Commentary | 06 August 2012

How the RBA should manage a high $A

In an opinion piece in The Australian Financial Review, Lowy Institute Professorial Fellow Warwick McKibbin argues that when markets are distorted by foreign central banks, it might be better to deal directly with the distortion.

  • Warwick McKibbin

In an opinion piece in The Australian Financial Review, Lowy Institute Professorial Fellow Warwick McKibbin argues that when markets are distorted by foreign central banks, it might be better to deal directly with the distortion.

  • Warwick McKibbin

Executive Summary

How the RBA should manage a high $A

Warwick McKibbin

The Australian Financial Review

6 August 2012

P. 47


There is a danger in economic debate to believe that the correct response to one economic shock must be the same for all shocks. This is particularly true when considering the role of the strong Australian dollar and its impact on traded goods, but particularly on manufacturing.


It is clear that it is better to take the appreciation of the real exchange rate caused by a commodities boom through a stronger currency rather than through higher inflation. However, it is important to understand that whether or not to intervene in the foreign exchange market depends on the shock hitting the economy. Allowing a pure float is not always optimal, especially if you have information on the nature of the shock driving the exchange rate.


There are many factors driving the value of the Australian dollar. The case of foreign central banks buying Australian dollars is particularly unusual. It is difficult to find a song that could explain what to do in this case, but there is a large economics literature, particularly written during the 1980s, that gives important insight on the question as to whether there is ever a case for foreign exchange market intervention.


Some answers can be found in a paper I co-authored with Jeffrey Sachs on "Comparing the Global Performance of Alternative Exchange Arrangements" which was published in the Journal of International Money and Finance when I was at the RBA in 1988. This paper explored the question of when to allow a pure float and when to intervene in the currency market to either "lean with the wind" or "lean against the wind". That is, whether to hasten an exchange rate adjustment or to slow it down.


The exchange rate question was examined in a global context of redesigning the international monetary system and whether it is better to float exchange rates or have a global system of fixed exchange rates. It turns out the answer is "it depends". What it depends on is the nature of the economic shocks either at home or overseas that are moving the exchange rate. In the case of shifts in the demand for our exports or a shift in the supply of our domestic goods due to a productivity decline, it is better to allow the exchange rate to do the adjustment rather than inflation and other relative prices.


This insight can be applied to the current issue facing Australia where foreign central banks are increasing their holdings of Australian dollars. This is a pure portfolio shock generated in the financial market.


If foreigners want to hold more Australian dollars in order to park these dollars in foreign exchange reserves and will not be using these dollars to buy Australian goods and services, then the best response is for the Reserve Bank to print more Australian dollars. These additional dollars should be sold to foreign central banks in return for foreign assets. The foreign assets would appear on the RBA balance sheet exactly balancing the increase in money supply. There would be no effect on the domestic economy from this global shock if the RBA undertook this transaction.


This policy could be carried out either by direct transactions with foreign central banks or it could be done in the foreign exchange market by the Reserve Bank buying foreign currency equal to the amount they know foreign central banks are buying of Australian dollars. It is the case that the Australian money supply would increase but this would be by the amount of purchases by foreigners and thus domestic liquidity would be unchanged and domestic monetary policy and economic activity would be unaffected.


It is also the case that the same outcome could be achieved if the RBA held the domestic interest rate constant in the face of tightening monetary conditions caused by foreign central banks purchasing Australian dollars and driving up the value of the currency. In order to fix the interest rate, the RBA could buy government bonds and sell Australian dollars. The outcome for domestic liquidity is similar except for the changes in the composition of the RBA balance sheet where bonds are acquired instead of foreign exchange.


However, with a scarcity of high quality government debt globally and with the tendency for exchange rates to overshoot, a case can be made that it is better to use the direct intervention path and purchase foreign reserves directly in the foreign exchange market to the extent that foreign central banks are buying Australian dollars.


It is difficult to argue against this strategy when the shock is known. However, if the shock is not observed completely but it is expected that foreign portfolio purchases of this type are occurring, then the insight of the McKibbin and Sachs paper is that it may still pay to intervene. The degree of intervention depends on how confident the central bank is of the particular shock to the economy.


However, that paper also showed that if the central bank thought the shock was of a particular type and is turned out to be a completely different shock then intervention in the exchange market could do great harm to the economy. This complexity is what makes monetary policy so difficult.


In the current situation, where the RBA has clear information on the portfolio purchases by foreign central banks, it seems building up foreign exchange reserves and satisfying some of that greater demand for Australian dollars would be a prudent step in preventing an unnecessary appreciation of the currency. It would, however, need to be explained carefully. The general point is that even though allowing markets to work is usually the best policy, when markets are distorted by the policies of foreign central banks, it might be better to deal directly with the distortion rather than allowing markets to propagate the shock unnecessarily.


The question of whether the RBA should intervene in the foreign exchange market is completely independent of the question of whether the RBA should change interest rates at its next meeting.


The argument for intervention is to eliminate a foreign shock that is distorting the debate about monetary policy. It is completely consistent to argue that there should be intervention as argued above to take out excessive exchange rate appreciation and still believe that domestic interest rates should be raised to assist in managing a major restructuring of the Australian economy.




Warwick McKibbin is a professor of economics at ANU, a Lowy Institute professorial fellow and a senior fellow at the Brookings Institution in Washington.