The 1990s saw a series of financial crises affecting emerging market countries which highlighted the vulnerability of these countries to sudden changes in market perceptions and their highly destabilising consequences. This in turn focussed attention on the need for preventive action aimed at forestalling crises or at least reducing their intensity. In this context, there has been a great deal of emphasis on the need to strengthen the fiscal-financial system in emerging markets and to increase the degree of transparency in the system. The emphasis on transparency reflects the perception that information asymmetries are an important cause of financial market failures, which in turn lead to crises.

Following this approach, emerging market countries are increasingly expected to comply with various international Codes and standards applicable to different parts of the fiscal-financial system. These Codes are not mandatory, but countries interested in accessing financial markets will be expected to conform to them and the extent of compliance is increasingly regarded as an indicator of fiscal-financial soundness. IMF surveillance has already begun to focus on the extent of compliance as is evident in the preparation of Reports on Standards and Codes (ROSC) by IMF staff, documenting the extent of compliance by individual developing countries with a core set of standards.

In this background, the Governor of the Reserve Bank of India established a Standing Committee on International Financial Standards and Codes under the co-chairmanship of Dr. Y.V. Reddy, Deputy Governor and Dr. E.A.S. Sarma, then Secretary, Economic Affairs, to review developments regarding international standards and Codes that have come up in various areas and to assess the relevance of these Codes for India. The Standing Committee in turn established ten separate Advisory Groups to look into existing Codes and standards in specific areas. The Advisory Group on Fiscal Transparency is one of these groups.

The composition of the Advisory Group is given in Appendix I. The Group has held several meetings. It has benefited from discussions with representatives from the Office of the Comptroller and Auditor General. It has also met with several Finance Secretaries of the states on matters related to fiscal transparency at the state level.

Fiscal transparency is important because fiscal soundness is one of the core requirements for financial stability and transparency is needed for markets to be able to assess fiscal soundness accurately. The first step in defining international standards for fiscal transparency was the preparation by IMF Staff of a 'Code of Good Practices on Fiscal Transparency - Declaration of Principles'. This document was discussed in the Executive Board of the Fund and by the Interim Committee of the Board of Governors of the IMF and adopted by the Committee in 1998. India is a member of the Interim Committee and was represented in the Committee when the Code was adopted, and it is therefore especially important to monitor the extent to which our practices comply with the requirements of the Code.

The original Code has been revised on the basis of experience gained in order to clarify and elaborate various aspects. The revised 'Code of Good Practices on Fiscal Transparency' (hereafter referred to as the Code) was approved by the Executive Board of the Fund in May 2001. We have used this version as the benchmark against which the degree of fiscal transparency in India has been evaluated.

The Code is based on four general principles each of which is further elaborated into more detailed guidelines and practices. The four principles are:

• Clarity of roles and responsibilities within government and between government and the rest of the economy.
• Public availability of information on fiscal outcomes.
• Open and transparent budget preparation, execution and reporting.
• Assurances of integrity, including those relating to the quality of fiscal data and the need for independent scrutiny of fiscal information.

The rationale of the individual guidelines specified in the Code and the practical problems that are likely to arise in implementing the Code, are discussed in detail in the Revised Manual on Fiscal Transparency (hereafter referred to as the Manual).

The Group has sought to examine the extent to which our fiscal practices comply with each of the detailed guidelines on the Code keeping in mind the clarifications and elaborations in the Manual. Where our existing practices do not comply, or comply only partially, we have attempted to assess whether full compliance is desirable and feasible, and also indicated a time frame for compliance.

Several important issues related to fiscal transparency need to be clarified at the outset.

The first is the distinction between fiscal transparency and the other two ingredients of good fiscal management namely the soundness of public finances and the efficiency of fiscal policy instruments. The soundness of fiscal management relates to the macro-economic issue of the fiscal balance which is an essential pre-condition for stability. The efficiency of fiscal policy instruments relates to the micro-economic issues of the efficiency of particular expenditure programmes in achieving their objectives and the efficiency of tax policies in raising revenues with minimum economic distortion. Fiscal transparency is quite distinct from these two important aspects of fiscal policy since it focuses on the more limited issue of whether sufficient information on the fiscal situation is being provided in a timely fashion to enable observers, including especially financial market participants, to make an accurate assessment of the underlying fiscal position.

It is perfectly possible to envisage a situation where the underlying fiscal position is sound but the system is not adequately transparent. In such situations, a country may not be able to derive the full benefit due from fiscal soundness because this source of strength of the economy is not well recognized. Equally, one can envisage a situation where the underlying fiscal situation is weak but the extent of weakness is not evident because the system is not transparent. It is important to emphasise that non-transparency in such situations is not necessarily an advantage because hiding the underlying weakness in the fiscal situation can create complacency and delay corrective action. Such delays encourage an accumulation of problems which lead to a sudden crisis when the situation has sufficiently worsened and the full facts become known. Transparency is also quite distinct from efficiency since high levels of transparency could exist, the fiscal situation could also be macro-economically sound, and yet fiscal policy could be characterized by inefficiency in tax policy and/or expenditure policies. Ideally, the objective of fiscal policy should be to achieve high standards on all three counts: fiscal soundness, efficiency and transparency. This report focuses only on the limited issue of transparency.

A second issue relates to whether the Code should be applied to all levels of government. Universal applicability is certainly desirable in principle, especially where a significant volume of general government activity takes place at sub-national (state and local government) levels as is the case in India. However, it must be recognised that the central government may not be in a position to enforce transparency at sub-national levels. It is therefore recognised in the Manual that the application of the Code may have to be limited to the central government at least in the first instance. Our assessment of compliance in this report therefore focuses on transparency at the central government level, but we also indicate where possible, the extent of compliance at state government levels and also make recommendations in this regard.

A third issue relates to the fact that although the Code provides fairly detailed guidelines about the various elements that go into transparency, the issue of whether a particular institutional arrangement in a country actually functions in a manner which meets the requirements of the Code will inevitably depend upon judgement. This is especially so since institutions vary considerably across countries, and it is not easy to judge whether a particular set of institutions is as effective in one country as similar institutions in another. The Manual provides useful guidance on how to interpret and apply the Code and it also identifies 'best practice' in many areas. In making our assessment, we have sought to evaluate current Indian practice against the requirements of the Code as clarified and elaborated in the Manual, keeping in mind current practices in other countries.

The report is structured as follows. Chapter II presents our detailed assessment of the position regarding compliance with respect to each item in the IMF Code. Chapter III provides an overview of our findings and highlights the areas where our current practices are not in compliance with the Code of Good Practices on Fiscal Transparency and the further measures, which should be taken to enhance transparency in the Indian context.

In this chapter we examine the extent to which our current practices are in conformity with the prescriptions in the revised version of the Code. As pointed out in Chapter 1 the Code deals with four general principles with detailed guidelines enumerated under each principle. (The guidelines as contained in the Code are reproduced in italics). We assess the degree of compliance with each of these detailed guidelines. Where compliance is inadequate we make recommendations on the extent to which compliance can be increased, and over what time period, keeping in mind Indian conditions.

This part of the Code, which is further sub-divided into 5 separate components, prescribes that the structure of government and the responsibilities of different parts of the government should be clearly defined and that the relationship between government and the rest of the economy should be based on clear and transparent rules. Such demarcation is clearly essential to ensure transparency on the extent of government activity in the economy, and also to achieve accountability.

Our overall assessment is that our current practice is broadly compliant with these principles, though there are some areas where steps should be taken to increase the degree of transparency.

This principle is fully complied with.

