint equilibrium for producers and consumers is indicated
by the tangency of the indifference curve and the production possibility frontier.
The tangency of the indifference curve and production frontier is jointly determined
with the relative price of tradables in terms of nontradables, P = Pt/Pn. This relative
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 549
price, P, is one way to define the real exchange rate (RER), one of the important
innovations of the Australian model.4 (Box 15–1 provides a more detailed explanation
of the real exchange rate.) This formulation separates out prices that are under the
influence of monetary and fiscal policy and domestic market forces, Pn, from prices
that can be changed only by adjustments of the nominal exchange rate, Pt = ePt*.
Note that the slope of the price line that is tangent to the production possibility curve
and the consumer indifference curve is the only real exchange rate consistent with
equilibrium in the model.
If P rises (the price line becomes steeper in the diagram), tradables become
more expensive relative to nontradables. Producers then attempt to switch along
the production frontier away from N goods, toward T goods. Consumers attempt to
switch in the opposite direction, up along the indifference curve to consume fewer T
goods and more N goods. Therefore, a rise in P should increase the surplus of T-good
production over consumption.
If the production of T goods exceeds consumption of T goods, there is an exter-
nal surplus, which is identical to a surplus in the balance of trade. To see this, start
with the definition of the trade balance as
Bt = X - M [15–1]
where X and M are exports and imports. Because exports are the surplus of sup-
ply over demand for exportable goods, while imports are the opposite, a surplus of
demand over supply, we can write the balance of trade as
4
Chapters 18 and 19 will define the real exchange rate index as RER = RoPw /Pd. The term Ro is an
index of the nominal exchange rate; in this chapter we use e, the nominal exchange rate itself. The term
Pw is an index of world prices, often the U.S. consumer or wholesale price index and is similar or identical
to P* as measured in practice. But Pd is a domestic consumer or wholesale price index that includes both
tradable and nontradable prices, whereas Pn is an index of nontradable prices only. Thus, the Australian
formulation of the real exchange rate is a more-precise definition than those given in the later chapters.
5 5 0 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
B O X 15–1 R E A L V E R S U S N O M I N A L E X C H A N G E R ATE S
Most anyone who has traveled outside their home country has experience with
nominal exchange rates. The nominal exchange rate is simply the number of units
of local currency you can buy from the country you visit with one unit of your own
currency. For instance, in August 2010, you could buy just over 46 Indian rupees,
or 12.6 Mexican pesos, for US$1. (Note, this implies that you could also have
bought 3.65 Indian rupees with one Mexican peso.) But when you bring a dollar
into India or Mexico what you really care about is not how many rupees or pesos
you can buy with that dollar but rather the quantity of actual goods and services
that you can buy with that dollar. This depends on the prices of goods and ser-
vices in the host country in addition to the price of its currency. This distinction is
the key idea underlying the concept of the real exchange rate (RER).
In the most general sense, RER is the relative price of foreign goods in
terms of domestic goods. In practice, economists have developed a range of
approaches to quantifying this idea. Economists working on developed econo-
mies typically measure the RER as the relative price of domestic and foreign
goods. In contrast, economists working on developing countries typically mea-
sure the RER as the relative price of tradables and nontradables. Tradable com-
modities are goods that are or could be e traded internationally, in contrast to
nontradable goods (such as housing and many services), which are not traded
internationally. A key practical distinction between these categories of goods
is that there are world market prices for tradable goods, but the prices of non-
tradables are determined purely by local supply and demand conditions in each
country.
The nominal and real exchange rates are linked together by the require-
ment that the relative prices of tradable and nontradable goods in the RER
be expressed in the same currency units. The prices of tradables are typically
expressed in U.S. dollars, whereas the prices of nontradables are expressed in
units of the local currency. Thus we need to use the nominal exchange rate to
convert the dollar-denominated price of tradables into units of the local currency
to calculate the RER.
We can construct this RER as follows. Expressing the nominal exchange
rate, e, in terms of the number of local currency units per dollar, and expressing
the prices of tradables and nontradables as Pt and Pn , respectively, we can con-
struct the RER as
ePt* Pt
RER = =
Pn Pn
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 551
In this equation, the asterisk on Pt in the second term indicates that the price of
tradables is expressed in dollars. Multiplying this dollar-denominated price by e,
the nominal exchange rate as defined above, converts the price of tradables into
the same local currency units as the price of nontradables.
It is important to consider several issues related to this concept of the RER. First,
note that the RER is based on price indices rather than actual nominal price lev-
els. The RER is thus expressed relative to a base year, and changes in the relative
price of tradables to nontradables, say from 1 to 1.2, would indicate percentage
changes (in this case, 20 percent) relative to the base year. We refer to an increase
in the RER as a depreciation n of the local currency, and a decrease in the RER as
an appreciation n of the local currency. While this may sound counterintuitive, the
rationale for these terms is that when a currency depreciates in real terms, a given
quantity of foreign goods can be exchanged for a greater quantity of that country’s
domestic goods (and vice versa in the case of an appreciation). It is for this reason
that the RER is often thought of an as an indicator of a country’s international com-
petitiveness: When a country’s currency depreciates in real terms relative to its trad-
ing partners’ currencies, that country’s goods become less expensive to foreigners.
An important practical challenge in constructing an RER lies in the need to
choose price indices for tradables and nontradables. Pt and Pn are indices of the
prices of entire categories of goods. Thus constructing these price indices first
requires deciding which goods (and services) belong in which category. Specific
price data may also be lacking. One short cut for addressing these challenges
may be to use the U.S. consumer price index in place of Pt and a similar indicator
from the home country (with the same base year) in place of Pn .
