The rupee is now a big market. Summing across both spot and derivatives, perhaps $30 billion a day of onshore trading and $40 billion of offshore trading takes place. Both these markets are tightly linked by arbitrage. In other words, for all practical purposes, it’s . . . → Read More: The rupee: Frequently asked questions]]> q: How big is the market for the rupee?

The rupee is now a big market. Summing across both spot and derivatives, perhaps $30 billion a day of onshore trading and $40 billion of offshore trading takes place. Both these markets are tightly linked by arbitrage. In other words, for all practical purposes, it’s like NSE and BSE which are a single market unified by arbitrage. If you place a small order to buy 100 shares on either NSE or BSE, you get essentially the same price, and arbitrageurs are constantly at work equalising the price across both markets. It is a similar state of affairs between the onshore and the offshore rupee. Both markets are tightly integrated by arbitrage.

The offshore market for the rupee, and a large part of the onshore market, is OTC trading. Hence, the efficiencies of algorithmic trading and algorithmic arbitrage cannot be brought to bear on onshore/offshore arbitrage. So the arbitrage is done by manual labour. Still, it gets done. Both markets are tightly linked and show the same price. We should think of them as one market. It’s one big market, it is one of the big currencies of the world, it’s roughly $70 billion.

q: How might RBI do manipulation of this market?

If RBI wants to hit the market with orders big enough to make a difference, they have to be ready to do fairly big orders and to be able to do it on a sustained basis. As a rough thumb-rule, I might say that in order to make a material difference to a market with daily volume of $70 billion, they have to be in the market with atleast $2 to $3 billion a day.

q: What would go wrong if they tried this?

Three things would go wrong.

First, foreign exchange reserves are $275 billion. If RBI sells off $2.75 billion a day, the reserves would be quickly gone.

Second, when RBI sells dollars and buys rupees, this sucks liquidity out of the market. The side effect of selling dollars would be a sharp rise in domestic interest rates. In other words, monetary policy would get hijacked by currency policy. This would not be wise. Monetary policy should be focused on delivering low and stable inflation: it should have no ulterior motives.

Third, suppose you and I saw a fake market price of Rs.45 per dollar, which is created by RBI and not a market reality. We would know that in time, the truth will out, that the price will go back to Rs.52 a dollar. The rational trading strategy for each of us would be: To sell any domestic assets, and to shift money out of the country. This would trigger off an asset price collapse in India. We would take the money out, and wait for RBI to give up on these adventures. At that point (perhaps Rs.52 a dollar, perhaps worse) we would bring the money back to India and buy back our assets. We might make two returns here: first, on the move of the INR/USD from 45 to 52 (or worse) and on the drop in asset prices.

q: Isn’t it hard to take money out of India in this fashion?

It’s easier than we think. Remember September 2008? The mythology in our heads was: India is crouching safely behind a wall of capital controls. In truth, the wall wasn’t there.

q: But until recently, Mother RBI used to give us a pegged INR/USD exchange rate! What changed?

In late 2003, RBI ran out of bonds for sterilisation. Associated with that, there was a first structural break in the rupee exchange rate regime, with a doubling of volatility. A short while later, in March 2007, there was another doubling of volatility. From April 2009 onwards, RBI’s trading in the market has gone to roughly zero. Mother RBI stopped managing the exchange rate a while ago.

The exchange rate is the most important price of the economy. The decontrol of this exchange rate is the biggest achievement of the UPA in economic reforms. The credit for this goes to Y. V. Reddy and Rakesh Mohan (who took the first two steps of doubling exchange rate flexibility) and to Dr. Subbarao (who got out of trading on the currency market, which did remarkably little to INR/USD volatility).

q: Why did nobody tell me that something changed in the exchange rate regime?

RBI should be talking more transparently about what is going on. But they are not transparent about what they do. Even though hundreds of millions of people are affected by their trading on the currency market (or the lack thereof), the manual which governs their currency trading at any point in time (i.e., the documentation of the prevailing exchange rate regime) is not transparently disclosed to the people of India. We have to decipher what is going on by statistically analysing exchange rate data.

q: So what might happen to the rupee next? Is there a `law of gravity’ which will pull it back to erstwhile values of Rs.45 or Rs.50?