The general government sector, as defined in the United Nations (UN) System of National Accounts 1993 and in the GFS Manual, encompasses 'all institutions performing government functions as their primary activity'. It should therefore include all national and sub-national government units including extra-budgetary funds. It should also include non-profit institutions that provide non-market services of a governmental nature.

The structure of government is clearly specified in the Constitution which distinguishes between the central and state government level and, following the 73^rd and 74^th Amendments, also explicitly recognizes a third level i.e. the Panchayats in rural areas and urban local bodies in urban areas. The functions of the central and state governments are clearly separated based on a central list of subjects (List I of the Seventh Schedule) where only the Parliament can legislate, an exclusive States List (List II) where only the state assemblies can legislate, and a Concurrent List (List III) where both the central legislature and state assemblies can legislate, but in the event of a conflict the central legislation prevails. Following the 73^rd and 74^th Amendments, the local bodies in urban and rural areas have also been recognized as an independent level of government to which additional responsibilities and resources must be transferred. The extent of devolution of responsibilities and resources to the local level is not specified by the Constitution but is left to the states to delegate.

A basic requirement of fiscal transparency identified in the Manual is that an institutional table should be published showing the structure of government and the rest of the public sector along the lines of the tables included in the Government Financial Statistics Yearbook of the IMF. Box 1 shows the institutional table for India as it appears in the GFS and also the entry for Brazil. We note that the information provided for India is extremely sketchy compared to the information provided for Brazil. However, information is available in other documents/Reports about other public bodies similar to the examples shown for Brazil.

We recommend that the institutional table for India in the GFS should be revised to make it comparably detailed. For example, the Employees Provident Fund, which is similar to a Social Security Fund, could be shown separately. There are also a number of central government units which have their own budgets outside the central government budget but which perform a non-market non-profit function, e.g. central universities, the Indian Institutes of Technology, central research institutions, etc.

The Manual clarifies that the allocation of expenditure responsibilities and tax powers between different levels of government should be clearly defined, and at each level the fiscal role of the executive and legislative branches should be clear. This requirement is met because the allocation of expenditure responsibilities follows the division of legislative power between the central and state government level laid down in the Constitution. Each level of government has its area of exclusive jurisdiction and the concurrent list in the Constitution defines an area of overlap between the centre and the states. While the overlap is substantial, the areas are clearly defined and the transparency requirement is therefore met.

Taxation powers are also clearly divided between the central and state governments with the Constitution specifically listing the taxes which can be levied by the centre and those that can be levied by the states. All residual powers of taxation which are not specifically assigned to the centre or the states lie with the centre. The resulting division of taxation power is heavily weighted towards the centre for which reason the Constitution provides an explicit mechanism for sharing central tax revenues with the states. A Finance Commission has to be appointed every five years which recommends the extent of the central tax revenue to be shared with the states and also the formula for sharing. Thus far, the recommendations of the Finance Commission on the sharing of revenues between the centre and the states have been accepted without amendment.

In addition to the sharing of revenues through the Finance Commission award, the central government also makes discretionary transfers in the form of central assistance to the states to finance state plans. The size of these transfers is determined for each Plan period as part of the approval of five-year plans and the distribution of the total amount of central assistance among states is determined on the basis of an established formula. The system of transfers is complex but it is on the whole transparent.

The role of the executive, legislative and judicial branches is also clearly defined. The executive branch is responsible for preparing the budget consisting of expenditure proposals and taxation proposals and presenting it to the legislature for approval. The executive is also responsible for implementing the budget once approved.

The executive is constrained to limit expenditure under various heads to the levels approved by the legislature. It is recognized that circumstances may arise when the executive needs to incur expenditure without explicit budgetary approval and for this purpose, it is permitted to draw from the Contingency Fund subject to ex post facto approval of the expenditure by the legislature. However the size of such expenditure is normally limited by the size of the Contingency Fund, which in the case of the central government is Rs.50 crores. The size of the corresponding contingency funds at the state level varies from state to state.

On the taxation side, no taxes can be levied either by the centre or the states without explicit legislative sanction. Indirect taxes can be lowered by the executive below the levels authorised by the legislature, but they cannot exceed these levels. In recent years, the government has declared that it will not use this power to give any tax exemptions or reliefs for individual cases.

The role of the judiciary is also substantial. Tax payers are free to challenge the legality of taxation by appeal to the courts, a feature specifically mentioned as desirable in the Manual. Apart from cases of individual tax liability, where there is a well-defined judicial appeal structure, citizens can also challenge any fiscal action by the executive in the Courts under the general provision of judicial review if it is felt to lack legal sanction or if it is ultra vires of the Constitution.

The Manual states that there should be an active committee of the legislature that oversees the conduct of fiscal policy and which facilitates civil society input into the budget. There are several mechanisms in India through which the legislature performs an active role.

1. Following the presentation of the budget and before its approval the Demands for Grants of each Ministry are referred to the Standing Committee of the Parliament for the Ministry concerned. This Committee examines past expenditure levels along with proposals for the next year and submits a report to the Parliament as part of the normal process of budgetary approval. It is open to the Standing Committee to invite inputs from civil society, but this is not yet an established practice.
2. The Public Accounts Committee (PAC) is an important permanent Committee of Parliament with a rotating membership. The Chairman is normally a Member of Parliament from the Opposition. This Committee examines all reports of the Comptroller and Auditor General (CAG), an independent Constitutional authority which submits detailed reports to Parliament examining various aspects of government expenditure and also revenues. The PAC can also examine various aspects of the budget suo moto. Reports of the PAC are laid on the Table of the House and the convention is that they are taken to be recommendations of the House to which the government must then react. The government is, however, free to take such action as it sees fit.
3. In addition to (i) and (ii), there is a Consultative Committee for the Ministry of Finance which discusses matters related to the Finance Ministry. This Committee determines its own agenda and covers all aspects of fiscal policy in its purview. However, it does not submit reports to Parliament and its proceedings are not made public.
4. A separate permanent Parliamentary Committee, called the Committee on Public Enterprises, exercises Parliamentary oversight on the non financial public sector enterprises. This Committee examines reports of the CAG on public sector enterprises and can also examine any matter relating to the public sector suo moto. In view of the importance of the public sector and its inter-relationship with government and the prevalence of quasi-fiscal activities in the public sector, this is an important institution for involving the legislature in fiscal management. Reports of this Committee are laid on the Table of both Houses and discussed in Parliament.

State government legislatures do not have a system of Standing Committees and the budget is generally presented and discussed in the Assembly. However, they do have a Public Accounts Committee and a Committee on Public Enterprises performing broadly the same functions as in the central legislature.

The Manual points to several different issues that are relevant in evaluating compliance with this principle. Our assessment on each count is as follows:

The responsibility of fiscal management is clearly specified.

The responsibility for macro-fiscal management lies squarely with the Finance Ministry while the responsibility for managing individual programmes in terms of their efficiency and effectiveness lies with the administrative ministry concerned.

The Finance Ministry prepares the central government budget and is technically the Ministry responsible for introducing mid-course corrections should they become necessary, either in the form of expenditure cuts or in the form of tax increases, though the latter require parliamentary approval. However, while the technical responsibility for managing the budget lies with the Finance Ministry, the existing incentive structure does not put strong pressure on the Ministry to implement mid-course corrections should these become necessary. For example, there is no obligation at present to report to Parliament if the revenue out-turn deviates significantly from the levels anticipated in the budget on the basis of which expenditure plans were approved. The Fiscal Responsibility and Budget Management Bill recently introduced in Parliament will make a significant difference in this respect if passed into law (see 3.1.1 and Box 7).

