The Ghost of FCNR(A) Was Invited Through the Swap Window

Public policy rarely returns under its old name. That would invite comparison, memory, and, in extreme cases, accountability. It reappears instead as a temporary facility, with a new acronym and distinctions between legal form and economic substance.

The Reserve Bank of India’s latest FCNR(B) window demands more than admiration for its ingenuity. It offers an at-par dollar-rupee swap—effectively bearing the full hedging cost on principal—against fresh three-to-five-year deposits mobilised until September 30. It has also temporarily withdrawn the interest-rate ceiling. The purpose is straightforward: attract foreign currency, strengthen the balance of payments and support the rupee without formally targeting an exchange-rate level. 

Why the arrangement attracts money is simple. Remove a high cost from a deposit product, and the product becomes more attractive. Banks can share some of the benefit with depositors while retaining a cheaper source of term funding. Non-resident depositors earn higher foreign-currency return without taking rupee risk. The RBI receives dollars, while banks avoid the market cost of hedging principal.

This is commonly described as a win-win arrangement. That description is almost complete. It leaves out the institution providing the win.

The academic provocation is that FCNR(A) has returned through FCNR(B).

Legally, it has not. Under FCNR(A), banks mobilised foreign currency but the RBI managed the funds and carried the exchange-loss burden; the Government later inherited that burden. FCNR(B), introduced in 1993, was the institutional correction: banks reclaimed the liability and discipline of pricing its risks.

The depositor’s claim is on the bank, not the RBI. Banks still carry liquidity, credit, operational and residual interest-rate risks. The present facility is limited by maturity and time; it does not convert the entire FCNR(B) book into a central-bank liability. A legal opinion declaring FCNR(A) restored would therefore be brief and wrong.

An economic opinion can afford to be more troublesome.

The defining difference between the regimes was not typography but the allocation of foreign-exchange risk. Once the RBI bears the full hedging cost on qualifying principal, the bank remains the contractual borrower, but the least attractive part moves to the central bank.

A zero-cost swap does not guarantee an RBI cash loss. But the subsidy exists at inception: the difference between the market cost of the hedge and the price charged to the bank. That value has not disappeared. It sits on the public balance sheet.

This is where the ghost of FCNR(A) begins to look employed rather than merely remembered.

The 1992 crisis exposed the fragility of confidence-sensitive NRI funding. Writing in the IMF’s Finance & Development, economist Ranjit Teja noted that NRIs withdrew deposits, bank borrowing dried up, and India’s readily usable foreign-exchange reserves faced near-complete depletion. Teja was describing the wider NRI run, not FCNR(A) alone. FCNR(A), however, added a more dangerous asymmetry: the depositor could run; the exchange guarantee could not.

The 1991 devaluation showed what that allocation could have meant. An RBI 2003 Occasional Paper records that FCNR(A) losses drew down the Exchange Fluctuation Reserve and reduced the Contingency Reserve to ₹8.59 billion by end-June 1993. The exchange-loss liability shifted to the Government, followed by several billion rupees of related transfers. FCNR(A) was retired not because it failed to attract deposits, but because the guarantee became expensive for the guarantor.

The lesson was not merely that exchange guarantees required better footnotes. It was that the foreign-exchange regulator should not write exchange-rate protection on bank liabilities. As RBI lore has it, S. S. Tarapore called the exchange guarantee one of the cardinal sins. The sin was institutional before it became fiscal: the RBI was simultaneously monetary authority, reserve manager, banking intermediary and bearer of currency risk.

Hidden Costs
The present facility is narrower—temporary, principal-only and confined to fresh deposits. These limits constrain scale but not direction. Central-bank swaps can legitimately provide liquidity at appropriate prices. This is different: a three-to-five-year off-market hedge designed to favour one funding channel. That is risk transfer and credit allocation through the central-bank balance sheet.

It also creates an uncomfortable combination of roles. The RBI regulates the foreign-exchange market while assuring the public that the rupee is market-determined. It is now a counterparty whose balance-sheet exposure is affected by that exchange rate. The point is not that the RBI will manipulate the rupee. The point is that sound institutional design does not create such an incentive and then ask that it be ignored.

