Inflation and Exchange Rate Dynamics
In this section we tie together our short-run and long-run findings about the effects of
monetary changes by examining the process through which the price level adjusts to its
long-run position. An economy experiences inflation when its price level is rising and
deflation when its price level is falling. Our examination of inflation will give us a deeper
understanding of how the exchange rate adjusts to monetary disturbances in the economy.
Short-Run Price Rigidity versus Long-Run Price Flexibility
Our analysis of the short-run effects of monetary changes assumed that a country’s price level,
unlike its exchange rate, does not jump immediately. This assumption cannot be exactly correct,
because many commodities, such as agricultural products, are traded in markets where
prices adjust sharply every day as supply or demand conditions shift. In addition, exchange
rate changes themselves may affect the prices of some tradable goods and services that enter
into the commodity basket defining the price level.
Many prices in the economy, however, are written into long-term contracts and cannot
be changed immediately when changes in the money supply occur. The most important
prices of this type are workers’ wages, which are negotiated only periodically in many industries.
Wages do not enter indices of the price level directly, but they make up a large
fraction of the cost of producing goods and services. Since output prices depend heavily
on production costs, the behavior of the overall price level is influenced by the sluggishness
of wage movements. The short-run “stickiness” of price levels is illustrated by
Figure 15-11, which compares data on month-to-month percentage changes in the dollar/
yen exchange rate, , with data on month-to-month percentage changes in the ratio of
money price levels in the United States and Japan, . As you can see, the exchange
rate is much more variable than relative price levels, a fact consistent with the view that
price levels are relatively rigid in the short run. The pattern shown in the figure applies to
all of the main industrial countries in recent decades. In light of this and other evidence,
we will therefore continue to assume that the price level is given in the short run and does
not make significant jumps in response to policy changes.
This assumption would not be reasonable, however, for all countries at all times. In extremely
inflationary conditions, such as those seen in the 1980s in some Latin American
countries, long-term contracts specifying domestic money payments may go out of use.
Automatic price level indexation of wage payments may also be widespread under highly
inflationary conditions. Such developments make the price level much less rigid than it
would be under moderate inflation, and large price level jumps become possible. Some
price rigidity can remain, however, even in the face of inflation rates that would be high by
everyday industrial-country standards. For example, Turkey’s 30 percent inflation rate for
2002 seems high until it is compared with the 114 percent depreciation of the Turkish lira
against the U.S. dollar over the same year.
PUS/PJ
E$/ ¥
CHAPTER 15 Money, Interest Rates, and Exchange Rates 373
Our analysis assuming short-run price rigidity is nonetheless most applicable to countries
with histories of comparative price level stability, such as the United States. Even in
the cases of low-inflation countries, there is a lively academic debate over the possibility
that seemingly sticky wages and prices are in reality quite flexible.8
Although the price levels appear to display short-run stickiness in many countries, a
change in the money supply creates immediate demand and cost pressures that eventually
lead to future increases in the price level. These pressures come from three main sources:
1. Excess demand for output and labor. An increase in the money supply has an expansionary
effect on the economy, raising the total demand for goods and services. To
meet this demand, producers of goods and services must employ workers overtime and
make new hires. Even if wages are given in the short run, the additional demand for labor
allows workers to ask for higher wages in the next round of wage negotiations. Producers
8For a discussion of this debate, and empirical evidence that U.S. aggregate prices and wages show significant
rigidity, see the book by Hall and Papell listed in Further Readings. Other summaries of U.S. evidence are given
by Mark A. Wynne, “Sticky Prices: What Is the Evidence?” Federal Reserve Bank of Dallas Economic Review
(First Quarter 1995), pp. 1–12; and by Mark J. Bils and Peter J. Klenow, “Some Evidence of the Importance of
Sticky Prices,” Journal of Political Economy 112 (October 2004), pp. 947–985.
Price level ratio change
Exchange rate change
–15
–10
–5
0
5
10
15
20
Changes in exchange rates and price level
ratios–U.S./Japan (percent per month)
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008
Figure 15-11
Month-to-Month Variability of the Dollar/Yen Exchange Rate and of the U.S./Japan Price Level Ratio,
1980–2009
The much greater month-to-month variability of the exchange rate suggests that price levels are relatively
sticky in the short run.
Source: International Monetary Fund, International Financial Statistics.
374 PART THREE Exchange Rates and Open-Economy Macroeconomics
step, as the theory in the text would predict.
Moreover, the trends in the price level and exchange
rate are of the same order of magnitude: The price
level rose by 22,908 percent between April 1984
and July 1985 and the peso price of dollars rose by
24,662 percent over the same period. These percentage
changes actually are greater than the corresponding
percentage increase in the money supply
(which is “only” 17,433 percent), but the difference
is to be expected. Exploding inflation causes real
money demand to fall over time, and this additional
monetary change makes money prices rise even
more quickly than the money supply itself rises.
We chose July 1985 as the endpoint for the comparison
because the Bolivian government introduced
a dramatic stabilization plan near the end of
August 1985. You can see in the data how the
money supply and, more dramatically, the price
level and exchange rate all began to level out in the
two months after August.
Money Supply Growth and Hyperinflation in Bolivia
In 1984 and 1985 the small Latin American country
of Bolivia experienced hyperinflation—an explosive
and seemingly uncontrollable inflation in which
money loses value rapidly and may even go out of
use.* During hyperinflations the magnitudes of monetary
changes are so enormous that the “long-run”
effects of money on the price level can occur very
quickly. These episodes therefore provide laboratory
conditions well suited for testing long-run theories
about the effects of money supplies on prices.
On the next page we show data on Bolivia’s
money supply and price level during the hyperinflation.
An official exchange rate between the Bolivian
peso and the U.S. dollar was controlled by the
Bolivian government during this period, so we list
instead values for an exchange rate that better reflected
market forces, the price of dollars in terms of
pesos on the La Paz black market.
The data show a clear tendency for the money
supply, price level, and exchange rate to move in
*In a classic paper, Columbia University economist Phillip Cagan drew the line between inflation and hyperinflation at an
inflation rate of 50 percent per month (which, through the power of compounding, comes out to 12,875 percent per year).
See “The Monetary Dynamics of Hyperinflation,” in Milton Friedman, ed., Studies in the Quantity Theory of Money.
Chicago: University of Chicago Press, 1956, pp. 25–117.
are willing to pay these higher wages, for they know that in a booming economy, it will
not be hard to pass higher wage costs on to consumers through higher product prices.
2. Inflationary expectations. If everyone expects the price level to rise in the future,
their expectation will increase the pace of inflation today. Workers bargaining
over wage contracts will insist on higher money wages to counteract the effect on
their real wages of the anticipated general increase in prices. Producers, once again,
will give in to these wage demands if they expect product prices to rise and cover the
additional wage costs.
3. Raw materials prices. Many raw materials used in the production of final goods,
for example, petroleum products and metals, are sold in markets where prices adjust
sharply even in the short run. By causing the prices of such materials to jump upward, a
money supply increase raises production costs in materials-using industries. Eventually,
producers in those industries will raise product prices to cover their higher costs.
Permanent Money Supply Changes and the Exchange Rate
We now apply our analysis of inflation to study the adjustment of the dollar/euro exchange
rate following a permanent increase in the U.S. money supply. Figure 15-12 shows both
the short-run (Figure 15-12a) and the long-run (Figure 15-12b) effects of this disturbance.
We suppose that the economy starts with all variables at their long-run levels and that output
remains constant as the economy adjusts to the money supply change.
Figure 15-12a assumes the U.S. price level is initially given at An increase in the
nominal money supply from to therefore raises the real money supply from
to in the short run, lowering the interest rate from (point 1) to
(point 2). So far, our analysis proceeds exactly as it did earlier in this chapter.