Indian current practice is broadly compliant in this respect.

There are very few extra-budgetary funds in India (see Box 2). Revenue inflows into these funds which do not enter the Consolidated Fund but are held in what is called the Public Account as deposits. However, the transactions in these funds are fully accounted for in the budget. The transactions are shown in the aggregate in the budget and in detail in the Finance Accounts and the Final Audited Accounts. Some of the funds are relatively small, e.g., the Steel Development Fund or the Sugar Development Fund while others such as the recently created National Savings Fund are very large. However, in all cases, transactions are fully recorded in the budget.

The most important extra-budgetary arrangement which has a significant fiscal impact but which is not included in the budget is the Oil Pool Account operated by the Oil Coordination Committee. The operation of the Oil Pool Account is described in Appendix III. It is difficult to determine whether this is an extra-budgetary fund, or a form of Quasi-Fiscal Activity (QFA) [see 1.1.4(c)]. It was originally meant to be a self-balancing account, created to manage the administered pricing mechanism (APM) in the petroleum sector. Under the APM, crude oil producers and refineries were paid administered prices. Consumers of petroleum products were charged administered prices and the revenue flow from consumers was expected to cover costs paid to domestic producers plus payments made for imports of crude oil and products. Administered prices for consumers were meant to be adjusted periodically to ensure that over time the revenue from sale of products covered payments to domestic producers and payments for imports. In earlier years, the Pool ran large surpluses which were not reflected in lowering of prices but were held as special deposits with the government, thus providing fiscal support for the central government. More recently, the Pool has run large deficits because of the failure to adjust consumer prices when international prices increased. These deficits have had to be made good by the government by issuing bonds to the oil companies, which were later cancelled from subsequent surpluses generated in the oil pool.

The scale of deficits in the oil pool account in recent years has been very large and its non-inclusion in the budget clearly introduces a significant element of non-transparency in the system. Fortunately, the Oil Pool Account is expected to be dissolved in April 2002 when the APM is dismantled. When this happens, it will remove the only major extra-budgetary operation which is not fully accounted in the budget.

Indian practice is fully compliant on this score.

Under Article 266 of the Constitution all revenues of the government including fees and charges levied by government departments engaged in the provision of services (e.g. the revenue receipts of the railways and, prior to its recent corporatisation, the receipt of the Department of Telecommunications) are shown as receipts in the budget and credited to the Consolidated Fund of India. Expenditure therefrom is shown in gross terms and is part of the budget requiring Parliamentary approval.

This is fully complied with.

External assistance to the government for financing development projects and programmes is routed through the budget and recorded as an inflow in budget documents with a corresponding expenditure item reflecting the use of resources.

This is a critical issue where our current practices are not fully compliant as indicated below for each type of non-government institutions.

Transparency requires that if QFAs are carried out by non-government organizations they should be carried out under clearly specified and open arrangements with a quantification of the extent of the QFA. Over the years, the extent of quasi-fiscal activities has been reduced but the scale remains significant. They are not always based on clear and transparent arrangements and more importantly, they are not quantified.

The Manual lists a number of QFAs which central banks typically engage in i.e., administered interest rates, directed lending, multiple exchange rates and exchange guarantees. The position in India with respect to each of these can be summarized as follows:

1. The range of interest rates control exercised by the central bank over commercial banks has been greatly reduced over the past ten years. Nevertheless, commercial bank interest rates are still not fully market determined. Interest rates on loans upto Rs.2 lakhs are fixed, by direction of the central Bank, at a level not exceeding PLR. It was also recently announced that bank loans for education will be extended at the PLR level. Fixing an interest rate at the PLR level is less rigid than the earlier practice of specifying the actual interest rate for different types of loans, since the PLR itself can be varied by the Bank. Nevertheless, this is an example of non-market fixation of interest rates since the PLR is normally a rate offered only to customers who meet the highest standards of credit standing.
2. Directed credit requirements in the banking system also continue to be substantial. At present, banks are required to ensure that 40 per cent of their advances are to the priority sector of which 18% should be to agriculture and 10 per cent to small-scale industries. It is difficult to determine the quantitative implication of this particular QFA since it implies estimation of the implicit subsidy in such loans.
3. The Reserve Bank does not engage in multiple exchange practices.
4. The Reserve Bank also does not engage in provision of exchange guarantees in the manner in which it did earlier, when non-resident foreign currency deposits benefited from an exchange guarantee by the RBI. The recent resort to a maintenance of value account to cover the exchange risk on Resurgent India Bonds is not an example of an exchange guarantee given by the Reserve Bank but of an exchange guarantee given by the Government of India. This issue and its implication for fiscal transparency is examined subsequently under 2.1.3.

The Manual also states that links between fiscal and monetary operations of the central Bank should meet the requirements of Sections 1.2 and 1.3 of the Monetary and Financial Transparency Code.

Section 1.2 of the Monetary and Financial Transparency Code relates to the need to define clearly the institutional relationship between monetary and fiscal operations. This relationship has evolved over time. Earlier, government deficits were financed automatically by the RBI through creation of ad hoc treasury bills at very low rate of interest. This practice was finally ended in 1997 on the basis of an agreement reached between the RBI and the Ministry of Finance that a shortage of funds in the government account with the RBI would not lead to monetisation of the deficit as in the past but to flotation of marketable government securities by the RBI. However, since the Reserve Bank of India as the debt manager is responsible for the success of the government securities issue, and is also allowed to pick up government securities at the primary issue stage, there is a conflict between this role and its role as the monetary authority. As a debt manager the RBI wishes to minimize the cost of raising debt and is therefore often tempted to pick up substantial amounts of government securities to avoid creating interest rate spikes. Transparency requires that the RBI should not intervene in the primary issue of government securities but only in the secondary market where it can independently determine the trade-off between risking monetary expansion and tolerating high interest rates.

We note that the Fiscal Responsibility Bill seeks to prohibit RBI support of government securities in the primary market after 3 years. In our view, this is a desirable step. It should be accompanied by the government taking over the debt management function from the RBI as has been recommended separately by the Advisory Group on Monetary and Financial Transparency.

Section 1.3 of the Monetary and Financial Transparency Code requires that agency roles performed by the central bank on behalf of government should be clearly defined. At present, the RBI makes payments to commercial banks of commissions on certain government transactions, collections for the Public Provident Fund and other agency functions rendered by the banks. These costs are clearly recognised in the books of accounts of the central bank but they are not reflected in government budget data.

An important aspect of the relationship between the government and the Reserve Bank of India is the treatment of Reserve Bank profits. Current practice involves transfer of RBI profits to the Budget in a transparent manner, with RBI profits being shown as current receipts. Details regarding the basis for the profit amount are published in the Annual Report of the Reserve Bank of India.

The Manual proposes that it is a basic requirement of fiscal transparency that the annual reports of public financial institutions should indicate the non-commercial services that the government requires them to provide. As pointed out above, commercial banks are subjected to certain interest rate controls and directed lending requirements that amount to QFAs as defined in the Manual. While the nature of these directions and controls is transparent, the annual reports of these institutions do not quantify the extent to which they are engaging in activity which is non-commercial.

We note however that it is not easy to quantify the extent of QFA involved in these interventions since it is difficult to determine how far the directed credit requirement actually leads to lending which would not take place otherwise because the directed credit target is only an aggregative target which the bank must meet, leaving the bank free to decide on individual credit decisions.

