On the insulation properties of flexible exchange rates - article ; n°4 ; vol.30, pg 719-746

Stephen Turnovsky

Description

Revue économique - Année 1979 - Volume 30 - Numéro 4 - Pages 719-746
du taux d'échange flexible crée dans le long terme une protection totale contre la deuxième catégorie de perturbations, en général seulement une protection partielle pourra être réalisée dans le court terme. En fait, le degré de protection variera entre 0 et 100 % et un certain nombre de cas différents sont présentés.
This paper considers the insulation properties of flexible exchange rates in the face of two types of foreign price disturbance : (i) An increase in the (steady) foreign rate of inflation ; (ii) A once-and for-all increase in the foreign price level.
With respect to the former, the general conclusion to emerge is that perfect insulation will resuit in both the short run and steady state if and only if those real foreign variables which impinge on the domestic economy are invariant with respect to the exogenous foreign rate of inflation. Although flexible rates provide complete long-run insulation against the second type of disturbance, in general only partial insulation in the short run will be achieved. Indeed the degree of insulation can vary between 0 % and 100 % and varions cases are discussed.
28 pages
Source : Persée ; Ministère de la jeunesse, de l’éducation nationale et de la recherche, Direction de l’enseignement supérieur, Sous-direction des bibliothèques et de la documentation.

Sujets

• Sociologie, société et politique

Informations

• Publié par: REVUE_ECONOMIQUE
• Publié le: 01 janvier 1979
• Nombre de lectures: 197
• Langue: English
• Poids de l'ouvrage: 1 Mo

