Budget and RBI : New Directions Address By Dr. Y.V. Reddy Deputy Governor Reserve Bank of India at Administrative Staff College of India, Hyderabad on March 8, 1997

Friends,

1. It is a pleasure for me to be with you in Administrative Staff College today. I have my Gurus here, and I also have many friends. I spent a year on full time basis about a decade ago as Visiting Faculty here. I am thankful to the Principal for giving me an opportunity to speak on an important subject of contemporary relevance.

2. Mr.Chidambaram's Budget of 1997 has been widely acclaimed as a path breaking budget. From the point of view of Reserve Bank of India, it has very significant implications. The Budget has acknowledged that interest tax is a bad tax and has commenced its phasing out. It has removed Tax Deduction at Source on Government securities which was an irritant in developing an active second ary market for Government Securities. Government has accepted RBI's proposal to launch a Capital Indexed Bond. In the external sector, policy of capital outflows has been liberalised, acknowledging that globalisation is not a one-way street. Replacing Foreign Exchange Regulation Act and designing a process of capital account convertibility are two of the announcements in the Budget which are very dear subjects to me, but, I will not comment on them. A Working Group and a Committee in RBI are working on them and I cannot preempt them. There is, however, one subject which has generated debate, sometimes based on misinformation if not misunderstanding. That is the elimination ad hoc and introduction of a new system of Ways and Means Ad vances (WMA). My predecessor, Dr. S.S.Tarapore who played a significant part in pursuing this idea has described this change over as a watershed in Indian monetary history.

   Origin of Ad-hocs

3. The intellectual origins of ad-hocs in its present form can be traced to the First Five Year Plan. It said, 'Judicious credit creation somewhat in anticipation of the increase in production and availability of genuine savings has also a part to play'. Thus, deficit financing which in our context meant RBI credit to Government was assigned a place in the financing of the Plan though its quantum was to be limited to the extent it was non-inflationary.

4. The ad hoc Treasury Bills thus emerged as a mode of financing Central Government's deficit in mid-1950's. For smooth conduct of Government business, it was mutually agreed between Central Government and RBI that a minimum cash balance of Rs.50 crore on Fridays and Rs.4 crore on other days would be held by the Central Government. To adhere to this administrative arrangement, it was agreed that Reserve Bank would replenish Government's cash bal ances by creation of ad hoc Treasury Bills in favour of the Reserve Bank. The ad hoc Treasury Bills, which were meant to be temporary, gained a permanent as well as a cumulative character. Indeed, it became an attractive source of financing Government expenditures since it was available at an interest rate pegged at 4.6 per cent per annum since 1974, i.e., actually at a negative real interest rate.

   Assault on Ad hocs

5. The Committee to Review the Working of the Monetary System recommended that the Reserve Bank may adopt a system of monetary targeting with feedback. The target is to be fixed taking into account the growth of real sector and an acceptable order of increase in prices. The adherence to target implied a limit to monetisation of Government deficit. After all, any investment in Central Government securities by the RBI results in monetisation of Government's deficit, as RBI's credit to Government is a source of reserve money generation.

6. Dr.Rangarajan, our Governor brought out the ill-effects of automatic monetisation of Government deficit through ad hocs in his Presidential Address at the Annual Conference of the Indian Economic Association in December, 1988. He has been pursuing this subject with a missionary zeal since then. Analysing fiscal and monetary interactions, he brought out the following issues. First, the scale of Government borrowing was maintained at a high level and surpluses on revenue account have given way to deficits.

   Second, such borrowing has been at less than market rates. Third, even the growing captive investors in Government securities such as commercial banks were not willing to subscribe beyond their statutory obligations.

   Fourth, RBI as a manager of public debt became the residual subscriber.

   Fifth, since Government incurred deficits year after year, the question of retirement of Treasury Bills did not arise. Sixth, the expansionary impact of these deficits, after taking into account the movements in net foreign exchange assets, had to be countered.

   Seventh, to counter the impact, open market operations were not o73 possible given the interest rate structure. Eighth, this meant that Cash Reserve Ratio had to be increased to absorb the excess liquidity.

   Finally, when ad hoc Treasury Bills were created automatically to finance Central Government deficit without any limit, the system contributed to rapid monetary expansion when the budget deficit became very large.