When you don’t manipulate a financial market, the price time-series comes out to something close to a random walk. In the ideal random walk, all changes are permanent. The random walk never forgets; there is no law of gravity which takes it back to recent values. Your best estimator of what it will be tomorrow is: what you see today.

In order to get a sense of what will come next, go through the following steps. First, go to INR/USD options trading at NSE, and pluck out the implied volatility for the four at-the-money options. I just did that, and the values are: 10.43, 10.32, 10.33 and 10.08. Calculate the average of these. With the above four values, the average is: 10.3. (This is a quick and dirty method; here is one which is much better).

This tells a very important thing: The options market believes that in the future, the volatility of the INR/USD rate will be 10.3 per cent per year.

In order to re-express this as uncertainty per month, we divide by sqrt(12). This gives the volatility for a month as : 3% per month.

Roughly speaking, the 95% confidence interval for what might happen over a month, then, runs from -6% to +6% (this is twice the standard deviation, which we just worked out was 3% per month).

The INR/USD is now Rs.51.62. By the above calculation, we can be 95% certain that one month from today, it will lie somewhere between 48.5 and 54.7.

These trivial calculations have been done by equity market participants for the longest time. It is a standard and trivial idea: To read the implied volatility off the Nifty options market, and to do such calculations to get a sense of what might come next with Nifty. But on the currency market, this is relatively novel. Only recently have we got a nice currency options market, and only recently have we got to a genuine market. Now these skills can be brought to bear on the currency market.

q: What changed in imports and exports which gave us the big recent move of the rupee?

The current account (goods, services, and then some) adds up to a mere buying and selling of $4 billion a day. The bulk of currency trading is about the capital account. The currency is a financial object; the exchange rate is defined by financial considerations and not by current account considerations.

q: What happens to the Indian economy when the rupee depreciates?

This has been the source of a great deal of confusion and it’s important to think straight about this. There are exactly three important effects in play:

1. Some people had borrowed in dollars, and left is unhedged since they were speculating that the INR would appreciate. They have got burned. That’s okay – in a market economy, many people place bets about future fluctuations of financial prices, and half the time the speculator loses money. (If the rupee had not depreciated sharply, these speculators would have been truly joyous).
2. When the rupee depreciates, imports become costlier and India’s exports become more competitive. So exports (X) gradually start going up and imports (M) gradually start going down. The net gain in X-M is increased demand in the local economy. In this fashion, INR depreciation is good for aggregate demand (and conversely INR appreciation pulls back demand). However, we have to bear in mind that these effects are small and take place with long lags.
3. Many things in India are tradeable. It is important to focus on the things that are tradeable and not imported. As an example, there are many transactions between a domestic producer of steel and a domestic buyer of steel. Both buyer and seller are in India. But the price at which they transact is the world price of steel (which is quoted in dollars) multiplied by the INR/USD exchange rate. Through this, the domestic prices of tradeables go up when the rupee depreciates.

When change comes to a stabilised currency, as it must, that change is painful. Change in the long term is inevitable. The random walk doles out a little change every day, which is . . . → Read More: How to decontrol the price of oil]]> We know a lot about price controls from the field of exchange rates. Here’s an argument from way back, in 1998:

When change comes to a stabilised currency, as it must, that change is painful. Change in the long term is inevitable. The random walk doles out a little change every day, which is less painful than sudden large changes.

…

Currencies which are random walks yield a deeper sort of stability. The steady pace of small changes every day generates realistic expectations about currency risk and continual realignment in production processes in the economy. It avoids sudden changes, and keeps the currency out of the domain of politics. The random walk regime is sustainable without incurring serious distortions in the economy.

In the field of exchange rates, India understood these arguments, and moved to a floating exchange rate. In March 2007, the INR/USD volatility moved up to roughly 9% and from early 2009 onwards, RBI stopped trading in the currency market. This was the biggest achievement of the UPA in economic reforms: In the 2007-2009 period, we got to a market determined rate on the most important price of the economy.

These same ideas are useful in thinking about the price of petrol. A large jump of Rs.1.8 per litre attracts attention. It is far better to let the price fluctuate every day. Ultimately, the price has to adjust. We suffer a lower political cost by letting it adjust every day (through the depoliticised market process). If we bottle up the small changes, then we have to make large changes. These are a bad use of political capital.

