Economic Analysis Series No.97
The Insulation and Transmission Mechanisms of
Floating Exchange Rates Analyzed by the
EPA World Econometric Model*

March, 1985
(Deputy Director General)
Masaru Yoshitomi
(North American Unit)
Yasuko Niimura, Katsuki Oda, Masaru Ishio, Mamoru Ohbayashi
(European Unit)
Masataka Hirano, Hruko Hirose, Norikazu Yasuhara, Motozumi Miwa
(Pacific Unit)
Takuma Nishida, Noriki Hirose, Tetsushi Ueda, Junichi Kawanishi
(Trade-Linkage Model)
Masahiro Matsuda
(System Management Unit)
Seiichi Itoh, Takahiro Sasaki, Keiichi Yahagi

( Part I )

I. Main Purpose of the Paper and the Method of Analysis

The main purpose of this paper is to analyze by the use of the EPA World Econometric** Model the insulation and transmission mechanisms of macroeconomic disturbances under flexible exchange rates. An issue of central importance in this paper is the extent to which monetary and fiscal policies adopted by one government generate and transmit disturbances to the rest of the world. Another focus of concern is the extent to which floating exchange rates insulates an economy from disturbances in the rest of the world, including petroleum price increases.

It has been argued by advocates of flexible exchange rates that the flexible system would insulate the domestic economy from fluctuation in the world economy ("insulation" effects), while it would free monetary policy for using domestic stabilization (policy autonomy), thanks to the function of flexible rates for balancing international payments ("external balancing" function). Furthermore, it has been argued that there would be no need for policy coordination under flexible exchange rates since national policies would have only national effects (i.e., "bottling-up"effects).(Please see kouri 1982.)

In sharp contrast to these arguments in favour of floating exchange rates, however, the EPA World Econometric Model shows many channels through which external disturbances are transmitted ("transmission" effects) despite or because of the flexibility of exchange rates .

In order to empirically identify possible channels of international transmission mechanism of economic disturbances in the present world economy, we need a multi-country model with exchange rates endogenized. The EPA World Econometric Model has been constructed exactly for this purpose. It is a quarterly macroeconometric model, linking nine individual country models("Summit" seven plus Australia and Korea) and six regional models via an international trade linkage submodel. Exchange rates of major currencies are endogenized by the structural balance of payments approach to the determination of the exchange rate.

Using the revised EPA model of February 1984 (please see Appendix Table 1 for in-sample dynamic simulation tests), we have performed "full-link" simulation exercises of the ordinary dynamic multiplier analysis to identify insulation and transmission channels. Each model was first run within a sample period (from the first quarter of 1976 up to the last quarter of 1981) to produce the standard solution. Then each model was shocked by changing one exogenous variable (a policy variable or the oil price)to produce deviations from the standard solution, that is, multipliers for a change in an exogenous variable.

This paper examines the abovementioned insulation effects and transmission mechanisms under floating exchange rates. Our main interest concerns (1) bottling-up effect, (2) external balancing effects, (3) insulation effects, and (4) transmission effects. Our analysis of these effects will have important but complicated implications for domestic macroeconomic policies or the alleged policy autonomy as well as for international macroeconomic coordination. In this paper, however, such policy issues will not be discussed explicitly.

This paper is composed of Parts I and II. Part I contains a background description of the EPA World Econometric Model, focusing on main differences in assumptions between a simple model of perfect insulation and the Mundell-Fleming model on the one hand and the EPA model on the other. (For a more general description of the EAP model, please see Amano 1982.) In order to provide a somewhat concrete understandeing of the major differences in assumptions between the theories and the EPA model, we will present several simulation exercises in Part I. They include (1) the feedbacks of domestic prices, import prices and the exchange rate changes, (2) the J-curve phenomena, (3) the Phillips curves, (4) the IS-LM curves and (5) the external financing of the J-curve. (For more detailed explanations of these and other properties of the EPA model, please refer to our separate report of the EPA World Model Group 1984.)

Part II provides analyses of the multipliers of the EPA World Econometric Model in order to clarify channels for the insulation from and transmission of various shocks under floating exchange rates as compared with those under fixed exchange rates. We will examine simulations for the following three shocks: a monetary policy shock (Section III), a fiscal policy shock (Section IV) and an oil shock (Section V).

The last section provides a very brief concluding remark of the insulation and transmission mechanisms of floating exchange rates analyzed through multiplier exercises performed by using the EPA model.


*  This paper was presented at the EPA International Symposium on "The Mechanisum for Transmission of and Insulation from Economic Disturbances under Floating Exchange Rates - Comparisons between Theories and Multipliers of the EPA World Econometric Model -" , held on March 13 - 15, 1984 in Tokyo

**  The econometric estimations and model simulations were performed by using our own multi-purpose computer system called SIGMA, which was developed by the Economic Research Insutitute of the EPA in cooperation with Fujitsu Limited.


Structure of the whole text

( Part I )

  1. page2
    I.  Main Purpose of thePaper and the Method of Analysis
  2. page5
    II. The Main Differences Theoretical Assumptions between Theories and the EPA Model
    1. page6
      (A) A Simple Model
    2. page9
      (B) Jcurve Phenomena
    3. page16
      (C) Introducing International Capital Movements into a Model
      1. page16
        (i)  A Case for full insulation even under internatinal capital movements
      2. page18
        (ii) The Mundell-Fleming Model
    4. page25
      (D) Ambiguous aspects of the transmission of monetary shock
      1. page25
        (i)  Expansionary or contractionary transmission of tight monetary policy?
      2. page26
        (ii) Transmission of foreign interest rates
      3. page27
        (iii)Effects of sterilization under fixed exchange rates
    5. page28
      (E) Fiscal Shock and the Exchange Rate

( Part II )

  1. page33
    III.Multiplier Analysis of the Insulation and Transmission Mechanisms
    1. page33
      (A) Effects of Contractionary Monetary Policy
      1. page33
        (i)  Multipliers for sustained reduction in the U.S. high-powered money
      2. page39
        (ii) Effects of higher official discount rates
    2. page43
      (B) Effects of sterilization policy under fixed exchange rates - a comparison between Germany and Japan
  2. page48
    IV. Is the Transmission of Foreign Fiscal Policy Disturbance Magnified or Lessened by Floating Exchange Rates?
    1. page48
      (A) General Observations
    2. page51
      (B) US Fisical Policy
    3. page55
      (C) Effects of Non-US Fiscal Policy
      1. page55
        (i)  German fiscal policy
      2. page57
        (ii) Japan's fiscal policy
      3. page57
        (iii)UK fiscal policy
    4. page58
      (D) The International Transmission of Interest Rates
    5. page62
      (E) Brief Summary of the Insulation and Transmission Effects under Expansionary Fiscal Shocks
  3. page63
    V.  Oil Shocks, Exchange Rates and Transmission Mechanisms under Different Exchange Rate Regimes
    1. page63
      (A) Some Theoretical Issues
    2. page65
      (B) The Impact of an Oil Shock under Floating Exchange Rates
      1. page65
        (i)  General observations
      2. page67
        (ii) Nearperfect insulation of Japan's GNP from a real shock
      3. page70
        (iii)The impact of an oil crisis on the world economy under floating rates
      4. page72
        (iv) Implications of oil price hike for US dollar
    3. page73
      (C) What Would Have Happened, If an Oil Shock Had Taken Place under Fixed Exchange Rates
  4. page75
    VI. Conclusions
  5. page78
    Appendix
  6. page79
    References
  • 1-6-1 Nagata-cho, Chiyoda-ku, Tokyo 100-8914, Japan.
    Tel: +81-3-5253-2111