외환론

Foreign Exchange Theory

Lecture 7

1
Part I: Real Exchange Rate and Purchasing Power
Parity

• Reading
– Schimitt-Grohe, Uribe, and Woodford, slides
for chapter 9
– Bekaert and Hodrick, chapter 8

2
Review: Law of One Price

• The law of one price (LOOP) states that a good costs the
same abroad and at home.
– 𝑃 = ℇ𝑃∗.
– Assumption: no arbitrage
• Types of goods for which the LOOP is more likely to hold
– Commodities, luxury consumer goods.
• Types of goods for which the LOOP is unlikely to hold
– personal services (e.g., health care, education, restaurant meals,
domestic services, and personal care like haircuts), housing,
transportation, and utilities.

3
Review: Big Mac and the LOOP

ℇ𝑃𝐵𝑖𝑔𝑀𝑎𝑐∗
𝑒 𝐵𝑖𝑔𝑀𝑎𝑐 = 𝐵𝑖𝑔𝑀𝑎𝑐
𝑃
• The LOOP holds for the Big Mac when 𝑒 𝐵𝑖𝑔𝑀𝑎𝑐 = 1.
• This means that a Big Mac should cost the same abroad
and at home when measured in a common currency.

4
Review: Changes in Big Mac RER from 2006 to
2019

5
From a single good to a consumption basket

ℇ𝑃∗
𝑒=
𝑃
• 𝑒 : Real exchange rate
• P* : the price level in a foreign country
• P : the price level in a domestic country
• ℇ is the nominal exchange rate as before.

• Price level : the nominal price level of a country’s “basket

of goods”
– Weighted average of goods and services (i.e., we spend 1% of
our income on shoes)
6
Real Exchange Rate

ℇ𝑃∗
𝑒=
𝑃
• The real exchange rate: e
– measures how expensive the foreign country is relative to the
home country
– indicates the relative price of a consumption basket in the foreign
country compared to the home country.
• If RER increases, foreign goods become relatively
expensive.
– domestic country's exports become more competitive.
• RER affects international trade and investment decisions
7
Purchasing Power Parity (PPP)
• The LOOP for broad baskets of goods representative of
households’ actual consumption, as opposed to a single
good
• In the next,
– Absolute PPP
– PPP Exchange rate
– Relative PPP

8
Absolute PPP
ℇ𝑃∗
𝑒=
𝑃
• RER and PPP
– When 𝑒 = 1, we say that absolute purchasing power parity holds.
• Testing Absolute PPP
– Requires data on prices of P* and P
– Data collection is difficult, time-consuming, and expensive
– World Bank's International Comparison Program (ICP) is the
only available data source
– ICP collects price level data of over 1,000 goods for 199
countries every 6 years
– Using the data, ICP estimates P* and P for each country
– Data for ℇ is readily available from financial markets. 9
10
Comparing the ICP and Big-Mac RER

Big-Mac RER is a good proxy measure of RER

11
PPP Exchange Rate
• The PPP exchange rate is the nominal exchange rate
that would make the consumption basket equally
expensive in two countries.
ℇ𝑃𝑃𝑃 𝑃 ∗ = 𝑃.
• ℇ𝑃𝑃𝑃 > ℇ : the domestic country is more expensive than
the foreign country (𝑃 > ℇ𝑃∗ )
– the foreign currency is undervalued.

• ℇ𝑃𝑃𝑃 < ℇ: the foreign currency is overvalued.

12
Actual USD/EUR and PPP Exchange Rates

13
PPP Exchange Rate and Standard of Living
Comparisons
• Prices of similar goods vary widely across borders
– Comparing GDP per capita across countries can be misleading
• Example
– US GDP per capita is 32 times higher than India's in 2011
– However, this doesn't account for purchasing power differences
– One way to compare is by calculating how many burgers (e.g.
Big Macs) can be bought with each GDP
– Big Mac costs $5.58 in the US but only $2.55 in India
– Americans are 15 times richer than Indians in terms of Big Macs
• Income gap is still large, but not as significant as
suggested by simple ratio of dollar GDPs
14
Per Capita GDP at PPP Exchange Rate
• Per Capita GDP at PPP Exchange Rate reflects
differences in the cost of living between countries.
• Convert each country’s GDP using PPP exchange rate
instead of nominal exchange rate

ℇ𝑃𝑃𝑃 × 𝐺𝐷𝑃𝑆𝑜𝑢𝑡ℎ 𝑘𝑜𝑟𝑒𝑎

15
16
Relative Purchasing Power Parity
• Most studies of PPP focus on changes in RER, rather
than on its level.
– Consumer Price Indices (CPIs) are readily available for many
countries at a relatively high frequency (typically monthly)
– CPIs provide information about the change in the price of a
basket of goods, not about the price level.
– CPIs are used to calculate changes in the real exchange rate

