We next turn to the key issue of the current debate over the process
of international adjustment: the role of real exchange rates in the adjustment
process. In the standard view fiscal imbalances work through the real exchange
rate: A budget deficit leads to a real appreciation, which reduces the
competitiveness of a country's industry and thus leads to a trade deficit.
American critics of the conventional wisdom, however, have argued that
no real exchange rate change is necessary and that a shift in savings rates
can change the trade balance at constant relative prices. European and
Japanese commentators often go further, seeming to argue that deficit
correction is an alternative to real depreciation and that the deficit needs to
fall in order to keep the dollar from declining further. Thus in a recent article
Wakasugi (1987) writes:
The fundamentalcausesof the dollar'sdepreciationare the U.S.budget deficitand
an unfavorablebalance of payments which shows no sign of improving. Only
the U.S. itself can recover the dollar's status as an intemational key currency.
Therefore,in the long run, decreasingthe budget deficitand enhancingproductivity
are vital steps.
The fact that the United States advocates of the view that real depreciation
is unnecessary are more or less monetarist in their views on macroeconomic
policy, and that their scepticism of the need for real depreciation
is tied to a denial of real effects of nominal depreciation, makes it seem to
casual observers that this dispute is yet another monetarist-Keynesian
argument that hinges on the issue of price flexibility. However, this is a
misperception. This is a replay of an old debate, but it is Keynes versus
Ohlin, not Tobin versus Friedman. It is the old question of the relative price
effects of an international transfer of resources.
To see the nature of the issue, it is useful to consider a rudimentary
model that reveals the conditions under which a real depreciation is or
is not necessary as part of current account adjustment. (A more formal
treatment of this model is given in appendix A at the end of this chapter.)
We can then examine the empirical evidence that bears on the question.
Redistributing Expenditure and the Real Exchange Rate
Suppose that the world consists of only two countries, US and ROW.
US is assumed initially to be running an undesirable current account deficit.
We initially suppose that each of the countries produces only a single good,
so that the real exchange rate may be defined as the price of the US
good relative to the ROW good. Finally, suppose for the sake of argument
that both countries are initially at full employment so that a balanceof-
payments adjustment cannot involve an expansion in either country's output.
Now let us try to reduce US's current account deficit. Can we do
this at a constant real exchange rate? It is useful here to write the balanceof-
payments identity in its alternate form
X-M= Y-E;
that is, the external balance is the difference between income and expenditure.
Since the real exchange rate is being held constant, we can measure
the terms in this equation in terms of either good. More conveniently stilL
it does no harm to suppose that nominal prices are held constant, so we can
simply measure income and expenditure in nominal terms.
The first point to notice is that there is no channel that links the
budget deficit to the trade deficit other than through its effect on expenditure.
A shift in fiscal policy reduces US's expenditure and raises ROW's
expenditure, and that is all. There is no direct way in which it makes the
US good more competitive. The issue then is whether it is possible to
reduce US's expenditure and raise ROW's expenditure, while keeping the
relative price of the US and the ROW's goods constant.
Suppose that through fiscal contraction US reduces its expenditure by
$100 billion, while ROW increases its expenditure by the same amount. The
fall in US's expenditure will directly reduce spending on the US good by
100(1 - m) billion dollars, where m is the fraction of a marginal dollar of
US's spending that is spent on imports. On the other hand, the rise in
ROW's spending will raise spending on the US good by IOOm~billion
dollars, where m~ is the fraction of a marginal dollar spending that falls
on imports. The net change on spending on the US 'food is therefore
IOO(m+ m~ - 1) billion dollars. Ifm + m~ < I-which we will see below
is certainly the case in practice-then the redistribution of world expenditure
will reduce the demand for the US good and increase the demand for
the ROW good. To correct the excess supply of the US good and the
excess demand for the ROW good, the relative price of the US good must
fall:The correction of the current account deficit must be effected via a real
depreciation.
The key criterion here is a familiar one: It is the criterion for a terms-oftrade
effect of a transfer. A redistribution of world expenditure must be
accompanied by a change in relative prices unless the marginal spending
pattern of the countries increasing their expenditure is the same as that
of the country reducing its spending. If the United States has a higher
marginal propensity to spending on its own goods than other countries do,
which is the case where m + m~ < 1, then a fall in the U.S. share of world
spending must be accompanied by a fall in the U.S. real exchange rate.
It is important to avoid two confusions that can obscure this point.
First is the idea that the issue is somehow tied to the degree of capital
mobility. McKinnon (1984) has argued strongly that the real exchange
rate needs to change to adjust the trade balance only when an economy
is "insular," that is, closed to capital movement. He argues that when
capital is mobile, savings-investment gaps are directly reflected in trade
balances, with no need for relative price changes. 'With smoothly functioning
capital markets, little or no change in the 'real' exchange rate is necessary
to transfer saving from one country to another" (1984, 14).
What is wrong with this argument should be immediately clear. It
confuses the question of whether a change in the savings rate will be
reflected in a change in the distribution of world expenditure with the
question of whether a change in that distribution necessitates a change
in relative prices. The latter question is a question about goods markets,
not capital markets. No matter how mobile capital may be, if Japanese
residents spend much less on U.S. goods at the margin than do U.S.
residents, a redistribution of world spending from the United States to
Japan will reduce the demand for U.S. goods at constant relative prices.
The other confusion that can obscure the issue is to mix up the necessity
for a change in relative prices with the question of whether changes in
nominal exchange rates help produce such changes. If prices are flexible, a
currency depreciation by itself has no relative price effects, and a real
depreciation can be achieved with a constant nominal exchange rate via
deflation in one country and/or inflation in the others. However, this has
nothing to do with the question of whether the real exchange rate needs
to change.
Read More:
Do Real Exchange Rates Have to Change?