This makes you think: What countries land up in this display, and how bad are the spreads?

Let’s . . . → Read More: Watching markets work: Spreads at a money changer]]> I was at a money changer in London and saw a tariff card, for purchase and sale of a few currencies (all to the GBP). (This was a while ago: It was on 12 May 2011).

This makes you think: What countries land up in this display, and how bad are the spreads?

Let’s start with the tightest spreads: USD and EUR. The spread — 19.98% for the Euro and 20.87% for the USD — is a pretty huge one
compared with the transaction efficiencies that we’re used to seeing on NSE and BSE. (Don’t miss the GBP 3 charge that’s also tacked onto transactions). The bid/offer spread on the wholesale market for the USD/GBP and the EUR/GBP are roughly zero. The inventory risk carried by the money changing firm must also be quite low given that many customers are likely to come by with such orders. Hence, the values of the spread seen there represent the pure cost of the retail front-end: paying rent, paying salaries, the cost of capital etc.

It’s hence interesting to subtract out the lowest value (19.98% for the Euro) and sort the remainder:

Japan and Australia are floating rates with full convertibility. There is no illegality involved. But the inventory risk is greater given that these are smaller countries; there would be fewer buyers/sellers of their currencies to the GBP. The vol is much like GBP/USD or GBP/EUR, so the enhanced spread reflects purely the greater inventory risk.

Saudi Arabia and UAE have credible hard pegs to the USD. Their vol to the GBP is exactly the vol of the USD to the GBP. (And, they are as convertible as the US). But their spreads are much bigger than that seen for the USD. It must reflect a small number of
transactions and hence inventory risk. They are small countries and even fewer transactions would be taking place. Many of their
nationals would probably hold the bulk of their liquid wealth in USD so the question of transacting through the local currency might not even arise.

Russia has full capital account convertibility, so there is no illegality. But it’s a highly volatile currency and the transaction flow is small. So we get the next step up in the spread, to 6.18%.

China has near-zero volatility to the USD, which means they are a high volatility rate to the GBP. It is a big country so there must
be quite a bit of traffic; there would be low inventory risk. The real issue is the illegality. The enhanced spread is the price paid
by people undertaking these transactions, for the capital controls of China.

And then we have India, the fattest spread in this group of countries, where I reckon it’s a combination of illegality (akin to
China), low volume of transactions (since India is a much smaller economy than China) and currency volatility (since India floats
while China does not).

I wasn’t able to make any sense of the list of countries that showed up in the list. Why Kenya and Sri Lanka, and why not Nigeria
or Indonesia?

After the rupee became a float, so far, it has . . . → Read More: What in the world is happening to the rupee?]]> The INR/USD rate is now nudging Rs.50 to the dollar. This is a big move over a short period: a depreciation of 12.1 per cent over the 84 days from 1 July till 23 September.

What fluctuations of the INR/USD can we reasonably expect?

After the rupee became a float, so far, it has had average volatility of roughly 9 per cent annualised. Roughly speaking, this means that over a one year horizon, the movement over a year would range between -18 per cent and +18 percent, with a 95 per cent probability. More extreme movements would happen with a 5 per cent probability.

Over a period of 84 days, roughly speaking, we’d have expected this 95 per cent range to run from -8.6 per cent to +8.6 per cent. Compared with that, a 12.1 per cent move is a bit unusual.

It’s only a bit unusual because the historical volatility of the INR/USD, in the period of the float, was rather low. The USD/EUR rate,
which is perhaps the world’s most liquid market, has had an annualised volatility from January 1999 onwards of 10.3 per cent. The INR/USD has got to surely be more volatile than this, given the inferior liquidity of the INR and given the greater macroeconomic volatility in India. Hence, I think we should consider the 9 per cent vol, that was seen in the early days of the float, as relatively unusual. The future will most likely hold bigger values for this vol.

The implied volatility of the INR/USD at the NSE has reared up to values like 14 per cent annualised. That sounds more sensible to me.

What about other currencies?

We tend to do wrong by focusing too much on the bilateral INR/USD rate. In the recent days of distress, as fear has resurged, people
have taken money out of everything under the sun and put it into US Treasury bills. This has given a strong dollar at the expense of
essentially every other currency. Here’s the picture for the INR, against the four major currencies of the world, from 1 July till 22
September:

The picture of the rupee is much more complex than that implied by simply watching the bilateral rupee/dollar rate.