Extrait

Monsieur Stephen Turnovsky
On the insulation properties of flexible exchange rates
In: Revue économique. Volume 30, n°4, 1979. pp. 719-746.
Résumé
du taux d'échange flexible crée dans le long terme une protection totale contre la deuxième catégorie de perturbations, en
général seulement une protection partielle pourra être réalisée dans le court terme. En fait, le degré de protection variera entre 0
et 100 % et un certain nombre de cas différents sont présentés.
Abstract
This paper considers the insulation properties of flexible exchange rates in the face of two types of foreign price disturbance : (i)
An increase in the (steady) foreign rate of inflation ; (ii) A once-and for-all increase in the foreign price level.
With respect to the former, the general conclusion to emerge is that perfect insulation will resuit in both the short run and steady
state if and only if those real foreign variables which impinge on the domestic economy are invariant with respect to the
exogenous foreign rate of inflation. Although flexible rates provide complete long-run insulation against the second type of
disturbance, in general only partial insulation in the short run will be achieved. Indeed the degree of insulation can vary between 0
% and 100 % and varions cases are discussed.
Citer ce document / Cite this document :
Turnovsky Stephen. On the insulation properties of flexible exchange rates. In: Revue économique. Volume 30, n°4, 1979. pp.
719-746.
http://www.persee.fr/web/revues/home/prescript/article/reco_0035-2764_1979_num_30_4_408482THE INSULATION PROPERTIES ON
OF FLEXIBLE EXCHANGE RATES*
I. INTRODUCTION
As a result of the general world-wide move towards more flexible
exchange rates? the question of the insulation they provide from
inflationary pressures abroad has been receiving increased attention
by economists. One of the first formal analyses if this issue was by
Turnovsky and Kaspura [1974] who, using a simple IS-LM model,
concluded that a flexible exchange rate regime will provide complete
insulation against a foreign price disturbance. As they noted, this
conclusion followed directly from the fact that in their model the
foreign price level and the exchange rate always appear together
multiplicatively, with the result that any increase in the former is
offset by a corresponding reduction in the latter, leaving the rest of
the system unaffected. It is clear that any modification to the model
which destroys this symmetry between these two variables will cause
the the breakdown of these perfect insulation properties and a recent
comment by Casas [1977] has provided one simple example.
As discussed by Turnovsky ([1976, 1977], p. 236) the perfect insu
lation provided by flexible exchange rates in the Turnovsky-Kaspura
model is a consequence of the incomplete specification of the finan
cial sector and capital flow relationships. Once these are properly
specified, flexible rates need not provide complete insulation for
the domestic economy from foreign inflationary disturbances.
* An earlier version of this paper was presented at the Western Economic
Association meetings, Honolulu, Hawaii, June 1978. The paper was revised and
extended white the author was Visiting Professor at the Université de Paris IX
and Visiting Research Associate at CEPREMAP.
719
Revue économique — N" 4, juillet 1979. Revue économique
In contrast to the static flow approach underlying the IS-LM
analysis, recent advances in international macro economic theory have
moved towards the development of consistently specified portfolio
models, which explicitly incorporate the dynamic interaction between
stocks and flows in the economy. These papers are typically concerned
with investigating a number of issues, such as the effects of convent
ional monetary and fiscal policies, with the question of insulation
being given somewhat secondary treatment. Yet this aspect is an
important one, especially for a small country, and certainly merits
more detailed discussion. Accordingly, in this paper we address our
selves to the insulation question within the framework of a consis
tently specified dynamic model. We should make it clear that el
ements of our results have been obtained elsewhere in the literature.
Our objective here is to bring them together in an integrated fashion,
in the process giving a more unified analysis.
The question of the degree of insulation provided by flexible
exchange rates has several important dimensions to it. In the first
place, one should distinguish clearly between the insulation provided
against :
(i) Increases in the (steady) foreign rate of inflation ;
(ii) once-and-for-all increases in the foreign price level.
A will become clear from our subsequent analysis, the conditions
required for insulation against these two types of price disturbance
are not the same, and both will be studied in some detail.
Secondly, there is the question of the time horizon under consider
ation. The static literature focuses on the short run (one period),
whereas in a dynamic context both the short run and the long run
(steady state) are obviously of importance. This point is stressed
by Laidler (1977) who finds that for his model flexible rates provide
perfect insulation in the long run but not in the short run.
Thirdly and most importantly, although the foreign rate of inflation
(or price level) may be viewed as exogenous to a small economy, it
is an endogenous variable insofar as the rest of the world is concer
ned. As such, it is determined along with other endogenous foreign
variables such as the foreign nominal interest rate, level of income,
etc. in terms of various policy instruments and structural pa
rameters. Any change in the foreign inflation rate is thus a conse
quence of changes in these policy and parameters
and hence is likely to be accompanied by changes in some of these
720 Stephen J. Turnovsky
other endogenous foreign variables. Thus ideally, the insulation pro
perties of flexible rates should be analyzed in terms of the insulation
from the effects of the ultimate foreign policy variables, but this
would require a full two country analysis and would become extre
mely cumbersome. We shall therefore continue to treat the foreign
rate of inflation as being completely exogenous. This can be most
conveniently interpreted as referring to a situation where the foreign
monetary authorities are following a policy of steady expansion in
their nominal money supply, leading to an equal rate of infla
tion abroad. In this case any increase in the foreign rate of inflation
resulting from an expansionary monetary policy will typically be
accompanied by changes in other foreign variables, most notably the
foreign nominal interest rate. Indeed as well become evident below,
the insulation properties of flexible exchange rates to changes in this
form of (foreign) monetary policy depend critically upon the real
variables in the rest of the world being neutral with respect to the
growth in the foreign nominal money supply.
In Sections 2,3 we consider the insulation against the first type
of price disturbance in a world of secular inflation. We begin by
considering the case where capital is perfectly mobile internationally.
Not only is this becoming increasingly relevant empirically, but also
it is analytically most tractable. We then discuss a number of modif
ications to this basic model, and consider the extent to which the condi
tions for insulation, obtained for the simple model, continue to apply.
The insulation against a once-and-for-all increase in the foreign price
level is analyzed in Section 4. The concluding section summarizes
our main conclusions and indicates some directions in which the
analysis might by extended.
II. INSULATION UNDER PERFECT
CAPITAL MOBILITY
We shall assume that the country produces a single commodity,
part of which is consumed domestically, the remainder of which is
exported. The price of this commodity is determined in the market
for domestic output, so that the price of exports is endogenously deter
mined. On the other hand, we assume that the country is sufficiently
competitive in the market for its imports to take the price of imports
721
10 Revue économique
as given. While this is a limiting assumption, it is a fairly standard
one in the literature. For expositional simplicity we assume no domest
ic output of the imported good and no imported inputs. We are
therefore focusing our attention on the insulation provided against
disturbances in the price of foreign output.
The model underlying our analysis contains the following rela
tionships.
(la) Z = H (Y, r — tz, a, W) + X (a) + G domestic output sector
X' > 0 0 < Pi Hi/P < 1, Ha < 0, H3 > 0, H4 > 0,
l(QE)1-5 P = = P!: 0 < 5 < 1 (2a) (p (Pi, QE)
P = 5 pi + (1 - + (2b) -5) (q e)
cost-of-living a = QE/Pi (2c)
= ai-8 = (P (a) = P/Pi (2d)
a (Z - Z) Pi = + 7t domestic price equation (3)
M = L(PiZ/P, r — 71, — TC, W) (4a)
Li > 0, L2 < 0, L3 > 0, 0 < L4 < 1
-^-= J(PiZ/P,r-7C,-7C,W) (4b) financial
secto