Treasury Bills are financial instruments used by Government of India for short term financing. They are generally issued at some discount to face value and are therefore also called discounted securities. When we subtract the issue price from the maturity value we get the return on investment.
Basically te treasury bills are classified on the basis of their maturity period like, ad-hoc Treasury bills, 3 months, 12months Treasury bills etc. In India, there used to be 14-days, 91-days, 182-days and 364-days Treasury bills but at present only 91-days and 364-days T-bills are traded in market. They are traded at both primary as well as secondary markets.
1.1 Benefits of investing in T-bills
No tax deducted at source
Zero default risk being sovereign paper
Highly liquid money market instrument
Better returns especially in the short term
Simplified settlement
High degree of tradeability and active secondary market facilitates meeting unplanned fund requirements.
Market related yields
Ideal matching for funds management particularly for short term tenors of less than 15 days.
Transparency in operations as the transactions would be put through Reserve Bank of India's SGL or Client's Gilt account only.
Two way quotes offered by primary dealers for purchase and sale of treasury bills.
Certainty in terms of availability, entry & exit.
When to switch away from T-bills
But treasury bills are not always the best investment option to park the surplus funds. A study done by Ruben Trevino and Barbara Yates in 2003 examines the question of when it might be advisable to switch from money market instruments (such as Treasury bills) to the typically higher yielding short-term bonds. Specifically, the study tests whether current yield levels and current yield spreads are useful in predicting subsequent differences in one-year returns between investing in five-year T-bonds and three-month T-bills.
Using the period of 1954 to 2003, the study finds that while the prior yield level is not consistently related to the relative performance of short-term T bonds compared with T-bills, the size of the prior yield spread does predict the odds of short-term T-bond returns outperforming T-bill returns.
When the spread is positive and greater than 200 basis points, there is a 70 percent chance that short-term T-bond returns will be higher than T-bill returns.
When the yield spread is negative (inverted yield curve), the odds are clearly in favour of T-bills; there is a 75 percent chance that T-bill returns will be higher than short-term T-bond returns. For spreads between 0 and 200 basis points, the odds are 50-50 that short term T-bonds or T-bills will provide higher returns.
The historical evidence also suggests that the reason spreads can predict relative performance between short term T-bonds and T-bills is because they anticipate, to some extent future changes in interest rates.
1.3 Portfolio Management Strategies
Strategies for managing a portfolio can broadly be classified as active or passive strategies.
Buy And Hold: A buy and hold strategy can be described as a passive strategy since the Treasury bills once purchased, would be held till its maturity. The salient features of this strategy are:
Return is fixed or locked in at the time of investment itself.
The exposure to price variations due to secondary market fluctuations is eliminated.
There is no risk of default on maturity.
Buy And Trade: This strategy can also be described as an active market strategy. The returns on this strategy are higher than the buy and hold strategy as the yield can be optimised by actively trading the treasury bills in the secondary market before maturity. This strategy involves altering the average duration of t-bills in a portfolio depending upon the portfolio manager's expectations regarding the direction of interest rates. If T-bill yields are expected to fall, the portfolio manager would buy the T-bills with longer duration and sell T-bills with shorter duration, until the portfolio's average duration becomes longer than the market's average duration.
2. Auctioning of T-Bills
2.1 Primary Market
In the primary market, treasury bills are issued by auction technique. Salient features of the auction technique are shown in the table below.
T-Bill
Notified Amount (Rs. Crore)
Day of Auction
Day of Payment
91 day
500
Every Wednesday
Following Friday
364 day
1000
Wednesday to coincide with reporting Friday
Following Friday
The auction of treasury bills is done only at Reserve Bank of India, Mumbai.
Bids are to be submitted on NDS by 2:30 PM on Wednesday. If Wednesday happens to be a holiday then bids are to submitted on Tuesday.
Bids are submitted in terms of price per Rs 100. For example, a bid for 91-day Treasury bill auction could be for Rs 97.50.
Auction committee of Reserve Bank of India decides the cut-off price and results are announced on the same day.
Bids above the cut-off price receive full allotment; bids at cut-off price may receive full or partial allotment and bids below the cut-off price are rejected.
2.2 Auction of T-Bills in India
Auction, in a greater sense is a price building method by allowing competitive bidding.In this process the seller gets the general market acceptance on the price / value of the asset to be sold. Generally the most active participation comes from banks and insurance companies because they need these to comply with their statutory requirements but primary dealers also participate in the auction not for the purpose of statutory requirement but for the purpose of market making and for trading in the secondary market.
