4/18/2018

International Parity Conditions

Assoc. Prof. Dr. Mai Thu Hien

Faculty of Banking and Finance
Foreign Trade University

International Parity Conditions

• Some fundamental questions managers of MNEs,
international portfolio investors, importers, exporters and
government officials must deal with every day are:
 What are the determinants of exchange rates?
 Are changes in exchange rates predictable?
• The economic theories that link exchange rates, price
levels, and interest rates together are called international
parity conditions.
• These international parity conditions form the core of the
financial theory that is unique to international finance.
• For reading: Eiteman 12e chapter 7, Pilbeam 3e chapter
6, Levi 4e chapter 8 2

International Parity Conditions

• These theories do not always work out to be “true” when
compared to what students and practitioners observe in
the real world, but they are central to any understanding
of how multinational business is conducted and funded
in the world today.
• The mistake is often not with the theory itself, but with
the interpretation and application of said theories.

1
 4/18/2018

Theories of exchange rate

determination

1. Purchasing power parity

2. Interest rate parity interest rate vs exchange rate in money market
3. International Fisher effect non-exchange rate valuables

1. absolute and relative

2. cover and uncover
3. non-exchange valuables and inflation rate

Price-based Methodologies for Equilibrium

Exchange Rate Assessment

• Assessing the equilibrium level of the exchange rate is a

critically important policy objective
• Consequences of substantial currency misalignments
extremely costly “Currency Wars” Sept- Nov 2010
• Here we examine certain price-based methodologies for
assessing equilibrium exchange rates
• Collectively known as purchasing power parity, PPP theories
of exchange rate determination

inflation rate vs exchange rate

1. Purchasing power parity

• Provides the link between price levels and the exchange

rate between two countries:
 The law of one price
 Purchasing power parity by homogeneous goods
(absolute and relative)
 Purchasing power parity by heterogenous goods
 A generalized version of PPP
- a distinction between traded and non-trade goods-
 Productivity Differentials in the Traded Goods Sector:
Balassa-Samuelson Effects

2
 4/18/2018

The law of one price

• Assumes that:
 Identical products or services can be produced and
sold in two different markets (homogeneity of goods).
 No restrictions exist on the sale (trade barriers) or
transportation costs of moving the product between
markets.
 Competitive markets to establish this condition
• States that: Consider the price of a good j, at time t in the
domestic, Pj, and foreign country, Pj*. The Law of One
Price states:
Pj,t = St . Pj,t*
 The products prices should be the same in both
markets when prices are expressed in terms of the
same (domestic) currency. 7

The law of one price

If NER = $1.5/£, a sweater that sells for $45 in the US must sell for
£30 in the UK, the dollar price of the sweater in the UK is ($1.5/£) *
(£30) = $45
If NER = $1.55/£ then a sweater is sold:
In the UK: £30 or ($1.55/£) * (£30) = $46.5
In the US: $45
 Buy sweaters in the US and ship them to the UK (until prices
are equal in the two locations)
 In the forex market, the demand for $↑ when UK traders
exchange £ for $  the dollar appreciates
 In the US, the demand for sweaters↑  price of sweaters↑
demand for $↑  the dollar appreciates
 In the UK, the supply of sweater increases  price of sweaters
in the UK↓  demand for £↓  the pound depreciates
 Thus, the dollar appreciates relative to the pound until the
exchange rate falls to the original level.
8

The law of one price

If NER = $1.45/£ then a sweater is sold:
In the UK: £30 or ($1.45/£) * (£30) = $43.5
In the US: $45
Buy sweaters in the UK and ship them to the US until
prices are equal in the two locations.
PUS = NER ($/£) * PUK
Law of one price (LOP) implies dollar price of good j is the
same wherever it is sold.

3
 4/18/2018

Raw Big Mac Index, Nominal Rates Economist 16/10/2010

1. In USA Big Mac
cost $3.71
2. In Brazil – Guido
Mantega casualty of
“Currency War” it
is $5.26: suggest
Real overvalued by
≈ 42%
3. Euro overvalued by
29%
4. In China it is $2.18
Yuan undervalued
by ≈ 40%
10

After adjusting for GDP per person (Income)

Economist 28/07/2011

Regression
Line

China less undervalued, but Euro and Real still overvalued relative to
USD (note USD is itself undervalued vs other currencies) 11

Why LOP may not hold

• Transaction Costs
– Let the price of a book (j) in the US be Pjt = USD 40 and the spot
exchange rate be St = USD/GBP 2.0. If the shipping and
insurance costs for exporting the book are USD 4, the price of the
book in the UK, Pjt* could be as low as GBP 18, or as high as GBP
22, before you consider an arbitrage transaction.
• Non-traded goods
• Quantitative or Tax restrictions (Quotas and Tariffs)
• Imperfect competition
– Exclusive dealerships => segmented markets.
– Manufacturers discourage parallel imports to profit from price
discrimination.
– Entry costs hinder arbitrage.
– Reluctance to change prices (menu costs).