Non-financial public enterprises in most countries engage in significant QFAs. Indeed, the very rationale for having non-financial public enterprises in the first place is linked to the fact that they can be made to act differently from market driven private enterprises. It is important to emphasise that not all departures from commercial behaviour should be described as a QFA. For example, non-financial public enterprises may suffer from inefficiency, poor work culture and poor management quality, all of which leads to certain types of 'non-market behaviour' which is reflected in lower profits or losses. Such losses should not be viewed as QFAs.

For our purposes, the term QFA should be restricted to non-market behaviour which is specifically mandated by the government, by virtue of its ownership and control, with the objective of meeting an explicit fiscal purpose such as providing a subsidy to a particular group in society.

Even if we adopt this restrictive definition of QFAs there is no doubt that non-financial public enterprises in India engage in significant QFAs and these are not adequately identified.

Indian Railways for example is a departmental enterprise of the government of India which engages in a substantial degree of cross-subsidisation of passenger traffic by freight and also engages in cross-subsidisation within freight categories by overcharging some categories and undercharging others. While some of this information is made public, it is not presented as part of the Railway budget. Nor are the implicit subsidies identified as such in the General budget. Similarly, Indian Airlines engages in substantial cross-subsidisation on uneconomic low traffic density routes, e.g., the northeast, and the extent of this particular QFA is not transparent. The Food Corporation of India (FCI) is another public enterprise with an explicit social rather than commercial objective. The FCI receives a subsidy to cover its losses and this amount is explicitly shown in the budget as a food subsidy which is further broken down into the subsidy on account of distribution of subsidised food and the subsidy on account of the carrying cost of the buffer stock. However, the budget does not distinguish between the subsidy which arises because FCI carries stocks in excess of the normal buffer stock requirements. This part of the subsidy is not a consumer subsidy but a producer subsidy and should be separately identified.

More generally, all public sector enterprises follow a policy of price preference in purchases which operates in favour of public sector suppliers. The implicit subsidy involved is not explicitly quantified in the income Statements of the enterprises, nor is an aggregate amount reported in the budget documents.

The Oil Pool Account operated by the Oil Coordination Committee (OCC) can also be described as a QFA. As pointed out in 1.1.3(b), the absence of any information on the scale of deficit in this account is an important source of fiscal non-transparency at present.

The stated policy of the government is to reduce government ownership in public sector enterprises to a minority, in which case they would cease to be public sector units. However, as long as there is a significant public sector it is desirable that the annual accounts of these undertakings should explicitly state the costs borne by these units out of QFA obligations.

The problem of lack of transparency about QFAs is even more important at the state level where there are many public sector organisations which act in a non-commercial manner. The State Electricity Boards are statutory bodies which make large losses which have been increasing over time. While a large part of these losses is due to operational inefficiency and cannot be called a QFA, a substantial part is also due to deliberate underpricing of power supplied to certain categories of consumers, mainly farmers but also households for domestic use, which are clearly QFAs. Road transport corporations are another example where deliberate underpricing accounts for a substantial part of the losses. The states also have a large number of public sector organisations which engage in substantial QFAs.

This guideline is substantially met.

The interface between the Government and the rest of the economy is governed by an elaborate Constitutional and legal framework. Government regulation of the non-government sector is based on specific powers conferred by law and is exercised on the basis of well-defined rules and procedures. These powers are subject to challenge in Courts as to their legality.

Earlier, government intervention in the private sector was extensive, but this has been very substantially reduced as part of the process of economic reforms (see Box 3). Nevertheless it can be argued that many government controls, though exercised on the basis of clear legal authority are not always based on a clear identification of the criteria to be used in the exercise of executive authority. These deficiencies need to be addressed as part of the ongoing reforms process.

The Manual cites the OECD recommendations relating to the characteristics of transparent regulations which are reproduced in Box 4. We would recommend that these be treated as indicators of best practices and existing rules and regulations should be systematically reviewed in the light of these guidelines.

An important new development over the past few years has been the establishment of independent regulatory authorities in key infrastructure sectors, such as electricity, telecommunications, major ports and insurance. Many of the regulatory functions earlier performed by the Ministries have been transferred to statutory regulators e.g. (CERC, TRAI, SEBI, IRDA etc.) in order to introduce greater transparency in operation and in particular to ensure a level playing field between different players. This is an important step forward in achieving transparency though there have been 'teething' problems in the process. For example, questions have been raised about the lack of clarity in the demarcation of responsibilities of the regulators vis-a-vis the government. There has also been criticism of the unwillingness of government owned organisations to accept certain rulings of the regulator and to challenge them in the courts. However, these problems can be regarded as transitional and the process of introducing independent regulators can be said to be a substantial move towards greater transparency in regulating the private sector.

The need for simplification of control mechanisms and the introduction of clarity and transparency is much greater at the state government levels.

This part of the Code focuses on the need for a legal framework governing the management of budgetary and extra-budgetary transactions, levying of taxes and enforcing ethical standards of behaviour for public servants.

This guideline is substantially met.

Expenditure of public funds is possible only by a process laid down in the Constitution and within limits clearly authorized by the legislature. The Manual elaborates that the purpose of this guideline is to ensure that this authorisation should be comprehensive and should include all government transactions, including those through extra-budgetary funds. Individual agencies should be held accountable for funds they collect and use. Contingency reserve provisions should specify clear and stringent conditions for use of such funds and there should be independent audits of use of funds. All these conditions are met.

The Manual also states that the basic principles of budget management should be embodied in a general budget system law which should have constitutional or near constitutional status. There is no budget system law in India but the scope of the budget has been specified in the Constitution and has been refined through the recommendations of various Parliamentary Committees. In conformity with countries governed by common law traditions (as distinct from Civil Law), budget activity is deemed to be within the realm of delegated legislation and is a part of the administrative process. However, there are clearly established rules (e.g. General Financial Rules, Treasury Rules etc.) governing each aspect of budget preparation. Nevertheless, these rules fall short of the discipline that would be entailed by legislation. We recommend that government could consider amplifying the scope of the Fiscal Responsibility and Budget Management Bill (FRBMB) to include elements of a budget law.

The requirement that management of public funds should ensure accountability is fully met. The permanent secretary of the each department is also its Principal Accounting Officer and is accountable for the prudent use of the budgetary appropriations.

One of the aspects of Indian practice thus far, which is not compliant with expected standards of transparency and sound budget preparation is the practice of presenting supplementary budgets in the course of the year to seek legislative authority for additional expenditure. Supplementary budgets may appear to be transparent in the sense that they are formally presented to the Parliament and therefore receive Parliamentary approval. However, they also impart significant non-transparency to the process because the impact on the overall fiscal situation as a result of these demands is usually not reviewed at the stage of consideration by the legislature. This problem would be less serious once the Fiscal Responsibility Bill, which is currently under consideration, is approved. The Bill would require review of the overall fiscal situation before supplementary grants are approved.

The Manual also refers to possible non-transparency arising from the use of contingency or reserve funds. The conditions for use of contingency funds and reserve funds are clearly specified at the central government level. Central government practice is fully compliant with the Code but the situation varies in the states. There have been occasions in some states where the size of the contingency fund has been increased through an ordinance when the Assembly is not in session to accommodate large expenditures not authorised by the legislature. The ordinance is then allowed to lapse when the Assembly reconvenes thus avoiding discussion of the reasons for the increase. The expenditure incurred during the life of the ordinance is subsequently 'regularised' with little or no discussion as a post facto event. Such use of the contingency fund introduces non-transparency in the system.

The wide scope for administrative discretion is a major area of concern.

The first part of the guideline, relating to the legal basis for taxes and duties, is fully met since taxation is only possible under the authority of law. The application of tax laws is also subject to procedural and legal safeguards. Tax payers can contest tax liability through a well defined appeal structure involving quasi-legal tribunals and ultimately the courts.

