The concept of purchasing-power parity (PPP) has two applications: it was originally developed as a

theory of exchange rate determination, but it is now primarily used to compare living standards across
countries.

• From the perspective of exchange rate determination, PPP is useful as a reminder that monetary
policy has no long-run impact on the real exchange rate. Thus, countries with different inflation rates
should expect their bilateral exchange rate to adjust to offset these differentials in the long run. The
exchange rate, however, can deviate persistently from its PPP value in response to real shocks.

• To compare living standards across countries, PPP exchange rates are constructed by comparing the
national prices for a large basket of goods and services. These rates are used to translate different
currencies into a common currency to measure the purchasing power of per capita income in different
countries. A PPP exchange rate constructed in this manner is not, however, an accurate measure of the
equilibrium value of the market-determined exchange rate.

it has been argued that the Canadian dollar is undervalued because its current market value is below the
purchasing-power-parity (PPP) exchange rate calculated by the Organisation for Economic Co-operation
and Development (OECD) and Statistics Canada (Chart 1). While the deviation of the value of the
Canadian dollar from its purchasingpower-parity rate has been growing in recent years,

this article argues that this deviation cannot be interpreted as implying that the Canadian dollar is
undervalued by a comparable amount. Instead, this deviation indicates that the prices of goods and
services are, on average, lower in Canada than in the United States, when measured in the same
currency at the prevailing exchange rate.

Chart 1

Canada’s PPP Exchange Rat

PPP as a Theory of Exchange Rate

Determination

While the origins of the PPP concept can be traced back to the Salamanca School in 16th-century Spain,

its modern use as a theory of exchange rate determination begins with the work of Gustav Cassel

(1918), who proposed PPP as a means of adjusting pre–World War I exchange rates or parities for
countries intending to return to the gold standard system after hostilities ended.1 Some adjustment was
necessary because countries that left the gold standard in

1914 experienced significantly different rates of inflation during and after the war.

As a theory of exchange rate determination, the simplest and strongest form of PPP (absolute PPP) is
based on an international multi-good version of the law of one price (Box 1). Absolute PPP predicts that
the exchange rate should adjust to equate the prices of national baskets of goods and services between
two countries because of market forces driven by arbitrage.

Under absolute PPP, the exchange rate is simply equal to the ratio of the domestic to the foreign price of
a given aggregate bundle of commodities, but this implies that the real exchange rate is constant.3

In practice, however, absolute PPP does not hold for a number of reasons, and these undermine its
usefulness as a theory of the determination of the level of the exchange rate.4 The most important are

• the existence of non-traded goods and services that preclude arbitrage

• the presence of significant transactions costs for traded goods, including transport costs, tariffs, taxes,
information costs, and other non-tariff trade barriers that make arbitrage costly

1. Dornbusch (1987) provides a historical overview and insightful discussion of PPP.

2. Like Cassel, Keynes believed that the exchange rate needed to be adjusted for inflation differentials
because he recognized that wages and prices were too sticky to adjust. Unfortunately, Winston
Churchill, as Chancellor of the Exchequer, decided to return the United Kingdom to its pre-war parity in

1925. This move proved disastrous; exports fell and unemployment increased sharply.

3. The real exchange rate is defined here as the exchange rate deflated by the ratio of the domestic to
the foreign price index.

4. Perhaps the strongest criticism of the absolute version of PPP is by Paul Samuelson (1964, p. 153),
“Unless very sophisticated indeed, PPP is a misleadingly pretentious doctrine, promising us what is rare
in economics, detailed numerical prediction.”

5. Rogoff (1996) surveys evidence that the law of one price does not hold for most traded goods and
services; he concludes that international markets for these items are much less integrated than
domestic markets.

• the composition of the basket of goods and services included in measures of national price levels
differs across countries, especially for producer-based as opposed to consumer-based price indexes, and

• the fact that the real exchange rate is not constant in the short run because aggregate price levels are
sticky and the exchange rate is affected by money or asset market shocks, or in the long run because of
persistent real shocks

In practice . . . absolute PPP does not hold for a number of reasons, and these undermine its usefulness
as a theory of the determination of the level of the exchange rate.

A weaker version of PPP, known as relative PPP, implies that the exchange rate between two countries
should eventually adjust to account for differences in their inflation rates. That is, countries that follow
monetary policies with different inflation-rate objectives should expect to see this difference manifest
itself in an exchange rate movement. To illustrate the circumstances where relative PPP may provide
useful explanatory power, consider Table 1, which compares the cumulative inflation and exchange rate
experiences of
Canada and Mexico relative to the United States over the period 1975–2001. The table clearly shows
that, for

Canada, the exchange rate movement over the period is largely due to a depreciation of the underlying
real exchange rate because the cumulative inflation differential with the United States is only 6 per cent
of the total exchange rate movement. The opposite is true for

Canada

Mexico

United States

Table 1

Relative Inflation and Exchange Rates, 1975–2001

Country (1) (2) (3) Contribution

CPI in Price Exchange of relative

2001 ratiosa ratesb inflationc

1975=1 1975=1

3.38 1.03 1.52 6%

2260 687 747 92%

3.29 1 1 na

a. Canadian and Mexican price levels relative to the U.S. price level in 2001

b. Units of domestic currency per U.S. dollar

c. Proportion of exchange rate depreciation relative to the U.S. dollar that is explained by higher rates of
inflation in Canada and Mexico ns

To compare living standards between countries, it is necessary to translate per capita income or
expenditure values measured in the local currency into a common currency, normally the U.S. dollar.
This presents the problem of determining the appropriate exchange rate to use for the currency
translation. One could use

the nominal bilateral exchange rate with the U.S. dollar, but this ignores the often large differences in
the prices of a broad set of goods and services that are not reflected in the value of the exchange rate

Absolute PPP is obtained by extending the law of one price to multiple commodities in an international
setting. This “law” implies that, in the absence of transactions costs, competitive arbitrage should force
the same good to sell for the same price, expressed in a given currency, across countries.

To illustrate the law of one price, let and be the domestic and foreign currency prices of commodity (a
good or service) and the exchange rate (expressed as the price of foreign exchange).

Thus, the law of one price implies that

To extend this illustration to PPP, let and be the domestic and foreign price levels, which are constructed
by taking a weighted average of the prices of commodities in the national production or consumption
baskets:

, (2) where and represent the weights of commodity in the basket. If it is further assumed that the
weights are identical and the law of one price holds for all commodities, then

(3)

or

. (4)

Pi Pi

iE

Pi EPi = ∗

P P∗

P wi

ii=

∑ Pi and P∗ wi

ii=

∑ Pi = = ∗

wi wi

EP∗ = P

E PP = ⁄ ∗

As a theory of exchange rate determination, absolute PPP, given by equation (4), predicts that the
exchange rate will adjust to equalize price levels.
Note that absolute PPP assumes that the real exchange rate—the nominal exchange rate adjusted for
differences in national price levels—is constant:

In practice, absolute PPP does not hold because of obstacles to international trade. If these trade
frictions, denoted by , are assumed to be relatively constant, then (4) can be modified as

, (5) and taking the ratio between time and time gives

. (6) Equation (6) represents a weaker version of PPP (relative PPP) that predicts that the exchange rate
will adjust to offset inflation differentials between two countries over time. Thus, if most of the shocks
affecting the exchange rate are monetary rather than real, then relative PPP will be able to explain a
substantial portion of the exchange rate movement between two countries.

EP∗ ⁄ P = 1.

E kP = ⋅ ⁄ P∗

to

Et

Eo

------ Pt Po ⁄

Pt

∗ Po ⁄ ∗ = ---------------------