Capital controls and unhedged currency exposure among Indian firms

Currency mismatches have devastated corporate balance sheets in many countries, and thus precipitated macroeconomic crises. How can these be avoided? The Indian evidence shows that an elaborate system of capital controls did not prevent firms from taking on unhedged currency exposure when they desired it. A macroeconomic environment with a flexible exchange rate is the key to reshaping the incentives of firms and thus avoiding crisis.

The imposition of capital controls as a means for achieving financial stability has seen renewed debate among policy makers and researchers in recent times. One possible rationale for capital controls in emerging markets is the financial fragility of firms. It is argued that incomplete financial markets in emerging economies limit the capacity of firms to hedge their balance sheet mismatches. Hence, it is argued that government has to protect firms. Central banks address this phenomenon that can have economy wide effects, primarily in two ways. One, they impose capital controls in the form of restrictions on foreign currency borrowing, and two, they intervene in currency markets to reduce currency flexibility.  

In India, RBI has been doing both. Restrictions have been imposed on foreign currency borrowing by firms, who can only borrow after the approval of the RBI under rules specified for External Commercial Borrowing (ECB). Both the maturity of the loan and interest rates for loans are prescribed by the RBI. Short term borrowing is banned. There is an implicit understanding that when the aggregate borrowing by all Indian firms reaches a certain amount, say USD 20 billion or so, further approvals are not forthcoming during that year. In addition, the RBI has intervened in foreign exchange markets with a view to reducing rupee dollar volatility for most of the period since 1993, when the rupee became `market determined'.

When the central bank acts to reduce currency volatility then it could end up providing an incentive for agents to take currency exposure. When the central bank pegs the exchange rate, low volatility against a target currency gives firms an implicit guarantee against short term movements of the exchange rate. The optimisation of firms leads them to desire foreign borrowing under such conditions.

Once the incentives for taking on unhedged foreign currency exposure are in place, the question becomes one about the effectiveness of capital controls. Capital controls are generally imperfect. Once a firm wants to achieve a transaction, smart lawyers and financial engineers will generally find a way past the bureaucrats manning the capital controls. The key issue is the incentives of firms.

Indeed, the relationship may hold both ways. Unhedged firms are likely to engage in political lobbying in favour of free public sector risk management. A central bank could hence choose the exchange rate regime depending on the extent of currency mismatches. When pre-existing currency mismatches are in place, this could push a central bank towards pegging. In turn, the optimal response of firms to the implicit guarantee offered by the central bank is to carry unhedged currency exposure. These two processes could reinforce each other in an equilibrium with unhedged currency exposure of firms and pegged exchange rates.

Empirical studies using firm level data find some support for the moral hazard hypothesis for Latin America, Mexico and East Asia.  Moving to more flexible regimes reduced currency mismatches on balance sheets of firms in Latin America. Firms in East Asia were less hedged under pegged exchange rates.

In India the additional element in this story was capital controls. Even if low currency volatility provided the incentive to make bets on the rupee, the elaborate regime of Indian capital controls should have prevented firms from building up balance sheet mismatches. But has this happened? Have capital controls been able to prevent unhedged currency exposure among Indian firms? 

In a recent paper we examine this question (Patnaik and Shah, 2010). The rupee has been pegged to the USD in the period 1993-2008, but four distinct periods of significantly different currency volatility can be identified. The first period, beginning in 1993, when the rupee moved away from an administered exchange rate was a period of low currency volatility. This was followed by a period of high volatility during the Asian crisis. The third period witnessed low volatility, followed by the fourth of high volatility. The extent of market completeness, in terms of capital controls or access to currency derivatives, did not fundamentally change over the period 1993-2008.

Exchange rate exposure of firms is measured using standard market models employing stock price data. We find that in Period 1, when there was negligible currency volatility, firms had large exposures. This gave way to the high currency volatility of Period 2:  we find that currency exposure of firms became very low. In Period 3, currency volatility was low and firms responded with increased unhedged exposure. Period 4 witnessed an increase in volatility and a reduction in unhedged currency exposure.

In other words, firms responded powerfully to the implicit guarantees in the currency regime. When the government tightly managed the exchange rate, firms carried substantial currency exposure. When there was greater currency flexibility, firms switched to lower unhedged exposures.

Firms responded to the incentives offered by free public sector risk management. It is not clear through which legal mechanisms this was done considering the elaborate system of capital controls in place. Yet, it demonstrates that capital controls were not able to prevent firms from betting on the rupee.

This evidence has important policy implications. If the goal of policy makers is to avoid macroeconomic crises owing to currency mismatches taken on by the firms, the path lies in having greater exchange rate flexibility and not in capital controls.

• Login or register to post comments
• 244 reads
• Calendar
• Printer-friendly version
• Email this page