Back to 2007: Fear of appreciation in emerging economies

Andrea Kiguel and  Eduardo Levy-Yeyati in this article argue that the fear of appreciation is back in emerging economies. While what they claim in this article is true to a large extent in the Indian case as well, what one doesn't understand is how much are the authorities willing to resist appreciation? Politically there is only a favourable climate to prevent appreciation and thus there is no reason to believe that the crisis will change the way emerging economies look at managing economies that are fast integrating with the rest of the world. However, there are a couple of statements that probably needs empirical evaluation:

Asian currencies are unlikely to lead the way insofar as the Chinese renminbi, against which many Asian currencies are implicitly anchored, remains stable. 

It will be interesting to see which of the currencies are implicitly anchored on the CNY. If the article talks about the HKD, a paper by Dong He et. al from the HKMA might be an interesting read. Insofar as other currencies are concerned, I think the DXY index provides a better idea as the CNY is anyway pegged to the US dollar. 

On the other hand, leaning against the wind policy of exchange rate management will not only harm the global rebalancing initiative (is there one underway?), but also continue to highlight the credibility of Central Banks of the respective economies regarding inflation and domestic monetary management. In specific, Brazil and Philippines, following an Inflation Target regime (on paper and to a large extent except for exchange rate intervention) will be under severe stress. In short, it might well be the case that short-term goals of export stabilisation will be given greater priority than learning from the current crisis. 

Throwing ATMs open.

The Reserve Bank of India had thrown open ATMs owned by all banks for cash withdrawals since April, 2009. What one doesn't understand is the rationale behind why the RBI chose to single out and micro-manage this service provision provided by banks to its own customers. Here's why I say so:


For sometime now, I've tracked down the transaction cost of withdrawing money using my SBI ATM card. Least surprised, I have found that the unit cost of withdrawing Rs. 1000 has only been increasing and peaked during the weekend after the Bank strike on the 6th and 7th of August, 2009. This is a case in point that doesn't seek to address another problem with the nationalised banks: They don't stock up their ATMs with money! There are 8 ATMs near my workplace out of which only one is that of a private bank, ICICI. The rest are SBI (4), Canara Bank (2), and Indian Bank (1). The maximum it would cost me if I were to withdraw money in a user charge regime is a five minute walk +/- the Rs.20 which other banks charged me for using their ATM! Let us see why this is so:


All the blue balloons (but for one) in Qutub Institutional Area refer to ATMs

View NIPFP in a larger map

Controlling for extraneous factors (of not knowing hindi and being heckled to pay more for an auto ride), and for different times of the day (it was almost always around 8pm on Fridays), the following graph plots the cost of withdrawing a sum of Rs.1000. True, one could have withdrawn Rs.2000 instead and the marginal cost of withdrawal goes down. But wait a minute, aren't ATMs supposed to help you avoid carrying so much money around all the time other than of course reduce the burden of going to a bank and standing in long queues to write yourself a cheque?

Also, what one often does not understand is that there is no prior information that an ATM does not have money. Therefore, I have always managed to visit at least three ATMs before getting money. This estimate of course does not include the opportunity cost of time spent in looking for an ATM with money.

So now, what do we see? We see that the least I have spent in the past one month is Rs. 20 to withdraw money from an ATM that isn't a service provided by SBI as I withdrew from Indian Bank ATM. Why did I have to spend this amount of money? SBI has the largest network of ATMs in the country and in all probability, being the biggest operator when it comes to this service provision, there was no incentive for SBI to stock money in the ATMs given that everybody else is going to use the same.
The case in point is that the RBI, as a banking regulator and a monetary policy authority, shouldn't be interfering with what the banks know best and what the banks do better: serve customers. If one goes through the notification carefully, it says:

"3. International experience indicates that in countries such as UK, Germany and France, bank customers have access to all ATMs in the country, free of charge except when cash is withdrawn from white label ATMs or from ATMs managed by non-bank entities. There is also a move, internationally, to regulate the fee structure by the regulator from the public policy angle. The ideal situation is that a customer should be able to access any ATM installed in the country free of charge through an equitable cooperative initiative by banks.

4. In view of this, RBI had placed on its website an Approach paper and sought public comments. The comments received have been analysed. Based on the feed back a framework of service charges would be implemented by all banks as under:

Sr.No.	Service	Charges
(i)	For use of own ATMs for any purpose	Free (with immediate effect)
(2)	For use of other bank ATMs for balance enquiries	Free (with immediate effect)
(3)	For use of other bank ATMs for cash withdrawals	No bank shall increase the charges prevailing as on December 23, 2007 (i.e. the date of release of Approach Paper on RBI website) Banks which are charging more than Rs.20 per transaction shall reduce the charges to a maximum of Rs.20 per transaction by March 31, 2008 Free - with effect from April 1, 2009.

5. For the services at (1) and (2) above, the customer will not be levied any charge under any other head and the service will be totally free.

6. For the service number (3) the charge of Rs.20/- indicated will be all inclusive and no other charges will be levied to the customers under any other head irrespective of the amount of withdrawal.

7. The service charges for the following types of cash withdrawal transactions may be determined by the banks themselves:

(a) cash withdrawal with the use of credit cards

(b) cash withdrawal in an ATM located abroad.

Firstly, this was a move based on "international experience". Elsewhere in the world, withdrawing money from other ATMs may not be a problem because there are other norms issued by banks that regulate withdrawals. Eg: Customers cannot withdraw more than thrice (and in some cases twice) a day. Secondly, banks know money demand and velocity and based on that decide location, and facilities provided to customers. Thirdly, in most cases one would find an ATM of every bank within a mall or where there is large movement of population. Fourthly, ATM facility isn't a service provision on which banks compete abroad. They compete on different provisions and make money on other things.

In India, and more specifically using Delhi as a case in point, main market places at the most have two different banks with ATM facilities, and why would those two banks need to incur the cost of reloading these machines for a bank that has chosen not to provide that facility for its customers in the same market? Secondly, one has permanently altered the incentive structure for banks to open up new ATM centers. And last but not the least, the customer (me in this example) is having to incur more cost in withdrawing money than what I used to before.

The current RBI stance of restricting the number of transactions by a single customer using ATM centers provided by other banks is not going to help either. Simply put, the scale and number of people with ATM cards is large and the restriction fails to address basic issues with such notifications: altering incentives and future investment propositions by banks into ATM centers.