Lecture Twenty
International Economics; the Mundell-Fleming
Model
(Economics 100b; Spring 1996)
Professor of Economics J. Bradford DeLong
601 Evans, University of California at Berkeley
Berkeley, CA 94720
(510) 643-4027 phone (510) 642-6615 fax
delong@econ.berkeley.edu
http://www.j-bradford-delong.net
March 13, 1996
Administration
Small Open Economy
The Mundell-Fleming Model
MF Under Flexible Exchange Rates
MF Under Fixed Exchange Rates
Administration
Makeup exam: Evans 639 between 2:30 and 3:30 p.m. on Friday.
Small Open Economy
Up until now we have dealt only with "closed economies": economies
that do not trade with the rest of the world. This may be a mistake,
because there are few "closed economies" left in the world:
international trade and international finance are at the heart of
most of economic policy, and much of macroeconomic events even in the
United States--the most closed economy in the world, by some
measures.
But even in the United States:
- 1995 exports: some $804 billion
- 1995 imports: some $916 billion
- 1995 net capital inflow: some $112 billion (which is also the
excess of domestic investment over saving)
All out of a 1995 GDP of about $7,246 billion
and with some $2.77 trillion of U.S.-owned assets abroad, and some
$3.35 trillion of foreign-owned assets here in the U.S.
What we do first is a "small open economy" version of the IS-LM
model. Key assumptions: the country being analyzed is "small" in the
sense that it doesn't affect world prices but is affected by them.
And we are going to start out in a "timeless" fashion: just look at
short-run equilibria; don't think about how such equilibria will
evolve over time; assume domestic (and international) inflation are
zero; assume no expected exchange rate changes
The Mundell-Fleming Model
Start with our IS curve:
(1) Y = C(Y-T) + I(r) + G
and add a "net exports" (NX) term:
(2) Y = C(Y-T) + I(r) + G + NX(e)
where e is the exchange rate, and the lower is e, the higher are net
exports--the greater are exports relative to imports.
Brief digression on Lerner stability conditions and the J-curve. How
do we know that NX(e) slopes the "right" way? Answer: we don't--in
the short run. But we are certain it does in the long run.
We also have:
(3) M/P = L(i, Y)
(4) i = r (from no expected inflation; no shifts in interest rates
expected to mess up the term structure)
and finally:
(5) r = r*, where r* is the real interest rate "out there" in the
world, set by forces removed from the domestic economy.
Note that if you remember that on the savings-expenditure side:
(6) Y = C + Sp + T
(1) is equivalent to the loanable-funds equilibrium
condition:
(7) Sp - I(r) - DEF = NX(e)
Now let's fix everything domestic--fix G, fix T, fix the parameters
of the consumption function, and substitute r* in for r in (1) and
(2) to obtain what Mankiw calls the IS* and LM* curves:
(8) Y = C(Y-T) + I(r*) + G +NX(e)
downward sloping in Y-e space, with e on the vertical axis,
and:
(9) M/P = L(r*, Y)
a vertical curve at equilibrium output... Why? What makes the LM
curve upward sloping is that as income rises (and transactions demand
for money rises along with it), a rise in the interest rate causes
households and businesses to economize on their stocks of liquid
assets.
Here the interest rate doesn't rise--pegged by international markets.
Hence output is pinned by the LM* curve--the exchange rate rises
instead.
That there is some truth in this can be seen in the reaction to the
Reagan tax cuts; 63% increase in dollar.
- Short methodological digressions: schizoid; sometimes U.S.
more like a closed economy (a large chunk of the world). Sometimes
it is more like an open economy (constrained by world equilibrium
conditons)
MF Under Flexible Exchange Rates
A fiscal expansion under flexible exchange rates shifts the IS* curve
outwards on the Y-e diagram..
A monetary expansion under flexible exchange rates causes a
fall in the exchange rate, and an increase in output.
- In a small open economy, a monetary expansion puts downward
pressure on the interest rate. But as soon as this downward
pressure makes itself felt, capital flows out--and the exchange
rate falls. Falling exchange rate raises the demand for goods, and
boosts exports and GDP.
A tariff to alter the trade balance?
- Doesn't work: NX(e) = Y - C(Y-T) - I(r) - G; a tariff doesn't
shift anything on the right-hand side, hence it can't shift NX.
- What it does do is shift the function NX(e). Reducing imports
at any given exchange rate through the tariff moves the IS* curve
to the right, but the exchange rate rises--and you are back at
your old NX, but at a higher exchange rate.
- May be worth doing--if no other country responded--because of
terms-of-trade effects. I tend to think that trying to exploit
terms-of-trade effects is asking for a lot of trouble: "the
President must carry McComb county"
MF Under Fixed Exchange Rates
Suppose that you adopt a fixed exchange rate system: e = e*; buy and
sell dollars for yen, marks, whatever to keep the exchange rate
perfectly fixed. What happens?
If equilibrium exchange rate is higher than fixed rate, then
dollars are a good deal: everyone tries to sell yen for dollars, wait
for the exchange rate to rise, and then collect a profit. But as long
as the Federal Reserve maintains the fixed parity, it can't refuse to
sell dollars for yen--hence the money stock rises. And as the money
stock rises the LM* curve shifts out. Where does it end? Where all
three curves cross.
So: maintaining a fixed exchange rate means you give up control over
your money supply.
- What happens if the Federal Reserve tries to carry out an
expansionary monetary policy? Either (a) it finds that capital
flows immediately undo its actions, or (b) a devaluation...
- What happens if the government tries to carry out an
expansionary fiscal policy? Either (a) accomodated by monetary
expansion, or (b) revaluation.
- What happens if the government places high tariffs on imports
IS* curve shifts out--and to maintain exchange rate LM* curve
shifts out as well. So a tariff can be a job creator, but only
under these particular conditions (and why not adjust your
exchange rate directly) (possible digression on Keynes and the
tariff)
So should exchange rates be floating or fixed. Fixed--because it
makes international trade easier (and keeps monetary authorities from
messing up). Floating--because then you can use monetary policy to
stabilize the economy. And floating--because no one believes that
fixed exchange rates will be maintained, and an erratic system of
"punctuated equilibrium" has all the disadvantages of both.
>