The first change in our analysis comes when we ask how the American money supply
change (shown in the bottom part of panel (a)) affects the foreign exchange market (shown
in the top part of panel (a)). As before, the fall in the U.S. interest rate is shown as a leftward
shift in the vertical schedule giving the dollar return on dollar deposits. This is no
longer the whole story, however, for the money supply increase now affects exchange rate
expectations. Because the U.S. money supply change is permanent, people expect a longrun
increase in all dollar prices, including the exchange rate, which is the dollar price of euros.
As you will recall from Chapter 14, a rise in the expected future dollar/euro exchange
rate (a future dollar depreciation) raises the expected dollar return on euro deposits; it thus
shifts the downward-sloping schedule in the top part of Figure 15-12a to the right. The dollar
depreciates against the euro, moving from an exchange rate of (point ) to
(point ). Notice that the dollar depreciation is greater than it would be if the expected
future dollar/euro exchange rate stayed fixed (as it might if the money supply increase were
temporary rather than permanent). If the expectation did not change, the new short-run
equilibrium would be at point 3oe rather than at point 2oe.
E$/€
e
2oe
E$/€
E 1oe 2 $/€
1
R$
R 2 $
M 1 US
2 /PUS
M 1 US
1 /PUS
1
MUS
M 2 US
1
PUS
1 .
CHAPTER 15 Money, Interest Rates, and Exchange Rates 375
Macroeconomic Data for Bolivia, April 1984–October 1985
Month
Money Supply
(Billions of Pesos)
Price Level
(Relative to 1982 Average 1
Exchange Rate
(Pesos per Dollar)
1984
April 270 21.1 3,576
May 330 31.1 3,512
June 440 32.3 3,342
July 599 34.0 3,570
August 718 39.1 7,038
September 889 53.7 13,685
October 1,194 85.5 15,205
November 1,495 112.4 18,469
December 3,296 180.9 24,515
1985
January 4,630 305.3 73,016
February 6,455 863.3 141,101
March 9,089 1,078.6 128,137
April 12,885 1,205.7 167,428
May 21,309 1,635.7 272,375
June 27,778 2,919.1 481,756
July 47,341 4,854.6 885,476
August 74,306 8,081.0 1,182,300
September 103,272 12,647.6 1,087,440
October 132,550 12,411.8 1,120,210
Source: Juan-Antonio Morales, “Inflation Stabilization in Bolivia,” in Michael Bruno et al., eds., Inflation Stabilization: The
Experience of Israel, Argentina, Brazil, Bolivia, and Mexico. Cambridge: MIT Press, 1988, table 7A-1. Money supply is M1.
376 PART THREE Exchange Rates and Open-Economy Macroeconomics
Figure 15-12b shows how the interest rate and exchange rate behave as the price level
rises during the economy’s adjustment to its long-run equilibrium. The price level begins
to rise from the initially given level eventually reaching Because the
long-run increase in the price level must be proportional to the increase in the money
supply, the final real money supply, is shown equal to the initial real money
supply, . Since output is given and the real money supply has returned to its
original level, the equilibrium interest rate must again equal in the long run (point 4).
The interest rate therefore rises from (point 2) to (point 4) as the price level rises
from to .
The rising U.S. interest rate has exchange rate effects that can also be seen in
Figure 15-12b: The dollar appreciates against the euro in the process of adjustment. If
exchange rate expectations do not change further during the adjustment process, the
foreign exchange market moves to its long-run position along the downward-sloping
schedule defining the dollar return on euro deposits. The market’s path is just the path
traced out by the vertical dollar interest rate schedule as it moves rightward because of
PUS
P 2 US
1
R$
R 1 $
2
R$
1
MUS
1 /PUS
1
MUS
2 /PUS
2 ,
PUS
P 2 . US
1 ,
PUS
R$
2 R$
1 R$
2 R$
1
MUS
PUS
2
MUS
2
2
1
Dollar/euro exchange
rate, E$/€
Dollar return
Expected
euro return
Rates of
return (in
dollar terms)
2'
3'
1'
0
2
1
U.S. real
money holdings
U.S. real
money holdings
(a) Short-run effects (b) Adjustment to long-run equilibrium
0
2
4
Dollar return
Expected
euro return
2'
4'
E$/€
1
MUS
PUS
L(R$,YUS)
E$/€
2
1
MUS
2
1
P1 US
L(R$,YUS)
E$/€
2
E$/€
3
E$/€
U.S. real
money supply
Figure 15-12
Short-Run and Long-Run Effects of an Increase in the U.S. Money Supply (Given Real Output, Y)
(a) Short-run adjustment of the asset markets. (b) How the interest rate, price level, and exchange rate
move over time as the economy approaches its long-run equilibrium.
CHAPTER 15 Money, Interest Rates, and Exchange Rates 377
the price level’s gradual rise. In the long run (point ), the equilibrium exchange rate,
is higher than at the original equilibrium, point . Like the price level, the dollar/
euro exchange rate has risen in proportion to the increase in the money supply.
Figure 15-13 shows time paths like the ones just described for the U.S. money supply,
the dollar interest rate, the U.S. price level, and the dollar/euro exchange rate. The figure is
drawn so that the long-run increases in the price level (Figure 15-13c) and exchange rate
(Figure 15-13d) are proportional to the increase in the money supply (Figure 15-13a).
Exchange Rate Overshooting
In its initial depreciation after a money supply rise, the exchange rate jumps from up
to , a depreciation greater than its long-run depreciation from to (see
Figure 15-13d). The exchange rate is said to overshoot when its immediate response to
a disturbance is greater than its long-run response. Exchange rate overshooting is an
important phenomenon because it helps explain why exchange rates move so sharply
from day to day.
The economic explanation of overshooting comes from the interest parity condition.
The explanation is easiest to grasp if we assume that before the money supply increase
first occurs, no change in the dollar/euro exchange rate is expected, so that equals
the given interest rate on euro deposits. A permanent increase in the U.S. money supply
doesn’t affect so it causes to fall below R and remain below that interest rate € R$
R 1 €,
R€R$ ,
1
E$/€
E 3 $/€
E 1 $/€
2
E$/€
1
E$/€ 1oe
3 ,
4oe
MUS
1
MUS
2
PUS
2
E$/€
3
E$/€
1
R$
2
U.S. money
supply, MUS
Dollar
interest rate, R$
U.S. price
level, PUS
Dollar/euro
exchange rate, E$/€
t Time 0
t0 Time t0 Time
t Time 0
R$
1
E$/€
2
(d)
PUS
1
(b)
(c)
(a)
Figure 15-13
Time Paths of U.S. Economic Variables After a Permanent Increase in the U.S. Money Supply
After the money supply increases at in panel (a), the interest rate (in panel (b)), price level (in panel
(c)), and exchange rate (in panel (d)) move as shown toward their long-run levels. As indicated in
panel (d) by the initial jump from to , the exchange rate overshoots in the short run before
settling down to its long-run level, E$/€.
3
E$/€
E 2 $/€
1
t0
378 PART THREE Exchange Rates and Open-Economy Macroeconomics
(Figure 15-13b) until the U.S. price level has completed the long-run adjustment to
shown in Figure 15-13c. For the foreign exchange market to be in equilibrium during this
adjustment process, however, the interest difference in favor of euro deposits must be offset
by an expected appreciation of the dollar against the euro, that is, by an expected fall in
Only if the dollar/euro exchange rate overshoots initially will market participants
expect a subsequent appreciation of the dollar against the euro.
Overshooting is a direct consequence of the short-run rigidity of the price level. In a
hypothetical world where the price level could adjust immediately to its new, long-run level
after a money supply increase, the dollar interest rate would not fall because prices would
adjust immediately and prevent the real money supply from rising. Thus, there would be no
need for overshooting to maintain equilibrium in the foreign exchange market. The exchange
rate would maintain equilibrium simply by jumping to its new, long-run level right away.