Although the availability of such price indicators as the consumer price index
makes them convenient, their use in constructing RERs is problematic. Theory
calls for an index of nontradables prices, but the consumer price index (CPI) is
typically constructed to reflect the price of a basket of consumption goods that
includes both tradable and nontradable goods. The larger the share of tradable
goods in that basket, the greater the divergence between what the RER tells
us in theory and what we actually measure if we construct an RER using those
broad price indices. In practice, aggregate price indices purely for nontradables
rarely exist. Similar problems exist in choosing a price index to represent trad-
ables prices (for use in the numerator of the RER for a given country). In this
case, in which the goal is to choose a price index based to the greatest extent
possible on tradables, many authors use the wholesale price index (WPI) from
the United States or from a given country’s trading partners. Yet this approach
too is problematic because (as Lawrence Hinkle and Peter Montiel notea) for-
aLawrence E. Hinkle and Peter Montiel, Exchange rate misalignment: concepts and measurement
eign WPIs may not provide a very close indication of the tradables prices actually
faced by consumers in the home country. There is no perfect match between the
theoretical requirements and practical data availability in constructing empirical
RERs. A common compromise is to use the foreign WPI to represent Pt and the
domestic CPI to represent Pn .
An additional question is whether the relevant RER is purely between two
specific countries (the home country and a single trading partner—that is, the
bilateral RER) or between the home country and multiple trading partner coun-
tries. In general, policy makers in a given country will be more concerned with
how their currency relates in real terms with all of their trading partners. In that
case, it is necessary to take a (trade-weighted) average of all the bilateral RERs
between the home country and its trading partners. This average is called the
real effective exchange rate (REER).
The central challenge for policy makers concerned with their country’s inter-
national competitiveness is whether the level of the REER at any given time
reflects its equilibrium value or whether it is overvalued or undervalued rela-
tive to that equilibrium. Equlibrium in this setting generally refers to the level of
the REER at which a country’s internal market (that is, its supply and demand
for nontrable goods and labor) and its external market (that is, its supply and
demand for tradable goods) are in balance.
intersection of price line P from point 1 to the T axis.5 In national income accounting,
we distinguish two concepts. Gross domestic product, a measure of the value of out-
put, is given by
GDP = C + I + X - M [15–3]
where C and I are consumption and investment by both the government and the pri-
vate sector. Gross domestic expenditure, often called absorption, is
A = C + I = GDP + M - X [15–4]
Along the T axis, Y1 is measured in prices of the T good, so PtY1 = PtT1 + PnN1 or Y1 = T1 + (Pn/Pt)N1.
5
But Pn/Pt = T/N, with N = N1 and T = Y1 - T1, the distance along the T axis from T1 to Y1. Thus
the value of both goods in T prices is T1 + Y1 - T1 = Y1.
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 553
that any excess of expenditures over income implies a negative trade balance of equal
amount. (The appendix to this chapter provides a more detailed review of balance of
payments accounting.)
This exploration of the Australian model yields three results. First, macroeco-
nomic equilibrium is defined as a balance between supply and demand in two mar-
kets: nontradable goods (internal balance) and tradable goods (external balance).
Second, to achieve equilibrium in both markets, two conditions must be satisfied:
Expenditure (absorption) must equal income, and the relative price of tradables
(the real exchange rate) must be at a level that equates demand and supply in both
markets (the slope of P in Figure 15–1). Third, this also suggests two remedies for an
economy that is out of balance: A government can achieve equilibrium (stabilize the
economy) by adjusting absorption, the nominal exchange rate, or both. Generally,
both instruments must be used to achieve internal and external balance.
Sn
ΔA > 0 St
Dn 1
2 Dn 2
P2 ΔA > 0
Price (P)
Price (P)
1 1
P1 P1
Dt 2 3
P3
Dt 1
T1 T2 N1 N3
Quantity ( N) Quantity ( N)
(a) (b)
Figure 15–3 does this. It puts the real exchange rate, Pt = ePt*/Pn, on the verti-
cal axis and real absorption, A, on the horizontal axis. The diagram also contains two
curves, each representing equilibrium in one of the markets. Along the EB, or exter-
nal balance, curve, the T-goods market is in balance (St = Dt ). Along the IB, or inter-
nal balance, curve, the N-goods market is in balance (Sn = Dn).
The slopes of the two curves, EB and IB, can be derived from Figure 15–2. In the
tradables market, when absorption is A1, equilibrium is at P1, where T1 is produced
and consumed. This equilibrium point 1 also is shown in Figure 15–3a. If absorption
increases to A2 in Figure 15–2a, the demand curve moves outward and shifts equi-
librium to point 2. Note that with higher absorption, A2, the real exchange rate, P2,
must be higher to restore equilibrium in the T-goods market. Increased absorption
raises the demand for T goods. To meet this demand, it is necessary to raise output,
which can be achieved only through a higher relative price of T goods, P2. This higher
price also helps regain balance by reducing the demand for T goods along the new
demand curve. Point 2 is transferred to Figure 15–3a at (P2, A2).
In the nontradables market, when absorption is A1, equilibrium is at P1, where
N1 is produced and consumed. This equilibrium point 1 also is shown in Figure
15–3b. If absorption increases to A3 in Figure 15–2b, the demand curve moves out-
ward and shifts equilibrium to point 3. In the N-goods market, higher absorption, A3,
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 555
Infl
2
atio
Une
P2
n
mp
Real exchange rate
loy
ΔP > 0
me
nt
1 ΔA > 0
P1 P1
ΔA > 0 1
us ΔP < 0
P3 3
rpl
Su
t
fici
De
A1 A2 A1 A3
Absorption Absorption
(a) (b)
Brazil or Argentina. Unemployment implies not only jobless workers but also idle
capital and other factors of production. In other words, there is unemployment
when an economy is inside the production frontier in Figure 15–1. A country may
have high levels of labor unemployment but be unable to increase output because it
is fully utilizing its capital or land.
al
A
ern
anc
Ext
Surplus +
e
e
anc
Inflation
bal
Real exchange rate (P = eP*t/Pn)
D B
Surplus + Deficit +
Unemployment Inflation
C
Deficit +
Unemployment
Absorption (A)
F I G U R E 1 5 – 4 Zones of Imbalance
The economy is in equilibrium only at the intersection of the external
balance (EB) and internal balance (IB) curves. Zones of imbalance are
labeled. For example in zone A, the supply of T goods exceeds demand,
so there is a surplus, and the demand for N goods exceeds supply, so
there is inflation.