However the situation regarding clarity of the law and the scope for administrative discretion leaves much to be desired. The Manual states that 'a corollary of requiring that taxes must be imposed under law is that administrative discretion in applying tax laws must be limited'. Unfortunately, the scope for administrative discretion is very large in practice. With the large number of taxes, numerous exemptions and other reliefs, and frequent changes to tax provisions, tax laws are relatively complex. The role of exemptions in rendering the tax system nontransparent is highlighted in Box 5.

The seriousness of the problem should be evident from the fact that any standard publication of the excise tariff structure runs to about 720 pages of which as many as 220 pages are devoted to exemptions. On the face of it, the exemptions number about 70 but individual exemptions have as many as 259 "entries" that amplify their scope enormously. Thus, the actual coverage of exemptions is far more extensive. For small-scale industries alone, there are five exemptions; for job work, there are four; for exports, twenty; and so on.

The exemptions (which include not only zero rates but also reduced rates) are applicable subject to various conditions. There are 52 conditions and 7 lists with several items in each list that specify the conditions under which an exemption would apply (e.g., "for use in leather industry"). The descriptions of the conditions under which the exemptions would apply are often opaque and subject to the interpretation and discretion of administrators, leading to corruption and loss to the exchequer.

An important issue specifically identified in the Manual is that 'case by case negotiation of tax liabilities between officials and tax payers should not be the general rule in any country both because of the non-transparency of such practices and the potential for corruption'. Our current practice can be faulted on this count. The administrative mechanism set up for both central and state level taxes often lead to unnecessary contact between tax-payer and tax administrator which leads to non-transparency, discretion, and has the potential for harassment and corruption (see Box 6).

The Manual points out that information technology can eliminate opportunities for discretionary action as well as providing for effective monitoring of arrears, exemptions, appeals, and payments. However, the use of IT has been initiated and is still in a rudimentary phase, and electronic filing is hardly prevalent in India. A shift to electronic filing is strongly recommended and could be implemented over a 3 to 4 year period.

The Manual also emphasizes that an internal audit system should be established to ensure accountability of tax collection staff and system, and adherence to tax administration policies and procedures in their dealings with taxpayers. There is no such internal audit system and this is an important lacuna in the Indian system.

There are statutory laws relating to the behaviour of civil servants, and prescribed penalties for misconduct. However, the enforcement of these laws, efficiency in identifying cases of malfeasance and timely disposal of the cases under the law, could be greatly improved.

The objective of transparency necessarily requires provision of comprehensive, accurate and timely information to the public. The Code prescribes a number of guidelines in this regard dealing with different aspects of the problem.

This broad guideline is further divided into 5 separate guidelines dealing with (1) the coverage of the annual budget, (2) information on out-turns and forecasts, (3) contingent liabilities, tax expenditures and quasi-fiscal activities, (4) debt and financial assets, and (5) the consolidated position of general government.

The central government budget documents cover all central government transactions. The budget documents report the Budget Estimates for the year for which the budget is being presented, the year proceeding the budget and the actual outcome of the previous year. The Annual Financial Statement, which is constitutionally the budget, provides aggregated information. Detailed information on allocation for individual Ministries is provided in the Demand for Grants, and the Expenditure budgets. The Receipts Budget provides information on revenue receipts. A summary Budget at a Glance provides a concise overview of the net fiscal position of the central government. A detailed list of the documents for the central government is given in Appendix IV-A.

The various state governments also provide extensive documentation related to the budget. A list of core state government documents is given in Appendix IV-B.

The Manual points out that detailed Statements should also be provided for all extra-budgetary funds. As pointed out in Box 2, the major funds are included in the budget. Although details are not always presented in the budget documents, they are fully accounted for in the Annual Finance Accounts prepared by the Controller General of Accounts which are audited by the Comptroller and Auditor General of India. The exclusion of the Oil Pool Account, which is a source of large hidden deficits, is the most important lacuna in this context.

The Manual also requires that similar information should be provided on the activities of the autonomous central government agencies established as separate legal entities to provide health, education and other service, or to conduct specialized regulatory or quasi-judicial function. Not only should grants or transfers received by such agencies from the central government be identified in the central budget, but information on their gross expenditure and revenue, including revenue from user charges, should be reported. In the Indian context, for example, this would mean that total expenditure on educational services provided by central universities and IITs should be reported as part of the central budget on a gross basis. This is not being done at present. However, the details of the services performed by these entities and their audited annual accounts are placed in Parliament separately.

This standard is only partially met.

The central budget provides the following information on the fiscal outcome of the two years preceding the budget. Information in both Budget Estimates and Revised Estimates is provided for the year immediately preceding the budget year. For the previous year only the Actuals are provided. No forecasts are provided for years following the budget.

The lack of forward projection is a major weakness in current practice. The Manual indicates that 'best practice' in this area involves providing aggregate fiscal projections 5-10 years ahead but a minimum requirement is two years following the budget. Forecasts for 5-10 years may not be feasible in our situation since such long-term projections would become mechanistic and correspondingly distant from reality. However, we recommend that a start should be made by giving a forward projection for two years ahead of the budget. The Fiscal Responsibility and Budget Management Bill when enacted will make this mandatory (see 3.1.3 below).

The provision of information on Budget Estimates and Revised Estimates for the year immediately preceding the budget provides some indication of the degree of fiscal marksmanship. However, transparency would be increased if, for the previous year, data were provided on both Revised Estimates and Actuals. We recommend that this additional information should be provided, at least in the Budget at a Glance which presents an aggregative picture.

Current practice at the central government level is partially compliant with this guideline in the case of contingent liabilities but is totally non-compliant in the case of tax expenditures and QFAs.

Some information is being provided for contingent liabilities but it falls short of the requirement.

Loan guarantees given by the central government are reported in the budget. However, we note that letters of comfort, which are in many ways similar to loan guarantees, are not included. A guarantee not listed relates to the obligation to meet dues to all Life Insurance Corporation (LIC) policy holders. The liability of the government arising out of the exchange guarantee given for the Resurgent India Bonds and the India Millennium Deposits has been met through the issue of non-negotiable non-interest bearing securities of infinite maturity to RBI. This has not been reflected as a contingent liability in the past nor was the issue of securities shown as a part of the fiscal deficit. However, for 2001-02 onwards the liability is being shown on a certain basis in the budget.

The Manual also states that information should also be provided on other contingent liabilities such as indemnities against different types of risks, sometimes given in connection with infrastructure projects, calls on callable capital contribution, and also possible liabilities from legal action. It is difficult to make an assessment of the possible liability arising from legal action but liabilities related to indemnities and calls on callable capital are quantifiable. They are not likely to be significant, but they should be included which is not the case at present.

A major aspect of non-transparency related to indemnities is the potential demand on the central government for recapitalising public sector financial institutions. The central government does not have any formal obligation to recapitalise these institutions. However, recent actions indicate that under-capitalised public sector banks will not be allowed to close down their operations or privatised but instead will be pushed into restructuring programmes in which capital injection by the government would be one element of the restructuring. Recently, the government also provided support to the Unit Trust of India (UTI) to protect investors’ interests. Acceptance of large capitalisation responsibilities in connection with maintaining the health of public financial institutions is not unusual. However, it does raise questions about the possible fiscal risks on this count.

The Manual also states that the public policy purposes of each provision which gives rise to a central government contingent liability, its duration and intended beneficiaries should also be clearly stated. This is not done at present. We recommend that this basic requirement be implemented.

At present no information is provided on tax expenditures.

