On
the surface, an early correction of the baseless expectations
of a few financial investors should not be cause for
worry. Few can demonstrate that India's till-quite-recently
inactive financial markets drive the economy. However,
the problem is that the crises that such changed expectations
resulted in elsewhere in East Asia, Latin America, Eastern
Europe and Turkey show that they tend to damage the
real economy as well. Moreover, when the crisis of finance
becomes a crisis of the real economy, its burden falls
disproportionately on the poor and the lower middle
classes, even though they do not participate in or often
are not even conscious of the workings of financial
markets. Having invited foreign financial investors
into their economies, countries find that ignoring their
sentiments and/or closing the door on them when they
retreat, involves painful adjustments that the people,
and therefore democratic governments, find hard to face
up to.
Thus, India seems to have a double problem on its balance
of payments. Net foreign exchange inflows are a problem
because they threaten an appreciation of the currency
and a decline in exports. Foreign exchange outflows
are a problem because they seem to result from even
non-economic factors, and could cumulate into a crisis
if they trigger a cycle of unfulfilled expectations.
The only redeeming feature is the high level of India's
reserves, which gives it space for manoeuvre.
The question is really the form these manoeuvres should
take. Clearly mere open market intervention to neutralize
the effects of excess foreign currency supply is inadequate.
What needs to be looked at is the possibility of managing
net foreign currency flows themselves. An examination
of the balance of payments offers some pointers on what
needs to be done in this regard.
The pressure on the rupee leading to its appreciation,
which is affecting export competitiveness adversely,
arises because India, which has recorded a current account
surplus since financial year 2001-02 (Chart 4), has
encouraged and attracted large inflows on its capital
account. India's current account surplus, we must note,
is not a reflection of its strong trade performance.
Rather, as Chart 1 shows, it is because net inflows
under what is called the ''invisibles'' head of the
current account of the balance of payments have been
more than adequate to finance a large and recently rising
merchandise trade deficit.
Chart
1 >>
The principal sources of current account inflows have
been buoyant remittance flows, captured under the ''Transfers''
head, and inflows on account of ''software services''
captured under the ''Miscellaneous Services'' head (Charts
2 and 3). That is, transfers made by Indian workers
abroad, either on short or long-term contracts and software
service exports, have helped overcome the adverse balance
of payments consequences of India's lack of competitiveness
reflected in a large trade deficit. Inflows on account
of software services rose from $5.75 billion in 2000-01
to $6.88 billion in 2001-02, $8.86 billion in 2002-03
and $9.09 billion over the first nine months of 2003-04,
while private transfers (mainly remittances) touched
$15.2 billion in 2002-03 and $14.6 billion during April-December
2003, after having fallen from $13.1 billion to $12.5
billion between 2000-01 and 2001-02. In an intensification
of this trend, during the first nine months of the recently
ended financial year 2003-04, net inflows on account
of invisibles was, at $18.22 billion, well above the
$15 billion deficit on the trade account.
Chart
2 >> Chart 3 >>
Even
while India's current account was relatively healthy
on account of the foreign exchange largesse of Indian
workers abroad and the software services boom, the country's
liberalized capital markets have attracted large inflows
of capital amounting to a net sum of $10.57 billion
in 2001-02, $12.11 billion in 2002-03 and a massive
$17.31 billion during the first nine months of 2003-04
(Chart 4). Expectations are that, because of huge portfolio
capital inflows during the last three months of 2003-04
encouraged by the government's privatization drive,
net capital account inflows during 2003-04 will be in
excess of $20 billion.
Chart
4 >>
There are two issues that arise in this context. The
first relates to the nature of the capital inflows during
these years. The second relates to the implications
of these inflows for the value of the rupee under India's
liberalized exchange rate management system. Three kinds
of inflows have dominated the capital account (Chart
5). An early and important source of inflow during the
years of financial liberalization has been in the form
of NRI deposits in lucrative, repatriable foreign currency
accounts. On a net basis, such inflows accounted for
$2.32 billion, $2.75 billion, $2.98 billion and $3.5
billion respectively in 2000-01, 2000-02, 2002-03 and
April-December 2003, respectively. They reflect the
attempt by richer non-residents to exploit arbitrage
opportunities offered by the higher (relative to international
rates) interest rates on repatriable, non-resident,
foreign exchange accounts, to earn relatively easy surpluses.
Chart
5 >>
A second important source of capital inflows has been
portfolio capital flows, reflecting investments by foreign
bodies, especially foreign institutional investors,
in India's stock and debt markets, encouraged more recently
by the disinvestment of shares in profitable public
sector undertakings. On a net basis, such inflows had
fallen from $2.8 billion in 2000-01 to $2.0 billion
in 2001-02 and just $979 million in 2002-03, but rose
sharply to $7.6 billion in the first nine months of
2003-04. As compared with this, net foreign direct investment
which rose from $4.0 billion in 2000-01 to $6.1 billion
in 2001-02 fell to 4.7 billion in 2002-03 and $3.2 billion
during April-December 2003.
The third important source of capital was a financial
liberalization-induced increase in the net liabilities
of commercial banks (other than in the form of NRI deposits),
which rose from a negative $1.43 billion in 2000-01
to $2.63 billion in 2001-02, $5.15 billion in 2002-03
and $2.56 billion during April-December 2003. This is
possibly explained by the expansion of the operations
of international banks in the country.
In sum, capital inflows that create new capacities either
in manufacturing or in the infrastructural sectors have
been limited. Much of the capital inflow has consisted
of financial investments that expect to earn higher
annual returns than available in international markets
or obtain windfall gains from the appreciation of the
value of such investments, as has recently been witnessed
in India's stock markets.
Given the determination of the exchange rate of the
rupee by supply and demand conditions in the market,
this large inflow of foreign capital in the context
of a current account surplus was bound to exert an upward
pressure on the rupee. When inflows contribute to an
appreciation of the rupee, foreign investors also gain
from the larger pay off in foreign currency that any
given return in rupees involves. This tends to increase
the volume of inflows. The real losers are exporters,
on the one hand, who find that the foreign exchange
prices of their products are rising, eroding their competitiveness,
and domestic producers, on the other, who find that
the prices of competing imports are falling or rising
less that their own costs of production.
If the government wants to manage these capital inflows,
it needs to control inflows on account of NRI deposits,
portfolio flows and banking, all of which are the results
of excessive financial liberalization, do not contribute
to enhancing productive investment, offer extremely
high returns that imply a net foreign exchange loss
to the country when matched by the low returns obtained
on accumulated reserves, and are extremely footloose
in character. India's current account position is such
that it does not need large capital inflows for a comfortable
balance of payments. In particular, it does not need
inflows that increase financial vulnerability without
contributing to any increase in productive investment
or exports. By doing away with the differential in interest
rates offered on non-resident external accounts and
international interest rates, the central bank has taken
the first step forward. It must now go further.
May
05, 2004.
|