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 557
Economies spend considerable time in one of the four zones of imbalance shown
in Figure 15–4. Zone A to the north is a region of external surplus and inflation, where
the exchange rate is undervalued. In zone B to the east of equilibrium, the economy
faces inflation and a foreign deficit, due principally to excessive expenditure (absorp-
tion is greater than income). To the south is zone C, where the exchange rate is over-
valued (too appreciated) and there is both unemployment and an external deficit.
And west of the bliss point the economy is in zone D, where, because of insufficient
absorption, there is unemployment of all resources but a foreign surplus.
Once in disequilibrium, economies have built-in tendencies to escape back into
balance. Figure 15–5 describes them separately for external balance (panel a) and
internal balance (panel b). Start with an external surplus, point 1 (Figure 15–5a). The
excess supply of tradables generates two self-correcting tendencies. First, the net
inflow of foreign exchange adds to international reserves. If the central bank takes no
countermeasures, the money supply increases and interest rates fall and induce both
consumers and investors to spend more. The increase in absorption moves the econ-
omy rightward, back toward external balance. Second, the inflow of foreign exchange
creates more demand for the local currency and, if the exchange rate is free to float,
forces an appreciation. This is a move downward in the diagram, also toward the
EB line. The net result of these two tendencies is the resultant, shown as a solid line
in the diagram, heading toward external balance. If, instead, the economy starts in
Inte
al
ern
rna
bala
Ext
nce
l
nce
bala
Absorption Absorption
(a) (b)
external deficit at point 2, the tendencies are the opposite but the result is the same,
a tendency to regain external balance.
The tendency to regain internal balance is shown in Figure 15–5b. When there
is inflation (point 3), it affects both the real exchange rate and real absorption. If the
nominal exchange rate remains fixed (or is not allowed to depreciate as fast as infla-
tion), the rise in Pn causes a real appreciation. At the same time, the rise in prices
can cause a fall in the real value of absorption, assuming that the central bank does
not take steps to increase the money supply to compensate for inflation. Under these
assumptions, the economy would move from inflation at point 3 back toward internal
balance. Unemployment (point 4) would be self-correcting also if prices are able to
fall as easily as they rise, but this seldom is the case.
Despite these self-correcting tendencies, in practice, they often fail to work smoothly
or quickly enough because of structural rigidities in the economy. For instance,
exchange-rate changes may take time to affect actual imports and exports, perhaps
as long as two years to have a full impact. In economies like Ghana and Zambia,
dominated by one or two export products such as cocoa, oil, and copper, with long
gestation periods for new investment, supply elasticities for tradables may be espe-
cially low, and foreign deficits can persist for a time despite real devaluations.
Nontradables prices probably rise very quickly when demand exceeds supply,
as in Figure 15–5b. But in many developing economies, inflation, once started, may
resist corrective policies, and prices do not fall so easily when there is unemploy-
ment: Unions strike wage bargains that try to maintain real wages by continually
raising nominal wages; banks use their market power to keep interest rates high; pro-
ducers depend on imports, the prices of which are responsive only to exchange rate
adjustments; and large firms with monopoly or oligopoly power keep prices up to
cover costs that resist downward pressures. Such rigidities have frequently been cited
to explain chronic trade deficits and inflation in Latin America, especially in Argen-
tina and Brazil.
However, arguments about structural rigidities can be overstated. There is some
flexibility in production for most export industries, even in the short term. And many
producer prices are quite flexible, including those of most farm products, those in the
large informal sector, and even those of some modern manufacturing firms. Neverthe-
less, the automatic tendencies toward external and internal balance depicted in Fig-
ure 15–5 are likely to be too slow and politically painful to satisfy most governments.
Not all the barriers to adjustment are structural. Sometimes, policies work against
adjustment. When foreign reserves fall, for example, the money supply also falls
automatically unless the central bank’s policy is to sterilize these shifts by expanding
domestic credit to compensate for the fall in reserves and keep the money supply from
falling. Sterilization prevents the move from points 1 or 2 of Figure 15–5a toward exter-
nal balance. And nominal exchange rates respond to changing market conditions only
if the exchange rate is allowed to float or the government makes frequent adjustments
in the nominal exchange rate to match changing economic conditions.
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 559
However, the opposite policy, a fixed nominal exchange rate, is needed if infla-
tion in nontradables prices is to cause a real exchange-rate appreciation, as depicted
at point 3 of Figure 15–5b. This fixed nominal rate is called an exchange rate anchor
because the fixed rate alone can halt the upward drift of prices as the economy
moves due south from point 3. Chile used such an anchor to slow inflation during
the late 1970s (Box 15–2). If government devalues the rate to keep up with inflation,
B O X 15 – 2 P I O N E E R I N G S TA B I L I Z AT I O N :
C H I L E , 1973 – 8 4
In the last year of the Salvadore Allende regime in Chile, when the public sec-
tor deficit soared to 30 percent of the gross domestic product (GDP) and was
financed mostly by printing money, inflation exceeded 500 percent a year. In
1973, General Augusto Pinochet overthrew Allende and established an auto-
cratic regime. An early goal of his government was to stabilize the economy. It
proved to be a difficult task of many years, with important lessons for later stabi-
lizations in Latin America.
