Doing everything wrong on debt inflows

Business Standard, 5 April 2006

There is a deeply held belief that debt inflows are bad, that all policy levers should be used to block debt inflows. These positions are incompatible with economic logic, and India's policies on debt flows are riddled with mistakes.

The "original sin" of many developing countries is the issuance of bonds in foreign currency, also called "liability dollarisation". Such a borrower suffers from "currency mismatch" - earning in rupees but repaying the debt in dollars. A sharp rupee depreciation can induce bankruptcy through unexpectedly high cost of debt servicing. The same happens for a government that earns taxes in rupees and repays foreign loans in dollars.

Currency mismatch is not innately bad. What leads to trouble is bad decision making in the context of currency mismatch. Governments are notorious for making mistakes in the quest for short-term political gains. Hence, the issuance of foreign currency debt by a government is considered a bad idea. The next prominent mistake is dollar borrowing by banks. Banks are highly leveraged, with a debt-equity ratio of 20:1. The slightest mistake leads to a bankrupt bank. Banks in immature countries are poorly regulated and underwritten by the government, so foreign currency borrowing by banks is fraught with problems.

A key source of bad decisions is the currency regime. When a central bank tries to create an illusion that the currency is not volatile, it encourages firms, banks and the government to be reckless in dollar borrowing. Borrowers tend to not hedge currency risk. But no central bank is able to sustain this illusion. Every now and then, the illusion of calm inevitably breaks down, with sharp currency movements which have been pent up. At this time, borrowers in dollars can run into trouble.

Most currency crises in history have been spawned by this toxic combination of a fixed / pegged exchange rate, which gives out false perceptions of currency risk, coupled with unhedged dollar borrowing by governments, banks or firms. This story - about how "original sin", coupled with an inflexible exchange rate leads to damage - is more nuanced than a dogmatic insistence that all debt inflows are bad. In India, we have policy mistakes on all aspects of debt inflows.

Sound policies require that the government should not have dollar liabilities. But de facto sovereign dollar borrowing is taking place through devices such as bond issues by SBI, or NRI deposits of banks. No large bank in India has been allowed to go bust, so NRI deposits with banks are effectively a sovereign dollar liability.

FII investment into rupee-denominated bonds is perfectly safe. The foreign investor bears all the currency risk! There is no "original sin" when an FII buys a rupee bond. It is the dream of every emerging market, to graduate out of "original sin", with local currency bonds accepted into global portfolios. E.g. in 9/2005, Brazil graduated into local currency globalised issuance of government bonds, to a round of global applause. But in India, we have emphasised NRI deposits and imposed harsh restrictions on FII ownership of rupee bonds.

Sound policies require that firms should be discouraged from dollar borrowing until the currency regime and the currency derivatives are in good shape. But we do this exactly wrong. FII investment into rupee corporate bonds - which is not original sin - has been harshly restricted. The rules have pushed firms to borrow abroad through ECB, which is liability dollarisation.

Further, ECB often involves banks as intermediaries. But banks are acutely leveraged and poorly regulated. It is better if a firm borrows directly - by having an FII buy corporate bonds in India - rather than borrowing through a bank.

Our policy framework induces firms, banks and GOI to make mistakes on hedging of dollar debt, because the rupee is pegged to the US dollar. A prodigous effort is made to generate a false perception of currency risk. Borrowers are encouraged to believe this risk is not there. But as we know from international and Indian experience, this risk is very much there. The INR/USD will regularly lurch into episodes of high volatility, causing pain to unhedged dollar borrowers. RBI tries to have "rules requiring hedging" but this sort of micro-management is neither desirable nor feasible.

Our policy framework has blocked the development of sound currency risk management markets, which requires futures and options that are traded on exchanges. India has a world-class futures and options market on the NSE-50 index. But there is no high quality market for futures and options on exchange rates. This damages the currency risk management which is feasible in the country.

Every element of this policy framework is imprudent. Economic agents in India are being pushed into risky avenues, being given false signals encouraging them to not hedge, and being deprived of sound tools for hedging. Risk-averse policy making, which genuinely improves "financial stability", involves five pieces:

  1. A currency regime with greater flexibility, which does not emit a false perception of safety.
  2. Elimination of NRI deposits, and programs like MIB.
  3. Tight limits for foreign-currency borrowing by firms (ECB). A limit on the gross inflow at 1% of GDP every year might make sense.
  4. Removal of all limits for FII investment into rupee-denominated government bonds or corporate bonds.
  5. Launch of rupee-dollar futures and options at NSE and BSE.

This remarkable situation - where a dogmatic insistence that debt inflows are destabilising actually conceals a stream of destabilising policy mistakes - illustrates why removing controls makes sense, and why India should have capital account convertibility. As a general principle, high minded arguments can be made about market failures which should justify controls. Sophisticated arguments about market failures can be trotted out on trade, industry or capital flows. Nuanced policy packages such as points 1 to 5 can be concocted.

But the practical reality is that governments are incapable of nuance. The controls that are actually found in the country are seldom illuminated by economic logic. As we have seen with trade, industry and capital flows, the controls found in the real world reflect some mixture of politics, rent-seeking and ignorance. Liberal policies are superior not because the market is always right, but because the mistakes of the market are smaller than the mistakes of the control raj.

Back up to Ajay Shah's 2006 media page
Back up to Ajay Shah's home page