Tax expenditures include exemptions from the tax base, allowances deducted from gross income, tax credits etc. They serve the same purpose as expenditure programmes directed at the same end, but they are much less transparent since the extent of revenue loss is not quantified or explicitly approved by the legislature every year. The Manual states that inclusion of a statement of the main central government tax expenditure with a statement of the public policy purpose, intended duration and benefit is a basic requirement of transparency.

The treatment of tax expenditures varies considerably across countries (see Appendix V). While many industrialised countries have been reporting tax expenditures for some time, the practice is not limited to industrialised countries alone. Brazil, for example, reports some tax expenditures in its budget following the requirements of its Fiscal Responsibility Law. We recommend that a beginning should be made in systematic reporting of the revenue loss from major existing and new tax concessions and incentives. Initially, the report on tax expenditures need not be part of the budget for proposals in the budget but could be presented in the Economic Survey for the years preceding the budget. As confidence is gained in the method of estimation, the reporting of tax expenditures could be incorporated in budget documents.

A related issue is the system of refunds of tax and interest payments. The treatment of these refunds is not transparent at present as the tax receipts are shown net of expenditure on refunds and interest thereon. There is a need to reflect these refunds and interest paid on these as tax expenditure, which then can be subject to parliamentary and budgetary scrutiny.

The issue of tax expenditure is equally important at the state government level. State governments in India have also entered into an understanding that no fresh tax incentives will be given following the agreement on floor rates in sales tax structures. However, some states have not met this understanding drawn up by a high powered committee of finance ministers of states. It is important for states also to report tax expenditures from their concessions and incentives.

No information on QFAs is provided in the budget.

As already pointed out in 1.1.4 above, there are significant QFAs in the system arising from the activities of public financial institutions and non-financial public enterprises. The Manual states that it is important to 'identify and where possible to quantify, and to report information on quasi-fiscal activities’ but it also recognises that this is a complex matter. It therefore concludes that 'some pragmatism is called for in deciding on the range of quasi-fiscal activities to be reported and financial magnitude is certainly a criterion that should be applied in making this decision.' The Manual states that it is a basic requirement of fiscal transparency that a Statement on quasi-fiscal activities be included in the budget which indicates the public policy purpose of each QFA, its duration and the intended beneficiaries.

We recommend that a start may be made towards this objective by listing the major QFAs in the system. Initially, these estimates could be provided in the Economic Survey for earlier years.

Contingent liabilities, tax expenditures and QFAs also present major problems at the state governments level. All three take place on a significant scale at state government levels but state government budgets do not provide information on these items.

There has been an explosion of contingent liabilities of state governments arising from guarantees given for loans of state public sector enterprises, local authorities, statutory boards and other special purpose vehicles which is alarming. The Constitution requires that any state government that has outstanding dues to the centre, (which in effect means all state governments), cannot 'raise any loan' from the market without the express approval of the central government. On the face of it this provision should help provide a basis for fiscal discipline through the exercise of central government control over state borrowing. In practice, it has been widely evaded because no approval from the central government is required for giving a guarantee and state governments have relied extensively on this mechanism to mobilise resources. In some cases, the funds thus mobilised are deposited in the state’s public account and therefore become available to finance the state government’s normal expenditure.

The total outstanding guarantees extended by 17 major state governments now exceed Rs.100,000 crores. Contingent liabilities amount to about one fourth of the total liabilities of all 26 states. Some states have decided to impose a ceiling on the total quantum of government guarantees which is within the power of the state legislature. This is a commendable step that would improve fiscal discipline. It would necessarily entail monitoring of total state guarantees and publication of the outstanding position which is required for transparency. A start has been made in some states in reporting contingent liabilities, but the quality of reporting needs to be evaluated.

The state governments also fund loss making public sector organisations which are ostensibly performing essential social fuctions and are therefore engaged in quasi-fiscal activities. They also have implicit liabilities arising out of possible capitalisation requirements of state level financial institutions. The extent of QFAs is not identified or quantified. This is clearly an area where greater transparency is needed.

Our assessment of the present position regarding contingent liabilities, tax expenditures and QFAs at the central and state level and our recommendations regarding each are summarised below.

The guideline on debt reporting is substantially met though there are gaps that need to be filled.

A comprehensive report on central government liabilities is provided in the Economic Survey which is traditionally presented to Parliament in February, a few days before the budget, and can be treated as a budget related document. The information given in the Survey distinguishes between internal debt and external debt of the central government. Internal debt in turn is broken down into 'market borrowings' (securities issued by the Government of India as part of its market borrowing programme), 'other internal debt' (mainly small savings certificates) and other internal liabilities (deposits in the public account).

The budget documents also provide a comprehensive statement on liabilities in the Receipts Budget. However, unlike the Economic Survey, external liabilities reported in the Receipts Budget are valued at historical exchange rates. This is potentially misleading and we recommend that the basis of reporting should be changed to the market exchange rate.

The extent of correspondence between debt reporting at present and the standards outlined in the Manual may be summarised as follows:

• Debt should be reported by maturity, distinguishing between short-term (less than 12 months), medium-term and long-term. The existing system falls short of this guideline in some areas. The Economic Survey, which provides a comprehensive statement of total debt, does not provide any maturity breakdown. The Receipts Budget provides details of the maturity breakdown of outstanding government securities, but it does not provide any information on the maturity breakdown of other domestic liabilities or of external debt. Small saving liabilities are an important part of total internal debt and they have a defined maturity structure but information on this is not readily available anywhere. As regards external debt, the budget documents do not classify the debt into short, medium or long-term debt. The Ministry of Finance publishes an annual report on external debt 'India’s External Debt: A Status Report' but this publication does not provide information on the maturity structure of external government debt. The government generally does not incur short-term external debt, but part of its long-term external debt is short and medium-term on the basis of residual maturity which should be the relevant basis for classification. However, information on this aspect is not readily available in the public domain.
• Debt should be broken down by currency of issue. In the Indian context, all internal debt is rupee denominated while the external debt is denominated in various foreign currencies. The currency composition of total external debt is reported in the Ministry of Finance Status Report on External Debt.
• Debt should be broken down by residence of holder. In the past, rupee denominated debt was held by residents and external debt was invariably held by non-residents. Since Foreign Institutional Investors (FIIs) have been allowed to invest (within limits) in debt instruments, a part of the domestically issued rupee debt of the Government of India can be held by non-residents. Information is not available on the amount of government securities held by FIIs.
• Debt arrears should be disclosed with interest and principal arrears separately noted. There have been no debt defaults by the central government, so the issue has not arisen but the data reporting system could be elaborated to recognize this possibility and report nil arrears.

Best practice in debt reporting is represented by the SDDS, which in addition requires reporting of guaranteed debt and also 'encourages' reporting of debt service projections. There are some cases of central government guaranteeing of external loans routed to public sector entities either in the form of explicit guarantees or through Letters of Comfort. At present, loans covered by an explicit guarantee are not shown as part of central government debt in the budget documents but they are shown in the Ministry of Finance document 'Status Report on External Debt'. As regards debt service projections on existing debt for the next few years, these are provided in the 'Status Report on External Debt'.

As regards financial assets, the budget provides information on the government’s opening cash balance. Government holdings of equity in public sector companies are not given in the budget but this information is incorporated in the Public Enterprises Survey brought out annually with a lag of about two years by the Department of Public Enterprises. Total outstanding loans to public enterprises is also not given in the budget but is available in the Public Enterprises Survey.

The Manual states that best practice in this area requires publication of a government balance sheet as part of budget documentation, which should cover financial liabilities and assets and also non-financial assets. Moving to a government balance sheet may not be feasible in the near future, but we recommend a start should be made by making the listing of financial liabilities and assets more comprehensive. Information for the Public Enterprises Survey on government equity and loans to PSEs could be indicated in the Economic Survey.

