Click Here to View .pdf file

Over the past few months, "NDF arbitrage" has been the easiest whipping boy in the forex market whenever the rupee loses ground, and, indeed, there appears to be considerable evidence - both anecdotal and analytic - that this may well be the case.

The chart overleaf shows how the difference between the 1-year NDF rate and the domestic 1-year forward has been moving relative to the rupee since 2003. The left hand axis shows the arbitrage gain achievable by buying [dollars] domestically and selling offshore; the right hand axis shows the onshore spot rate. An even cursory perusal of the chart shows that in recent months there has been a very high correlation (nearly 95%, in fact) between the two variables. [Interestingly, the correlation is this strong only for the 1-year NDF; for shorter tenors - say, 1 month - the correlation is actually negative.] The chart suggests that a spread of around 30-35 paise seems to trigger domestic dollar buying and corresponding sales offshore. Other than the rupee's current move below 45, the last time this happened - end August early September - the NDG play pushed the rupee down from 43.75 to 44.30 in a couple of days; this brought the arbitrage spread down to around 20 paise, at which level the NDF activity appeared to close down.

The NDF market began in the early 1990s, as a result of investors' needs to hedge their risks in markets that had capital controls. The bulk of the liquidity was focused on Latin American markets, with the Mexican peso having the largest trading volumes. As investment flows into emerging markets increased, the NDF market spread to Asian and Eastern Europe gaining liquidity and importance, and there are some who believe it was a driver in the precipitation of the Asian crisis of 1997. However, volumes were relatively modest and concentrated in a few currencies like the S. Korean won and the Taiwan dollar. Liquidity in Indian rupee NDFs was virtually zero, with about $ 20-50 million being traded each week as recently as 2000. By 2003 (according to a study conducted by the U.S. Fed), the global NDF volume was about $ 4 bn a day, with the S. Korean won the most liquid, trading around $ 1.25 bn a day. The Chilean peso, Brazilian real and the Taiwanese dollar each traded about $ 700 mn a day, with the Chinese yuan next at about $ 250 mn a day. The Indian rupee was nowhere in sight; anecdotal evidence indicates that volumes at that time were not much more than $ 20-50 million a day.

Today, however, it is a different game. I met a banker the other day who told me that it is today quite easy to do volumes of $ 100-150 million of rupee NDFs in a single day, with a market impact of not much more than 4 or 5 paise; compared to the 2 paise or so impact such a transaction would have in the domestic market, this is pretty good liquidity. He also said there is a fledgling NDF options market, which has reasonable liquidity - not much less than the domestic options market. From this admittedly anecdotal information, it would seem that on active days, rupee NDFs today probably trade close to $ 800 mn to $1 bn a day, which is around 8-10% of the volume of the domestic USD/INR market! Of course, NDF liquidity is very volatile and much of the time the volumes are nowhere near this level. But when they are, they are certainly large enough to drive the domestic market, particularly with an increasing number of players having legs both onshore and offshore. The high correlation seen in the chart in recent months certainly seems to confirm this.

For instance, on October 7, the 1-year NDF market is quoting at 45.06 while the and the 1-year forward in the domestic market was at 44.57. You could buy domestically and sell offshore, provided, of course, you had a short dollar exposure domestically - in other words, if you had imports or loans or investments in the equity or debt markets. The 49 paise gain would be reduced by the buy/sell spread in both markets - typically about 1-2 paise on-shore and, reportedly 4-5 paise offshore. The trade also carries a basis risk, since the NDF transaction will be settled against the RBI reference rate on the due date, while the domestic forward purchase will need to be either utilized or cancelled in the inter-bank market at that time.

Nonetheless, it is clear that a net gain of more than 25 paise per dollar (over 0.5%) with very little risk is an attractive play for anybody. While much of the action is doubtless from financial players, like hedge funds who have investments in India, there are reports of several large corporates, particularly in commodity import businesses, who have outstandings in excess of $ 100 million in this market.