The debt is not invisible: non-resident deposits are recorded as external debt of deposit-taking corporations. What sits elsewhere is the subsidised public hedge. The bank liability is visible; the central bank insurance is not in the same amount. The first leg strengthens headline reserves; the second creates future dollar commitment. The RBI reports aggregate forward position, but not scheme-specific notional exposure, market value, maturity concentration or subsidy.

The debt is counted. The hedge is aggregated. The reserves are celebrated weekly.

Nor must one wait for maturity to identify the cost. The market charges for a three-to-five-year hedge; the RBI has reduced that price to zero. The subsidy is certain, even if the eventual accounting loss is not. A depreciating rupee makes the cost visible; a favourable outcome does not turn the concession into a market transaction.

Revaluation accounts and contingency buffers determine when the cost is recognised; they cannot change who bears it. Consequences may appear as weaker buffers, lower surplus transfers to the Government, or fiscal support. The RBI’s balance sheet is a national shock absorber. It should not be pledged casually to improve commercial-bank funding economics.

There is a monetary consequence too.

When the RBI takes dollars and supplies rupees, it injects liquidity. At roughly $10 billion of reported inflows, the gross rupee leg is already close to ₹1 trillion. Leave the liquidity in place and conditions ease; sterilise it, and the RBI bears a carrying cost. Either way, external-sector intervention acquires a monetary-policy footprint while bank credit already grows faster than deposits.

The balance-sheet exposure, liquidity injection, reserve presentation and moral hazard are not four unrelated objections. They follow from one decision: the RBI has reassumed a risk that FCNR(B) was designed to leave with banks.

The banks’ incentives deserve equal scrutiny. Removing the principal-side hedge lowers the rupee-equivalent cost of the liability. The benefit can be passed to depositors, retained as margin or used to price domestic credit more aggressively. The subsidy may end as credit whose price no longer reflects underlying funding costs.

The leverage provisions deepen concern. RBI clarification permits loans, liens, and standby letters of credit against qualifying deposits, while banks have sought to finance them through GIFT City branches. Some inflows may be leveraged carry structures rather than unencumbered diaspora savings. Before celebrating every dollar, the RBI should disclose the leveraged share, same-group financing, beneficial-ownership checks and safeguards against circular or fronted transactions.

Mandate Creep
The deeper moral hazard is straightforward. FCNR(B) was designed to make banks price and manage the currency risk of the deposits they mobilised. The present window rewards them for raising the same liability after removing its most consequential market cost. A regulator that repeatedly rescues one funding channel from its own price signals should expect banks to rely on that rescue again.

There is also a simpler question. The RBI had already withdrawn the interest-rate ceiling. Why was that insufficient? Banks could have paid more for deposits while bearing the cost of their own hedge. Before socialising a market risk, the central bank should show that less intrusive tools were inadequate. Instead, the market has been given higher coupons, hedges and permission to borrow the subscription.

The 2013 precedent will be offered as defence: the concessional window mobilised tens of billions of dollars and helped stabilise the rupee. The present facility may work too. But a subsidy designed to attract dollars is evidence of arithmetic, not necessarily policy wisdom. “It worked” answers mobilisation; it does not answer mandate, incidence, opportunity cost or precedent.

India’s external vulnerability does not arise merely from insufficient NRI deposit appetite. It reflects the current account, oil dependence, portfolio volatility, the composition of capital inflows and the credibility of the macroeconomic framework. FCNR money can buy time. It cannot substitute for reforms that make time valuable.

At a minimum, the RBI should publish scheme-specific notional exposure, market value, subsidy against prevailing forward rates, maturity schedule, concentration by bank, liquidity injected and sterilised, and the share supported by loans or standby letters of credit. Those disclosures would not make the intervention prudent. They would allow the public to decide whether it is.

The issue is no longer whether an FCNR(A)-like feature has reappeared. It is whether the RBI has crossed an institutional boundary that FCNR(B) was designed to establish, without demonstrating necessity, proportionality or safe limits.

FCNR(A) has not returned as law. It has returned as temptation: the belief that foreign-currency debt becomes safer by shifting its inconvenient risk to the central bank.

The letter B still identifies the bank that owes the depositor. The swap identifies the institution that made the liability commercially attractive. When the RBI picks up the hedge, the alphabet is not the most informative part of the scheme.

The ghost did not force its way into Mint Street. It was invited through the swap window.