12

4
 4/18/2018

Purchasing Power Parity by

homogenous goods
• Purchasing power parity is the application of the law of one price
across countries for all goods and services, or for representative
groups (“baskets”) of goods and services.
PUS = NER (USD/GBP) . PUK
PUS = level of average prices in the US (Price level of domestic
commodity basket measured in units of domestic currency)
PUK = level of average prices in the UK (Price level of foreign
commodity basket measured in units of foreign currency)
NER (USD/GBP) = USD - GBP nominal exchange rate
• Purchasing power parity states that the exchange rate between
two countries' currencies equals the ratio of the countries’ price
levels.
• Purchasing power parity asserts that all countries’ price levels are
equal when measured in terms of the same currency (assumptions
hold as the LOP).
13

Purchasing Power Parity by

homogenous goods
• Purchasing power parity comes in 2 forms:
 Absolute PPP: Exchange rates equal the level of relative average
prices across countries.
 Relative PPP: changes in exchange rates equal changes in prices
(inflation) between two periods

14

Absolute purchasing power parity

• APPP states that the spot exchange rate is determined
by the relative prices of similar baskets of goods or
exchange rates equal the level of relative average prices
across countries (application of law of one price to price
levels):
NER (USD/GBP) = PUS / PUK
• If Price Levels based on same basket of goods with
identical weights, then APPP follows from LOP
• If goods baskets are not identical APPP follows only as
an approximation (RPPP) even if LOP held exactly for all
goods
• By comparing the prices of identical products
denominated in different currencies, we could determine
the “real” or PPP exchange rate that should exist if
markets were efficient. 15

5
 4/18/2018

Problems with absolute PPP

• APPP is unlike to hold because:

 All goods not identical in both countries.
 In fact, restrictions exist on transaction costs and trade barriers.
 Statistical problems: different basket of goods are used in
different countries to compute price indices and difference in
proportion of goods and services consumption in different
countries (because of different tastes and needs)
 RPPP can be expected to hold even in the presence of
such distortions.

16

Relative purchasing power parity

• RPPP is a weaker form of PPP, and relaxes assumptions of
APPP
• RPPP holds that PPP is not particularly helpful in determining
what the spot rate is today, but that the relative change in
prices between two countries over a period of time determines
the change in the exchange rate over that period, or changes
in exchange rates equal changes in prices (inflation) between
two countries over period of times or if domestic price level
increases by x%, domestic currency depreciates by x%.
(NERUSD/GBP,t+1 - NERUSD/GBP,t) / NERUSD/GBP,t = p US - p UK
where p = inflation rate from period t to t+1
∆NER ≈ ∆P - ∆P*
∆NER >0 (∆P > ∆P*): domestic currency depreciates/foreign currency
appreciates
∆NER <0 (∆P < ∆P*): domestic currency appreciates/foreign currency
depreciates 17

Relative purchasing power parity

• In other words, if the spot exchange rate between two

countries starts in equilibrium, any change in the
differential rate of inflation between them tends to be
offset over the long run by an equal but opposite change
in the spot exchange rate.
• If not, these price changes lead to a deficit on the current
account in the BOP unless offset by capital and financial
flows (country experiences higher inflation than those of
its trading partners and its exchange rate doesn’t change
 more imports  deficit).

18

6
 4/18/2018

Exhibit 7.2 Relative Purchasing Power Parity

(PPP)

7-19

Relative purchasing power parity

• The absolute version of PPP: St = PUS,t / PUK,t
where PUS,t and PUK,t are the price index values (cost of the
baskets of goods) in USD and GBP at time t
• Annual percentage change in price levels in the US and the
UK at time t+1 (or inflation rate) p
• The spot exchange rate at time t+1 would be:
St+1 = PUS,t (1+ pUS)/ PUK,t(1+ pUK) = St (1+ pUS)/ (1+ pUK)
• Change in the spot rate between t and t+1:
(St+1 - St) / St = (pUS - pUK) / (1 + pUK)
• Approximated by dropping the denominator (1 + pUK) if it is
consider to be small, then (St+1 - St) / St = pUS - pUK
or ∆S = pUS – pUK

20

Example
So if Pt = 100 and inflation at home is 6% per annum, and Pt *
is 200 and inflation is 3%. If St = 2 initially:
• This implies: Pt+1 = 106 and P*t+1 = 206
 St+1/2 = 1.06/1.03 = 1.0291
• The RPPP condition predicts that St+1, the exchange rate
at t+1 will be St+1 = 2.0582 (= 2 x 1.0291).
• This is a depreciation of domestic currency of 2.91%
• In practice, under RPPP the percentage change in the
exchange rate (= 2.91%) is approximated by the inflation
differential (6 - 3)% = 3% ≈ 2.91%

