chapter 6 (2) Flashcards
When PPP is violated the real exchange rate will
Deviate from unity.
Suppose for example the annual inflation rate is 5 percent in the united states and 3.5 percent in the UK and the pound appreciated against the dollar by 4.5 percent then the real exchange rate is q = 0.97
q = 1
q > 1
q < 1
q = 1: competitiveness of the domestic country is unaltered
q > 1: Competitiveness of the domestic country deteriorates
q < 1: Competitiveness of the domestic country improves
nontradables
commodities that never enter into international trade
e.g. haircuts, housing etc
Fisher effects
Fisher effect holds that an increase in the expected inflation rate in a country will cause a proportionate increase in the interest rate in the country
Fisher effect can be written for the U>S as follows
i$ = P$ + E(pi$) + P$*E(pi$) == P$ + E(pi$)
P$ = the equilibrium expected real interest rate int the united states
The Fisher effect implies that the expected inflation rate is the difference between the nominal and real interest rates in each country, that is
E(pi$) = (i$ - P$)/(1+P$) == i$-P$
E(Ppound) = (ipound -Ppound)/(1 + Ppound) == ipound - Ppound
International Fisher effect
Suggests that the nominal interest rate differential reflects the expected change in exchange rate. For instance if the interest rate is 5 percent per year in the united states and 7 percent per year in the UK the dollar is expected to appreciate against the british pound by about 5 percent per year
When the international fisher effect is combined with IRP that is
(F-S)/S = (i$ - ipound)/(1 + ipound) we obtain
(F - S)/S = E(e)
Forward expectation parity
(F-S)/S = E(e)
States that any forward premium or discount is equal to the expected change in the exchange rate. When investors are risk neutral, forward parity will hold as long as the foreign exchange market is informationally efficient
Most forecasting techniques can be classified into three distinct approaches
Efficient market approach
Fundamental approach
Technical approach
Efficient market hypothesis (EMH)
Suppose that foreign exchange markets are efficient. This means that the current exchange rate has already relected all relevant information, such as money supplies, inflation rates, trade balances and output growth. The exchange rate will then change only when the market receives new information. in a word, incremental changes in the exchange rate will be independent of the past history of the exchange rate. If the exchange rate indeed follows a random walk, the future exchange rate is expected to be the same as the current exchange rate, that is,
> Sl = E(St+1)
Random walk hypothesis
suggests that todays exchange rate is the best predictor of tomorrows exchange rate
The fundamental approach to exchange rate forcasting
Uses various models for example ethe onetary approach to exchange rate determination. Suggests that the exchange rate is determined by three independent explanatory variables
1 Relative money supplies
2 relative velocity of monies
3 relative ational outputs
One can formulate the omnetary approach in the following empirical form
s = alpha + beta1(m - m) + beta2(V-v) + beta3(y*-y) + u
Where
s = naturla logarithm of the spot exchange rate
m - m* = natural logarithm of domestic/foregin money supply
v-v* = natural logarithm of domestic/foregin velocity of money
y*-y = natural logarithm of foreign/domestic output
u = random error term, with mean zero
alpha, beta = model parameters
Generating forecasts using the fundamental approach would involve three steps
1) estimation of the structural model to determine the numerical values for the parameters such as alpha and beta
2) estimation of future values of the independent variables like (M - m) (v-v) and (y*-y»)
3) substituting the estimated values of the independent variables into the estiated structural model to generate the exchagne rate forecasts
If for example the forecaster would like to predict the exchange rate one year into the future, he or she has to estimate the values that the independent variable will assume in one year. These values will then be substituted in the structural model that was fitted to historical data