Faced by rapid inflation and unsustainable external deficits, the government
imposed a fiscal and monetary shock on the economy. The budget deficit was
cut to 10.6 percent of GDP in 1974 and again to 2.7 percent in 1975. Monetary
policy was tight: From the second quarter of 1975 through the middle of 1976, it
has since been estimated, households and firms were willing to hold more money
than was in circulation. But inflation persisted; consumer prices nearly doubled in
1977.
Despite draconian measures, prices continued to rise for two reasons. First,
the peso was aggressively devalued to improve the foreign balance, the more so
because of the 40 percent fall in copper prices in 1975. In 1977, the peso was
worth about one-80th its 1973 value against the dollar. Second, wages in the
formal sector were determined by rules that permitted adjustments based on the
previous year’s rate of inflation, a rule that helped perpetuate the higher rates of
earlier years. It also was argued by some that the monetary policy was not strin-
gent enough.
In 1978, the government switched gears and began using the exchange rate
as its main anti-inflation weapon. At first a crawling peg was adopted with prean-
nounced rates, the tablita a, that did not fully adjust to domestic inflation. In 1979,
the rate was fixed at 39 pesos to the dollar for three years. The appreciating real
exchange rate, or anchorr, helped control inflation, which was down to 10 percent
by 1982. But it also discouraged export growth and contributed to a growing
5 6 0 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
current-account deficit. At the same time, Chile liberalized its controls over for-
eign capital flows and attracted large inflows of loans: Net long-term capital rose
from negligible amounts before 1978 to average over $2 billion a year in the fol-
lowing five years, equivalent to 8 percent of GDP in 1980. This inflow not only
financed the growing current deficit but contributed to the real appreciation of
the exchange rate.
Not until after 1984 did Chile finally achieve a semblance of both internal and
external balance. It did so through a large real devaluation, approaching 50 per-
cent, supported by tighter fiscal and monetary policies. After a decade and a half
of falling income per capita, Chilean incomes grew by 5.8 percent a year from
1985 to 1991.
Source: Based on the account by Vittorio Corbo and Andrés Solimano, “Chile’s Experience with
Stabilization Revisited,” in Michael Bruno et al., eds., Lessons of Economic Stabilization and Its After-
math (Cambridge, MA: MIT Press, 1991).
Brazil’s practice for many years, then real appreciation is thwarted and there is no
anchor. Similarly, real absorption falls with inflation only if the government fixes its
expenditure and its deficit in nominal terms and allows inflation to erode the real
value of the expenditure and if the central bank restrains the money supply to grow
more slowly than inflation. More typically, the fiscal authorities adjust the expendi-
ture, while the monetary authorities adjust both the money supply and the nominal
exchange rate, to fully compensate for inflation. In that case, rising prices have no
impact on the real exchange rate or real absorption and an inflationary economy
remains at point 3 in Figure 15–5b.
STABILIZATION POLICIES
Whether the barriers to rapid automatic adjustment are inherent in the economic
structure or created by policy contradictions, in most cases, governments need to
take an active role to stabilize their economies. They have three basic instruments for
doing so: exchange-rate management, fiscal policy, and monetary policy.
Alternative exchange-rate regimes were introduced in Chapter 12. Govern-
ments can vary the exchange rate by having the central bank offer to buy and sell
foreign currency at a predetermined or fixed official exchange rate (e in our nomen-
clature) that nevertheless can be changed from time to time or by allowing the rate to
float in the currency market, although the central bank sometimes may intervene to
influence the price. An intermediate case is the crawling peg, under which the central
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 561
bank determines the rate but changes it frequently, as often as daily, to ensure that
the official rate stays in line with domestic and world inflation; this results in a con-
stant or slowly adjusting real exchange rate (P).
Governments have two policies that can influence the level of absorption. Fiscal
policy, adjusting levels of government expenditure and taxation, directly affects the
government’s components of consumption and investment. It also influences pri-
vate expenditure, especially consumption, which depends on disposable income, or
income net of taxes. Monetary policy also affects private expenditure. If the central
bank acts to increase the money supply, as described in Chapter 12, it increases the
liquidity of households and firms, lowers interest rates, and stimulates private con-
sumption and investment.
The power of the phase diagram is that it indicates the necessary directions for
these policies, depending on the state of the economy. Figure 15–6 provides such a
policy map. It shows the same external and internal balance lines as in the previous
diagrams but adds a new element: four policy quadrants, I to IV, within which the
policy prescription always is the same.
Take, for example, point 1, which has been placed on the external balance line
but in the inflationary zone. For many years, Brazil was in this situation, with buoy-
ant exports and balance in foreign payments but chronic inflation running from 40 to
well over 100 percent a year. Because the demand for nontradables exceeds supply,
we know that one necessary correction is a reduction in real absorption, monetary
and fiscal austerity, that would reduce demand and move the economy due west
from point 1. But, if that is the only policy taken, the economy would not reach inter-
nal balance until point 4, in the zone of external surplus. One imbalance is exchanged
for another. To avoid generating a surplus, reduced absorption needs to be accompa-
nied by an appreciation of the exchange rate, a move due south from point 1. The
result would be a move approximately toward the equilibrium or bliss point, 0.