21

7
 4/18/2018

PPP and the Real Exchange Rate

The real exchange rate, Q, measures the relative price
level of foreign to domestic goods (both expressed in units
of domestic currency), namely (ignoring time subscripts):
1. Under APPP Q = S.P*/P (1)
or, in logarithm: q = s + p* - p
The Real Exchange Rate is constant:
S = P / P* → Q = 1 (q = 0)
Q >1 (Q rises): domestic currency depreciates
Q <1 (Q falls): domestic currency appreciates
2. Under RPPP Q = k.S.P*/P
where k is a constant
or, in logarithms: q = c + s + p* - p (2)
Where: c = log (k) = 0 under APPP
A change in the ratio of price levels implies an equal
proportionate change in the exchange rate
22

PPP and the Real Exchange Rate

1


CA  RER  Y *   Y   RER  CA  Y *   Y 




 is very high ( = ) → RER = 1

where : real exchang rate elasticity of CA ( > 0):
measuring the responsiveness of the trade balance to a
change in real exchange rate.
The effects of RER on CA depend on:
• : RER↑→CA↑
• Value of :
 If domestic and foreign goods are homogenous or perfectly
substitutable:  is high ( = ): RER↑(→1)→ CA↑ (high): PPP
 If domestic and foreign goods are heterogenous or imperfectly
substitutable:  small: RER↑→CA↑ (small)

23

Exchange Rates Pass-through

• The degree to which the prices of imported and exported goods
change as a result of exchange rate changes is termed pass-
through.
• Although PPP implies that all exchange rate changes are passed
through by equivalent changes in prices to trading partners,
empirical research in the 1980s questioned this long-held
assumption.
• For example, a car manufacturer may or may not adjust pricing of
its cars sold in a foreign country if exchange rates alter the
manufacturer’s cost structure in comparison to the foreign market.
• Pass-through can also be partial as there are many mechanisms by
which companies can compartmentalize or absorb the impact of
exchange rate changes.
• The degree of Pass-through is measured by the proportion if the
exchange rate change reflected in foreign currency price.
• Price elasticity of demand is an important factor when determining
pass-through levels. 24

8
 4/18/2018

Exchange Rates Pass-through

BMW automobile is produced in Germany and sold in the
US: S(USD/EUR) = 1, PBMW, EUR = EUR35,000
 PBMW, USD = EUR35,000 * USD1,000/EUR = USD35,000
• Exchange rate pass-through: If EUR appreciates 20%
versus USD, price of BMW should theoretically increase by
20% to:
PBMW, USD=EUR35,000*USD1,200/EUR = USD42,000
• Exchange rate pass-through is partial: USD price of BMW
rises only to
USD40,000 by 40,000/35000 = 1.1429 = 14.29%, not by 20%.
 degree of pass-though is 14.29%/20% = 0.71 = 71%:
Only 71% of the exchange rate change was pass through to
the USD price. The remaining 29% has been absorbed by
25
BMW.

Empirical Evidence on PPP

Theories

26
University of Manchester

Nominal Exchange Rates changes and

Inflation Differentials: 1974-2006
Infln
Diffs
vs
USA
CPIs

45º = PPP line

Changes in average annual exchange rates: 21 countries for 32 years vs USD

1. On the left: 672 1-year changes (32 years x 21 countries)
2. On the right: 84 (non-overlapping) 8-year changes, 1974-82,
1983-90, 1991-1998, 1999-2006.
27

9
 4/18/2018

Nominal Exchange Rates changes and

Inflation Differentials: 1974-2006
Infln
Diffs
vs
USA
CPIs

Changes in average annual exchange rates: 21 countries for 32 years vs USD

1. On the left: 42 (non-overlapping) 16-year changes

2. On the right: 21 (non-overlapping) 32-year changes

28

Nominal Exchange Rates changes and

Inflation Differentials: 1974-2006
Infln
Diffs
vs
USA
CPIs 672 1-year changes

29
Changes in average annual exchange rates: 21 countries for 32 years vs USD

Summary of empirical evidence on PPP

• Empirical studies have found that violations of purchasing

power parity over the short to medium term are extremely
common – see the 1-year evidence
• Shocks can push the exchange rate away from its
equilibrium (mean) PPP value for quite long periods
• The rate of re-adjustment back to the PPP value is slow –
half-life of 3 years (plus) in many studies.
• This may be due to a combination of factors, including lack
of integration of the world’s goods markets (because of
protectionism, taxes, transportation and transaction costs,
‘sticky’ goods market prices, etc).
30

10
 4/18/2018

Summary of empirical evidence on PPP

• Relative PPP may not hold in the short-run, but there is

strong evidence it is a reasonably good approximation in
the long-run.
• However, it may take 3-5 years to reach long-run
equilibrium (in the sense PPP holds).
• Thus, for several years, Q may be rising or falling,
affecting competitiveness, trade balances (net exports)
and output levels.

