Yesterday I tried to explain the decision of the Central Bank of the Republic of Turkey (CBRT) on a new FX purchasing system. Now it is time to talk on what this decision implies for the exchange rate. To begin with, I have to underline that the new system allows surprise interventions in the FX market, if desired.

In order to have a better understanding of this, let us take a quick look at how surprise interventions (interventions about which the FX market is not informed in advance) are carried out. In this method, the FX market cannot be informed of the intervention until it is actually carried out. In fact, to avoid the possibility that some actors cannot acknowledge it; central banks might issue press releases simultaneously with the intervention.

The system in Turkey functions as follows, roughly: banks engaging in transactions in the FX market remark their dollar purchase and sale price for the period when the markets are open. These prices apply for a fixed FX quantity (let us call it US$5 million). The prices are displayed in a screen in descending order and seen by all banks as well as the CBRT.

If the CBRT decides to make an intervention and purchase FX, it picks up the phone (where the conversation is recorded) and tells the bank which set the lowest FX sale price 'You have sold US$5 million to the CBRT'. There is a team at the CBRT responsible with this transaction. So, after a couple of seconds a second person calls the bank which had the second cheapest FX sale price to declare the same order.

Simultaneously, news agencies are informed that the CBRT intervened in the FX market. Even if they are not notified, banks sense from the immediate movements (as the bid of the bank receiving the order disappears) in the screen that something is going on and either scamper away or mark up prices, if they are not willing to sell FX at a low price. Therefore, exchange rate goes up during the intervention. The mechanism works in the opposite direction if the CBRT tends to sell FX.

Central banks which apply 'pure-ish' floating exchange rate regime are not a big fan of such operations. They have several reasons for this: first, it is quite likely that such interventions give the signal that the central bank seeks to maintain exchange rate at a certain level. But this implies loss of credibility for the bank which claims to be implementing a floating exchange rate regime. The bank would most likely face unpleasant criticisms reading "You say one thing but do another thing."

Second, even if the exchange rate changes during the intervention, it is not necessarily permanent. For instance, if accompanied with strong FX inflows, the rise in exchange rate due to such FX purchase intervention might disappear even the day after.

Third, when you purchase FX, you inject domestic currency to the market. This liquidity might be a matter of concern for central banks due to the inflationary risk it poses. To eliminate this risk, you have to sell bonds and draw back the currency from the market. But it is most likely that the return on the bonds you sell is higher than that on foreign bonds you are to buy with the FX you acquired as a result of the intervention. This would reduce the asset quality.

More importantly, after withdrawing currency from the market interest rate reaches a level higher than that in the case where not intervention was made. However, given that low exchange rate was one reason for the FX purchase intervention, you do not eliminate but intensify the problem (high interest rate - capital inflow - low exchange rate).

The CBRT's decision might lead to convergence to this new interventionist system, if desired. I do not think that the CBRT has such an intention; but the flexibility of the newly announced system creates this possibility as it is. Particularly the fact that the quantity of additional FX to be purchased will be known at the beginning of the week gives way to a considerable uncertainty in the FX market for the first day of the week.

In the former system where the interventions in the FX marker were occasional, daily intervention was least possible. But now every Monday it will be well acknowledged that intervention in the market for high quantities is quite likely. Actors will be informed of this only on Monday. But this will not eliminate the uncertainty, either since it will not be certain what proportion of the announced quantity will be purchased in the rest of the week.

When I first read the relevant decision, I thought that it will not affect the exchange rate. But with further elaboration I feel that this new mechanism might have an effect on the exchange rate if implemented aggressively. I believe the CBRT pulled an exotic rabbit out of the hat. This rabbit can gain the appreciation of exporters depending on how it is 'decorated (implemented)'.

This commentary was published in Radikal daily on 04.10.2010