So, what to make of all of this?

First off, if you are a user of the market - a corporate that has forex risk to hedge - it is clear that you need to keep careful track of NDF prices. The chart shows that the current collapse of the rupee - to below 45 to the dollar - was also driven, at least partly, by NDF activity, triggered by the 1-year arbitrage having reached 50 paise in the first week of October. Interestingly, even though the rupee has recovered a small part of its losses, the NDF arbitrage is still around 45 paise; and, unless it collapses over the next few days, we could see the rupee come under pressure again, once the quarterly monetary policy (Oct 25) is out of the way.

Granted the Finance Minister has put his - as yet, not inconsiderable - weight behind the rupee, but jawboning can't stand in the way of risk-free money for too long. The louder message for the government, and, in particular, the RBI, is that THE MARKET IS CRYING OUT FOR MORE DEREGULATION. In the bad old days, when the forex market was completely controlled, we had a flush and thriving havala market. Now, while there's no denying that the forex market is much less controlled, the fact that NDF liquidity has been growing dramatically to the point where it is now noticeably impacting the domestic market (and the risk management efforts of its users) makes it clear that it is time for further opening up. Global experience has shown that NDF liquidity shrivels away in currencies of countries that have allowed greater capital account convertibility to the point where currency hedging is fully available onshore.

On a broader basis, however, the NDF show raises further questions. Why, indeed, is this arbitrage available? Why are offshore investors bidding the rupee weaker than the domestic market? It [the offshore market] was quite happy holding rupees in the 43.50-43.80 range for a long time, judging from the fact that the arbitrage spread remained less than 20 paise for nearly six months since March.

Of course, by late August, the near-certainty of steadily higher U.S. rates started influencing global risk appetites negatively, which, coupled with India's steadily rising current account deficit - not a bad thing at all, mind you - could have created a short-term softness in investors' attitudes to India. Ergo, several global analysts started looking at a weaker rupee, which, of course, started driving perceptions in the NDF market.

So, where does it go from here? Well, as already pointed out, the arbitrage window is still quite wide - 1 year NDFs are at 45.95, while the 1-year forward is at just 45.50. Thus, unless the arbitrage suddenly collapses - and it is critical to watch this carefully - the downward pressure on the rupee will remain. Further, since markets always overshoot and since the market is still heavily short dollars - the Economic Times estimates that over $ 30 bn of imports remain uncovered [over what horizon?] - it is possible that some time in the next few months, we could well see even the 46 level taken out. Companies with short dollar positions should use any dip (to or below 45) as an opportunity to hedge out to 3 months.

However, it is hard to see these levels sustaining - note that not even the most pessimistic of global analysts are looking at any level beyond 46, and on a medium term, everybody, but everybody, is still bullish on the rupee. Thus, we would see any level north of 46 as a strong opportunity to buy rupees (sell dollars); in fact, it would be an excellent plan to set up ECBs and push the button if and when the rupee approaches/crosses 46.

currency markets view: rupee and majors

INR
Fortnightly movement: O-44.45 H- 45.45 L-44.45 C-45.08
Sentiment - nervous
Expected range for 1 Month: 44.75-45.75
Expected range for 3 Months: 44.50-46.25

All hell broke loose in the USD/INR market last week as the NDF arbitrage swiftly pushed the rupee below 45 to the dollar. It hit a 11-1/2 month low of 45.45 in frantic trading on the 10th, with most short dollar holders simply watching - significantly, the forwards did not move up sharply, confirming the fact that the rupee weakness was not being driven by domestic buying.

The finance minster stepped in verbally and there have been scattered reports of RBI intervention, which has cooled the market to around 45.20; however, the rupee remains vulnerable since the NDF arbitrage spread remains high.