   Dr.Rangarajan concluded his analysis with a solution, i.e. an agreement between Government and Reserve Bank of India setting a limit, each year, on the extent of fiscal deficit and its monetisation. This agreement would provide greater maneuver ability to the RBI to regulate the volume of money. Monetary targeting would then be meaningful. What we have arrived at today in the Budget of 1997 is exactly what was envisaged in 1988.

   Process of Elimination

7. The process of elimination of ad hoc has been designed in three stages :

   First, through limits on creation of ad hoc Treasury Bills which operated between 1994-95 and 1996-97.

   Second, through a transition period of two years beginning April 1, 1997, when ad hocs are eliminated and the new system of Ways and Means is introduced; but, overdraft above Ways and Means would be permissible beyond ten continuous working days, though at a cost.

   Third, the fullfledged system of Ways and Means operates effective April 1999.

   We have completed the first phase, agreed on the second and third phases and are commencing the transition period viz., the second phase from April 1997. The second and third phases would become operational through a fresh agreement between Government of India and Reserve Bank of India. Incidentally, the arrange ments for WMA to State governments are also based on agreements between Reserve Bank of India with State Governments.

   Experience with First Phase

8. In September 1994, it was mutually agreed between Government and RBI that limits will be placed on net issue of ad hoc Treasury Bills for year end and also within-the-year. The year-end limits of Rs.6,000 crore for 1994-95 and Rs.5,000 crore for the next two years were fixed. The within-the-year limit during this period was kept at Rs.9,000 crore.

   During 1994-95, the net issue of ad hoc Treasury Bills during the most part of the year stood negative and even at the end of the year, the net issue was only Rs.1750 crore, much lower than the year-end cap. During the years 1995-96 and 1996-97, though there was some difficulty of staying below the within-the-year limits, it may be noted that during 1996-97, since August 14, 1996, the net issues of ad hocs remained mostly below the within-the-year limit. The most significant part of this agreement was that the practice of issuing ad hoc Treasury Bills to finance Central Government's deficit would be totally discontinued by 1997-98. In the last year's Budget, Government committed to announce concrete proposals for phasing out ad hoc Treasury Bills. This year, Finance Minister did exactly what he promised - to comply with the agreement.

9. A question that is often asked is what happens to the ad hocs outstanding ? The outstanding level of ad hoc Treasury Bills as at end-March 1997 will have to be funded into special securities at an interest rate of 4.6 per cent per annum. These securities, in addition to the volume of funded securities of Rs.71,000 crore already in the books of RBI, will continue to be held by the Reserve Bank. The Reserve Bank at a later stage may have to use these securities for open market operations by converting them into marketable lots.

   New System

10. What will the proposed system be ?

    First, a scheme of WMA by Reserve Bank of India to Government of India (GOI) has been evolved to accommodate temporary mismatches in Government receipts and payments.

    Second, the limit for WMA and the rate of interest on WMA will be mutually agreed between the RBI and Government from time to time.

    Third, any drawals by Government from Reserve Bank of India in excess of the limit of WMA would be permissible only for ten consecutive working days.

    Fourth, when 75 per cent of WMA is utilised, the Reserve Bank would trigger fresh floatation of Government Securities.

    Fifth, consistent with the discontinuance of ad hoc Treasury Bills, the system of 91 Days Tap Treasury Bills will also be discontinued with effect from April 1, 1997.

    Sixth, with the discontinuance of ad hoc Treasury Bills and Tap Treasury Bills and the introduction of WMA, the concept of conventional Budget Deficit loses its relevance. Therefore, it is proposed to discontinue the practice of showing Budgetary Deficit; instead Gross Fiscal Deficit would become the key indicator of deficit. As a transparent way of reporting the monetised deficit or net RBI Credit to the Central Government, the likely extent of RBI support in respect of dated securities and auction Treasury Bills is separately shown as the 'Monetised Deficit' in the Budget document.

    Ad hocs and WMA : Differences

11. There are significant differences between ad hocs and WMA.

    First, WMA will not be a source of financing Budget Deficit. It is only a mechanism to cover day-to-day mismatches in receipts and payments of the Government. Therefore, the use of WMA will have to be periodically vacated. WMA will also not be shown as a source of financing in the Budget estimates.

    Secondly, limits on WMA will be fixed and any excess drawal by Government beyond the limit will become permissible for not more than 10 consecutive working days.

    Thirdly, WMA will be charged at market related interest rate.

