was characterised by targeting money growth. Interest rates were set to limit growth in monetary aggregates, which was viewed as the key to price stability. Since the exchange rate was not an explicit part of this strategy, foreign exchange interventions were not required.
The arrival of inflation targeting in the 1990s significantly challenged the one-variable approach; instead, all variables that might influence future inflation were taken into account in setting monetary policy. In this context the degree of exchange rate pass-through to domestic prices determines the extent to which the central bank will have to incorporate exchange rate movements in their decision process. If the exchange rate is important for future inflation (i.e., the pass-through effects of exchange rate changes on inflation occur faster than the interest rate effects on inflation) then it follows that intervention might be a useful instrument.
How monetary policy is conducted
Monetary policy can operate through monetary targeting, exchange rate targeting or inflation targeting. A monetary targeting strategy will have an implicit inflation target, which is used to determine the optimal growth of the monetary aggregate. Central banks that pursue exchange rate targeting do not require a target rate for inflation. Ideally, the exchange rate would be pegged to a low inflation currency with the aim of mirroring their inflation performance; as such there is no need to specify an inflation target rate.
This is quite the opposite of inflation targeting, where there is a specific figure announced and widely communicated. In this case there is no interim target for the public to observe, which of course raises the profile of the final target. Practically, all three strategies are managed through short-term interest rates.
A recent approach has been to take account of all indicators, known as a look at everything strategy, rather than a single variable. This is very much allied to improved communications (signalling) between the policy makers and the public, and genuine accountability. In the UK this is seen in the letter that is sent from the Bank of England Governor to the Chancellor if inflation deviates over 1% from target.
One common feature in this transparent approach is the publication of minutes of monetary meetings, albeit with a time lag. It is important that the public believe in the policy makers’ commitment; anchoring inflation expectations is viewed as critical by all central bankers in public utterances and in official reports. The implicit assumption in inflation targeting is that low and stable inflation will promote macroeconomic goals such as economic growth and employment. Inflation targets are generally around 2% and are either published as a single figure or as a target range.
Events of 2008 and 2009 highlighted that monetary policy is not exclusively about preventing excessive inflation; many central banks at this time were looking to counter deflationary pressures (a decline in prices and weak growth) and in some cases sold their currency. Another good example of this is the Japanese policy response in 2001 to a decade long period of near zero growth and deflation. The policy conclusion was that the key to recovery was an expansion of the money supply. This was started in March 2001 by the adoption of quantitative easing (a term which has come into the public domain following the 2007-09 financial crisis). This was followed up by massive unsterilised intervention from the end of 2002 to 16 March 2004. To give some idea of the scale of intervention, on 5 March 2004 the Japanese purchased USD 11.2bn and sold yen.
By the end of 2003 there were signs of Japanese economic recovery and in March 2004 Alan Greenspan, the US Federal Reserve Bank Chairman, indicated that the intervention strategy had worked. He said:
“Partially unsterilised intervention is perceived as a means of expanding the monetary base of Japan, a basic element of monetary policy. In time, however, as the present deflationary situation abates, the monetary consequences of continued intervention could become problematic. The current performance of the Japanese economy suggests that we are getting closer to the point where continued intervention at the present scale will no longer meet the monetary policy needs of Japan.”
It should be noted that intervention had the full support of the US administration, which felt that economic stagnation was not in the interests of the US. [6] By not registering objections to the intervention the US effectively made it possible for the Japanese to operate in the market.
The efficacy of intervention
The effects of sterilised intervention are somewhat debatable. The standard approach is to view the potential impact through two routes: the portfolio channel and the signalling channel. In the former an intervention that changes the relative outstanding supply of assets denominated in domestic and foreign currencies will require a change in expected relative returns on the asset whose outstanding stock has increased, thereby leading to a change in the exchange rate. This is based on the assumption that investors consider foreign and domestic assets to be imperfect substitutes. The portfolio channel approach no longer carries much weight because the scale of possible intervention has declined relative to the size of the foreign exchange market. It may, however, have greater relevance in emerging markets where central bank reserve holdings are large relative to local market turnover.
It is widely thought that intervention operates mainly through the signalling channel. This may convey to the market future changes in monetary or exchange rate policy or that the authorities view the exchange rate to be out of line with economic fundamentals. A sterilised purchase (sale) of domestic currency reflects a desire for a stronger (weaker) domestic currency and this desire eventually leads to a tighter (looser) monetary policy. However, monetary policy may cause the exchange rate to appreciate or depreciate too much and prompts intervention to moderate or even reverse the trend of exchange rate movements. Dealers’ reactions (and success) will much depend on the perceived credibility of the central bank.
A signal may be used to reduce market expectations of current and future volatility. In recent years this has been the dominant theme – to reduce excessive exchange rate volatility. That being said, central banks may desire an increase in short-term volatility if they are faced with an undesirable exchange rate trend. The central bank will attempt to remove or reduce the one-way bet mentality by restoring two-way risk. Signalling intentions can of course be made clear via verbal commentaries, or ‘jawboning’ as it is described in the market.
There is secret intervention which fits neither the portfolio channel nor the signalling channel. By definition it is virtually impossible to get a handle on this. It is difficult to see this changing a trend but it may well slow down the process, providing a two-way risk dimension.
One way that intervention can be made more visible is through concerted efforts and this can indicate a strong commitment to exchange rate changes. The difficulty arises when too many central banks get involved. The market loses faith in the message in so far as a wide range of countries with skewed economic fundamentals may indicate a different policy requirement.
Coordinated intervention, however, is rare. A recent case of coordinated intervention was on 18 March 2011 when the Japanese were joined by the Group of Seven (G7) major industrialised countries to stall the surge in the yen after the tsunami and nuclear incident in Japan prompted market chaos. The G7 statement said that this action was to stabilise “excess volatility” and “disorderly movements in exchange rates”. The last previous coordinated intervention was in 2000 when the euro was bought.
Models of exchange rate behaviour assume that currency prices are efficient aggregators of information and market expectations are rational. In practice the foreign exchange markets may not be efficient and intervention signals may not always be credible or unambiguous. The question of whether intervention policy influences exchange rate volatility obviously depends on the definition of ‘volatility’. According to Dominguez and Frankel (1993) unanticipated and