Case Study
Can Higher Inflation Lead to Currency Appreciation?
The Implications of Inflation Targeting
In the overshooting model that we have just examined, an increase in the money supply
leads to higher inflation and currency depreciation, as shown in Figure 15–13. It may
seem puzzling, then, that readers of the financial press often see headlines such as the
following one from the Financial Times of May 24, 2007: “Inflation Drives Canadian
Dollar Higher.” In light of the seemingly reasonable model set out in this chapter, can
such statements possibly make sense?
A clue comes from reading further in the Financial Times news story on Canadian
inflation. According to the FT:
[A]nalysts said that the main driver of the recent bout of Canadian dollar appreciation
was higher-than-expected April inflation data, which saw the bond market fully
price in a 25 basis point rise in Canadian interest rates by the end of the year.
If central banks act to raise interest rates when inflation rises, then because higher interest
rates cause currency appreciation, it might be possible to resolve the apparent contradiction
to our model. To do so fully, however, we must consider two aspects of the way
in which modern central banks actually formulate and implement monetary policy.
1. The interest rate, not the money supply, is the prime instrument of monetary
policy. Nowadays, most central banks do not actually target the money supply in order
to control inflation. They instead target a benchmark short-term rate of interest (such
as the overnight “federal funds” rate in the United States). How does our discussion of
money market equilibrium help us to understand this process? Consider Figure 15-3,
and assume that the central bank wishes to set an interest rate of . It can do so
simply by agreeing to provide or take up all of the cash that the market wishes to trade
at that rate of interest. If the money supply is initially , for example, there will be an
excess demand for money at the interest rate , so people will sell bonds to the central
bank for money (in effect, borrowing) until the money supply has expanded to
and the excess demand is gone. Central banks tend to set an interest rate, rather than
the money supply, because the money demand schedule shifts around frequently
in practice. If the central bank were to fix the money supply, the result would
L(R, Y)
Q1
R1
Q2
R1
E$/€
E 3 $/€.
PUS
2
CHAPTER 15 Money, Interest Rates, and Exchange Rates 379
be high and possibly damaging interest rate volatility; it is thus more practical to fix
the interest rate and let the money supply adjust automatically when necessary.9
Our discussion above of the positive relationship between the money supply and
price level will tip you off, however, to one potential problem of an interest rate instrument.
If the money supply is free to grow or shrink as markets collectively desire,
how can the price level and inflation be kept under control? For example, if
market actors doubt the central bank’s resolve to control inflation, and suddenly
push the price level up because they expect higher prices in the future, they could
simply borrow more money from the central bank, thereby bringing about the
money supply increase needed to maintain higher prices in the long run. This worrisome
possibility brings us to the second pillar of modern monetary policy.
2. Most central banks adjust their policy interest rates expressly so as to keep inflation
in check. A central bank can keep inflation from getting too high or too low
by raising the interest rate when it learns that inflation is running higher than expected,
and lowering it when inflation is running lower. As we will see more fully in
Chapter 17, a rise in the interest rate, which causes the currency to appreciate, puts a
damper on demand for a country’s products by making them more expensive compared
to foreign goods. This fall in demand, in turn, promotes lower domestic prices.
A fall in the interest rate, symmetrically, supports domestic prices. Indeed, many
central banks now follow formal strategies of inflation targeting, under which they
announce a target (or target range) for the inflation rate and adjust the interest rate to
keep inflation on target. Central banks generally target so-called core inflation,
which is inflation in the price level excluding volatile components such as energy
prices, rather than headline inflation, which is inflation in the total consumer price
index. The formal practice of inflation targeting was initiated by New Zealand’s
central bank in 1989, and the central banks of many other developed and developing
areas, including Canada, Chile, Mexico, South Africa, Sweden, Thailand, the United
Kingdom, and the euro zone, have followed suit. The central banks of the United
States and Japan, while strongly averse to inflation, have so far been reluctant to
announce definite inflation targets.10
We can now understand the “paradox” of higher-than-expected inflation causing
currency appreciation rather than depreciation. Suppose market participants unexpectedly
push up prices and borrow to enlarge the money supply. Thus, when the
Canadian government releases new price data, the data show a price level higher than
9For a nontechnical account of modern central bank policy implementation, see Michael Woodford, “Monetary
Policy in a World Without Money,” International Finance 3 (July 2000), pp. 229–260. Woodford’s provocative
title points to another advantage of the interest rate instrument for central banks: It is possible to conduct monetary
policy even if checking deposits pay interest at competitive rates. For many purposes, however, it is reasonable
to ignore the variability of the L(R, Y) schedule and simply assume that the central bank directly sets the
money supply. In the rest of the book we shall, for the most part, make that simplifying assumption. The major
exception will be when we introduce fixed exchange rates in Chapter 18.
10The International Monetary Fund (IMF), which we will discuss in Chapter 19, maintains a useful classification
of its member countries with regard to their monetary policy frameworks as well as their exchange rate systems;
see the documentation and data at http://www.imf.org/external/np/mfd/er/2008/eng/0408.htm. The IMF does not
consider Switzerland to be an inflation targeter, but the difference between its actual procedures and inflation targeting
is small, and this Case Study therefore includes it among the inflation-targeting countries. On inflationtargeting
practices and the theory behind them, see the books by Bernanke et al. and by Truman in Further
Readings. For a critique of the idea of targeting core rather than headline inflation, see Stephen Cecchetti, “Core
Inflation Is an Unreliable Guide,” Financial Times, September 12, 2006.
380 PART THREE Exchange Rates and Open-Economy Macroeconomics
what market participants had previously predicted. If the Bank of Canada is expected
to raise interest rates quickly so as to push the price level and money supply back on
course, there is no reason for the future expected exchange rate to change. But with
higher Canadian interest rates, interest parity requires an expected future
depreciation of the Canadian dollar, which is consistent with an unchanged future
exchange rate only if the Canadian dollar appreciates immediately. The picture of the
economy’s adjustment after the unexpected increase in money and prices would look
like Figure 15–13 in reverse (that is, constructed to reflect a monetary contraction
rather than an expansion)—with the added assumption that the Bank of Canada gradually
moves interest rates back to their initial level as the price level returns to its
targeted path.11
Economists Richard Clarida of Columbia University and Daniel Waldman of
Barclays Capital offer striking statistical evidence consistent with this explanation.12
These writers measure unexpected inflation as the inflation rate estimate initially announced
by a government, prior to any data revisions, less the median of inflation forecasts
for that period previously published by a set of banking industry analysts. For a
sample of ten countries—Australia, Britain, Canada, the euro area, Japan, New Zealand,
Norway, Sweden, Switzerland, and the United States—Clarida and Waldman examine
the exchange rate changes that occur in the period lasting from five minutes prior to an
inflation announcement to five minutes afterward. Their key findings are these:
1. On average for the ten currencies that they study, news that inflation is unexpectedly
high does indeed lead a currency to appreciate, not depreciate.
2. The effect is stronger for core than for headline inflation.
3. The effect is much stronger for the inflation-targeting countries than for the United
States and Japan, the two countries that do not announce inflation targets. In the
case of Canada, for example, the announcement of an annual core inflation rate that
is 1 percent per year above the market expectation leads the Canadian dollar to
appreciate immediately by about 3 percent against the U.S. dollar. The corresponding
effect for the U.S. dollar/euro exchange rate, while in the same direction, is
only about one-quarter as big.
4. For countries where sufficiently long data series are available, the strengthening
effect of unexpected inflation on the currency is present after the introduction of
inflation targeting, but not before.
Scientific theories can be conclusively disproved, of course, but never conclusively
proved. So far, however, the theory that strict inflation targeting makes bad news on
inflation good news for the currency looks quite persuasive.
11Strictly speaking, the narrative in the text describes a setting with price level rather than inflation rate targeting.
