The Open Economy

Chapter 6 of Macroeconomics, 8th

edition, by N. Gregory Mankiw
ECO62 Udayan Roy
 Chapter Outline
• In chapter 2, we saw that Y = C + I + G + NX
when Y, C, I, G, and NX are interpreted as data
• In chapter 3, we saw
– a long-run theory of Y and
– a long-run theory of how Y is split between C, I,
and G in a closed economy
• In this chapter, we will see
– a long-run theory of how Y is split between C, I, G
and NX in an open economy
 Saving – Investment = Net Exports
• In chapter 2, we saw that Y = C + I + G + NX
• Therefore, Y − C − G − I = NX
• In Ch. 3, Y − C − G was defined as national saving
(S)
• Therefore, S − I = NX

• But in Chs. 3 and 4, we had assumed a closed

economy (that is, NX = 0)
• Consequently, we had S = I
• That’s no longer true in an open economy
 Saving, investment, and the trade balance (percent of GDP)
1960-2007
24% 8%

22% investment
6%
20%
4%
18%

16% 2%
saving
14% 0%

12%
-2%
10%

8%
trade balance -4%
(right scale)
6% -6%
1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
 Chapter 3 Recap
• 𝑌 = 𝐴𝐾 0.3 𝐿0.7
• 𝐶 = 𝐶0 + 𝐶𝑦 ∙ (𝑌 − 𝑇)
• 𝑆 =𝑌−𝐶−𝐺

Predictions Grid
Y C S
K, L, A (Technology) + + +
Net Taxes, T − +
Co + −
Now for the new stuff!
Govt Spending, G −
 Perfect Capital Mobility
• Assumption: people are free to lend to or
borrow from anyone anywhere in the world
• Assumption: lending to foreign borrowers is
in no way different from lending to domestic
borrowers

• The real interest rate is r for domestic loans

and r* for loans to foreigners
• Our two assumptions imply r = r*
 Real Interest Rate: predictions
• Our assumption of perfect capital mobility
implies that real interest rates will be the
same both at home and abroad: r = r*
• Further, the foreign real interest rate is
assumed exogenous
• Therefore, we already have a complete (and
trivial!) theory of the domestic real interest
rate Predictions Grid
r
r* r
r* +
 “Small Country”
• The assumption that the foreign real interest
(r*) rate is exogenous and that it determines
the domestic real interest rate (r = r*)
represents the idea that the domestic
economy is affected by the foreign economy
and is unable to affect the foreign economy
• In other words, we assume that the domestic
economy is a “small country”
The Real Interest Rate: predictions
• r = r* Predictions Grid
Y C S r
K, L, Technology + + +
Taxes, T − +
Co + −
Govt, G −
r* +
 Investment and the real interest rate
• Assumption: investment spending is inversely
related to the real interest rate
• I = I(r), such that r↑⇒ I↓
r

r* r
I
I(r)

I (r )

I
Investment and the real interest rate
r Investment is still a
downward-sloping function
of the interest rate,
but the exogenous
world interest rate…
r*
…determines the
country’s level of
investment.
I (r )

I (r* ) I
 Investment and the real interest rate
r
• Algebraically, I = Io − Irr
– Here Ir is the effect of
r on I and r*B Io2 − Irr

– Io represents all other r*A

Io1 − Irr
factors that also affect
business investment I
spending
• such as business
optimism,
technological progress,
etc.
 Investment: example
• Suppose r* = 7 percent
• Then, r = r* = 7 percent
• Suppose I = 16 – 2r is the investment function
• Then, I = 16 – 2 ✕ 7 = 2
r* r
I
I(r)
 Investment: predictions
• I = Io − Irr = Io − Irr* Predictions Grid
Y C S r I
– Note that this expresses K, L, Technology + + +
investment (which is Taxes, T − +
endogenous) entirely in Co + −
terms of an exogenous Govt, G −
variable (r*) and two r* + −
parameters (Io and Ir) Io +
– So, this tells us all we
can say about
investment spending
 Net Exports: predictions
Predictions Grid
• We saw earlier that Y C S r I NX
NX = S – I K, L, Technology + + + +