Can RBI block such a large depreciation?

Let’s think through the steps which would follow if RBI tried to sell dollars in trying to prop up the INR:

• Global trading in the INR stands at roughly $75 billion a day. If you want to manipulate this market, you need a big stick. Small trades will do nothing. If preventing INR depreciation is the goal, RBI has to go into this with trades of $2 to $5 billion a day, with the willingness to stick it out for the long run. With reserves of $281 billion, there is not much hope here. Specifically, if RBI sells $80 billion in reserves, the market will see that. They will know that further rupee defence is now going to be hard (since $200 billion of reserves is starting to look like a small hoard), and speculators across the world will start betting that RBI’s defence of the rupee will fail.
• Reserve money is only $275 billion. For each $27.5 billion that RBI sells, reserve money drops by 10%. At a difficult time like
  this, a sharp and sudden monetary tightening will be an unpleasant side effect of defending the rupee. (This trading can be sterilised, but that has its own problems. I just want to emphasise that selling reserves is not easy and is not a free lunch).
• The rational speculator knows that the exchange rate will eventually find its level. When RBI prevents a large INR depreciation today, they are giving a free lunch to the speculator, who would take a bet that INR would depreciate in the future. Specifically, it would be efficient for domestic and foreign investors to dump assets in India, take money out at (say) Rs.45 to the dollar which is the artificial price, wait for the gradual depreciation to Rs.50 to the dollar, and come back into India to buy back the same assets. This trade generates 11% returns over a short period and is thus very attractive. In other words, a defence of the
  rupee would trigger off an asset price collapse in India.

Meddling in the affairs of the currency market is thus highly ill-advised for a central bank.

Should RBI try to block INR depreciation, even if they could?

Let us play a thought experiment where RBI had $2810 billion, i.e. 10x larger than what’s with us today. In that case, RBI could
play in the currency market, selling $2 to $5 billion a day for a year without serious distress. Is this a good idea?

I would argue that this is not a good idea. When times are bad, the rupee should depreciate. This drives up the profit rates of all
Indian tradeables firms and thus bolsters the economy.

Under a floating rate, in good times, the INR appreciates (which pulls back the exuberance of tradeables) and in bad times, the INR
depreciates (which fuels profits and thus the physical investment in tradeables). This is arguably the only element of stabilisation
in Indian macroeconomic policy.

RBI is playing this mostly right

From early 2007 onwards, the INR has been quite flexible. In particular, after early 2009, RBI’s trading on the market has tailed
off. There have been a few months with minor amounts of trading by RBI. This trading has mystified me, since these small trades can do nothing to influence the price. In practice, the INR has been a float.

A floating exchange rate is exactly the right stance for difficult times like this. In bad times, the best thing that can happen for
India is a big INR depreciation, thus bolstering the tradeables sector.

Let’s evaluate an alternative policy platform: To peg the INR in normal times but to let go in difficult times. Is this feasible?
Yes. But this is very disruptive: if economic agents have been given an implicit promise that the INR will not move, then the large move (which will surely come) would cause pain. It is far better to stay out of the market all the time, and create a trustworthy structure of expectations in the minds of economic agents about what the future holds.

We had a large depreciation in the crisis of 2008, and that served India well. In similar fashion, we should welcome the INR depreciation that is accompanying global gloom.

The only element of RBI policy where I have a major disagreement is communication. RBI has never used the words floating  exchange rate. RBI needs to clearly communicate to the economy that the rupee is now a market determined exchange rate, and RBI is no longer in the business of trading in this market. There is greater clarity of thought at RBI as compared with the quality of communciation; the speech writing still suffers from twinges of 1960s economics.

What is the collateral damage of a large INR depreciation?

There are three things that go wrong alongside a big INR depreciation:

1. Firms who have unhedged foreign currency borrowing get hurt, because they have to pay back more than anticipated. A person who borrowed Rs.100 (in unhedged USD) has to pay back Rs.110, owing to the 10 per cent INR depreciation. The stock market is doing a fine job of identifying these firms and beating down their stock prices.Of crucial importance is the fact that from early 2009 onwards, the INR had already moved to a float with a 9 per cent annualised vol. So CEOs and CFOs knew that the INR/USD rate was going to fluctuate. They were not lulled into complacence thinking that the exchange rate was going to be stable. By avoiding this moral hazard associated with pegged exchange rates, RBI’s decision to float in early 2009 laid a good foundation for the structure of firm borrowing as of July 2011.