• Relative PPP holds if

ℇ𝑃∗
∆𝑒 = ∆
𝑃
where ∆= change over time.
17
Relative Purchasing Power Parity
• Relative PPP holds if
ℇ𝑃∗
∆𝑒 = ∆
𝑃
• When ∆𝑒 < 0, RER appreciates and the domestic
country becomes more expensive
• When ∆𝑒 > 0, RER depreciates and the domestic
country becomes less expensive

18
Does relative PPP hold in the long run?

19
Does relative PPP hold in the long run?
• Let us define RER
ℇ𝑡 𝑃𝑡𝑈𝑆
𝑒𝑡 =
𝑃𝑡
– If 𝑒𝑡 increases, there is a real depreciation of the country’s
currency against the U.S. dollar

20
• 𝜖𝑡𝑟 ∶ the real depreciation rate of the country’s currency
against the U.S. dollar

ℇ𝑡
• 𝜖𝑡 = − 1: the nominal depreciation rate of the
ℇ𝑡−1
country’s currency against the U.S. dollar
𝑃𝑡
• 𝜋𝑡 = − 1: inflation rate
𝑃𝑡−1

21
• Taking the natural log and using the approximation
ln(1 + x) ≈ 𝑥

𝜖𝑡𝑟 = 𝜖𝑡 + 𝜋𝑡𝑈𝑆 − 𝜋𝑡

• Relative PPP holds if RER does not change over time,

𝜖𝑡𝑟 = 0:
𝜖𝑡 = 𝜋𝑡 − 𝜋𝑡𝑈𝑆

• Relative PPP holds if the rate of depreciation of the

country’s currency against the dollar is equal to the
inflation differential between the country
22
Does relative PPP hold in the long run?

• Let 𝜖 , 𝜋 ,and 𝜋 𝑈𝑆 be the averages of 𝜖𝑡 , 𝜋𝑡 , 𝜋𝑡𝑈𝑆 over a

long period of time.
• Then relative PPP holds in the long run if
𝜖 = 𝜋 − 𝜋 𝑈𝑆

• The next slide suggests that this expression holds for

many countries.

23
Average Inflation Differentials and Depreciation
Rates, 1960-2017

24
Does relative PPP hold in the short run?
• RER moves around quite a bit.
– The standard deviation of RER is 9.3 percent.
– This means that typically, from one year to the next, the
United States becomes almost 10 percent more expensive
or cheaper than the United Kingdom.
• This suggests that relative PPP does not hold in the
short run.

25
Year-over-year percent change in the Dollar-
Pound RER, 1870-2018

26
Explaining the short-run deviations from relative
PPP
• Two factors that can explain failures in relative PPP
across cities
– Transportation costs: households may be less willing to travel
farther distances to take advantage of price differences
– International border: tariffs, quotas, government regulation
• How Wide is the Border? (Engel and Rogers, 1996)
– Use data on prices of 14 baskets of goods (apparel, footwear,
medical care, etc), 14 US citieis, and 9 Canadian cities from
1978-1994.

– Short-run deviations from relative PPP are increasing in distance

and in the existence of an international border.
27
How Wide is the Border?
• Engel and Rogers estimate the following regression:

– 𝜎 𝑔 : standard deviation of 𝜖𝑡𝑟 between cities

– 𝑑 ∶ distance in miles between cities
– 𝐵: B=0 if both cities are within the same country, otherwise, B=1
– 𝜇: regression residual
• Suppose c1 and c2 are separated by a border. If we
were to remove the border, by how much should we
increase the distance between c1 and c2 so that
𝜎 𝑔 stays the same?
– 12,000 miles! 28
Short-run deviations from PPP: Nontradable
• High transportation costs can make some goods non-
traded
• Example: a haircut – no one would fly to India just
because a haircut there is 10 times cheaper.
• This type of goods and services are called nontradable
goods.
– nontradable goods: services (haircuts, restaurant meals,
housing, primary health, education).
– tradable goods: agricultural commodities and many
manufactured goods
• Nontradables make up a significant share output,
typically above 50 percent. 29
Short-run deviations from PPP: Home Bias
• Different weights for goods in consumption baskets
across countries can lead to PPP failure.
– These variations in weights reflect differences in national tastes
acquired via resource abundance or production specialization.
– Ex: Americans spend more on burgers, while Japanese spend
more on Sushi.
– Ex2: Argentines spend more on beef, while Germans may spend
more on cars.
• This preference for domestic goods is known as home
bias.
• When beef prices rise relative to car prices, the price
index in Argentina will increase more than in Germany
30
because Argentines spend more on beef.