(Can you see the difference?) The reasoning in the case of inflation targeting is nearly identical, however,
provided that the central bank’s interest rate response to unexpectedly high inflation is sufficiently strong.
12See Clarida and Waldman, “Is Bad News About Inflation Good News for the Exchange Rate? And If So, Can
That Tell Us Anything about the Conduct of Monetary Policy?” in John Y. Campbell, ed., Asset Prices and
Monetary Policy (Chicago: University of Chicago Press, 2008). Michael W. Klein of Tufts University and Linda
S. Goldberg of the Federal Reserve Bank of New York used a related approach to investigate changing market
perceptions of the European Central Bank’s inflation aversion after its founding in 1999; see “Establishing
Credibility: Evolving Perceptions of the European Central Bank,” Institute for International Integration Studies
Discussion Paper 194, Trinity College, Dublin, December 2006.
CHAPTER 15 Money, Interest Rates, and Exchange Rates 381
aggregate money demand, p. 358
deflation, p. 372
exchange rate overshooting,
p. 377
inflation, p. 372
long run, p. 363
long-run equilibrium,
p. 368
money supply, p. 356
price level, p. 358
short run, p. 363
SUMMARY
1. Money is held because of its liquidity. When considered in real terms, aggregate
money demand is not a demand for a certain number of currency units but is instead a
demand for a certain amount of purchasing power. Aggregate real money demand depends
negatively on the opportunity cost of holding money (measured by the domestic
interest rate) and positively on the volume of transactions in the economy (measured
by real GNP).
2. The money market is in equilibrium when the real money supply equals aggregate real
money demand. With the price level and real output given, a rise in the money supply
lowers the interest rate and a fall in the money supply raises the interest rate. A rise in
real output raises the interest rate, given the price level, while a fall in real output has
the opposite effect.
3. By lowering the domestic interest rate, an increase in the money supply causes the domestic
currency to depreciate in the foreign exchange market (even when expectations
of future exchange rates do not change). Similarly, a fall in the domestic money supply
causes the domestic currency to appreciate against foreign currencies.
4. The assumption that the price level is given in the short run is a good approximation to
reality in countries with moderate inflation, but it is a misleading assumption over the
long run. Permanent changes in the money supply push the long-run equilibrium price
level proportionally in the same direction but do not influence the long-run values of
output, the interest rate, or any relative prices. One important money price whose longrun
equilibrium level rises in proportion to a permanent money supply increase is the
exchange rate, the domestic currency price of foreign currency.
5. An increase in the money supply can cause the exchange rate to overshoot its long-run
level in the short run. If output is given, a permanent money supply increase, for example,
causes a more-than-proportional short-run depreciation of the currency, followed
by an appreciation of the currency to its long-run exchange rate. Exchange rate overshooting,
which heightens the volatility of exchange rates, is a direct result of sluggish
short-run price level adjustment and the interest parity condition.
KEY TERMS
PROBLEMS
1. Suppose there is a reduction in aggregate real money demand, that is, a negative shift
in the aggregate real money demand function. Trace the short-run and long-run effects
on the exchange rate, interest rate, and price level.
2. How would you expect a fall in a country’s population to alter its aggregate money
demand function? Would it matter if the fall in population were due to a fall in the
number of households or to a fall in the size of the average household?
3. The velocity of money, V, is defined as the ratio of real GNP to real money holdings,
in this chapter’s notation. Use equation (15-4) to derive an expression
for velocity and explain how velocity varies with changes in R and in Y. (Hint: The
V = Y/(M/P)
382 PART THREE Exchange Rates and Open-Economy Macroeconomics
effect of output changes on V depends on the elasticity of aggregate money demand
with respect to real output, which economists believe to be less than unity.) What is
the relationship between velocity and the exchange rate?
4. What is the short-run effect on the exchange rate of an increase in domestic real GNP,
given expectations about future exchange rates?
5. Does our discussion of money’s usefulness as a medium of exchange and unit of account
suggest reasons why some currencies become vehicle currencies for foreign exchange
transactions? (The concept of a vehicle currency was discussed in Chapter 14.)
6. If a currency reform has no effects on the economy’s real variables, why do governments
typically institute currency reforms in connection with broader programs aimed
at halting runaway inflation? (There are many instances in addition to the Turkish
case mentioned in the text. Other examples include Israel’s switch from the pound to
the shekel, Argentina’s switches from the peso to the austral and back to the peso, and
Brazil’s switches from the cruzeiro to the cruzado, from the cruzado to the cruzeiro,
from the cruzeiro to the cruzeiro real, and from the cruzeiro real to the real, the current
currency, which was introduced in 1994.)
7. Imagine that the central bank of an economy with unemployment doubles its money
supply. In the long run, full employment is restored and output returns to its fullemployment
level. On the (admittedly unlikely) assumption that the interest rate before
the money supply increase equals the long-run interest rate, is the long-run increase in
the price level more than proportional or less than proportional to the money supply
change? What if (as is more likely) the interest rate is initially below its long-run level?
8. Between 1984 and 1985, the money supply in the United States increased to
billion from billion, while that of Brazil increased to 106.1 billion cruzados
from 24.4 billion. Over the same period, the U.S. consumer price index rose to 100
from a level of 96.6, while the corresponding index for Brazil rose to 100 from a level
of only 31. Calculate the 1984–1985 rates of money supply growth and inflation for
the United States and Brazil, respectively. Assuming that other factors affecting the
money markets did not change too dramatically, how do these numbers match up with
the predictions of this chapter’s model? How would you explain the apparently different
responses of U.S. compared with Brazilian prices?
9. Continuing with the preceding question, note that the monetary value of output in
1985 was billion in the United States and 1,418 billion cruzados in Brazil.
Refer back to question 3 and calculate velocity for the two countries in 1985. Why do
you think velocity was so much higher in Brazil?
10. In our discussion of short-run exchange rate overshooting, we assumed that real output
was given. Assume instead that an increase in the money supply raises real output in the
short run (an assumption that will be justified in Chapter 17). How does this affect the
extent to which the exchange rate overshoots when the money supply first increases? Is
it likely that the exchange rate undershoots? (Hint: In Figure 15-12a, allow the aggregate
real money demand schedule to shift in response to the increase in output.)
11. Figure 14-2 shows that Japan’s short-term interest rates have had periods during
which they are near or equal to zero. Is the fact that the yen interest rates shown never
drop below zero a coincidence, or can you think of some reason why interest rates
might be bounded below by zero?
12. How might a zero interest rate complicate the task of monetary policy? (Hint: At a
zero rate of interest, there is no advantage in switching from money to bonds.)
13. As we observed in this chapter, central banks, rather than purposefully setting the
level of the money supply, usually set a target level for a short-term interest rate
by standing ready to lend or borrow whatever money people wish to hold at that
interest rate. (When people need more money for a reason other than a change in
$4,010
$570.3
$641.0
CHAPTER 15 Money, Interest Rates, and Exchange Rates 383
the interest rate, the money supply therefore expands, and it contracts when they
wish to hold less.)
a. Describe the problems that might arise if a central bank sets monetary policy by
holding the market interest rate constant. (First, consider the flexible-price case,
and ask yourself if you can find a unique equilibrium price level when the central
bank simply gives people all the money they wish to hold at the pegged interest
rate. Then consider the sticky-price case.)
b. Does the situation change if the central bank raises the interest rate when prices are
high, according to a formula such as where a is a positive
constant and a target price level?
c. Suppose the central bank’s policy rule is where u is a
random movement in the policy interest rate. In the overshooting model shown in
Figure 15-12, describe how the economy would adjust to a permanent one-time
unexpected fall in the random factor u, and say why. You can interpret the fall in u
as an interest rate cut by the central bank, and therefore as an expansionary monetary
action. Compare your story with the one depicted in Figure 15-13.