• So, we can predict Taxes, T

Co
−
+
+
−
+
−
changes in net Govt, G − −
exports (NX) from r* + − +
Io + −
what we already
know about saving (S)
and investment (I)
 NX = S – I
• So far, we have seen how to calculate saving
(S) and investment (I)
• The difference gives us net exports: NX = S – I

r* r
I
I(r)
NX = S – I
G
K, L, F(K, L) Y S=Y–C–G
C
C(Y – T), T
 Net Exports: example
• Suppose F(K, L) = 5K0.3L0.7 and K = 2 and L = 10.
Then Y = 30.85. Suppose T = 0.85. Therefore,
disposable income is Y – T = 30.
• Now, suppose C = 2 + 0.8(Y – T). Then, C = 2 + 0.8
✕ 30 = 26
• Suppose G = 3. Then, S = Y – C – G = 30.85 – 26 –
3 = 1.85
• Suppose r* = 7 percent. Then, r = r* = 7 percent.
Suppose I = 16 – 2r is the investment function.
Then, I = 16 – 2 ✕ 7 = 2
• Then NX = S – I = 1.85 – 2 = – 0.15
 The Story So Far
• 𝑌 = 𝐴𝐾 0.3 𝐿0.7
• 𝐶 = 𝐶0 + 𝐶𝑦 ∙ (𝑌 − 𝑇)
• 𝑆 =𝑌−𝐶−𝐺
• 𝑟 = 𝑟∗ Predictions Grid
Y C S r I NX
• 𝐼 = 𝐼0 − 𝐼𝑟 ∙ 𝑟 K, L, Technology + + + +
• 𝑁𝑋 = 𝑆 − 𝐼 Taxes, T − + +
Co + − −
Govt, G − −
r* + − +
Io + −
 If the economy were closed…
r S
…the interest
rate would
adjust to
equate
investment
and saving: rc
I (r )

I (rc ) S, I
S
 But in a small open economy…
r
the exogenous S
world interest
rate determines
investment… NX
r*
…and the
difference rc
between saving
and investment I (r )
determines net
capital outflow I1 S, I
and net exports
 Next, four experiments:
1. Fiscal policy at Predictions Grid
Y C S r I NX
home (G and T) K, L, Technology + + + +
Taxes, T − + +
2. Fiscal policy abroad Co + − −
(r*) Govt, G − −
r* + − +
3. An increase in Io + −

investment
demand (Io)
4. Trade restrictions
 1. Fiscal policy at home
r S2 S1
An increase in G
or decrease in T NX2
reduces saving. r
1
*

NX1
Results:
I  0
NX  S  0 I (r )

I1 S, I
 NX and the federal budget deficit
(% of GDP), 1965-2009
8%
Budget deficit 2%
6% (left scale)

4% 0%

2%
-2%

0%

-4%
-2% Net exports
(right scale)
-4% -6%
1965 1970 1975 1980 1985 1990 1995 2000 2005 2010
 2. Fiscal policy abroad
r
Expansionary S1
NX2
fiscal policy
abroad raises r2*
NX1
the world
interest rate. r1
*

Results:
I  0 I (r )
NX  I  0
S, I
I (r )
2
*
I (r1* )
 NOW YOU TRY:
3. An increase in investment demand
r
Use the S
model to
determine r*
the impact of
an increase
NX1
in investment
demand on
NX, S, I, and I (r )1
net capital
outflow. I1 S, I
 ANSWERS:
3. An increase in investment demand
r
S
I > 0, NX2
S = 0, r*
net capital
outflow and
NX fall NX1
by the I (r )2
amount I
I (r )1

I1 I2 S, I
Nominal and Real

EXCHANGE RATES
 The nominal exchange rate

e = nominal exchange rate,

the relative price of
domestic currency
in terms of foreign currency
(e.g. Yen per Dollar)
A few exchange rates, as of 6/24/2009

country exchange rate

Euro area 0.72 Euro/$
Indonesia 10,337 Rupiahs/$
Japan 95.9 Yen/$
Mexico 13.3 Pesos/$
Russia 31.4 Rubles/$
South Africa 8.1 Rand/$
U.K. 0.61 Pounds/$
 The real exchange rate