2.1.5 Where sub-national levels of government are significant, their combined fiscal position and the consolidated fiscal position of general government should be published.

This requirement is fully met for central and state government budgets taken together. However, information on state government budgets is typically only available with a lag and information on local governments’ budgets is generally not available.

The Code recognises that it may not always be possible to present a consolidated picture of general government at all levels at the time of presenting the central government budget. This is certainly the case in India because some of the information on central government transfers to state governments, which is necessary to finalise state budgets, only becomes available at the time the central government budget is finalised, with the result that state government budgets are typically presented some months after the central government budget. The Code therefore accepts that the requirement for consolidation may only be feasible ex post.

The Reserve Bank of India consolidates data from the state budgets and presents a consolidated picture of the central and state government budgets in the RBI’s Handbook of Statistics, which is usually published in December. The Reserve Bank of India also publishes a consolidated and comprehensive study of state finances in December. This publication provides information on state government total debt and also on guarantees.

Although the consolidated position of the general government is available in the public domain it is generally not presented or discussed at the time of presenting the budget to Parliament. The Economic Survey includes a table in the statistical appendix which provides information on the consolidated position of the central and state governments (including the financing of Plan expenditure by public sector enterprises at both levels of government) but it does not actually discuss or comment on recent developments in the consolidated fiscal position. Given the scale of fiscal activity at the state government level and the size of state level fiscal imbalances, it is important to highlight the consolidated position at the time of discussion of the budget in the legislature. We recommend that the Economic Survey should incorporate a fuller discussion of recent trends in the consolidated position of central and state governments especially regarding trends in capital expenditures and in the basic fiscal balance measures (fiscal deficit, primary deficit, revenue deficit, etc.)

This guideline is further sub-divided into two elements, one relating to the legal basis for publication of fiscal information and the other to the commitment to announcing an advance release date calendar for publication of fiscal information.

The publication of fiscal information is not based on a legal obligation arising out of a budget law, but there is an established tradition of timely publication of all budgetary data by the central government.

The newly introduced Fiscal Responsibility Bill will, if enacted, create a legal commitment to provide certain information to be included in the Medium-term Fiscal Policy Statement, Fiscal Policy Strategy Statement and Macro-economic Framework Statement by the central government before the Parliament, along with the annual budget. However, the Bill as it stands does not cover all relevant aspects for disclosure of fiscal obligation [see our recommendation regarding 1.2.1].

India subscribes to the Special Data Dissemination Standards (SDDS), and in that context advance release calendar dates have been announced for release of data on central government operations. Compliance with these dates can be readily tracked. No dates, however, are specified in the states.

Transparency requires that the budget preparation and execution should be an open process, with adequate information being made available on how budgets are prepared and executed and also on the rationale underlying the budget strategy. In this context, the Code specifies requirements for budget documentation, budget presentation, procedures for budget execution and fiscal reporting.

A serious evaluation of the budget can only be made in the light of a clear statement of fiscal objectives, a quantification of the macro-economic framework within which the budget has to operate, any new policies being introduced by the budget and the fiscal risks which may derail the budget and call for possible mid-course corrections. The Code provides specific guidelines on each of these.

The Manual states that at the very least the budget documentation should indicate how the annual central budget fits in with the government’s broader objectives regarding government or public sector finances and its longer term deficit and debt targets. It recognises that statements about the longer term perspective could be qualitative but it points out that they are most helpful if they include quantitative detail on the longer term fiscal outlook.

At present, the Budget Speech of the Finance Minister outlines broad fiscal policy objectives in terms of fiscal deficits and levels of expenditure in critical areas, but the fiscal picture for the year to which the budget relates is not set in a quantified longer term framework of fiscal objectives, or analysis of fiscal sustainability. The only official medium-term projections are those made by the Planning Commission in the context of the five-year plans, but these projections are made once every five years and focus largely on the real sectors, and provide only a very aggregative constant price perspective on fiscal magnitudes. They do not provide a continuous rolling perspective of the fiscal position against which current fiscal performance and targets can be evaluated.

The treatment of fiscal sustainability at present is also unsatisfactory. Although the importance of fiscal sustainability has been recognised in general terms in budget speeches in recent years, there has been no attempt to quantify the relevant criteria for determining fiscal sustainability so that current fiscal policies can be judged in that context. Of course, assessing fiscal sustainability is extremely difficult and inevitably, as the Manual points out, there is little alternative but to resort to arbitrary rules.

The recently introduced Fiscal Responsibility and Budget Management Bill would, when passed into law, represent a major improvement in stating fiscal objectives and assessing fiscal sustainability. It would commit the government to an explicit phased reduction in the fiscal deficit and the revenue deficit over a five-year period (see Box 7). The Bill provides that the Government shall at the time of presenting the budget, also present the following:

• A Medium-term Fiscal Policy Statement which would set forth a three-year rolling target for prescribed fiscal indicators with specification of underlying assumptions. This statement would also include an assessment of sustainability relating to the balance between revenue receipts and expenditures and use of capital receipts including market borrowings for generating productive assets.
• A Fiscal Policy Strategy Statement, which would contain the policies of the Government for the ensuing year in relation to taxation, expenditures, market borrowings and other liabilities, lending and investment pricing of administered goods and services and description of other activities which have potential budgetary implications. It would indicate the key fiscal measures and rationale for any major deviation in fiscal measures pertaining to taxation, subsidy expenditure, administered pricing and borrowings. It would also indicate how current policies are consistent with the Medium Term Fiscal Policy Statement.
• A Macro-Economic Framework Statement. The content of this statement has not been spelt out.

The need to evaluate annual budgets in the context of an explicit medium-term framework is well established. The OECD Best Practice guidelines on budget presentation and documentation provide that there should be a pre-budget report, no later than one month prior to the tabling of the annual budget in Parliament which would state the Government’s medium-term economic and fiscal intentions highlighting total revenue and expenditure, fiscal deficit and debt. This could facilitate debate on the overall fiscal policy objectives and strategy prior to finalisation of the budget and presentation of detailed expenditure and revenue proposals to the legislature. The OECD guidelines also indicate that the draft budget should be presented no less than three months before the start of the relevant fiscal year, with the budget being approved before the fiscal year begins.

It is important to note that it is not developed countries alone that are complying with such guidelines. In South Africa, a Medium-term Budget Policy Statement is presented to Parliament three months before budget day. The OECD best practice guidelines also suggest that a long-term report assessing fiscal sustainability of current fiscal policies should be put out once every five years. Such a report should take account of the outstanding debt and also of the growth of future liabilities on pensions reflecting the ageing of the population.

In our view, much greater clarity and quantification of the medium-term framework in which a budget is framed and an analysis of fiscal sustainability is urgently needed. We would therefore recommend that the proposals regarding the placing of a Medium Term Fiscal Policy Statement, A Fiscal Policy Strategy Statement and a Macro-Economic Framework Statement should be adopted as part of normal budget practice with effect from the budget for the year 2002-03. This would lead to a substantial improvement in clarifying the macro-economic context and medium term fiscal framework in which the budget is formulated.

Until recently, budgeting in India has not been constrained by any fiscal rules so the question of specifying the rules did not arise. However, as indicated in Box 7, the Fiscal Responsibility Bill proposes a major change in the area because of the following:

1. Fiscal deficit of the government to be reduced by ½ per cent of GDP per year to reach 2 per cent of GDP by March 2006.
2. Revenue deficit to be reduced by ½ per cent of GDP per year and entirely eliminated by the year ending March 2006.
3. Total liabilities of the central government not to exceed 50 per cent of GDP by April 2011.