31

Purchasing power parity: Empirical

findings conclusions
• Two general conclusions can be made from these tests:
 PPP holds up well over the very long run but poorly for shorter
time periods; and,
 The theory holds better for countries with relatively high rates of
inflation and underdeveloped capital markets.
• Empirical testing of PPP and the law of one price has
been done, but has not proved PPP to be accurate in
predicting future exchange rates.

32

Mean Reversion in Real Exchange Rate

• The fact that over time scatter plot converges to 45° PPP line
in long-run means
• Q, the Real Exchange Rate has a tendency to revert to its
own average or mean value
This property is called MEAN REVERSION
• Moreover, in the long-run this equilibrium value is consistent
with PPP

33

11
 4/18/2018

Issues with PPP as a Theory of the

Real Equilibrium Exchange Rate
Two important points to note
• The scatter plots produced using CPI consumer price
indices: we could equally have used
 WPI (wholesale Price indices)
 Unit labour cost indices
 GDP deflators
 Export Price indices
• Applying PPP with a different choice of price indices can
yield very different measures of the real equilibrium
exchange rate, Q. In other words, our value of the
equilibrium Q is dependent upon which price indices we
choose to undertake our calculation

34

Case Study: Real Exchange Rate:

UK and Germany 1970-2000
Consider following hypothetical question
• What conversion rate for GBP to Euro would have been most
appropriate if UK had chosen to adopt the Euro in Jan 2001?
• Plausible Answer: Its real equilibrium exchange rate with
Germany…over what time period?
– post-Bretton Woods (post 1970)
– Post 1990?
• In other words, the average value to which Q tends to mean
revert in a particular time period

35

Case Study: Real Exchange Rate: UK and

Germany 1970-2000

CPI GDP Defl

WPI Unit Lab Cost

Export prices
1970-2000 1990-2000

36
Source: IMF Calculations

12
 4/18/2018

Case Study: Real Exchange Rate: UK

and Germany 1970-2000
• Each of the 5 panels shows a different measure of Real
Exchange Rate with 5 different price indices
• Also takes average level since 1970 and since 1990
• So, 10 different measures of how much sterling GBP
misaligned against the DEM (hence the euro) at the end of
2001
• All 10 values suggest GBP overvalued (euro excessively
weak) – useful for Policymakers
• However, estimates range from an overvaluation of 10%
to more than 40%
• Issue…which do you pick?

37

PPP: Graphical findings

1973-2002: the British price level rose 99% relative to the U.S price
level, and as the theory of PPP predicts, the dollar appreciated
against the pound by 73%, smaller than the 99% increase predicted
by PPP. price level is affected not only by CPI but other factors that make these things happen
1985-87: the British price level rose relative to that of the U.S. Instead
of appreciating as PPP theory predicts, the U.S. Dollar actually 38
depreciated by 40% against the pound.

Trick: If a factor increases the demand for

domestic goods relative to foreign goods, the Source: Mishkin (2006), chapter 19
domestic currency will appreciate and vice versa. 39

13
 4/18/2018

The Links Between LOP, APPP and RPPP

• If absolute PPP holds, then so will relative PPP.

• Relative PPP may hold even if there are persistent
deviations in the average absolute price levels across
countries, so…..
If prices in London are persistently 80% higher than in
Brussels, relative PPP holds.
• The deviation from relative PPP can always be computed -
even in cases where the consumption bundles differ
across countries.
• Even if one cannot compare different consumption bundles
directly, it is still possible to compare changes in the prices
of these baskets of goods.
• Even when LOOP holds for every single good, APPP will
not generally hold if the consumption bundles (weights)
differ across countries.
40

Shortcomings of PPP
• Do not consider interest rates, government‘s intervention, statistical
problem, productivity differentials (Balassa-Samuelson Model) among
the factors that affect the exchange rates.
• Do not consider substitution effect: increase in price level of
domestically-produced goods can be substituted by other goods rather
than imported goods.
• Do not consider changes in technology and natural resources.
• Works in the long run, not the short run.
• Many goods and services are not traded: e.g. housing, land, services
such as restaurant meals, haircuts, and golf lessons. Thus change in
the demand of those may cause domestic price to change relative to
foreign price of the same basket.
• Based only on the international exchange of goods and services and
do not consider financial flows and money stocks.

41

The application of PPP

• Defining that a country‘s exchange rate is overvalued or

undervalued with respect to PPP.
• Making cross country comparisons of income, wages, or
GDP.
• Forecasting the exchange rate in the long run.

42

14
 4/18/2018

Overvaluation or undervaluation
• Individual national currencies often need to be evaluated against
other currency values to determine relative purchasing power.
• The objective is to discover whether a nation’s exchange rate is
“overvalued” or “undervalued” in terms of PPP.
• This problem is often dealt with through the calculation of exchange
rate indices such as the nominal effective exchange rate index.
• Nominal effective exchange rate index uses actual exchange rate to
create an index of the value of the subject currency overtime on a
weighted average basis. It does not really indicate anything about
the true value of the currency, or any thing related to PPP. It
calculates how the currency value relates to some arbitrarily chosen
base period.
• Real effective exchange rate index indicates how the weighted
average purchasing power of the currency has changed relative to
some arbitrarily selected base period.