Note three things about this result. First, this combination of policies, austerity
and appreciation, would work from any point within quadrant I to return the econ-
omy to equilibrium. That is, the same combination is needed whether the economy
had inflation with a moderate external surplus or inflation with a moderate deficit,
either just above or just below the EB line. If the economy starts just below external
balance, with a moderate deficit, it may seem strange (counterintuitive) to recom-
mend an appreciation that, on its own, would worsen the deficit. But the reduction
in absorption, needed to reduce inflation, also reduces the deficit because it also low-
ers the demand for tradables. Indeed, it reduces the demand for tradables too much
and throws the economy into surplus; this is the reason an appreciation is needed. Of
course, the relative intensity of each policy is different, depending where in quadrant I
the economy starts. But the basic principle holds: Anywhere in quadrant I, the right
combination of policies is austerity and appreciation, the combination that moves
the economy toward point 0.
5 6 2 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
IV I
IB EB
4 ΔA > 0 6 ΔA < 0 1
Δe < 0 Δe < 0
Real exchange rate (P = ePt *IPn )
0 5
Δe > 0 Δe > 0
2
3
ΔA < 0
ΔA > 0
III II
Absorption (A)
Second, in general, two policy adjustments are required to move toward equi-
librium. This is a simple example of the general rule enunciated by Dutch economist
Jan Tinbergen: To achieve a given number of policy goals, it generally is necessary to
employ the same number of policy instruments. Here we have two goals, internal
and external balance, and need adjustments in both absorption (austerity) and the
real exchange rate (appreciation) to reach them both. It is not always necessary to
use two goals, however. If the economy lies just to the east of equilibrium at point
5, then a reduction in absorption achieves internal and external balance simultane-
ously. And, if the initial situation is point 6, due north of 0, then appreciation alone
does the job.
Third, we could view the policy prescription in either of two ways. Austerity is
needed to reduce inflation (move west) and appreciation is used to avoid surplus
(move south). Or appreciation can be targeted on internal balance (move south toward
point 2) but alone would cause a deficit, so that austerity then is required to restore
external balance. Therefore, no logic in macroeconomics suggests that one particular
policy should be assigned to one particular goal. Economic institutions often do this
anyway. In practice, the central bank might use the exchange rate to achieve external
EQUILIBRIUM IN A SMALL, OPEN ECONOMY 563
balance while the finance ministry uses the budget for internal balance. But if these
two approaches are not coordinated, they may well fail to reach equilibrium.
With these principles established for quadrant I, it is fairly routine to go around
the map in Figure 15–6 and see what policy responses are required:
Throughout this book we refer to different kinds of economic problems that are asso-
ciated with developing countries, including the Dutch disease, debt crises, terms-
of-trade shocks, foreign-exchange shortages, destructive inflation, and droughts or
other natural catastrophes. The Australian model and its phase diagram can be used
to show how these and other shocks affect macroeconomic balance and how they
should be handled.
DUTCH DISEASE
In Chapter 18 we will discuss the strange phenomenon of the Dutch disease, in
which a country that receives higher export prices or a larger inflow of foreign capital
may end up worse off than without the windfall. The Dutch disease was first analyzed
by Australian economists Max Corden and Peter Neary, using a version of the open-
economy model.6 Figure 15–7 traces the impact of a windfall gain using the phase
diagram. (Box 18–1 provides an alternative exposition of Dutch disease.)
An economy in equilibrium at point 1 suddenly begins to receive higher prices
for its major export or is favored by foreign aid donors or foreign investors. All the
oil producers, from Saudi Arabia to Indonesia to Mexico, were in this position in the
1970s, as were coffee (and many other commodities) exporters during the boom of
the mid-1970s. Egypt and Israel were rewarded with large aid programs by the United
States after the Camp David accord of 1978, as was Ghana by the World Bank and
others during its stabilization of the 1980s (Box 15–3). Both Chile in the late 1970s
and Mexico after its stabilization in the late 1980s received large inflows of private
capital, much of it a return of previous flight capital. Foreign exchange windfalls are
more frequent than sometimes is supposed. In some cases, these windfalls result
from new discoveries of natural resources, such as the major offshore oil reserves dis-
covered by Ghana in 2007.
When the windfall occurs, the supply of tradable goods rises at any given price.
This can be shown as a rightward shift in the supply curve in Figure 15–2a. In the
phase diagram of Figure 15–7, there is a rightward shift in the EB curve. At point 1,
for example, which had been in external equilibrium along EB1, the economy now is
in surplus, so the new EB curve must be to the right—for example, at EB2. The econ-
omy cannot remain at point 1 because the inflow of reserves increases the money
supply; this adds to demand and, because the windfall increases private income
and government revenue, leads to greater expenditure. So absorption rises, a move
W. Max Corden and J. Peter Neary, “Booming Sector and Deindustrialisation in a Small Open
6
EB1
IB
EB2
ΔF > 0
Real exchange rate (P = ePt */Pn )
1 Δ Reserves > 0 2
P1
Δ Pn > 0
P3 3
A1 A3
Absorption (A)
from point 1 toward point 2. This moves the economy off its internal balance, into
inflation.7
The resulting rise in Pn has two effects: a reduction in real absorption that par-
tially corrects the initial rise in A and, assuming the official rate is fixed, a real appreci-
ation of the exchange rate. (The real rate also appreciates if the nominal rate is floating
because the greater supply of foreign currency drives down the price of foreign cur-
rency.) Therefore, the economy first moves from point 1 toward point 2 in Figure 15–7,
then begins to head in the general direction of the new equilibrium, point 3. In this
case, market forces are likely to be sufficient to reach the new equilibrium, unless the
7
If the windfall is an inflow of capital, this treatment is precise. In the case of a rise in export prices,
however, the move from point 1 to point 2 is an approximation. Strictly speaking, a rise in export prices
should raise Pt*, a depreciation of the real exchange rate that moves the economy upward from point 1,
after which the economy moves east toward EB2.
