Jaimini Bhagwati: Is the rupee correctly priced? - I

It is opportune to examine the feasibility of setting up a public debt office in Delhi and later in the state capitals

It is usually agreed that getting prices right is a crucial step towards a market-driven cost-efficient economy. In this context, this two-part article will argue that we need to improve the price determination of our money, i.e. the Indian rupee (INR). Of course, it is readily conceded that in addition to getting prices right it is important that subsidies are explicitly directed towards the disadvantaged. In the first part of this article the internal price of the INR is discussed and the second part deals with the external price of the INR.

Government of India (GOI) and state government internal liabilities stem from market borrowings, small savings, provident funds, insurance and pension funds, trusts and endowments, reserve funds, deposits and advances, loans against special securities and other accounts. The outstanding stock of the Centre's internal liabilities as of end March 2005 amounted to 62.1% of GDP. Of this fraction, market borrowings amounted to 40.9% of GDP. The corresponding numbers, as of the same date, for state governments were 33% and 6.8% (Source: Budget Documents and Combined Finance and Revenue Accounts of GOI and State Governments). That is, the total stock of internal liabilities for the Centre and state governments amounted to 95.1% of GDP as of end March 2005 and out of this number market borrowings totalled 47.7%. Consequently, the remaining stock of the Centre's and state government liabilities (i.e. small savings, provident and pension funds, trusts), which were contracted at administered interest rates amounted to 47.4% of GDP.

The secondary markets for Indian fixed income securities are predominantly made up of central and state government securities. For example, during 2004-05, transactions in corporate debt securities in the wholesale debt market (WDM) segment of the National Stock Exchange (NSE) amounted to only about 2% of the total trading turnover. In terms of trading in government securities during 2004-05, GOI dated securities constituted 92% of total trading while treasury bills amounted to 7% and the remaining 1% was the share of state government securities. It follows that if GOI debt securities are assumed to be free of default risk, sovereign domestic interest rates are the closest to market measure of the internal price of the INR. Of course, these interest rates are more a reflection of "lease" prices rather than outright purchase.

The primary market for GOI debt has come a long way since 1992-93, when we began moving away from an entirely administered interest rate system in which the government's debt was automatically monetised by the Reserve Bank of India (RBI). Currently, primary dealers (PDs) participate in bidding at GOI debt auctions, thereby setting "market-determined" interest rates for dated securities. However, the dominant participants in these auctions are the majority publicly owned institutions such the Discount Finance House of India (DFHI), now the SBI-DFHI, public sector banks, and public financial institutions. The RBI has promoted secondary market trading in GOI securities by setting up an electronic platform called the Negotiated Dealing System (NDS). However, membership in NDS is restricted to entities which have Special General Ledger (SGL) or current accounts with the RBI. To that extent direct retail participation through brokers is excluded.

A basic pre-requisite for interest rates to be market-determined is for the corresponding yield curve to be arbitrage-free. The reliability of government interest rates, as a proxy for the price of the INR, is, therefore, dependent on whether the sovereign yield curve is arbitrage-free.

The GOI's yield curve would be arbitrage-free if riskless profits cannot be made by simultaneously buying and selling GOI dated securities across the maturity spectrum. For example, a 3-year forward starting interest rate can be locked in by buying and short selling 5-year and 2-year bonds, respectively. If the 3-year forward combined with a 2-year spot starting security provides a higher/lower yield as compared to a 5-year spot, starting bond arbitrage profits would be possible. The GOI's yield curve exhibits such arbitrage profit opportunities from time to time. However, short selling GOI securities, except for intra-day transactions, is not permitted and hence even if there are opportunities for arbitrage profits these cannot be realised. It could, therefore, be argued that effectively the GOI yield curve is arbitrage-free. The flaw in this reasoning is that if arbitrage opportunities remain embedded in longer maturity interest rates, these rates do not reflect the market price of the INR.

Bond yields and therefore implied INR prices can be locked in over long periods of time only if the long-term secondary bond markets are adequately liquid. That is, there is regular and sizeable issuance at maturities up to 30 years. The GOI's bond offerings above 20 years' maturity are invariably lower in volume than those issued at shorter maturities. This is a reflection of lower investor appetite at the 30-year maturity, which means we need to look for additional investor categories. A potentially new class of investors could be tapped through issuance of Eurobonds in INR. INR Eurobonds could be marketed internationally to investors with directional views on Indian interest and exchange rates and help, at the margin, towards completing our long-term bond markets.

In most OECD countries the functions of sovereign debt issuance and liability management are entrusted to a specialised body separate from the central bank and government, e.g. a public debt office (PDO). In India, the RBI is responsible for issuance and liability management of the GOI's internal debt. The underlying reasoning is that in our current fiscally stressed situation the issuance-management of GOI liabilities cannot be assigned to an autonomous body.

In the context of the benefits of getting the price of the INR right, it is opportune to examine the feasibility of first setting up a PDO in Delhi and later in state capitals. These PDOs could start with back-offices with data recording and accounting systems. Subsequently, middle-offices and finally front offices could be added to analyse, issue and manage GOI and state government liabilities. A significant spin-off benefit would be that the implicit subsidies in our administered interest rates would be made explicit. Additionally, specialised capacity would be built and expertise would be decentralised. To begin with, the PDOs could be directly accountable to the RBI and government. Over time the PDOs could then be distanced to add value through "independent" evaluation of the technical aspects of sovereign liability management.