Page 183 - DECO503_INTERNATIONAL_TRADE_AND_FINANCE_ENGLISH

Unit 15 : Theories of Determination of Exchange Rate (PPP, Monetary)

Monetary models of exchange rate determination have been criticized because of the inadequacy of Notes
the assumptions used to derive equation (1). In particular, the assumption of the purchasing power
parity has been criticized as not consistent with the facts, especially the facts of the 1970s. The collapse
of purchasing power parity in the 1970s, however, is not (in my judgment) adequate reason for
rejecting equation (1) as a model (albeit an incomplete model) of exchange rate determination. This
equation can be derived without explicit reference to purchasing power parity; indeed, it can be
derived from a model that allows explicity for divergences from purchasing power parity. Moreover,
some empirical studies employing equation (1) have noted that there are divergences from purchasing
power parity and have argued that the conditions of money market equilibrium are more immediately
relevant for determining the exchange rate (which is a freely adjusting asset price) than they are for
determining national price levels. This, of course, leaves open the important question of what
determines the behavior of national price levels, which in turn is an important element in explaining
the behavior of real exchange rates. Nevertheless, if equation (1) worked well in explaining the behavior
of nominal exchange rates, this form of monetary model of exchange rate determination would clearly
make a sub-stantial contribution to our understanding of the economic forces influencing the behavior
of exchange rates.
The principal empirical difficulty with this form of monetary model is that equation (1) does not
work well in explaining actual movements in nominal exchange rates, unless we take into account
shifts in the demands to hold different national monies that are difficult to explain in terms of traditional
arguments appearing in money demand functions. An example illustrates this difficulty as well as a
set of regressions. Between October 1976 and October 1980, the British pound appreciated by 50% in
terms of the United States dollar, from $1.60 to $2.40. During this same period, monetary aggregates
in Britain grew more rapidly than corresponding monetary aggregates in the United States, while
real income (a key variable affecting the demand for money) grew less rapidly in Britain than in the
United States. Of course, the increase in dollar value of sterling might be explained by an increase in
the demand to hold sterling combined with a decrease in the demand to hold dollars, resulting from
increased confidence in the future value of sterling (due to North Sea oil and the policies of Prime
Minister Thatcher) and from increased concern about the inflationary consequences of the policies of
the Carter administration. However, it is difficult to take these effects into account in a rigorous and
disciplined fashion in an empirical version of equation (1).
Another important deficiency of equation (1) as a model of exchange rate determination is that it
does not explicitly reveal the critical role of expectations of future economic conditions in determining
the current exchange rate. From equation (1), there is no immediately apparent reason why changes
in exchange rates should be largely random and unpredictable, or why new information that alters
expectations about future economic conditions (including supplies and demands for national monies)
should induce such random and unpredictable changes in exchange rates.
The second general class of monetary models of exchange rate determination does not suffer from
this deficiency. These models usually treat a small or moderate size economy that takes conditions in
the rest of the world as given. The critical condition determining the exchange rate for this country is
the requirement of money market equilibrium;

m = k ζ+⋅ e η ⋅ D e () e , ζ , η > 0 , (2)
−
where m is the logarithm of the domestic money supply, e is the logarithm of the price of foreign
money in terms of domestic money, k summarizes all exogenous factors affecting the logarithm of the
e
()
demand for domestic money, and D e = E (e (t + 1); t) – e (t) is the expected rate of change of the
exchange rate. Equation (2) should be thought of as a reduced-form equilibrium condition derived
from a more basic model of goods and asset market equilibrium. In this reduced form, the parameter
ζ captures all of the mechanisms through which an increase in the price of foreign money increases
the demand for domestic money, and the parameter η captures all of the mechanisms through which
an increase in the expected rate of change of the price of foreign money affects the demand for domestic
money.

LOVELY PROFESSIONAL UNIVERSITY 177