43

Exhibit 7.3 IMF’s Real Effective Exchange

Rate Indexes for the United States, Japan,
and the Euro Area (2000 = 100)

An index value above 100 would suggest an overvalued currency

7-44
from a competitive perspective, or vice versa.

Exhibit 7.1 The McCurrency Menu – the

Hamburger Standard

Implied PPP = china/us (in local currency)

2 ways to calculate the undervaluation or overvaluation of currencies Actual exchange rate = yuan/in dollars
versus USD:
• Implied PPP rate/Actual rate -1 = 3.5/6.83 – 1 = -0.4879 ≈ -49%
• Price in China in USD/Price in the US in USD = 1.83/3.57 -1 = -0.4874 ≈
-49%

market price > theory price => overvaluation

VD: 6.83 > 3.5 => US overvalue; Yuan undervalue
45

15
 4/18/2018

Example

pUS pUK $/£2000 $/£2004

8.80% 9.13% 1.639 1.871

Which currency is overvalued or undervalued?

GBP overvalued; USD undervalued

46

Example

47

Cross-country comparisons

48

16
 4/18/2018

Exercise

• P7.1-7.5, 7.6, 7.10, 7.23

49

A version of PPP by heterogenous goods

If domestic goods and foreign goods are heterogenous,

price levels in domestic and foreign countries are
different P ≠ NER . P*
 Apply the LOOP to the consumption prices of the
baskets of commodities provided that domestic and
foreign consumers have identical preference on those
baskets, so that ratios of domestic and foreign
commodities in the domestic basket of commodities are
the same as those in the foreign basket of commodities.

50

A version of PPP by heterogenous goods


PC  P α  NER  P* 
1 α

α
 P 
PC *     P 
* 1 α

 NER 
PC
 NER   P C  NER  PC *
PC *

where a, 1- a: weight of domestic and foreign commodities

PC, PC*: domestic and foreign consumption price
measured in domestic and foreign currency

51

17
 4/18/2018

A generalized version of PPP

- a distinction between traded and non-trade goods-
• Make a distinction between traded and non-trade goods (non-
tradable goods are those that cannot be traded internationally at a
profit, such as houses and certain services such as a haircut, or
restaurant food): PN, P*N  PC = Pa T . P1-a N
• Tradable goods of the two countries are homogenous PT, P*T
where PT = NER . P*T  PC* = (P*Ta . (P*N1-a

RER  NER 
   
α
P C* PT PT*  PN*
 
1 α

P C PT* PTα  PN1α

1 α
 P* P* 
 RER   N T 
 PN PT 
weights (a and 1- a) indicate the (%) importance of tradable (T) and non-
tradable (N) goods’ prices in the economies price level.
In logarithm: q = (1- a) . (pN*- pT*) - (1- a) . (pN - pT)
RER = 1 if the relationship between traded and non-traded goods in 52
domestic and foreign country is the same.

Productivity Differentials in the Traded Goods

Sector: Balassa-Samuelson Effects

• Balassa and Samuelson argue that labour productivity in

rich countries is higher than that in poor countries.
• This productivity differential occurs predominately in the
tradables rather than non-tradables sector.
• Assumptions:
– Wages are the same in the tradables and non-tradables sector
within each economy and positively related to productivity
– Prices are determined positively by wages and inversely by
productivity.

53

Productivity Differentials in the Traded Goods

Sector: Balassa-Samuelson Effects

Relationships:
• In the poor economy: PN = W N/QN and PT = W T/QT
• In the rich economy: P*N = W*N/Q*N and P*T = W*T/Q*T
• Wages are the same in the tradables and non-tradables
sector: W N = W T and W*N = W*T
• Productivity is higher in rich country’s tradables sector
than that in poor country, but the same in nontradables
sector: Q*T > QT and Q*N = QN
• The price ratios of traded goods to non-traded goods in
each country: PN/PT = s and P*N/P*T = s*
• PPP holds for the tradables sector: S.P*T = PT

18
 4/18/2018

Productivity Differentials in the Traded Goods

Sector: Balassa-Samuelson Effects
Results:
• P*N/P*T =! PN/PT
 (W*N/Q*N).(Q*T/W*T) =! (W N/QN).(QT/WT)
 Q*T/Q*N =! QT/QN .
Since Q*T > QT and Q*N = QN
 Q*T/Q*N > QT/QN
 P*N/P*T > PN/PT
 s* > s : Relative price of nontrables to tradables will be
higher in rich country.
• From P*N/P*T > PN/PT
 S.P*N/SP*T > PN/PT .
Since S.P*T = PT
 S.P*N > PN : PPP doesn’t work for non-traded goods.
55