In this section we tie together our short-run and long-run findings about the effects of
monetary changes by examining the process through which the price level adjusts to its
long-run position. An economy experiences inflation when its price level is rising and
deflation when its price level is falling. Our examination of inflation will give us a deeper
understanding of how the exchange rate adjusts to monetary disturbances in the economy.
Short-Run Price Rigidity versus Long-Run Price Flexibility
Our analysis of the short-run effects of monetary changes assumed that a country’s price level,
unlike its exchange rate, does not jump immediately. This assumption cannot be exactly correct,
because many commodities, such as agricultural products, are traded in markets where
prices adjust sharply every day as supply or demand conditions shift. In addition, exchange
rate changes themselves may affect the prices of some tradable goods and services that enter
into the commodity basket defining the price level.
Many prices in the economy, however, are written into long-term contracts and cannot
be changed immediately when changes in the money supply occur. The most important
prices of this type are workers’ wages, which are negotiated only periodically in many industries.
Wages do not enter indices of the price level directly, but they make up a large
fraction of the cost of producing goods and services. Since output prices depend heavily
on production costs, the behavior of the overall price level is influenced by the sluggishness
of wage movements. The short-run “stickiness” of price levels is illustrated by
Figure 15-11, which compares data on month-to-month percentage changes in the dollar/
yen exchange rate, , with data on month-to-month percentage changes in the ratio of
money price levels in the United States and Japan, . As you can see, the exchange
rate is much more variable than relative price levels, a fact consistent with the view that
price levels are relatively rigid in the short run. The pattern shown in the figure applies to
all of the main industrial countries in recent decades. In light of this and other evidence,
we will therefore continue to assume that the price level is given in the short run and does
not make significant jumps in response to policy changes.
This assumption would not be reasonable, however, for all countries at all times. In extremely
inflationary conditions, such as those seen in the 1980s in some Latin American
countries, long-term contracts specifying domestic money payments may go out of use.
Automatic price level indexation of wage payments may also be widespread under highly
inflationary conditions. Such developments make the price level much less rigid than it
would be under moderate inflation, and large price level jumps become possible. Some
price rigidity can remain, however, even in the face of inflation rates that would be high by
everyday industrial-country standards. For example, Turkey’s 30 percent inflation rate for
2002 seems high until it is compared with the 114 percent depreciation of the Turkish lira
against the U.S. dollar over the same year.
PUS/PJ
E$/ ¥
CHAPTER 15 Money, Interest Rates, and Exchange Rates 373
Our analysis assuming short-run price rigidity is nonetheless most applicable to countries
with histories of comparative price level stability, such as the United States. Even in
the cases of low-inflation countries, there is a lively academic debate over the possibility
that seemingly sticky wages and prices are in reality quite flexible.8
Although the price levels appear to display short-run stickiness in many countries, a
change in the money supply creates immediate demand and cost pressures that eventually
lead to future increases in the price level. These pressures come from three main sources:
1. Excess demand for output and labor. An increase in the money supply has an expansionary
effect on the economy, raising the total demand for goods and services. To
meet this demand, producers of goods and services must employ workers overtime and
make new hires. Even if wages are given in the short run, the additional demand for labor
allows workers to ask for higher wages in the next round of wage negotiations. Producers
8For a discussion of this debate, and empirical evidence that U.S. aggregate prices and wages show significant
rigidity, see the book by Hall and Papell listed in Further Readings. Other summaries of U.S. evidence are given
by Mark A. Wynne, “Sticky Prices: What Is the Evidence?” Federal Reserve Bank of Dallas Economic Review
(First Quarter 1995), pp. 1–12; and by Mark J. Bils and Peter J. Klenow, “Some Evidence of the Importance of
Sticky Prices,” Journal of Political Economy 112 (October 2004), pp. 947–985.
Price level ratio change
Exchange rate change
–15
–10
–5
0
5
10
15
20
Changes in exchange rates and price level
ratios–U.S./Japan (percent per month)
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008
Figure 15-11
Month-to-Month Variability of the Dollar/Yen Exchange Rate and of the U.S./Japan Price Level Ratio,
1980–2009
The much greater month-to-month variability of the exchange rate suggests that price levels are relatively
sticky in the short run.
Source: International Monetary Fund, International Financial Statistics.
374 PART THREE Exchange Rates and Open-Economy Macroeconomics
step, as the theory in the text would predict.
Moreover, the trends in the price level and exchange
rate are of the same order of magnitude: The price
level rose by 22,908 percent between April 1984
and July 1985 and the peso price of dollars rose by
24,662 percent over the same period. These percentage
changes actually are greater than the corresponding
percentage increase in the money supply
(which is “only” 17,433 percent), but the difference
is to be expected. Exploding inflation causes real
money demand to fall over time, and this additional
monetary change makes money prices rise even
more quickly than the money supply itself rises.
We chose July 1985 as the endpoint for the comparison
because the Bolivian government introduced
a dramatic stabilization plan near the end of
August 1985. You can see in the data how the
money supply and, more dramatically, the price
level and exchange rate all began to level out in the
two months after August.
Money Supply Growth and Hyperinflation in Bolivia
In 1984 and 1985 the small Latin American country
of Bolivia experienced hyperinflation—an explosive
and seemingly uncontrollable inflation in which
money loses value rapidly and may even go out of
use.* During hyperinflations the magnitudes of monetary
changes are so enormous that the “long-run”
effects of money on the price level can occur very
quickly. These episodes therefore provide laboratory
conditions well suited for testing long-run theories
about the effects of money supplies on prices.
On the next page we show data on Bolivia’s
money supply and price level during the hyperinflation.
An official exchange rate between the Bolivian
peso and the U.S. dollar was controlled by the
Bolivian government during this period, so we list
instead values for an exchange rate that better reflected
market forces, the price of dollars in terms of
pesos on the La Paz black market.
The data show a clear tendency for the money
supply, price level, and exchange rate to move in
*In a classic paper, Columbia University economist Phillip Cagan drew the line between inflation and hyperinflation at an
inflation rate of 50 percent per month (which, through the power of compounding, comes out to 12,875 percent per year).
See “The Monetary Dynamics of Hyperinflation,” in Milton Friedman, ed., Studies in the Quantity Theory of Money.
Chicago: University of Chicago Press, 1956, pp. 25–117.
are willing to pay these higher wages, for they know that in a booming economy, it will
not be hard to pass higher wage costs on to consumers through higher product prices.
2. Inflationary expectations. If everyone expects the price level to rise in the future,
their expectation will increase the pace of inflation today. Workers bargaining
over wage contracts will insist on higher money wages to counteract the effect on
their real wages of the anticipated general increase in prices. Producers, once again,
will give in to these wage demands if they expect product prices to rise and cover the
additional wage costs.
3. Raw materials prices. Many raw materials used in the production of final goods,
for example, petroleum products and metals, are sold in markets where prices adjust
sharply even in the short run. By causing the prices of such materials to jump upward, a
money supply increase raises production costs in materials-using industries. Eventually,
producers in those industries will raise product prices to cover their higher costs.
Permanent Money Supply Changes and the Exchange Rate
We now apply our analysis of inflation to study the adjustment of the dollar/euro exchange
rate following a permanent increase in the U.S. money supply. Figure 15-12 shows both
the short-run (Figure 15-12a) and the long-run (Figure 15-12b) effects of this disturbance.
We suppose that the economy starts with all variables at their long-run levels and that output
remains constant as the economy adjusts to the money supply change.
Figure 15-12a assumes the U.S. price level is initially given at An increase in the
nominal money supply from to therefore raises the real money supply from
to in the short run, lowering the interest rate from (point 1) to
(point 2). So far, our analysis proceeds exactly as it did earlier in this chapter.