    Transition period

12. It is possible to argue that the period of agreement between 1994-97 itself was a transition and further transition now is not necessary. However, introduction of a fundamental change in a public system requires that functionaries in a variety of organisations in Government change procedures and cash management practices. This simply cannot happen like a dress rehearsal before the new system. In fact, some minor modifications may be needed as we go along. To say that transition is meant to dilute implies a predisposition to suspect the motives. After all, the change over from the old system to new system should be smooth. Government should evolve expenditure control mechanism and improve its day-to-day cash management. It may be necessary for the Government to discuss with major ministries for this purpose. Reserve Bank would also attempt streamlining its debt management techniques. The RBI will make efforts to develop the Treasury Bills market by proactive participation and issue of Treasury Bills of different maturities. Therefore, a transition period of two years has been proposed. During the transition period, the Government would be allowed to exceed the limit of WMA for periods beyond 10 consecutive working days. But, additional interest will be applied for such periods on the amount of overdraft.

    Current Debate

13. Consequent upon the announcement of the new system and presentation of the Budget incorporating the new system there has been a debate on the features of the system as well as its functioning in 1997-98.

    First, that it is only a new name for the old system.

    Second, that without statutory ceiling on debt, the system will serve no purpose.

    Third, that the degree of autonomy for monetary policy is still limited.

    Fourth, that the budgeted figures show that the new system will involve large borrowings and more monetisation and, therefore, is inflationary. Let me analyse each one of the issues.

14. As already explained, the new system implies vacation of advances made and not accumulation year after year. The limits are fixed and exceeding the limits is simply not permitted beyond ten days once the system becomes fully functional. There is a cost of using such advances which is market related. In addition, there is also the important advantage of providing mechanisms for open market operations to take care of surges in in flows of foreign capital. That is difficult under a system of ad hocs but is possible when auction Treasury Bills and dated securities take the place of ad hocs. Market based instruments are absolutely essential as we move along the path of capital account liberalisation. Further, the system is far more binding and transparent since the level of borrowings and the pattern of financing, especially RBI's contribution, are made part of Budget documents. Finally, the mixed bag that Budget Deficit repre sents, is replaced by a more transparent method.

15. Perhaps it is necessary to clarify the three concepts viz. the Budget Deficit, the Monetised Deficit or the net RBI credit to the Government and RBI's support in primary issues of Central Government securities. The Budget Deficit is defined as the total of the net issue of 91-day Treasury Bills (at face value) - ad hoc, tap and auction - net of increase Central Government balances during the financial year. Therefore, conceptually Budget Deficit is the short term financing availed of by the Central Government both from the Reserve Bank and other entities which include banks, financial institutions, State Governments, corporates and other parties. In this set up, the Reserve Bank is only one of the financing units of Budget Deficit.

    Net Reserve Bank credit to the Centre, on the other hand, includes not only the Reserve Bank's holding of 91 - day T-Bills but also its holding of dated securities and rupee coins net of increase in Central Government's cash balances. In other words, the Reserve Bank's holding of 91-day T-Bills that forms a part of Budget Deficit is also a subset of the monetised deficit. However, it may be noted that while the constituents of Budget Deficit are measured at face value, those of the net Reserve Bank credit to the Centre are measured at book value. This valuation difference is, however, negligible particularly for 91-day T-Bills.

    The correlation between the Budget Deficit and the mone tised deficit gets weakened at times of easy liquidity such as during the current financial year as market absorption of Government securities including 91-day T- bills have gone up substan tially. The purchase of Government securities by non - RBI entities reduces the monetised deficit in two ways : (i) directly by reducing devolvement on the Reserve Bank in 91-day T-Bills and dated securities auctions as well as through the sale of dated securities including repos and ii) indirectly, as the market off-take of Tap T-Bills and fresh Government securities improves the Centre's cash balances and reduces the Centre's recourse to ad hocs.

16. The RBI's support to primary issues of Central Government securities reflects only its support in the primary offers and net RBI credit to Government could be very different due to other factors. Let me illustrate this with some figures. In 1993-94, RBI's support to Government was Rs.7014 crore, but, net RBI credit was only Rs.260 crores. In 1996-97, till February 14th, RBI support was Rs.12,099 crore, but, net RBI credit was only Rs.7,837 crore. Monetised deficit is one source of money supply and not the only factor affecting money supply.