   When a country has a pegged exchange rate, you tend to see a big buildup of unhedged currency exposure on corporate balance sheets. When the big depreciation comes, the big businessmen then queue up to the central bank begging for defence of the LCY. Prevention is better than cure: It is far better to have high exchange rate volatility all along, so that firms do not undertake such risks, and the toxic political economy does not come into play.

2. With an INR depreciation, tradeables become costlier. On one hand, this bolsters the profitability of tradeables firms, and
   thus their investment plans. But at the same time, this feeds into inflation. In recent months, tradeables inflation has been  sleeping while non-tradeables have contributed to the high CPI-IW inflation. We will now see a resurgence of tradeables
   inflation. This will exacerbate the inflation crisis. RBI will need to stay on the project of raising rates in order to combat this
   inflation.
3. The government’s subsidy program with petroleum products and fertilisers gets costlier when the INR depreciates. So India’s
   fiscal crisis gets a bit worse when the INR depreciates.

This logic is rooted in high levels of de facto capital account openness. Sometimes, policy analysts think that you can have your cake  and eat it too, and try to dodge these arguments by utilising capital controls. This has not worked in India, and the levels of de facto
openness have only grown through the years.

In summary, what should RBI be doing?

RBI should be focused on using the short-term interest rate as a tool to bring CPI-IW inflation under control, without distortions of
interest rate policy caused by trying to meddle in the currency market. This should be accompanied by liberalisation of the Bond-Currency-Derivatives Nexus so as to achieve an effective monetary policy transmission. These are the two things that RBI needs to focus on.

India shifted away from government interference in the currency market, from 2007 onwards but particularly after 2009. This is one of the biggest achievements in India’s economic liberalisation. This is a bigger issue in economic liberalisation than (say) decontrol of petroleum product prices. The INR is now a market. Nifty and INR are the two most important markets in the economy. It is time for all of us to analyse the INR as we analyse Nifty: as the outcome of a market process.

Is RBI back to trading the INR?

We don’t know. The data only comes out at monthly resolution, with a two month lag. But early signs that would show up would be unusual jumps in the weekly data about reserves, reserve money, etc. Greater transparency from their side would help greatly.

Join the forum discussion on this post - (1) Posts

But what about the world’s balance of payments? Many economists assume these must also . . . → Read More: Mythbusting: Balance of Payments Edition]]>

Imagine a world with two countries. If one country has a current account surplus, the other must have an equal and opposite current account deficit. More generally, the sum of the current account balance, of all countries, is zero.

But what about the world’s balance of payments? Many economists assume these must also sum to zero. For example, one often hears the claim that if one country is running a balance of payments surplus then others must be running deficits. Another argument often heard is that the RMB cannot become a reserve currency until China stops running a balance of payments surplus, because otherwise other central banks will not be able to acquire RMB assets.

This is wrong. In fact, if the right conditions come together, every country of the world can simultaneously run a balance of payments surplus.

Once a country starts trading on the currency market, the identity between the current account and the financial account breaks down. As an example, China runs a surplus on both the current and capital accounts. (That’s how it is piling up so much reserves). Thus, when even one country in the world is trading on its own currency market, it is no longer the case that the balance of payments of the world have to add up to zero.

Does the accumulation of reserves by one country imply a loss of reserves by another? Consider the following two country example. Let’s say the two countries are the US and China, and lets assume that the RMB and dollar are both reserve currencies. Let’s say that the currencies are pegged at 1:1, so it doesn’t matter if you are talking about RMB or dollars. And let’s say that trade is balanced, so we can ignore it.

The US government now sells a 100 bond to the PBOC. And the Chinese government sells a 100 bond to the Fed. This yields a balance of payments surplus of 100 in both countries. Reserves went up by 100 in both countries. In both countries the economy (outside the central bank) has imported 100 in capital by selling bonds. So, the financial account in each country shows an inflow of 100, creating a surplus of 100.