The first change in our analysis comes when we ask how the American money supply
change (shown in the bottom part of panel (a)) affects the foreign exchange market (shown
in the top part of panel (a)). As before, the fall in the U.S. interest rate is shown as a leftward
shift in the vertical schedule giving the dollar return on dollar deposits. This is no
longer the whole story, however, for the money supply increase now affects exchange rate
expectations. Because the U.S. money supply change is permanent, people expect a longrun
increase in all dollar prices, including the exchange rate, which is the dollar price of euros.
As you will recall from Chapter 14, a rise in the expected future dollar/euro exchange
rate (a future dollar depreciation) raises the expected dollar return on euro deposits; it thus
shifts the downward-sloping schedule in the top part of Figure 15-12a to the right. The dollar
depreciates against the euro, moving from an exchange rate of (point ) to
(point ). Notice that the dollar depreciation is greater than it would be if the expected
future dollar/euro exchange rate stayed fixed (as it might if the money supply increase were
temporary rather than permanent). If the expectation did not change, the new short-run
equilibrium would be at point 3oe rather than at point 2oe.
E$/€
e
2oe
E$/€
E 1oe 2 $/€
1
R$
R 2 $
M 1 US
2 /PUS
M 1 US
1 /PUS
1
MUS
M 2 US
1
PUS
1 .
CHAPTER 15 Money, Interest Rates, and Exchange Rates 375
Macroeconomic Data for Bolivia, April 1984–October 1985
Month
Money Supply
(Billions of Pesos)
Price Level
(Relative to 1982 Average 1
Exchange Rate
(Pesos per Dollar)
1984
April 270 21.1 3,576
May 330 31.1 3,512
June 440 32.3 3,342
July 599 34.0 3,570
August 718 39.1 7,038
September 889 53.7 13,685
October 1,194 85.5 15,205
November 1,495 112.4 18,469
December 3,296 180.9 24,515
1985
January 4,630 305.3 73,016
February 6,455 863.3 141,101
March 9,089 1,078.6 128,137
April 12,885 1,205.7 167,428
May 21,309 1,635.7 272,375
June 27,778 2,919.1 481,756
July 47,341 4,854.6 885,476
August 74,306 8,081.0 1,182,300
September 103,272 12,647.6 1,087,440
October 132,550 12,411.8 1,120,210
Source: Juan-Antonio Morales, “Inflation Stabilization in Bolivia,” in Michael Bruno et al., eds., Inflation Stabilization: The
Experience of Israel, Argentina, Brazil, Bolivia, and Mexico. Cambridge: MIT Press, 1988, table 7A-1. Money supply is M1.
376 PART THREE Exchange Rates and Open-Economy Macroeconomics
Figure 15-12b shows how the interest rate and exchange rate behave as the price level
rises during the economy’s adjustment to its long-run equilibrium. The price level begins
to rise from the initially given level eventually reaching Because the
long-run increase in the price level must be proportional to the increase in the money
supply, the final real money supply, is shown equal to the initial real money
supply, . Since output is given and the real money supply has returned to its
original level, the equilibrium interest rate must again equal in the long run (point 4).
The interest rate therefore rises from (point 2) to (point 4) as the price level rises
from to .
The rising U.S. interest rate has exchange rate effects that can also be seen in
Figure 15-12b: The dollar appreciates against the euro in the process of adjustment. If
exchange rate expectations do not change further during the adjustment process, the
foreign exchange market moves to its long-run position along the downward-sloping
schedule defining the dollar return on euro deposits. The market’s path is just the path
traced out by the vertical dollar interest rate schedule as it moves rightward because of
PUS
P 2 US
1
R$
R 1 $
2
R$
1
MUS
1 /PUS
1
MUS
2 /PUS
2 ,
PUS
P 2 . US
1 ,
PUS
R$
2 R$
1 R$
2 R$
1
MUS
PUS
2
MUS
2
2
1
Dollar/euro exchange
rate, E$/€
Dollar return
Expected
euro return
Rates of
return (in
dollar terms)
2'
3'
1'
0
2
1
U.S. real
money holdings
U.S. real
money holdings
(a) Short-run effects (b) Adjustment to long-run equilibrium
0
2
4
Dollar return
Expected
euro return
2'
4'
E$/€
1
MUS
PUS
L(R$,YUS)
E$/€
2
1
MUS
2
1
P1 US
L(R$,YUS)
E$/€
2
E$/€
3
E$/€
U.S. real
money supply
Figure 15-12
Short-Run and Long-Run Effects of an Increase in the U.S. Money Supply (Given Real Output, Y)
(a) Short-run adjustment of the asset markets. (b) How the interest rate, price level, and exchange rate
move over time as the economy approaches its long-run equilibrium.
CHAPTER 15 Money, Interest Rates, and Exchange Rates 377
the price level’s gradual rise. In the long run (point ), the equilibrium exchange rate,
is higher than at the original equilibrium, point . Like the price level, the dollar/
euro exchange rate has risen in proportion to the increase in the money supply.
Figure 15-13 shows time paths like the ones just described for the U.S. money supply,
the dollar interest rate, the U.S. price level, and the dollar/euro exchange rate. The figure is
drawn so that the long-run increases in the price level (Figure 15-13c) and exchange rate
(Figure 15-13d) are proportional to the increase in the money supply (Figure 15-13a).
Exchange Rate Overshooting
In its initial depreciation after a money supply rise, the exchange rate jumps from up
to , a depreciation greater than its long-run depreciation from to (see
Figure 15-13d). The exchange rate is said to overshoot when its immediate response to
a disturbance is greater than its long-run response. Exchange rate overshooting is an
important phenomenon because it helps explain why exchange rates move so sharply
from day to day.
The economic explanation of overshooting comes from the interest parity condition.
The explanation is easiest to grasp if we assume that before the money supply increase
first occurs, no change in the dollar/euro exchange rate is expected, so that equals
the given interest rate on euro deposits. A permanent increase in the U.S. money supply
doesn’t affect so it causes to fall below R and remain below that interest rate € R$
R 1 €,
R€R$ ,
1
E$/€
E 3 $/€
E 1 $/€
2
E$/€
1
E$/€ 1oe
3 ,
4oe
MUS
1
MUS
2
PUS
2
E$/€
3
E$/€
1
R$
2
U.S. money
supply, MUS
Dollar
interest rate, R$
U.S. price
level, PUS
Dollar/euro
exchange rate, E$/€
t Time 0
t0 Time t0 Time
t Time 0
R$
1
E$/€
2
(d)
PUS
1
(b)
(c)
(a)
Figure 15-13
Time Paths of U.S. Economic Variables After a Permanent Increase in the U.S. Money Supply
After the money supply increases at in panel (a), the interest rate (in panel (b)), price level (in panel
(c)), and exchange rate (in panel (d)) move as shown toward their long-run levels. As indicated in
panel (d) by the initial jump from to , the exchange rate overshoots in the short run before
settling down to its long-run level, E$/€.
3
E$/€
E 2 $/€
1
t0
378 PART THREE Exchange Rates and Open-Economy Macroeconomics
(Figure 15-13b) until the U.S. price level has completed the long-run adjustment to
shown in Figure 15-13c. For the foreign exchange market to be in equilibrium during this
adjustment process, however, the interest difference in favor of euro deposits must be offset
by an expected appreciation of the dollar against the euro, that is, by an expected fall in
Only if the dollar/euro exchange rate overshoots initially will market participants
expect a subsequent appreciation of the dollar against the euro.
Overshooting is a direct consequence of the short-run rigidity of the price level. In a
hypothetical world where the price level could adjust immediately to its new, long-run level
after a money supply increase, the dollar interest rate would not fall because prices would
adjust immediately and prevent the real money supply from rising. Thus, there would be no
need for overshooting to maintain equilibrium in the foreign exchange market. The exchange
rate would maintain equilibrium simply by jumping to its new, long-run level right away.