Productivity Differentials in the Traded Goods

Sector: Balassa-Samuelson Effects

• PC*/ PC = [(P*T a .(P*N1-a ] / [PaT . P1-aN]

 PC*/ PC = (P*T /PT )a . (P*N /PN )(1-a)
Since S.P*T = PT and SP*N > PN
 (P*T /PT )a . (P*N /PN )(1-a > (P*T /S.P*T )a . (P*N /S.P*N )(1-a
 (P*T /PT )a . (P*N /PN )(1-a > (1/S)a . (1/S )(1-a
 (P*T /PT )a . (P*N /PN )(1-a > 1
 PC*/ PC > 1
 PC*> PC : Prices are on average higher in rich countries
than in poor countries

56

Balassa-Samuelson Effects and the RER

q = (1- α)(pN*- pT*) - (1- α)(pN - pT)
<<< B
1. A will be smaller than B, thereby causing the real exchange
rate to appreciate.
2. In other words, the real exchange rate will appreciate if
productivity in the tradables sector is higher in the domestic
country than in the foreign country.
Exp: Since WWII, Japan has had higher productivity in its
tradables sector than the US, which leads to a fall in
Japanese traded goods prices relative to US traded goods
prices. To maintain the PPP for tradables goods the yen
should appreciate against the dollar in real terms (by 9% in
the period of 1973-1983, so-called Japanese Economic
57
Miracle, see Chung et al., 2004).

19
 4/18/2018

2. Interest rate parity

• Provides the link between interest rates and exchange

rates between two countries:
 Interest arbitrage (covered and uncovered)
 Interest rate parity (covered and uncovered)

58

An Introduction
• We address the issue of the FOREX market’s efficiency.
• Prices should fully reflect information available to market
participants.
• It follows that in an efficient market (equilibrium)
international investors will have no desire to switch
investments from domestic to foreign currency
denominated assets, or vice-versa.
• Efficient market hypothesis can be reduced to the joint
hypothesis that:
– FOREX market participants are risk neutral
– FOREX participants (in an aggregate sense) act as if they are
endowed with rational expectations (adjusted for risk and formed
rationally)

59

An Introduction

• Most often, FOREX market efficiency discussions have

taken place in the context of the link between Spot and
Forward FOREX exchange rates?
• Forward Rate also reflects “market” expectations
(adjusted for risk and formed rationally) of the level of the
future spot exchange rate which will prevail at the time
the forward contract matures:
• What is the basis for these claims of the spot-forward
exchange rate relationship? Arbitrage strategy

60

20
 4/18/2018

An Introduction

Arbitrage strategy
• Trading strategies involving no market risk (all prices
are locked-in) at the time the strategy is initiated.
– Buy low, sell high
– Transactions executed simultaneously
• These strategies also involve no initial investment (any
funds required can be borrowed at known rates, so again
no market risk)
• May involve other risks (counterparty credit risk)
• An Efficient Market implies: no profitable arbitrage
opportunities exist
61

Interest Arbitrage
Hypothesis
• Domestic and foreign financial assets are homogenous
(perfect substitution)
• Costless arbitrage (no transaction cost, no capital barriers,
no country risk)
• Perfect competitive market
• Arbitrageurs act in the market and their activities push
market back to parity conditions. Arbitrageurs are risk-
neutral. If they are risk-averse, the risk premium  should
be added to the formula i  i*  ENER  

62

Interest arbitrage
• Arbitrage can be defined as the act of simulteneously
buying and selling the same or equivalent assets or
commodities for the purpose of making certain,
guaranteed profits.
• Arbitrage is a process through which an investor can buy
an asset or combination of assets at one price and
occurently sell at a higher price, thereby earning a profit
without investing any money or being exposed to any
risk.
• As long as there are profitable arbitrage opportunities,
the market cannot be in equilibrium. The market can be
said in equilibrium when no profitable arbitrage
opportunities exist.
• Interest arbitrage is an operation that aims to benifit from
the short-term employment of liquid funds in the financial
centre where the yield is highest.
63

21
 4/18/2018

Interest arbitrage

• Covered investment or borrowing involves two foreign

exchange transaction costs, one on the spot market, the
other on the forward market.
• Covered interest arbitrage invoves the coverage of
economic agents (not speculators) against exchange risk
by having recourse to the forward exchange market.

forms: covered and uncovered

64

Covered Interest Arbitrage

Eurodollar rate = 8.00 % per annum

Start End
$1,000,000 x 1.04 $1,040,000 Arbitrage
$1,044,638 Potential
Dollar money market
gain interest = 4,638 => 4,638/1000 x 100% = interest rate = 9%