ε = real exchange rate,

the relative price of
the lowercase domestic goods
Greek letter in terms of foreign goods
epsilon
(e.g. Japanese Big Macs per
U.S. Big Mac)
~ McZample ~
• Big Mac
• price in Japan:
P* = 200 Yen
• price in USA:
P = $4.00
• nominal exchange rate
e = 100 Yen/$ To buy a U.S. Big Mac,
someone from Japan
𝑒×𝑃 would have to pay an
𝜀= amount that could buy
𝑃∗ 2 Japanese Big Macs.
Understanding the units of ε
e P
ε 
P *
(Yen per $)  ($ per unit U.S. goods)

Yen per unit Japanese goods

Yen per unit U.S. goods


Yen per unit Japanese goods

Units of Japanese goods


per unit of U.S. goods
 ε in the real world & our model
• In the real world:
We can think of ε as the relative price of
a basket of domestic goods in terms of a
basket of foreign goods
• In our macro model:
There’s just one good, “output.”
So ε is the relative price of one country’s
output in terms of the other country’s output
 Purchasing Power Parity
• This is the simplest theory of the real
exchange rate ε

• PPP assumption: ε = 1 ε=1

NX

• That’s it! NX

• The PPP assumption is also called the Law of

One Price (LOOP)
 Purchasing Power Parity
• Nothing can Predictions Grid (PPP)
affect the real Y C S r I NX ε

exchange rate, K, L, Technology + + + +

Taxes, T − + +
under PPP Co + − −
• because it is Govt, G − −
r* + − +
always ε = 1 Io + −
under PPP
PPP is too easy! Besides the facts do not
give it much support. So, next comes a
more sophisticated theory of the real
exchange rate.
 How NX depends on ε: approach 2

ε  U.S. goods become more expensive

relative to foreign goods
 EX, IM
 NX
 The NX curve for the U.S.
ε

so U.S. net
When ε is exports will
relatively low, be high
U.S. goods are
relatively ε1
inexpensive
NX (ε)
0
NX(ε1) NX
The NX curve for the U.S.
ε At high enough
values of ε,
ε2 U.S. goods become
so expensive that
we export
less than
we import

NX (ε)
NX(ε2) 0 NX
 U.S. net exports and the real exchange rate, 1973-2009
4% Trade-weighted real 140
exchange rate index
2% 120

(March 1973 = 100)

100
0%
(% of GDP)

80
-2%
NX

60

Index
-4%
40
Net exports
-6%
(left scale) 20

-8% 0
1970 1975 1980 1985 1990 1995 2000 2005 2010
 The Net Exports Function
• The net exports function reflects this inverse
relationship between NX and ε :
NX = NX(ε )
 The Net Exports Function
• NX = NX(ε)

• Specific form: NX = NXo – NXεε

– Here, NXo represents all factors—other than the
real exchange rate—that also affect net exports
• Examples: preferences, tariffs and other trade policy
variables, foreign GDP, etc.

• Example: NX = 19.85 – 2ε
 Net Exports: calculation
• We just saw that NX = NXo – NXεε
• Therefore, NXεε = NXo – NX
• Therefore, ε = (NXo – NX)/NXε
– In our numerical example, NX = –0.15 was shown
earlier
– Suppose NX = 19.85 – 2ε, as in the previous slide.
Then, NXo = 19.85 and NXε = 2
– Therefore, ε = (19.85 – (–0.15))/2 = 10 (Yeay!)
 The Story So Far
• 𝑌 = 𝐴𝐾 0.3 𝐿0.7 Predictions Grid
Y C S r I NX ε
• 𝐶 = 𝐶0 + 𝐶𝑦 ∙ (𝑌 − 𝑇) K, L, Technology + + + + −
Taxes, T − + + −
• 𝑆 =𝑌−𝐶−𝐺 Co + − − +