2.4 Secondary Market
The major participants in the secondary market are scheduled banks, financial Institutions, Primary dealers, mutual funds, insurance companies and corporate treasuries. Other entities like cooperative and regional rural banks, educational and religious trusts etc. have also begun investing their short term funds in treasury bills. The treasury bills are issued in the form of promissory note in physical form or by credit to Subsidiary General Ledger (SGL) account or Gilt account in dematerialised form. Minimum amount of bids for treasury bills are to be made for a minimum amount of Rs 25000/- only and in multiples thereof.
Treasury Bills serve as a very useful instrument when it comes to gaining return from short term excess liquidity. For example banks do not pay any interest on money in current accounts which can be used to buy short term treasury bills as compared to FDs for which banks do not pay any interest for less than 15 days. Here treasury bills which can be used helps to deployment idle funds even for very short periods. Also, every week a 91-day treasury bill matures and every fortnight a 364-day treasury bill matures. So treasury bills of different maturities can be purchased according to the requirements to match with the fund-flow requirements. Sometimes when there is a liquidity crunch, returns on treasury bills are higher than long term bank deposits as well.
Treasury Bills Outstanding (Face Value) (Rs. Crore)
3. T-Bill Index
The T-bills index aims to capture portfolio returns when a certain sum is invested in the short term instruments. The short term instruments have been gaining importance as market participants are increasingly using these instruments for their treasury operations. In 2002-03, the T-bills constituted only 3.48% of the total outright trade in the market but in 2004-05, the same increased substantially to 21.75%. The increasing activity at the shorter-end of the market highlights the importance of T-bills in the current scenario. However, the market reality is that all sub-time bucket segments of the short term market are not equally liquid. As shown in the table below, the trading concentration on the residual maturity bucket of 121 to 240 days is the least followed by 1 to 60days residual maturity bucket. The market concentrates more on 61 to 120days.
A fixed duration index would provide equal weightage to all residual maturity buckets and hence would not take into account the liquidity aspect of the market. However, market participants may have use of the same to compare the performance of portfolio against certain benchmarks. Whereas in liquidity weighted index the trading behaviour of T-bills in various maturity buckets is considered. Each bucket is assigned a weight on the basis of trading volume.
The T-bills yields/prices of 30 days, 90 days, 180 days, 300 days and 361 days residual maturities would be used in the construction of the CCIL liquidity weight T-bills index.
Total Return Index is given by:
It = It-1* (1+TRt)
Where, TRi is the total return of the day t and It-1 is the Index on day t-1.
The total return is calculated from the components. In each component, Total return has two components - the investment yield for the day assuming holding for the day [(redemption price - T-bills price) / T-bills price * number of days] and market return due to change in price as we move from day t-1 to day t (Pt - Pt-1/Pt-1).
4. Determinants of Interest Rates
Based on the various models developed in the empirical literature we selected certain macroeconomic variables for our analysis. The variables and their impact on the interest rates are being discussed below:
(+) Spot Oil Prices: It is expected that as the Spot Oil prices go up the import bill will increase meaning that dollars flow out of the Indian economy. This will result in Rupee falling and the interest rates going up based on interest rates parity theory.
(+) Wholesale Price Index: As per the Fisher's interest rate theory investors want to be compensated for inflation so that they get a constant real rate of return. Hence the interest rates will rise.
(+) Index of Industrial production: This is a surrogate measure of the GDP. Since this figure is reported monthly it was used in place of GDP. Higher industrial production (which coincides with higher production in other sectors) would mean a higher expected demand for goods leading to higher demand for money and hence higher interest rates.
(-) Real Effective Exchange Rate: As the exchange rate rises the interest rate parity theory implies that interest rates should fall.
(+) Yield on 10-year U.S. government security: The Indian markets if assumed to be integrated with global markets should move in a similar direction. To prevent financial capital from moving out of India, when the U.S. Gsec10 rises, Indian interest rates would also rise.
(+) Bombay Stock Exchange Index: As the stock market prices rise money will flow into the stock market and demand for bonds will reduce. A lower demand for bonds will result in lower prices which would imply higher interest rates.
(-) Money supply: As the money supply in the economy will increase, easy availability of credit will be associated with lower interest rates.
Dua and Pandit (2002) studied behavior of short-term and long term interest rates. They established both VECM as well as VAR models for the prediction of interest rates and found that VAR models outperform VECM models.