Case Study
Can Higher Inflation Lead to Currency Appreciation?
The Implications of Inflation Targeting
In the overshooting model that we have just examined, an increase in the money supply
leads to higher inflation and currency depreciation, as shown in Figure 15–13. It may
seem puzzling, then, that readers of the financial press often see headlines such as the
following one from the Financial Times of May 24, 2007: “Inflation Drives Canadian
Dollar Higher.” In light of the seemingly reasonable model set out in this chapter, can
such statements possibly make sense?
A clue comes from reading further in the Financial Times news story on Canadian
inflation. According to the FT:
[A]nalysts said that the main driver of the recent bout of Canadian dollar appreciation
was higher-than-expected April inflation data, which saw the bond market fully
price in a 25 basis point rise in Canadian interest rates by the end of the year.
If central banks act to raise interest rates when inflation rises, then because higher interest
rates cause currency appreciation, it might be possible to resolve the apparent contradiction
to our model. To do so fully, however, we must consider two aspects of the way
in which modern central banks actually formulate and implement monetary policy.
1. The interest rate, not the money supply, is the prime instrument of monetary
policy. Nowadays, most central banks do not actually target the money supply in order
to control inflation. They instead target a benchmark short-term rate of interest (such
as the overnight “federal funds” rate in the United States). How does our discussion of
money market equilibrium help us to understand this process? Consider Figure 15-3,
and assume that the central bank wishes to set an interest rate of . It can do so
simply by agreeing to provide or take up all of the cash that the market wishes to trade
at that rate of interest. If the money supply is initially , for example, there will be an
excess demand for money at the interest rate , so people will sell bonds to the central
bank for money (in effect, borrowing) until the money supply has expanded to
and the excess demand is gone. Central banks tend to set an interest rate, rather than
the money supply, because the money demand schedule shifts around frequently
in practice. If the central bank were to fix the money supply, the result would
L(R, Y)
Q1
R1
Q2
R1
E$/€
E 3 $/€.
PUS
2
CHAPTER 15 Money, Interest Rates, and Exchange Rates 379
be high and possibly damaging interest rate volatility; it is thus more practical to fix
the interest rate and let the money supply adjust automatically when necessary.9
Our discussion above of the positive relationship between the money supply and
price level will tip you off, however, to one potential problem of an interest rate instrument.
If the money supply is free to grow or shrink as markets collectively desire,
how can the price level and inflation be kept under control? For example, if
market actors doubt the central bank’s resolve to control inflation, and suddenly
push the price level up because they expect higher prices in the future, they could
simply borrow more money from the central bank, thereby bringing about the
money supply increase needed to maintain higher prices in the long run. This worrisome
possibility brings us to the second pillar of modern monetary policy.
2. Most central banks adjust their policy interest rates expressly so as to keep inflation
in check. A central bank can keep inflation from getting too high or too low
by raising the interest rate when it learns that inflation is running higher than expected,
and lowering it when inflation is running lower. As we will see more fully in
Chapter 17, a rise in the interest rate, which causes the currency to appreciate, puts a
damper on demand for a country’s products by making them more expensive compared
to foreign goods. This fall in demand, in turn, promotes lower domestic prices.
A fall in the interest rate, symmetrically, supports domestic prices. Indeed, many
central banks now follow formal strategies of inflation targeting, under which they
announce a target (or target range) for the inflation rate and adjust the interest rate to
keep inflation on target. Central banks generally target so-called core inflation,
which is inflation in the price level excluding volatile components such as energy
prices, rather than headline inflation, which is inflation in the total consumer price
index. The formal practice of inflation targeting was initiated by New Zealand’s
central bank in 1989, and the central banks of many other developed and developing
areas, including Canada, Chile, Mexico, South Africa, Sweden, Thailand, the United
Kingdom, and the euro zone, have followed suit. The central banks of the United
States and Japan, while strongly averse to inflation, have so far been reluctant to
announce definite inflation targets.10
We can now understand the “paradox” of higher-than-expected inflation causing
currency appreciation rather than depreciation. Suppose market participants unexpectedly
push up prices and borrow to enlarge the money supply. Thus, when the
Canadian government releases new price data, the data show a price level higher than
9For a nontechnical account of modern central bank policy implementation, see Michael Woodford, “Monetary
Policy in a World Without Money,” International Finance 3 (July 2000), pp. 229–260. Woodford’s provocative
title points to another advantage of the interest rate instrument for central banks: It is possible to conduct monetary
policy even if checking deposits pay interest at competitive rates. For many purposes, however, it is reasonable
to ignore the variability of the L(R, Y) schedule and simply assume that the central bank directly sets the
money supply. In the rest of the book we shall, for the most part, make that simplifying assumption. The major
exception will be when we introduce fixed exchange rates in Chapter 18.
10The International Monetary Fund (IMF), which we will discuss in Chapter 19, maintains a useful classification
of its member countries with regard to their monetary policy frameworks as well as their exchange rate systems;
see the documentation and data at http://www.imf.org/external/np/mfd/er/2008/eng/0408.htm. The IMF does not
consider Switzerland to be an inflation targeter, but the difference between its actual procedures and inflation targeting
is small, and this Case Study therefore includes it among the inflation-targeting countries. On inflationtargeting
practices and the theory behind them, see the books by Bernanke et al. and by Truman in Further
Readings. For a critique of the idea of targeting core rather than headline inflation, see Stephen Cecchetti, “Core
Inflation Is an Unreliable Guide,” Financial Times, September 12, 2006.
380 PART THREE Exchange Rates and Open-Economy Macroeconomics
what market participants had previously predicted. If the Bank of Canada is expected
to raise interest rates quickly so as to push the price level and money supply back on
course, there is no reason for the future expected exchange rate to change. But with
higher Canadian interest rates, interest parity requires an expected future
depreciation of the Canadian dollar, which is consistent with an unchanged future
exchange rate only if the Canadian dollar appreciates immediately. The picture of the
economy’s adjustment after the unexpected increase in money and prices would look
like Figure 15–13 in reverse (that is, constructed to reflect a monetary contraction
rather than an expansion)—with the added assumption that the Bank of Canada gradually
moves interest rates back to their initial level as the price level returns to its
targeted path.11
Economists Richard Clarida of Columbia University and Daniel Waldman of
Barclays Capital offer striking statistical evidence consistent with this explanation.12
These writers measure unexpected inflation as the inflation rate estimate initially announced
by a government, prior to any data revisions, less the median of inflation forecasts
for that period previously published by a set of banking industry analysts. For a
sample of ten countries—Australia, Britain, Canada, the euro area, Japan, New Zealand,
Norway, Sweden, Switzerland, and the United States—Clarida and Waldman examine
the exchange rate changes that occur in the period lasting from five minutes prior to an
inflation announcement to five minutes afterward. Their key findings are these:
1. On average for the ten currencies that they study, news that inflation is unexpectedly
high does indeed lead a currency to appreciate, not depreciate.
2. The effect is stronger for core than for headline inflation.
3. The effect is much stronger for the inflation-targeting countries than for the United
States and Japan, the two countries that do not announce inflation targets. In the
case of Canada, for example, the announcement of an annual core inflation rate that
is 1 percent per year above the market expectation leads the Canadian dollar to
appreciate immediately by about 3 percent against the U.S. dollar. The corresponding
effect for the U.S. dollar/euro exchange rate, while in the same direction, is
only about one-quarter as big.
4. For countries where sufficiently long data series are available, the strengthening
effect of unexpected inflation on the currency is present after the introduction of
inflation targeting, but not before.
Scientific theories can be conclusively disproved, of course, but never conclusively
proved. So far, however, the theory that strict inflation targeting makes bad news on
inflation good news for the currency looks quite persuasive.
11Strictly speaking, the narrative in the text describes a setting with price level rather than inflation rate targeting.