What is going on? In this example, the central banks are inflating reserves by exchanging assets — I buy your government’s bond and you buy mine. But we call this a balance of payments surplus (in both countries) because we draw an arbitrary line, above which we record the government part of the transaction (inflow of fx from the bond sale) and below which we show the offsetting central bank transaction (outward investment). Since the assets are accumulating to the central bank in each case, we say that both nations are running BOP surpluses.

When countries do this, all countries can run a balance of payments surplus at the same time. Admittedly, this will be difficult for countries running current account deficits and facing capital outflows. But it is, technically, possible. That’s why, say, China has been able to build up $3 trillion in reserves without any major country losing reserves at all.

In the case of RBI’s data about RBI’s trading on the currency market, such data revisions should ordinarily not arise.

But yesterday, data released by RBI modified the . . . → Read More: A Puzzling Data Revision]]> Ordinarily, official statistics get revised because at first, provisional estimates are released, and when the full data filings come in, then improved estimates are put out.

In the case of RBI’s data about RBI’s trading on the currency market, such data revisions should ordinarily not arise.

But yesterday, data released by RBI modified the previous information that had been put out about RBI’s trading on the currency market. Earlier, trading in June had been claimed to be 0. Now it shows purchase of $370 million and sale of $260 million. Earlier, trading in September had been claimed to be 0. Now it shows a purchase of $260 billion. I wonder why this data revision took place.

The newest data – for October – shows a purchase of $450 million on the spot and $450 million on the forwards. At a time when rupee trading is estimated at above $40 billion a day (worldwide), it is hard to see how such a small scale of trading can generate a significant impact upon the price; so I wonder what is going on in terms of the rationale and the thought process.

Currency options are obviously useful as a risk-management tool. I feel that futures are nice simple linear contracts: they ask the person to make only one decision — . . . → Read More: Who Will Make the Exchange-Traded Currency Options Market?]]> In a few minutes, NSE and USE will start trading in currency options. This will be the first exchange-traded options in India on a non-equity underlying.

Currency options are obviously useful as a risk-management tool. I feel that futures are nice simple linear contracts: they ask the person to make only one decision — are you long or are you short. But once a futures position is entered into, the person needs the ability to manage the position since daily marking-to-market is done, and since there can be large losses for either the futures long or the futures short.

Compared with this, long positions on call or put options appeal to the kind of person that is willing to think carefully about a position at the outset, but after that it is fire and forget. This better describes the life of many firms exposed to currency risk, particularly those with relatively weak treasuries.

Currency options have, of course, been traded OTC for some time now. But there are real problems with this market. Customers have sometimes been ripped off by banks on pricing, given the lack of a liquid and transparent comparison point. While currency options are offered by banks to customers, there is not much by way of an inter-bank market.

As far as I know, there is relatively little by way of a build-up of human and systems capability in the banks for currency options trading (whether OTC or on exchange).

In contrast, there is a remarkable build-up of human and systems capability in the world of Nifty options trading. Options on Nifty have shaped up as one of the biggest options markets in the world. This involves end-users who think and trade options, staff working for securities firms who understand options (and understand issues about their credit risk when their customer has an options position), analytical software systems, and (most importantly) algorithmic trading systems. Options trading inevitably involves trading in a large number of underlyings. Strong computer systems which are able to think about, and place orders in, all the underlyings at one shot are of essence in achieving option liquidity. Such capabilities are now found in the world of Nifty options, and are absent in banks or in the OTC currency options market.

It is fairly easy for a person trading Nifty options to move to trading currency options. Hence, the brainpower and systems that have made Nifty options one of the world’s top contracts will easily be able to move to currency options trading, and make it work. I expect that the securities firms who dominate Nifty options trading will now dominate currency options trading.

I think three kinds of stories will now kick in:

1. Liquidity in currency options will fuel liquidity in currency futures, and vice versa. Corporate hedgers will be more interested in either, given that the other is also a possibility.
2. Skills and systems from Nifty options will flow into currency options. Banks will be able to rapidly bulk up their options capabilities by recruiting from the world of Nifty options, and by purchasing the software systems that have sprung up in that space.
3. Conversely, trading in both currency options and Nifty options will generate an increased business size for people who build knowledge and systems for options; it will also improve knowledge of options trading through an understanding and comparison of the nuances of two different underlyings. The number of FRM and PRM certified people in India will go up.