S =¥ 106.00/$ 180 days F 180 = ¥ 103.50/$

Yen money market

¥ 106,000,000 x 1.02 ¥ 108,120,000

Euroyen rate = 4.00 % per annum

65
Cover: phòng v ri ro

Uncovered Interest Arbitrage (UIA):

The Yen Carry Trade
Investors borrow yen at 0.40% per annum
Start End
vay yên nht vs lãi sut thp -> u t vào ng tin nc khác có lãi sut cao hn
¥ 10,000,000 x 1.004 ¥ 10,040,000 Repay
¥ 10,500,000 Earn
Japanese yen money market ¥ 460,000 Profit

S =¥ 120.00/$ 360 days S360 = ¥ 120.00/$

US dollar money market

$ 83,333.33 x 1.05 $ 87,500.00

Invest dollars at 5.00% per annum

In the yen carry trade, the investor borrows Japanese yen at relatively low interest rates, converts the proceeds to another currency
such as the U.S. dollar where the funds are invested at a higher interest rate for a term. At the end of the period, the investor
exchanges the dollars back to yen to repay the loan, pocketing the difference as arbitrage profit. If the spot rate at the end of the
period is roughly the same as at the start, or the yen has fallen in value against the dollar, the investor profits. If, however, the yen
were to appreciate versus the dollar over the period, the investment may result in significant loss.
66

22
 4/18/2018

Determining the currency of investment

1 i ?
JPY0 JPYn 1 + i   F  1 S  1  i 
*

1 + i   F  1 S   1  i * :
1S F JPY deposit choosing
1 + i   F  1 S  1  i *  :
USD0 USDn USD deposit choosing
1  i*

67

Determining the currency

in which to borrow

1 i ?
JPY0 JPYn 1 + i   F  1 S   1  i 
*

1 + i   F  1 S   1  i*  :
1S F
USD borrowing
1 + i   F  1 S  1  i* :
USD0 USDn
1  i* JPY borrowing

68

Borrowing and investing

for arbitrage profit
1 i ?
JPY0 JPYn 1 + i   F  1 S  1  i 
*

1 + i   F  1 S   1  i  :
*

1S F
JPY investing, USD borrowing
1 + i   F  1 S   1  i  :
*

USD0 USDn
1  i*
JPY borrowing, USD investing

69

23
 4/18/2018

Interest rate parity

• Interest rate parity (IRP) is an arbitrage condition that
must hold when international financial markets are in
equilibrium (provides the linkage between the foreign
exchange markets and the international money markets).
• IRP theory states the difference in the national interest
rates for securities of similar risk and maturity should be
equal to, but opposite in sign to, the forward discount or
premium for the foreign currency, except for transaction
costs.
• IRP derives with the assumptions that there are no
transaction costs, political risks of investing abroad, and
taxes.
70

Interest Rate Parity

i $ = 8.00 % per annum
(2.00 % per 90 days)
Start End
$1,000,000 x 1.02 $1,020,000
$1,019,993*
Dollar money market

S = SF 1.4800/$ 90 days F 90 = SF 1.4655/$

Swiss franc money market

SF 1,480,000 x 1.01 SF 1,494,800

i SF= 4.00 % per annum

(1.00 % per 90 days)
•Note that the Swiss franc investment yields $1,019,993, $7 less on a $1 million investment.
71

Interest rate parity

ES: expected spot rate
1 + i  = F  1 S   1  i *  : IRP F: forward exchange rate

F  S i  i* F S
S

1  i*
 i  i* 
S
 1  i*  
i  i *  f  s : reduced form

i  i *  F  S  S : CIP :
interest differenti al equals the forward margin

i  i *  ( ES  S ) S : UIP
72

24
 4/18/2018

Interest rate parity

KX  KX ( RR )
KM  KM ( RR)

 i*  ENER 
NK  NK ( RR )  NK   
 i 
Where  : RR elasticity of NK ( > 0).
The effects of RR on NK depends on:
•  :  ↑→NK↑
• The value of  :
 If domestic and foreign financial assets are homogenous or perfectly
substitutable:  high: RR↑→NK↑ (high): IRP
 If domestic and foreign financial assets are heterogenous or imperfectly
substitutable :  small: RR↑→NK↑ (small)

73

Uncover interest rate parity

    RR  NK 1   1
i *  ENER
RR   i  i*  ENER  RR  1
i

74

Covered interest rate parity

it  it*  NERtT,t 1  NERt

hay
i  i*  f  s

75

25
 4/18/2018

Speculating on Anticipated Exchange Rates

Chicago Bank expects the exchange rate of the New Zealand dollar to
appreciate from its present level of $0.50 to $0.52 in 30 days.

Borrows at 7.20% for 30

days
1. Borrows $20 4. Holds
million $20,912,320
Returns $20,120,000
Profit of $792,320
Exchange at Exchange at
$0.50/NZ$ $0.52/NZ$

Lends at 6.48% for 30

days
2. Holds NZ$40 3. Receives
million NZ$40,216,000

Speculating on Anticipated Exchange Rates

Chicago Bank expects the exchange rate of the New Zealand dollar to
depreciate from its present level of $0.50 to $0.48 in 30 days.

