The Role for Long‐run Target Values of the Exchange Rate in the Bank of Japan's Policy Reaction Function

• DOI: http://doi.org/10.1111/twec.12448
• Author: Michael Kühl,Joscha Beckmann
• Date: 01 September 2017
• Published date: 01 September 2017

The Role for Long-run Target Values of

the Exchange Rate in the Bank of Japan’s

Policy Reaction Function

Joscha Beckmann

1

and Michael K€

uhl

2

1

Kiel Institute for the World Economy, University of Bochum, University of Duisburg-Essen, Essen,

Germany and

2

Economics, Deutsche Bundesbank, Frankfurt, Germany

1. INTRODUCTION

IN the recent past, fears that some of the world’s major economies would conduct a more

active exchange rate policy have been at the core of an extensive discussion. Through the

global financial crisis and the consequences of the Great Recession, some countries have still

been struggling to stabilise their domestic economies because they are faced with high rates

of unemployment. An unsound fiscal policy in the past has contributed by confining the scope

for current policy actions due to high public indebtedness. In addition, the unconventional

monetary policy measures have fuelled fears of a ‘beggar my neighbour’ policy in which, at

least, a weakening in the external value of the currency is considered as one element of fos-

tering that country’s external competitiveness. Both the G7 and G20 summits in February

2013 rejected the use of exchange rate policy to enhance competitiveness. We contribute to

this discussion by investigating the role of empirical (long-run) fundamentals in the foreign

exchange market intervention policy of the Bank of Japan (BoJ) since the 1990s.

A justified reason for interventions should be the diagnosis of a widely accepted misalign-

ment to prevent resulting spillovers to the real economy. Such a situation led to the Plaza

Agreement in September 1985 (e.g. Dominguez, 1991). The argument of preventing specula-

tors from driving the exchange rate above a fundamental value through active foreign

exchange market interventions is frequently scrutinised against the background of its effec-

tiveness (Sarno and Taylor, 2001; Reitz and Taylor, 2008, 2012). In this vein, the BoJ’s

exchange rate policy has not only been controversial since the breakdown of Bretton Woods,

but has also turned out to be the starting point for empirical search for different reasons.

There is a heated debate as to whether interventions are generally able to influence the path

of the exchange rate or to dampen exchange rate volatility (Sarno and Taylor, 2001). A great

deal of research has analysed the general motivation for interventions in the foreign exchange

market (e.g. Almekinders and Eijffinger, 1994, 1996; Baillie and Osterberg, 1997; Ito, 2003;

Frenkel et al., 2005; Ito and Yabu, 2007; Beine et al., 2009). A popular argument is that a

central bank should obey fundamental concepts in the evaluation of a potential misalignment

of the currency. In that case, a ‘fundamental value’ enters as a target in the policy function of

the central bank. Some authors have claimed that the BoJ generally followed a strategy of

‘leaning against the wind’ that reverts misalignments (Hutchison, 1984; Galati et al., 2005;

Kim, 2007; Suardi, 2008). Hence, it is important to examine whether the Bank of Japan was

responding to deviations from such a fundamental value when it decided to intervene. If not,

this could indicate that interventions are being conducted for other reasons.

The paper represents the personal opinion of the authors and does not necessarily reflect the views of

the Deutsche Bundesbank.

©2016 John Wiley & Sons Ltd

1836

The World Economy (2017)

doi: 10.1111/twec.12448

The World Economy

Ito (2003) identifies a very pragmatic approach for modelling the behaviour of Japanese

intervention in the 1990s. He sees evidence that interventions occur when the JPY/US$

exchange rate has deviated from a level of 125. Consequently, this level can be understood as

an ad hoc long-run target not necessarily related to a fundamental assessment. However, the

results concerning this ad hoc target are very mixed. While the corresponding variable turns

out to be significant in Frenkel et al. (2005), it is not in Fatum and Hutchison (2010). Ito and

Yabu (2007) define a broader set of target rates in which the one, three and five-year averages

enter with different weights. In addition, the purchasing power parity is often used in

microstructural approaches as a simple fundamental value (e.g. Reitz and Taylor, 2008,

2012). Based upon empirical evidence for the 1970s and 1980s with respect to PPP and inter-

ventions, Frenkel et al. (2005) also use it in their reaction functi on.

However, the literature is notably silent when it comes to the issue of establishing an ade-

quate long-run fundamental value of the exchange rate for the use in the intervention function

which bases upon a broader set of fundamentals. Moreover, a commonly accepted definition

of a fundamental value for the exchange rate does not exist.

1

Although a regression-based

approach would be a natural point of departure, a fundamental value of the exchange rate

based upon an estimated reduced form structural model is virtually disregarded –mainly

because the in-sample relationship between exchange rate and fundamentals is subject to dif-

ferent instabilities (Beckmann et al., 2011; Park and Park, 2013). Nevertheless, a careful

investigation is admittedly necessary because it is important to know about the concept being

used to evaluate a fundamental value. As mentioned above, there is no justified answer

regarding the question of which exchange rate is the ‘true’ fundamental rate, but this does not

automatically mean that a broader analysis is not conducted by the central bank. For this rea-

son, a more flexible approach is necessary when evaluating the BoJ’s intervention behaviour.

Thus, a natural extension of this strand of the literature is to use estimated fundamentally

based values for the JPY/US$ exchange rate stemming from economic theory.

Against this background, the aim of this paper is to analyse the question of which econ-

omic concepts can explain the BoJ’s intervention behaviour in terms of intervention probabil-

ity. To clarify this issue, we begin by using a variety of concepts to estimate fundamentally

based values and pay specific attention to the role of long-run targets. More precisely, we start

by considering different estimation techniques for establishing possible fundamentally based

values based, inter alia, on a monetary exchange rate model. Then, we estimate several policy

reaction functions to explain the occurrence of interventions in the Japanese case after 1991.

These estimations also account for asymmetric intervention behaviour by distinguishing

between different patterns of fundamental deviations: positive and negative deviations are

linked to the cases of positive and negative exchange rate changes. In addition, we conduct a

regression analysis in order to analyse different subsamples to detect structural changes. Over-

all, we are able to provide evidence that the BoJ’s intervention behaviour is slightly better

characterised by empirically established fundamental values. Nevertheless, short-term targets

seem to dominate.

The rest of this paper is organised as follows. Section 2 proceeds by summarising theoreti-

cal thoughts on target levels in the context of interventions and also motivates our choice of

economic fundamentals based on the monetary approach. Section 3 provides econometric

1

See, for example, the discussion in Reitz and Taylor (2012). They conclude that the purchasing power

parity is sufficient to reflect a fundamental value.

©2016 John Wiley & Sons Ltd

EXCHANGE RATE TARGETS AND JAPAN’S MONETARY POLICY 1837