(Can you see the difference?) The reasoning in the case of inflation targeting is nearly identical, however,
provided that the central bank’s interest rate response to unexpectedly high inflation is sufficiently strong.
12See Clarida and Waldman, “Is Bad News About Inflation Good News for the Exchange Rate? And If So, Can
That Tell Us Anything about the Conduct of Monetary Policy?” in John Y. Campbell, ed., Asset Prices and
Monetary Policy (Chicago: University of Chicago Press, 2008). Michael W. Klein of Tufts University and Linda
S. Goldberg of the Federal Reserve Bank of New York used a related approach to investigate changing market
perceptions of the European Central Bank’s inflation aversion after its founding in 1999; see “Establishing
Credibility: Evolving Perceptions of the European Central Bank,” Institute for International Integration Studies
Discussion Paper 194, Trinity College, Dublin, December 2006.
CHAPTER 15 Money, Interest Rates, and Exchange Rates 381
aggregate money demand, p. 358
deflation, p. 372
exchange rate overshooting,
p. 377
inflation, p. 372
long run, p. 363
long-run equilibrium,
p. 368
money supply, p. 356
price level, p. 358
short run, p. 363
SUMMARY
1. Money is held because of its liquidity. When considered in real terms, aggregate
money demand is not a demand for a certain number of currency units but is instead a
demand for a certain amount of purchasing power. Aggregate real money demand depends
negatively on the opportunity cost of holding money (measured by the domestic
interest rate) and positively on the volume of transactions in the economy (measured
by real GNP).
2. The money market is in equilibrium when the real money supply equals aggregate real
money demand. With the price level and real output given, a rise in the money supply
lowers the interest rate and a fall in the money supply raises the interest rate. A rise in
real output raises the interest rate, given the price level, while a fall in real output has
the opposite effect.
3. By lowering the domestic interest rate, an increase in the money supply causes the domestic
currency to depreciate in the foreign exchange market (even when expectations
of future exchange rates do not change). Similarly, a fall in the domestic money supply
causes the domestic currency to appreciate against foreign currencies.
4. The assumption that the price level is given in the short run is a good approximation to
reality in countries with moderate inflation, but it is a misleading assumption over the
long run. Permanent changes in the money supply push the long-run equilibrium price
level proportionally in the same direction but do not influence the long-run values of
output, the interest rate, or any relative prices. One important money price whose longrun
equilibrium level rises in proportion to a permanent money supply increase is the
exchange rate, the domestic currency price of foreign currency.
5. An increase in the money supply can cause the exchange rate to overshoot its long-run
level in the short run. If output is given, a permanent money supply increase, for example,
causes a more-than-proportional short-run depreciation of the currency, followed
by an appreciation of the currency to its long-run exchange rate. Exchange rate overshooting,
which heightens the volatility of exchange rates, is a direct result of sluggish
short-run price level adjustment and the interest parity condition.
KEY TERMS
PROBLEMS
1. Suppose there is a reduction in aggregate real money demand, that is, a negative shift
in the aggregate real money demand function. Trace the short-run and long-run effects
on the exchange rate, interest rate, and price level.
2. How would you expect a fall in a country’s population to alter its aggregate money
demand function? Would it matter if the fall in population were due to a fall in the
number of households or to a fall in the size of the average household?
3. The velocity of money, V, is defined as the ratio of real GNP to real money holdings,
in this chapter’s notation. Use equation (15-4) to derive an expression
for velocity and explain how velocity varies with changes in R and in Y. (Hint: The
V = Y/(M/P)
382 PART THREE Exchange Rates and Open-Economy Macroeconomics
effect of output changes on V depends on the elasticity of aggregate money demand
with respect to real output, which economists believe to be less than unity.) What is
the relationship between velocity and the exchange rate?
4. What is the short-run effect on the exchange rate of an increase in domestic real GNP,
given expectations about future exchange rates?
5. Does our discussion of money’s usefulness as a medium of exchange and unit of account
suggest reasons why some currencies become vehicle currencies for foreign exchange
transactions? (The concept of a vehicle currency was discussed in Chapter 14.)
6. If a currency reform has no effects on the economy’s real variables, why do governments
typically institute currency reforms in connection with broader programs aimed
at halting runaway inflation? (There are many instances in addition to the Turkish
case mentioned in the text. Other examples include Israel’s switch from the pound to
the shekel, Argentina’s switches from the peso to the austral and back to the peso, and
Brazil’s switches from the cruzeiro to the cruzado, from the cruzado to the cruzeiro,
from the cruzeiro to the cruzeiro real, and from the cruzeiro real to the real, the current
currency, which was introduced in 1994.)
7. Imagine that the central bank of an economy with unemployment doubles its money
supply. In the long run, full employment is restored and output returns to its fullemployment
level. On the (admittedly unlikely) assumption that the interest rate before
the money supply increase equals the long-run interest rate, is the long-run increase in
the price level more than proportional or less than proportional to the money supply
change? What if (as is more likely) the interest rate is initially below its long-run level?
8. Between 1984 and 1985, the money supply in the United States increased to
billion from billion, while that of Brazil increased to 106.1 billion cruzados
from 24.4 billion. Over the same period, the U.S. consumer price index rose to 100
from a level of 96.6, while the corresponding index for Brazil rose to 100 from a level
of only 31. Calculate the 1984–1985 rates of money supply growth and inflation for
the United States and Brazil, respectively. Assuming that other factors affecting the
money markets did not change too dramatically, how do these numbers match up with
the predictions of this chapter’s model? How would you explain the apparently different
responses of U.S. compared with Brazilian prices?
9. Continuing with the preceding question, note that the monetary value of output in
1985 was billion in the United States and 1,418 billion cruzados in Brazil.
Refer back to question 3 and calculate velocity for the two countries in 1985. Why do
you think velocity was so much higher in Brazil?
10. In our discussion of short-run exchange rate overshooting, we assumed that real output
was given. Assume instead that an increase in the money supply raises real output in the
short run (an assumption that will be justified in Chapter 17). How does this affect the
extent to which the exchange rate overshoots when the money supply first increases? Is
it likely that the exchange rate undershoots? (Hint: In Figure 15-12a, allow the aggregate
real money demand schedule to shift in response to the increase in output.)
11. Figure 14-2 shows that Japan’s short-term interest rates have had periods during
which they are near or equal to zero. Is the fact that the yen interest rates shown never
drop below zero a coincidence, or can you think of some reason why interest rates
might be bounded below by zero?
12. How might a zero interest rate complicate the task of monetary policy? (Hint: At a
zero rate of interest, there is no advantage in switching from money to bonds.)
13. As we observed in this chapter, central banks, rather than purposefully setting the
level of the money supply, usually set a target level for a short-term interest rate
by standing ready to lend or borrow whatever money people wish to hold at that
interest rate. (When people need more money for a reason other than a change in
$4,010
$570.3
$641.0
CHAPTER 15 Money, Interest Rates, and Exchange Rates 383
the interest rate, the money supply therefore expands, and it contracts when they
wish to hold less.)
a. Describe the problems that might arise if a central bank sets monetary policy by
holding the market interest rate constant. (First, consider the flexible-price case,
and ask yourself if you can find a unique equilibrium price level when the central
bank simply gives people all the money they wish to hold at the pegged interest
rate. Then consider the sticky-price case.)
b. Does the situation change if the central bank raises the interest rate when prices are
high, according to a formula such as where a is a positive
constant and a target price level?
c. Suppose the central bank’s policy rule is where u is a
random movement in the policy interest rate. In the overshooting model shown in
Figure 15-12, describe how the economy would adjust to a permanent one-time
unexpected fall in the random factor u, and say why. You can interpret the fall in u
as an interest rate cut by the central bank, and therefore as an expansionary monetary
action. Compare your story with the one depicted in Figure 15-13.
No comments:
Post a Comment