• 𝑟 = 𝑟∗ Govt, G − − +
r* + − + −
• 𝐼 = 𝐼0 − 𝐼𝑟 ∙ 𝑟 Io + − +

• 𝑁𝑋 = 𝑆 − 𝐼 NXo +

𝑁𝑋0 −𝑁𝑋
• 𝑁𝑋 = 𝑁𝑋0 − 𝑁𝑋𝜀 ∙ 𝜀 which yields 𝜀 =
𝑁𝑋𝜀
 Real Exchange Rate: example
• Suppose F(K, L) = 5K0.3L0.7 and K = 2 and L = 10. Then Y
= 30.85. Suppose T = 0.85. Therefore, disposable
income is Y – T = 30.
• Suppose C = 2 + 0.8(Y – T). Then, C = 2 + 0.8 ✕ 30 = 26
• Suppose G = 3. Then, S = Y – C – G = 30.85 – 26 – 3 =
1.85
• Suppose r* = 7 percent. Then, r = r* = 7 percent.
Suppose I = 16 – 2r is the investment function. Then, I =
16 – 2 ✕ 7 = 2
• Then NX = S – I = 1.85 – 2 = – 0.15
• As NX = 19.85 – 2ε is the net exports function, we get
NX = 19.85 – 2ε = – 0.15.
• Therefore, ε = 10
 Real Exchange Rate: calculation

r* r
I NX(ε)
I(r)
NX = S − I
G
K, L, F(K, L) Y S=Y–C–G
C
C(Y – T), T
 Real Exchange Rate: predictions
• As net exports Predictions Grid
(NX) and the real Y C S r I NX ε
exchange rate (ε) K, L, Technology + + + + −
are inversely Taxes, T − + + −
related, the NX Co + − − +
and ε columns
Govt, G − − +
are opposites
r* + − + −
• Note that an
Io + − +
increase in the
net exports NXo +
function has no
effect on net
exports
 Next, four experiments:
1. Fiscal policy at home Predictions Grid
(G and T) Y C S r I NX ε
K, L, Technology + + + + −
2. Fiscal policy abroad Taxes, T − + + −
(r*) Co + − − +
Govt, G − − +
3. An increase in r* + − + −
investment demand Io + − +
(Io) NXo +

4. Trade policy to restrict

imports (NXo)
 1. Fiscal policy at home

A fiscal expansion S 2  I (r *)
reduces national ε S 1  I (r *)
saving, net capital
outflow, and the
ε2
supply of dollars
in the foreign
exchange market… ε1

NX(ε )
…causing the real
NX
exchange rate to rise NX 2 NX 1
and NX to fall.
 2. Fiscal policy abroad
An increase in r* S 1  I (r1 *)
reduces
ε S 1  I (r2 *)
investment,
increasing net
capital outflow and ε1
the supply of
dollars in the
foreign exchange ε2
market…
NX(ε )

…causing the real NX

NX 1 NX 2
exchange rate to fall
and NX to rise.
 NOW YOU TRY:
3. Increase in investment demand

Determine the ε S1  I 1
impact of an
increase in
investment
demand on
net exports, ε1
net capital
outflow, NX(ε )
and the real NX
exchange rate NX 1
 ANSWERS:
3. Increase in investment demand
An increase in S1  I 2
investment ε S1  I 1
reduces net
capital outflow ε2
and the supply
of dollars in the
foreign ε1
exchange
market… NX(ε )

…causing the real NX

NX 2 NX 1
exchange rate to rise
and NX to fall.
 4. Trade policy to restrict imports

At any given value of ε,

an import quota ε S I
 IM  NX
 demand for ε2
dollars shifts
right ε1
NX (ε )2
Trade policy doesn’t
NX (ε )1
affect S or I , so
capital flows and the NX
NX1
supply of dollars
remain fixed.
 4. Trade policy to restrict imports

Results:
ε S I
ε > 0
(demand
increase) ε2
NX = 0
(supply fixed) ε1
IM < 0 NX (ε )2
(policy)
NX (ε )1
EX < 0
(rise in ε ) NX
NX1
Nominal interest rate, inflation rate, price level