17. It is sometimes argued that the real instrument for fiscal prudence should be a statutory ceiling on debt. True, our Constitution enables a Parliamentary legislation relating to a ceiling on debt. The Ministry of Finance is due to bring out a Discussion Paper on this. Let me not preempt that. But, clearly the new system has merit by itself and will co-exist more effectively with a ceiling on debt. It has merit in itself since it seeks to limit the extent of RBI's support. Technically, there is nothing to stop the Government from accessing all the debt within the ceiling from RBI if the new system does not exist. A ceiling on debt provides a constraint on the broader issue of debt management and impinges on fiscal policy more directly. The new system tackles the issue of fiscal-monetary interface which is a critical component of both fiscal and monetary policies.

18. It is sometimes argued that the degree of autonomy for monetary policy is still limited. To the extent the limits on WMA interest costs and RBI's support to Government are negotiable during the transition, there is room for Government influencing the RBI. The main difference is that under the earlier system of automatic monetisation Government could unilaterally determine the limits, while after the new system becomes fully functional, the pattern similar to State Governments would be followed. The data in the Budget for 1997-98 are quoted to show that the RBI has not gained in autonomy. That would lead me to an analysis of these figures and the inflationary potential of the borrowing programme and the pattern of financing.

19. In the current debate, some have presumed that as per Budget estimates, the increase in high powered money for 1997-98 would consist of Rs. 16,000 crore (the monetised deficit) plus Rs.7000 crore on account of increase in the average level of WMA. This presumption is wrong for two reasons:

    First, the Budget estimates for monetised deficit at Rs.16,000 crore would consist of increase in RBI's holdings of Government securities including dated Government securities and RBI's hold ings of auctioned Treasury Bills.

    Second, the Budget estimates for 1997-98 has not assumed any contribution to the budget from WMA. It may be mentioned that Rs.1,00,000 crore has been shown on both the receipts side and the disbursements side, implying that whatever WMA were taken by the Government would be vacated before the end of the year.

20. Some others have argued that there is a large increase in net borrowings of Government, of the order of 33 per cent negating the intentions of the new system. Here again, the comparison between the net market borrowing figures of Rs.25,498 crore for 1996-97 and Rs.33,820 crore for 1997-98 is misleading. The figure for 1996-97 excludes financing through Budget Deficit i.e., mainly through the instrument of 91 day Treasury Bills including ad hocs. The figures of 1997-98 under market borrowings would include borrowings through 91 day (Auction) Treasury Bills. Therefore, the comparable figures would be Rs.32,398 crore for 1996-97 [(Rs.25,498 crore of net market borrowing plus Rs.6,900 crore of Budget Deficit (RE)] and Rs.33,820 crore for 1997-98 (total net market borrowing), implying an increase of only 4.4 per cent.

21. Some others have argued that growth in money supply at 16 per cent would imply an inflation of 10 per cent if GDP grows at 6 per cent. The money supply growth indicated is in terms of broad money and not narrow money. The income elasticity of demand for broad money which has been proven to be reasonably stable and is in the range of 1. 5 to 1.6 in India should be reckoned. Even accepting that growth in GDP will be no more than 6 per cent, the implied inflation could be only between 7.0 per cent (16.0 - 9.0 ) and 6.4 per cent (16.0 - 9.6).

    Outlook

22. Let me conclude with the outlook for successful implementation of the new System.

    First, let us recognise that what we are doing is really adhering to what the First Five Year Plan visualised viz. Judicious Credit Creation.

    Second, we are trying to operationalise the monetary targetting that Prof. Sukhamoy Chakravarthy Committee wanted us to. No doubt, we have to move to indirect instruments of monetary policy, but, we cannot escape some sort of effective monetary target ing.

    Third, a process of thinking initiated by Dr. Rangarajan in 1988, further elaborated by him into a proposal in Kutty Memorial Lecture on 'Autonomy of Central Banks' in September 1993, and formalised through an agreement between him as Governor and Prof. Manmohan Singh as Finance Minister has been honoured by our present Finance Minister Shri P. Chidambaram. We, therefore, have reason to believe that there is a consensus on the new system.

    Fourthly, the critics have not questioned the need for closer fiscal monetary interaction or more autonomous monetary policy. If at all, the critics want to make sure that the new system works effectively towards its objectives.

    Finally, what has been put in place is a system and undoubtedly a better system than ad hocs and certainly more suited to our times as well as our policies in the external sector. How it operates depends on public opinion, economic compulsions and the good faith with which this is implemented by Government and the Reserve Bank of India