Follow us on

|

New signs of vulnerability

Print edition : Oct 07, 2005

Comments

T+T-

A combination of rising oil prices and dramatic increases in non-oil imports is widening the merchandise trade deficit. Without capital inflows being directed to build productive capacities, India risks facing the next Asian financial crisis.

SINCE the euphoria over the BSE Sensex breaching one more psychological barrier, the 8000 mark, preoccupies the media, new signs of economic vulnerability remain unflagged and ignored. According to the latest trade statistics released by the Directorate-General of Commercial Intelligence and Statistics relating to the first five months of this financial year (April-August), the deficit in India's merchandise trade stood at $17,431.2 million as compared with $9,728.5 million during the corresponding period of the previous year. If this 80 per cent increase in the deficit persists over the rest of the year, the trend could take India's trade deficit to close to $50 billion over the financial year 2005-06.

It could be argued that such an increase was inevitable, given the sharp increase in the international prices of oil, which was and is expected to increase India's oil import bill substantially. Indeed, over the first five months of this financial year, oil imports rose in value by close to 37 per cent from $12,002 million to $16,428 million. However, what is noteworthy is that over the same period non-oil imports also rose by a similar 37 per cent from $26,803 million to $37,763 million. In the event, despite a creditable 23 per cent increase in the dollar value of India's exports during April-August 2005, the trade deficit has widened substantially. Even if the increase in the oil import bill is seen as temporary because oil prices must moderate and even fall, the same cannot be said of the non-oil import bill. Clearly, import liberalisation has meant that any buoyancy in the economy, even if it is not focussed on the commodity producing sectors, results in import bill increases that match those generated by events like the current oil shock. If that increase has to be moderated or reversed for any reason, lower economic growth must be the price that has to be paid.

The full significance of this trend comes through when we note that one comforting feature of India's balance of trade between 2000-01 and 2003-04 has been the surplus on the non-oil merchandise trade account (see chart). That surplus helped partially moderate the effects of a rising oil trade deficit, which rose sharply between 2001-02 and 2004-05, partly because of a gradual increase in oil prices and partly as a result of dramatic increases in the domestic consumption of oil and oil products.

However, in 2004-05 the non-oil trade balance was once again negative, removing the partial cushion offered by the trade in non-oil products against the effects of a rising oil trade deficit at a time when the rise in oil prices was sharper. What is happening is that, in a period when oil prices have registered particularly sharp increases, the non-oil import bill has kept pace with the oil import bill, resulting in a massive widening of the deficit on the merchandise trade account. It is of course true that even during the previous two financial years, the widening deficit on the trade account was not a cause for concern because of significant inflows of foreign exchange on account of remittances and exports of software and IT-enabled services.

According to the Reserve Bank of India, private transfers brought in a net amount of $20.5 billion in 2004-05 and software services exports contributed another $16.6 billion. This net inflow went a long way towards financing India's foreign exchange requirement in that year on account of the merchandise trade deficit and the deficit under other items of what are termed "invisibles". As a result, the deficit on the current account of the balance of payments was relatively small. Since India has also been a net recipient of substantial capital inflows on account of debt and foreign direct and portfolio investment, this led to a huge accumulation of foreign exchange reserves that implied a comfortable balance of payments situation.

It now appears that India's relatively strong current account position is weakening rapidly. As noted above, a combination of rising oil prices and dramatic increases in non-oil imports is resulting in a substantial widening of the merchandise trade deficit. Simultaneously, there is evidence that recent increases in remittance inflows are tapering off. Net remittances, which rose from $16.4 billion in 2002-03 to $22.6 billion in 2003-04, was down to $20.5 billion in 2004-05. While net revenues from software services continue to increase, from $8.9 billion in 2002-03 to $11.8 billion in 2003-04 and $16.6 billion in 2004-05, the current account deficit can be expected to widen because of the other two developments.