NOMINAL VARIABLES: OPEN

ECONOMY
 Chapter 5 is still applicable!
• Go back to Chapter 5 and review the steps in
the calculations for the long-run values of the
nominal variables i, π, and P.
• You will notice that at no point was it assumed
that the economy is closed (NX = 0)
• Therefore, the results of Chapter 5 are true for
open economies
 Chapter 5 Results—true for an open
economy also
• 𝝅 = 𝑴𝒈 − 𝒀𝒈
• 𝒊 = 𝒓 + 𝑴𝒈 − 𝒀𝒈 = 𝒓∗ + 𝑴𝒈 − 𝒀𝒈
Predictions Grid (Long Run, Open Economy)
𝑴∙𝒊
• 𝑷= Y C S r I NX ε π i P
𝑳𝟎 ∙𝒀 K, L, Technology + + + + − −
Taxes, T − + + −
Co + − − +
Govt, G − − +
r* + − + − + +
Io + − +
NXo +
Mg − Yg + + +
M +
 The Nominal Exchange Rate
• Recall that the real exchange rate is
e P
ε 
P*
 Therefore, the nominal exchange rate is
P*
e  ε 
P
 The Nominal Exchange Rate:
predictions
• ε = eP/P* Predictions Grid (Long Run, Open Economy)
Y C S r I NX ε π i P e
• εP */P = e K, L, Technology + + + + − − ?
Taxes, T − + + − −
Co + − − + +
Govt, G − − + +
r* + − + − + + −
Io + − + +
NXo + +
Mg − Yg + + + −
M + −
 The Nominal Exchange Rate, Growth
Rate
• e = εP */P
• Recall from chapter 2
– Z = XY implies Zg = Xg + Yg
– Z = X/Y implies Zg = Xg − Yg
• e = εP */P implies eg = εg + π* − π
• Assumption: the real exchange rate is
constant in the long run: εg = 0
• Therefore, eg = π* − π
 The Nominal Exchange Rate , Growth
Rate
• eg = π * − π
• The value of the domestic currency grows at a
rate equal to the foreign inflation rate minus
the domestic inflation rate
– Example: if China’s annual inflation rate is 8
percent and the U.S. annual inflation rate is 2
percent, then the yuan per dollar exchange rate
will increase at the annual rate of 6 percent.
 Nominal Exchange Rates: predictions
• eg = π* − π Predictions Grid (Long Run, Open Economy)
• Assumption: Y C S r I NX ε π i P e eg
The foreign K, L, Technology + + + + − − ?
inflation Taxes, T − + + − −
rate (π*) is Co + − − + +
exogenous Govt, G − − + +
r* + − + − + + −
Io + − + +
NXo + +
Mg − Yg + + + − −
M + −
π* +
Long-Run Predictions—Open Economy

Predictions Grid (Long Run, Open Economy)

Y C S r I NX ε π i P e eg
K, L, Technology + + + + − − ?
Taxes, T − + + − −
Co + − − + +
Govt, G − − + +
r* + − + − + + −
Io + − + +
NXo + +
Mg − Yg + + + − −
M + −
π* +
 Inflation differentials and nominal exchange rates for a cross
section of countries
% change 30%
in nominal Mexico
25%
exchange
rate 20%
Iceland
15%
Pakistan
10%
Australia S. Africa
5% Canada S. Korea
Singapore
0% U.K.
Japan
-5%
-10% -5% 0% 5% 10% 15% 20% 25% 30%
inflation differential
 CASE STUDY:
The Reagan deficits revisited
actual closed small open
1970s 1980s
change economy economy
G–T 2.2 3.9   
S 19.6 17.4   
r 1.1 6.3   no change
I 19.9 19.4   no change
NX -0.3 -2.0  no change 
ε 115.1 129.4  no change 
Data: decade averages; all except r and ε are expressed as a percent of GDP;
ε is a trade-weighted index.
The U.S. as a large open economy
• So far, we’ve learned long-run models for
two extreme cases:
– closed economy (chap. 3)
– small open economy (chap. 5)
• A large open economy – like the U.S. – falls
between these two extremes.
• The results from large open economy analysis
are a mixture of the results for the
closed & small open economy cases.
• For example…
 A fiscal expansion in three models
A fiscal expansion causes national saving to fall.
The effects of this depend on openness & size:
closed large open small open
economy economy economy
rises, but not as much no
r rises
as in closed economy change
falls, but not as much no
I falls
as in closed economy change
no falls, but not as much as in
NX falls
change small open economy