This prediction is developed using a theoretical model based on Miller's (1977) original hypothesis and is tested using a sample of daily Treasury bill yields surrounding the May 1997 change in the capital gains tax rate.
Regressions on yields surrounding the effective date of the change in rates show that T-bill yields decreased with the increase in the capital gains tax rate for individuals. This decrease is consistent with the theoretical model and supports the prediction of an indirect effect.
These findings contribute to the accounting literature on taxes and investment pricing in several ways. Overall these results suggest shareholder-level taxes are impounded into prices in ways not previously considered and should be of interest to researchers investigating the effect shareholder-level taxes have on investment returns.
5. Forecasting of Interest Rates
Timely forecasts of interest rates can provide valuable information to financial market participants and policymakers. Forecasts of interest rates can also help to reduce interest rate risk faced by individuals and firms. Forecasting interest rates is also very useful to central banks in assessing the overall impact (including feedback and expectation effects) of its policy changes and taking appropriate corrective action, if necessary.
Univariate as well as multivariate models are estimated for each interest rate. Univariate models include Autoregressive Integrated Moving Average (ARIMA) models, and ARIMA models with Autoregressive Conditional Heteroscedasticity (ARCH)/Generalised Autoregressive Conditional Heteroscedasiticity (GARCH) effects while multivariate models include Vector Autoregressive (VAR) models specified in levels, Vector Error Correction Models (VECM), and Bayesian Vector Autoregressive (BVAR) models.
The following variables are included in the multivariate models for the Treasury Bill rate (15-91 days): inflation rate (year-on-year), bank rate, yield spread, liquidity, foreign interest rate (3- months Libor), and forward premium (3-months). In the case of the 15-91 day Treasury Bill rate, the VAR model in levels produces the most accurate short- and long-term forecasts.
It is also found that the multivariate models generally produce more accurate forecasts over longer forecast horizons. This is because interactions and dependencies between variables become stronger for longer horizons. For short forecast horizons, predictions that depend solely on the past history of a variable may yield satisfactory results.
ARMA model produces marginally more accurate forecasts compared to the ARMA-ARCH model. However, since the U statistic is greater than or close to 1 for all forecast horizons, the forecast performance is not superior to that of a random walk.
For all univariate models (including the random walk) there is deterioration in accuracy with an increase in the forecast horizon.
The LVAR model outperforms the VECM model consistently.
The LVAR model also beats the BVAR models in terms of forecast accuracy.
Performance of all BVAR models is reasonable and generally improves on loosening the prior. In the extreme case, with a very loose prior, the BVAR model converges to LVAR.
Therefore, for the 15-91 day Treasury Bill rate, the LVAR models produce the most accurate short- and long-term forecasts.
ARMA model is generally more accurate than ARMA-GARCH.
LVAR model almost consistently outperforms VECM forecasts.
Performance of BVAR forecasts is satisfactory for short- and long-term forecasts and is almost consistently better than that of LVAR.
Of the BVAR models, the model with w= 0.2, d=1 and k=0.5 performs best.
All models are inaccurate for forecasts 16 through 22 weeks ahead. This can be attributed to the fluctuations in the interest rate from March to May 2002 (from 5.37 to 7.22%).
It is of primary importance to develop a model to forecast at least the direction in which the interest rates are likely to move. The study further says that a parsimonious multivariate model based on various macroeconomic variables culled from the literature outperforms a naïve univariate model. An accurate model can tell the banks about the future expected inflows and outflows during each of the time periods and hence ensure a proper match between the same.
Dua, Raje and Sahoo had developed both VECM as well as VAR models for the prediction of interest rates and find that VAR models outperform VECM models.
6. Conclusion
Treasury Bills are issued by GoI and traded at RBI which acts as exchange terminal. As T-bills are government backed securities, they are the safest instruments for organisations to park their surplus money for a very short term.
In India T-Bills are auctioned at primary markets and traded in secondary markets. The most popular and liquid T-bills in India are 91-day and 364-day T-bills. ARIMA, ARCH and GARCH are the univariate models while VAR, BVAR, LVAR and VECM are the multivariate models available for forecasting of interest rates for 15-91 days T-bills over short term and long term. Of these LVAR is the most accurate model.
Inflation rate (year-on-year), bank rate, yield spread, liquidity, foreign interest rate (3- months Libor), and forward premium (3-months) are the variables used in multivariate models for forecasting of interest rates. In addition to these some other variables which significantly affect the interest rates are spot oil prices, BSE index value, money supply and real forex rate.