5 6 6 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
B O X 15–3 R E C O V E R I N G F R O M M I S M A N A G E M E N T :
G H A N A, 1 9 8 3 –9 1
Source: This account is based on Ishan Kapur et al., Ghana: Adjustment and Growth, 1983–91
(Washington, DC: International Monetary Fund, 1991).
A P P L I C AT I O N S O F T H E A U S T R A L I A N M O D E L 5 6 7
authorities prevent appreciation and maintain real absorption and so keep the econ-
omy in an inflationary posture like point 2.
What, then, is the problem? The economy is at a new equilibrium, its terms of
trade improved, its currency appreciated and so citizens have more command over
foreign resources, people spending and consuming more without having to work any
harder. There are two flaws in this otherwise idyllic picture. First, such windfalls gen-
erally are temporary. When export prices fall or the capital inflow dries up, the EB
curve shifts back and a costly adjustment is necessary. We analyze that process in the
next section.
The second problem is that, in shifting from the old to the new equilibrium,
adjustments in the economy must be made. The real exchange rate P is lower, so St
has fallen, while Sn has risen. Because the booming export sector does not retrench,
nonboom tradables bear the brunt of the adjustment. Frictions in the labor market
are likely to mean at least temporary unemployment as workers switch from tradable
to nontradable production. If the tradable sector includes modern manufacturing,
then long-term development may be set back because manufacturing is the sector
likely to yield the most rapid productivity growth in the future. And if tradable indus-
tries close, it is more difficult to make the inevitable adjustment back toward point 1
when the windfall is over. This decline in nonboom-tradable production turns a for-
eign exchange windfall into a “disease.”
What can be done to cure the disease? The government could try to move the
economy back toward the old (and probably future) equilibrium at point 1. Its tools
are the official exchange rate, which would have to be devalued against the tenden-
cies of market forces, and expenditure, which would have to be reduced through
restrictive fiscal and monetary policies that also reduce inflation (lower Pn or at least
its growth). The resulting buildup of reserves and bank balances have to be sterilized
through monetary policy so they are held as assets and not spent. It is a neat politi-
cal trick to manage an austere macroeconomic policy in the face of a boom because
all the popular pressures are for more spending. Not too many countries have man-
aged it. Indonesia is among the few that have.
as Indonesia, Nigeria, and Venezuela, faced a similar kind of crisis once oil prices
began falling in the 1980s. We can understand the similarity between a debt crisis
and a decline in the terms of trade more clearly by seeing them in their common con-
text in the balance of payments. A decline in the terms of trade implies deterioration
in the balance of trade (in which the excess of imports over exports increases). All
else equal, a declining trade balance adds directly to the current account deficit. As
detailed in the appendix to this chapter, one of the few ways in which countries can
finance current account deficits is by borrowing abroad.8 Indeed, many developing
countries financed chronic current account deficits by borrowing abroad, in the pro-
cess accumulating enormous stocks of debt (often to levels greater than their GDP).
Debt crises ensue (as discussed in Chapter 13) when current account deficits become
unsustainable and lenders want to be repaid.
Figure 15–8 captures this process. An economy in balance at point 1 needs to find
additional resources to repay its foreign debt or needs to adjust to falling terms of
trade. The supply of tradables therefore shifts to the left in Figure 15–2a; in the phase
diagram, the EB curve also shifts leftward to EB2.9 If the crisis leads to debt relief or
additional foreign aid, the curve moves less far and might settle at EB3.
Now in foreign deficit, the economy begins losing reserves. If the government
has to repay some of the debt or falling export prices cut into its revenues, the govern-
ment needs to reduce its expenditures as well. Both cause a reduction in absorption.
These actions move the economy toward external balance but also into unemploy-
ment. To gain the new equilibrium at point 3, it is also necessary to devalue the cur-
rency. This could be done by the central bank under a fixed rate or by the foreign
exchange market under a floating rate. At the new equilibrium, the country produces
more and consumes fewer tradables because P has risen. This, of course, is a loss of
welfare for the populace. The surplus of St over Dt is used to repay the debt or simply
compensates for reduced export prices.
Debt crises and the hardships they cause are not an inevitable consequence
of borrowing to finance development, as was discussed in Chapter 13. If the bor-
rowed resources are invested productively, they increase the potential output of
both tradables and nontradables. Added production increases income and gener-
ates the capacity to repay the debt out of additional income, without a crisis and
an austerity program. Countries such as Korea and Indonesia have been large
international borrowers, but before the financial crisis of 1997–99, they escaped
debt crises.
8
The only other ways (beyond foreign borrowing) to finance a current account deficit are to attract for-
eign investment and/or to run down the central bank’s stock of foreign reserves. See the appendix to this
chapter for a concise summary of balance of payments accounting.
9
Strictly speaking, we cannot analyze the fall in export prices this way, but it is a reasonable approxima-
tion for many situations. See note 7.
A P P L I C AT I O N S O F T H E A U S T R A L I A N M O D E L 5 6 9
EB 2
EB 3
IB
Debt
relief EB1
3
P3
ΔP > 0
P1 1
ΔA < 0
A3 A1
Absorption ( A )
IB EB1
EB2
IMF
package
4
Real exchange rate (P = ePt */Pn )
2
P2
IMF
3 targets Δ e < 0?
P3
ΔA < 0 1
A2 A3
Absorption ( A )
there is little or no need to devalue to move from 1 to 3. Donors and the IMF never-
theless frequently insist on devaluation. Sometimes, that may be a requirement just
to reach a point like 3. In other cases, donors and the IMF may have in mind a self-
sustaining stabilization that will be valid even after aid is reduced and the external
balance curve moves back toward EB1. Whatever the motive, it is important to real-
ize that aid itself is a partial substitute for both devaluation and austerity. In essence,
the aid does what higher production of tradables otherwise must do and it finances
expenditures that otherwise must be cut. Ghana’s experience, which fits this descrip-
tion, is discussed in Box 15–3. More recently, Greece has suffered a major debt crisis,
the analysis of which draws on elements of both Figures 15–8 and 15–9 (Box 15–4).