Likely as not, market is waiting for tomorrow's credit policy before taking any fresh action, so we could see volatility increase later in the week. We, along with several other analysts expect a 25 basis point hike in the reverse repo rate, although there is a rather wide range of views on the subject. In our view, the cards are stacked against the rupee, unless Dr. Reddy gets really hawkish and pushes rates up by 50 basis points. We see this as unlikely, since his general approach has always been softly, softly - at least, after the blowout he caused in the rupee when he was Deputy Governor.

Nonetheless, the longer term outlook is for renewed rupee strength, based on the fundamental strength of the Indian economy. True, a strong economy means a higher trade deficit and this could prevent the rupee from expressing any intrinsic strength - however, it does appear that a somewhat wekaer rupee would reopen the floodgates of capital, both for the real economy and for speculative purposes.

We must be close to the bottom.

EUR
Fortnightly movement: O-1.2126 H- 1.2152 L-1.1874 C-1.1948
Sentiment - neutral but could improve
Expected range for 1 Month: 1.1875-1.2350
Expected range for 3 Months: 1.1650-1.2650

Last fortnight, the dollar pushed the euro all the way down to about 1.1875 thanks to growing expectations of continued Fed tightening, lower than expected US trade deficit in August and very strong net capital inflows of $91.3 bn into the US in the same month. However, benign US core CPI (despite a hysterically high headline number), an unexpected decline in US consumer sentiment, lower than forecast rise in US retail sales, and sharp declines in US industrial production and index of leading economic indicators, helped the euro to rebound strongly last Wednesday thru' Friday to about 1.2075 before a late session slump on Friday to just under 1.1950 reportedly on USD-bound inflows under the Homeland Investments Act.

While sentiment for the dollar remains positive, in the short term the market is clearly directionless and jumpy.

Expectations of further hikes in the Fed funds target rate to 4.5% or even higher in the next 3-4 FOMC meetings, combined with repatriations by foreign subsidiaries of American companies, could perhaps push the euro to new multi-month lows upto 1.1650 over the next few months, but, before this happens, we believe that a strong rally of 7 cents or more seems to be in the offing amid increasingly hawkish remarks from ECB officials and bullish technical divergences.

GBP
Fortnightly movement: O-1.7598 H- 1.7798 L-1.7391 C-1.7674
Sentiment: positive and likely to improve
Expected range for 1 Month: 1.7500-1.8100
Expected range for 3 Months: 1.7300-1.8300

Sterling fell to about 1.7390 within the first 3 trading of the last fortnight but then staged a smart recovery to about 1.78 following a rise in UK CPI to 2.5% y/y, the October MPC meeting minutes showing that a rate hike could be considered if high oil prices fed into the UK economy and an upbeat 0.7% rise in September UK retail sales. Last Friday's retreat of over a cent was again reportedly due to profit repatriations by US companies.

From a technical perspective, sterling could find support in the 1.7550-1.7600 area before rallying above 1.78 towards 1.83. However, even a dip below 1.7550 is unlikely to overshoot beyond 1.73 before an upward correction towards 1.83.

JPY
Fortnightly movement: O-113.68 H- 115.99 L-113.51 C-115.90
Sentiment: very negative (for yen) but may begin to improve
Expected range for 1 Month: 113.00-117.00
Expected range for 3 Months: 108.50-117.00

The dollar's rally against the yen continued almost uninterrupted except for a short-lived dip from about 115 to 113.75. While Bank of Japan has been hinting at an early end to its policy of quantitative easing, such a policy change doesn't seem likely till next year. Moreover, Japanese money supply growth in September surpassed that in August while Japan's current account surplus dropped sharply in August. These negative Japanese data enabled the dollar to surge to a 2-year high above 115 yen and stay over the key 115 yen level for an entire week.

So long as the dollars stays over 115, it could move up quickly to 117 yen before a correction of 7-9 yen sets in. Such a sharp dollar correction against the yen and other majors might ensue in the event of a decline in the US ISM service sector index below 50 or another round of Chinese yuan revaluation before President Bush's visit to China next month!