Consequently, a greater share of the net capital flows that India attracts in the form of debt and foreign direct and portfolio investment would now be needed to finance the current account deficit. This would be perfectly acceptable if these capital inflows were being used to build productive capacities that can support exports and earn the foreign exchange needed to meet future foreign repayment commitments that today's inflows imply. That, however, is clearly not happening. Portfolio flows create no additional capacities, though Foreign Institutional Investors (FII) investments drive the current stockmarket boom and create the euphoria that explains the lack of concern about potential external vulnerability. And, to the extent that foreign debt and direct investment inflows are indeed creating new capacities, they are not generating export revenues to finance the rising non-oil and oil import bill.

This is not surprising. It has been clear for some time now that unlike what occurred in the late 1980s and early 1990s in second-tier East Asian industrialisers like Thailand and Malaysia, and very much unlike what has been happening in China for close to a decade-and-a half now, "non-financial" investments financed with foreign capital in India have not been directed at greenfield projects that contribute to an expansion of exports. Rather, they have principally been: (i) directed at increasing the share of foreigners in firms they already control as a consequence of the relaxation of ceilings on foreign holdings in domestic joint ventures catering to the domestic market; (ii) used for acquisitions of local firms that provide foreign investors with a share in the domestic market for a range of products; and (iii) concentrated in greenfield projects in infrastructural services such as power and telecommunications, which in any case are sectors that produce "non-tradables", or services that are not normally exported to foreign markets.

The only area in which an increase in foreign presence involves export revenues as a rule is the software and IT-enabled services sector. But, even though export revenues from this sector have been rising rapidly, the sector is still too small to make up for the foreign exchange profligacy of the rest of the economy. Overall import liberalisation, combined with a concentration of incomes in sections of the population with a significant pent-up demand for imported or import-intensive goods, has resulted in an excess of demand for foreign exchange relative to current account earnings.

THE incipient tendency towards external vulnerability that this entails has thus far been ignored for two reasons. First, India's exports have been performing better in recent years than they did in the past. Second, India has been such an attractive destination for foreign financial investors that inadequacy of foreign exchange has become a feature of a rarely remembered past.

Except in 2001-02, when India's exports declined marginally, exports in dollar terms have been rising at over 20 per cent per year over most years of this decade. This has been the focus of statements by Commerce Ministry spokespersons. As when any reference is made to import growth, a rise in the import bill is presented more as evidence of recovery in the industrial sector, rather than as a cause for concern because that rate has implications for the merchandise trade deficit.

Implicit in this view is the belief that a trade deficit does not matter, since invisible revenues ensure that a rise in the trade deficit does not automatically translate into a rise in the current account deficit and that, even if it does, capital flows are more than adequate to cover the likely increase in the current account deficit. Recent experience has shown that the import surge is such that even with reasonable export growth this view is no longer true. What is more, periodic currency crises elsewhere in the world suggest that reliance on purely hot money flows to finance such a current account deficit is by no means a sensible strategy.

There is a more fundamental problem here. The success of any liberalisation strategy depends in the final analysis on the realisation of a rate of export growth that can deliver growth without balance of payments problems that are structural. This makes comparisons of the rate of export growth over time meaningless. Allowing for a reasonable lag, what is needed is a rate of export growth that, at any point of time, covers the increase in imports that liberalisation involves as well as generates the revenues needed to meet commitments associated with capital inflows. It would be absurd to use more capital inflows to cover past capital flow commitments, since this involves a spiral of dependence on capital inflows. Such dependence implies even greater fragility if capital flows are of a kind that are footloose and investors can exit the country with as much enthusiasm as they showed when they entered.

What the evidence on India's trade trends suggests is that even as dependence on volatile capital flows increases, an export growth rate that is presented as creditable appears increasingly adequate to cover the surge in non-oil imports. Add on a surge in the oil import bill and that inadequacy is all the greater. This implies that the dependence on volatile flows to sustain the balance of payments is rising. If the current boom in the stock market reaches its inevitable peak, then not only will new capital flows dry up but past capital flows would seek to exit the country. That is a denouement that must be avoided if India is not to follow the example of "emerging markets" like Mexico, South Korea, Thailand, Indonesia, Malaysia, Brazil, Turkey and Argentina. If it does, then it could be the next case where a financial crisis can be the means to ensure neo-colonial conquest of a country whose elite sees itself as representing a rising global power.

Comments

Comments

Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide to our community guidelines for posting your comment