B O X 1 5 – 4 TH E G R E E K D E B T C R I S I S O F 2 0 1 0 – 1 2
Problems of unsustainable debt are not limited to the poorest countries. Start-
ing in 2010, Greece faced a debt crisis so severe as to threaten the stability of
the European Union (EU). As of September 2011, it remained uncertain whether
Greece would be forced to default on its foreign debts or whether other mem-
bers of the EU would provide its second bailout package for Greece in two years.
In May 2010, the EU and the International Monetary Fund (IMF) provided a pack-
age of loans and balance of payments support worth €750 billion (approximately
$938 billion). In return, the government of Greece committed itself to an aus-
terity program that included severe reductions in expenditures and wages, ter-
mination of tens of thousands of government jobs, and extensive tax increases.
This agreement sparked widespread civil unrest in Greece, which contributed to
a lack of confidence in the country’s ability to implement the promised reforms
and service its foreign debts.
Greece’s debt problem accumulated over many years of current account defi-
cits, the magnitude of which ballooned during the late 2000s. Greece’s current
account deficits as a share of GDP began mounting in the mid-1990s. Deficits
on the order of 3 percent of GDP in the late 1990s more than doubled relative
to GDP during 2000–05. By 2005, Greece’s current account deficit was equiva-
lent to 7.5 percent of GDP, a proportion that doubled again to nearly 15 percent
by 2008. Between 2008 and 2010, Greece succeeded in reducing its current
account deficit to 10.5 percent of GDP, mainly by reducing imports. Borrowing by
the Greek government to finance its deficits had also grown rapidly, as the gov-
ernment budget deficit as a share of GDP increased from 5.7 percent in 2006
to 15.4 percent in 2009. The government’s total debt shot up from €183.2 bil-
lion in 2004 (equivalent to 99 percent of GDP, or approximately $228 billion)
5 7 2 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
IB EB EB'
EU/IMF
bailout
Real exchange rate (P = eP*t /Pn)
3
P3
4
P4
1
Greece in 2005
2
Greece in 2010
A3 A4
Absorption (A)
Another kind of stabilization also can be illustrated with Figure 15–9, rapid infla-
tion (or hyperinflation). In Bolivia’s hyperinflation of the mid-1980s or the chronic
inflations in the past in Brazil and Argentina, external balance is a secondary con-
sideration or not a major problem. Point 4 in the diagram depicts this situation.
Austerity still is required to move toward equilibrium at point 2 or 3 (if there is an
aid package), but devaluation only intensifies inflation. Instead, the currency must
be appreciated, which also dampens inflation. One way to achieve this would be to
fix the nominal rate and let the continuing (if decreasing) inflation in nontradable
prices (Pn) appreciate the real rate P. This is the exchange-rate anchor, a device used
often in Latin America, especially in Chile during the late 1970s, in Bolivia during the
5 7 4 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
mid-1980s, and in Argentina during the 1990s. It has the disadvantage that a lower
real rate discourages export growth. Yet investment in new exports may be part of a
strategy to open the economy, diversify exports, and move the external balance curve
to the right.
EB2
IB1
Food EB3
aid
EB1
IB2
Real exchange rate ( P = ePt*/Pn )
3 ΔY < 0 1
ΔG > 0
Absorption ( A )
F I G U R E 1 5 – 10 Drought
Drought or another natural disaster reduces the capacity to produce both
nontradables and tradables, so the curves shift to the west. Disaster relief
from abroad augments the external balance curve and shifts it to EB3, with
equilibrium at point 3. Remaining temporarily at point 1, the economy
becomes inflationary. The reduction in output and therefore in income reduces
absorption, but the government’s need to spend more on relief tends to offset
this move toward equilibrium. The outcome could be continued inflation.
SUMMARY
Many macroeconomic problems and crises in developing countries have their roots
in the balance of payments. Chapter 10 opens the discussion of balance of payments
by describing the role of foreign savings in financing investment. Chapter 13 dis-
cusses debt crises in developing countries, and Chapter 15 provides a model for diag-
nosing and responding to macroeconomic crises—including debt crises. Balance of
payments accounting provides a framework within which to understand many of the
opportunities and risks created by opening economies to trade and financial flows.
A P P E N D I X T O C H A P T E R 15 : N A T I O N A L I N C O M E A N D T H E B A L A N C E O F P A Y M E N T S 5 7 7
This appendix provides a brief overview of some key aspects of balance of payments
accounting.10
We begin by placing the balance of payments in the broader context of national
income accounting. The value added from all goods and services produced domes-
tically sums to Gross Domestic Product (GDP). In a closed economy, this exactly
equals gross national expenditure, which consists of private consumption on final
goods and services (C), government consumption and investment in final goods and
services (G), and private investments to expand the country’s capital stock (I). That
is, GDP = C + I + G in a closed economy. In Chapter 15, we refer to gross national
expenditure as absorption (A), or A = C + I + G. Adopting this notation, we can
say that in a closed economy, GDP = A. In addition, in a closed economy all of gross
national expenditures (or absorption) is paid to domestic entities, so GDP also equals
gross national income (GNI).
When we allow for trade with other countries, it is no longer required that gross
domestic product exactly equals absorption or gross national income. This is where
the balance of payments comes in. Balance of payments accounting records all of the
(home) country’s transactions with the rest of the world, including both trade and
financial flows. The main components of the balance of payments are the current
account, the financial account, and the capital account.