Borrows at 6.96% for 30

days
1. Borrows NZ$40 4. Holds
million NZ$41,900,000
Returns NZ$40,232,000
Profit of NZ$1,668,000
or $800,640
Exchange at Exchange at
$0.50/NZ$ $0.48/NZ$

Lends at 6.72% for 30

days
2. Holds $20 3. Receives
million $20,112,000

Example

Suppose that the annual interest rate is 5 percent in the

United States and 8 percent in the U.K., and that the
spot exchange rate is $1.50/£ and the forward exchange
rate, with one-year maturity, is $1.48/£. In terms of our
notation, i$ = 5%, i£ = 8%, S = $1.50, and F = $1.48.
Assume that the arbitrager can borrow up to $1,000,000
or £666,667, which is equivalent to $1,000,000 at the
current spot exchange rate. Check if IRP is holding
under current market conditions?

78

26
 4/18/2018

Interest Rate Parity and Equilibrium

• The following exhibit illustrates the conditions necessary

for equilibrium between interest rates and exchange
rates.
• The disequilibrium situation, denoted by point A, is
located off the interest rate parity line.
• However, the situation represented by point A is
unstable because all investors have an incentive to
execute the same covered interest arbitrage, which is
virtually risk-free.

79

FS
S

 1  i* 
IP

A
0.04

0.03

0.02
B F

0.01

-0.04 -0.03 -0.02 -0.01 0 0.01 0.02 0.03 0.04

i  i*
-0.01

C
• Extra borrowing in JPY will put upward pressure
-0.02 E
on JPY interest rates.
-0.03 • The spot sale of JPY for USD will help bid up
the spot price of USD, that S(JPY/USD) will
-0.04 increase.
D
• The USD that were purchased will be used to
invest in USD securities. This will cause the
price of USD securities to increase, and
therefore cause dollar yields to decrease, that
is, i* will fall.
• The forward sale of USD will lower F(JPY/USD)
80

Forward rate as an unbiased predictor

of the future spot rate
Hypothesis: Market is efficient.
• The forward exchange rate today (Ft,t+1), time t, for
delivery at future time t +1, is used as a predictor of the
spot rate that will exist at that day in the future.
• The forward rate is termed an unbiased predictor of the
future spot rate (UFR),
• Unbiased prediction simply means that the forward rate
will, on average, overestimate and underestimate the
actual future spot rate in equal frequency and degree
(Et(St+1) = Ft,t+1). The sume of the errors equals zero.
• Intuitively this means that the distribution of possible
actual spot rates in the future is centered on the forward
rate. 81

27
 4/18/2018

Exhibit 4.10 Forward Rate as an Unbiased

Predictor for Future Spot Rate
Exchange rate
t1 t2 t3 t4

S2 F2

S1 Error F3
Error

Error
F1 S3

S4

Time
t1 t2 t3 t4
The forward rate available today (Ft,t+1), time t, for delivery at future time t+1, is used as a “predictor” of the
spot rate that will exist at that day in the future. Therefore, the forecast spot rate for time St2 is F 1; the actual spot
rate turns out to be S 2. The vertical distance between the prediction and the actual spot rate is the forecast error.
When the forward rate is termed an “unbiased predictor of the future spot rate,” it means that the forward rate
over or underestimates the future spot rate with relatively equal frequency and amount. It therefore “misses the
mark” in a regular and orderly manner. The sum of the errors equals zero. 82

Expected
change in Parity line
exchange rate
E∆S = E∆NER

-2 Forward premium (+)

hay forward discount (-)
f-s
-2

83

3. International Fisher Effect

Fisher equation
• The Fisher Effect states that nominal interest rates in each
country are equal to the required real rate of return plus
compensation for expected inflation. i = (1 + r).(1+p) - 1
This equation reduces to (in approximate form): i  r p
where i is the nominal interest rate, r is the real interest
rate and p is the inflation rate
• Expectation form: i  r  EP  r  i  EP
• This obtains from (1  i)  (1  r )(1  EΔP)
• Empirical tests (using ex-post) national inflation rates have
shown the Fisher effect usually exists for short-maturity
government securities (treasury bills and notes). 84

28
 4/18/2018

3. International Fisher Effect

IP with ES  S *
i  i*   i  ES  i *  ENER
ES = S.(1+E∆S) S
Expected PPP ENER  EP  EP*
International Fisher
effect (real interest i  i *  EP  EP *
parity condition) i  EP  i *  EP *
The percentage differential in the expected inflation rates
equals the differential between comparable interest rates in
different countries. In equilibrium, real interest rates are
equal in different countries.
85

Exercises

• P7.7, 7.8, 7.11-7.15, 7.17, 7.22, 7.24, 7.25

29