We begin by incorporating the current account into national income. The cur-
rent account records all foreign trade in goods (the trade balance) and services
(net factor income from abroad), along with net unilateral transfers (such as for-
eign aid or remittances from residents working abroad). The current account
also includes interest payments on foreign debts (debt service payments, but
not repayment of principal on the debt). In an open economy, part of absorption
may be spent on imports (M), and must be subtracted from A; conversely, some
domestic production may be exported abroad (X), and the resulting payments
to domestic firms are added to A. In this case, we can write the national income
accounting identity as
GDP = C + I + G + X - M [A15–1]
Substituting absorption (A) in for gross national expenditure, we can re-write equa-
tion A15–1 as
GDP = A + X - M [A15–2]
10
This discussion draws on Robert C. Feenstra and Alan M. Taylor, International Macroeconomics,
2nd edition (New York: Worth Publishers, 2012), Chapter 5 (to which the reader is referred for a more
detailed treatment), as well as the IMF’s Balance of Payments and International Investment Position
Manual, 6th edition (Washington, DC: International Monetary Fund, 2011).
5 7 8 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y
That is, gross domestic product equals domestic expenditure plus net exports. Rear-
ranging the terms of equation A15–2, we can see that A - GDP = M - X. In other
words, if domestic spending exceeds GDP, then our imports exceed our exports and
we run a negative trade balance. The trade balance is typically the major component
of the current account. As noted above, net factor income from abroad (NFIA), or
income from the net export of services (e.g., wages paid to a home country resident
working abroad for a foreign firm), is also part of the current account and is included
in gross national income (GNI). Thus, GNI = GDP + NFIA. The difference between
gross national income (GNI) and gross national disposable income (Y) is that the lat-
ter also includes the final component of the current account, net unilateral transfers
(NUT). Putting these steps together, we can write
where CA is the current account, which equals the sum of net exports, net factor
income from abroad, and net unilateral transfers. Equation A15–3 is important. It
implies that if the home country’s total expenditures exceed its total receipts, the dif-
ference must be provided by running a deficit on the current account. From equa-
tion A15–3, we can see that A - Y = -CA. The current account balance thus serves
as an indicator of whether a country’s spending exceeds its income.
As discussed in Chapter 10, we can also approach the current account from the
perspective of national savings. National savings is what is left over from total receipts
net of private and government consumption, or S = Y - C - G. If we expand A in
equation A15–3 to write Y = C + I + G + CA and subtract C + G from both sides,
we arrive at the current account identity,
S = I + CA [A15–4]
which says that we can only save more than we invest if there is a CA surplus, and we
can only invest more than we save if there is a CA deficit.
Domestic savings has two components—private savings (Sp) and government
savings (Sg). Income in the private sector must be consumed, saved, or paid to the
government as taxes. Thus,
Sp = Y - C - T [A15–5]
Similarly, government savings is the difference between the tax revenue collected by
the government (T) and its expenditures (G), or
Sg = T - G [A15–6]
which is the government’s fiscal surplus (or deficit, if G 7 T). The sum of these two
equations equals our previous observation that S = Y - C - G = Sp + Sg , or total
national savings is the sum of private savings and government savings. Substituting
this idea into equation A15–4 and rearranging, we can see that
CA = Sp + Sg - I [A15–7]
A P P E N D I X T O C H A P T E R 15 : N A T I O N A L I N C O M E A N D T H E B A L A N C E O F P A Y M E N T S 5 7 9
CA = (Sp - I) + (T - G) [A15–8]
which tells us that if investment exceeds private savings and/or the government runs
a fiscal deficit, the current account must be in deficit by an equivalent amount.
The rest of the balance of payments story is about how the current account is
financed. As noted above, in addition to the current account, the balance of pay-
ments includes the financial account (FA) and the capital account (KA). The financial
account records foreign transactions involving financial assets such as stocks, bonds,
and real estate ownership. The major components of the financial account (using
the IMF definition) are foreign direct investment, portfolio investment, and (impor-
tantly) reserve assets (which is the stock of foreign currency—usually U.S. dollars,
euros, or yen) held by the central bank. The capital account refers to the transfer of
assets such as gifts to and from abroad.11
A fundamental principle of balance of payments accounting is that bills must be
paid. This implies that the balance of payments must sum to zero, or
CA + FA + KA = 0 [A15–9]
This is the balance of payments identity. It implies that there are only three ways in
which a country can finance a current account deficit: 1) it can borrow from abroad
(from foreign governments, banks, or international lending bodies), 2) it can attract
capital inflows (in the form of direct or portfolio investment), or 3) it can run down its
stock of foreign reserves. If the current account is in deficit, then it must be balanced
by a surplus in the financial and capital accounts (which potentially includes running
down the country’s stock of foreign reserves). If the balance of payments must also
equal zero, what is meant when we hear about a “balance of payments deficit?” In
common use, a balance of payments deficit refers to a situation in which a CA deficit
is not fully balanced by a surplus in FA + KA where FA is defined narrowly to exclude
the foreign reserve assets of the central bank. In such a case, the total balance of pay-
ments must be brought to zero by a change in foreign reserves (running them down,
in this instance, to finance the remaining CA deficit).
Running a surplus in the (non-reserve) financial and capital accounts means
that the home country is either borrowing from abroad or selling domestic assets
to foreigners. In either case, the home country is increasing the claims by foreigners
11
This narrow definition of the capital account (and broad definition of the financial account) is in
accordance with the approach now taken by the IMF, along with OECD and the UN System of National
Accounts. Previously, the accepted tradition was to define the capital account broadly enough to include
everything that the IMF now assigns to the financial account, with the exception of the reserve assets of the
central bank.
5 8 0 [ C H . 1 5 ] M A N A G I N G S H O R T- R U N C R I S E S I N A N O P E N E C O N O M Y