Adoption of these targets would be a major improvement in fiscal discipline.

While the intention to specify the time path and pace of fiscal adjustment is welcome, it should be noted that this has to be done with due recognition to the quality of fiscal deficit reduction i.e. whether it is being achieved by cutting expenditures which are high or low priority and also the problem of cyclical variation. Cyclical factors may lead to short term increases in the deficit which need not lead to corrective action if the 'structural deficit' is on track. The medium-term targets should therefore be applied to the structural deficit. The Fiscal Responsibility Bill does not address these issues explicitly but they will need to be addressed as the proposals in the Bill are implemented.

The basic macro-economic building blocks on which a budget has to be based are the following:

• Projected GDP growth and its supply-side sustainability.
• Aggregate demand factors consistent with projected growth, i.e., the investment rate, the savings rate and the net import-export balance.
• Projected domestic inflation.
• Projected rate of interest.
• Projected capital inflows and their consistency with the net import–export balance and reserve movements.
• Projected exchange rate.

At present, the assumption or projections of these variables on the basis of which the budget is formulated are not spelt out and certainly not made public as implied by the guideline.

The Budget Speech of the Finance Minister provides some indication of the underlying assumptions but the information provided is very limited. It usually indicates a projected growth rate for the economy. There is also some indication of inflation objectives but this is usually qualitative or directional. The Monetary Policy Statements of the Governor of the RBI are more precise in that they indicate an inflation range on which the monetary policy is based. No such range is explicitly indicated as the basis for budget making though the implicit rate of inflation in the GDP deflator can be worked out from the real growth targets and the quantification of the fiscal deficit as a percentage of GDP. No information is provided on the other macro-economic variables listed above.

Many countries, both industrialised and developing, have some formal methodology for macro-economic forecasting and policy formulation. The U.K. Treasury is required by law to make its macro-economic model publicly available. In Australia, the Treasury macro-economic model can be viewed at the Treasury website. In Germany, there is no official model but the government interacts with several research institutions, which have developed models of the economy. The Ministry of Finance in Argentina has a macro-economic model that is utilised to make budget projections, after incorporating inputs from various departments in the Ministry.

We recognise that there are obvious difficulties in having an official forecasting model for the Indian economy developed by and operated by the Finance Ministry. It would require a substantial professional input, with continuous updating of data and parameters. There are also problems arising from the fact that models differ significantly in specifications and calibration which can lead to very different results and the Finance Ministry may not wish to commit itself to any particular specification. However, the Finance Ministry could finesse this problem by sponsoring two or three research institutes to develop alternative models capable of simulating the major macro-economic variables, e.g., (1) a macro-economic balance; (2) a tax equation with scope of buoyancy and elasticity considerations; (3) an expenditure equation with scope of reduction or expansion; (4) if possible, a representation of non-financial public sector operations; (5) private investment, constraints thereof; and, if possible, a recognition of private sector investment in infrastructure; (6) foreign trade and international finance; (7) foreign investment; (8) labour market; (9) price controls; (10) a monetary equation; and (11) financial sector characterisation. It might even be possible to build in some estimates of the impact on poverty also.

The structure of these models could be published and reports on their forecasts could also be published each year, with mid-year updates. Each of the models would produce different results, and it would be left to the Finance Ministry to use these results as inputs to prepare a consistent macro-economic framework for the budget going forward a few years. The Ministry would be under no obligation to reconcile each model results with its own assumed medium-term macro-economic framework, but the process of discussion and debate on these model results would improve the basis on which internally consistent short-term and medium-term forecasts are made.

The Planning Commission, which also has an interest in medium-term forecasting and uses its own models for its purposes, should also be interested in the exercise as also the Reserve Bank. These organisations should share the cost of such external model building. They would be in a position to use the results of the forecasting models in their own work independently.

We would recommend that a start should be made in the direction of making the macro-economic forecasts underlying budget formulation more explicit. It would help to prepare the kind of medium-term documentation that is needed for implementing the various commitments under the Fiscal Responsibility Bill.

Expenditure on new programs is distinguished from existing commitments. The Demands for Grants lists out the new items of expenditure. In an accounting sense, there should be expenditure provision for the new programs introduced in the budget. Generally, all new policies are envisaged as part of development plans which get covered for under allocation towards developmental heads, but this should be shown separately.

The budget documents do not provide details on fiscal risks, except in a minimal sense, in the form of contingent liabilities viz. loan guarantees. However, macro-economic risks in terms of the likely impact of changes in the underlying macro-economic forecasts are not quantified. Other specific sources of risk on the revenue side e.g., uncertainty about privatization receipts, or on the expenditure side e.g., implicit guarantees to public financial institutions which are seen to be fragile and which need support are also not quantified.

Best practice is that a statement of fiscal risks should be included in the budget documents. This is done for example in New Zealand and Greece. A detailed quantification of all risks may not be feasible but we recommend that a start should be made by the central government in quantifying major fiscal risks. Quantification of macro-economic risk would require a macro-economic model to take account of general equilibrium effects of a macro-economic shock.

The Code contains specific guidelines relating to data classification, program objectives and indicators of government fiscal position.

This guideline seeks to ensure that information in the budget is useful for analysis and control and evaluation of activity in particular areas. The manual elaborates that the classification system should be compatible with GFS standards. This provides a widely accepted basis for economic classification of revenue and expenditure and also facilitates international comparisons.

Budget data are recorded in gross terms and on a cash basis, in a manner consistent with the GFS standards and hence can be compared internationally. The transactions are classified by administrative agencies, economic category and functions and programs. A functional classification of expenditure is not made available at the time of budget presentation but is published separately some months later [see Appendix IV-A.IX].

In addition to the traditional classifications, Indian budget data also make a distinction between plan and non-plan expenditure [refer to Appendix VI].

The Budget Speech indicates the broad objectives to be achieved by major budget programs. Targets for individual programs in the social sector are also spelt out in the expenditure budget. The Manual emphasises the need to go beyond 'line-item budgeting' to specify programs, sub-program and activity. Categories, e.g., expenditure on 'public health' should be linked to government’s broad aims of promoting preventive health care and specific objectives could be given, e.g., an anti-malarial sub-program. This is generally done in India.

The Manual notes that the overall balance of the general government is the appropriate indicator of the impact of fiscal policy on aggregate demand but it also notes that there may be practical or constitutional reasons why the overall balance of the central government is the relevant standard measure of fiscal performance. In India, the states contribute significantly to the fiscal deficit and a consolidated measure, including the states component is therefore needed. However, information in the budgetary outcome in the states is only available with a considerable lag. Discussing the budget therefore focuses on the overall balance (fiscal deficit) of the central government in the background of movements in the consolidated balance. The practical difficulties of presenting a consolidated picture at the time of presentation of the central budget have been discussed earlier.

The Manual notes a number of conceptual issues related to measurement of the overall balance that are relevant for transparency.

This guideline is potentially relevant for India but is not met.

As pointed out earlier, various public sector agencies outside the general budget conduct significant quasi-fiscal activities and these are not reflected in the conventional measure of the overall balance. The Manual states that in such situations, identification and quantification of quasi-fiscal activity comprise a basic requirement of fiscal transparency. The Manual suggests that a system of reporting should be established which covers the broader public sector and the overall public sector balance (including the activities of the non-financial public sector enterprises) could be used as an additional measure of the government’s fiscal position.

Given the size of the public sector in India there can be no doubt that the public sector deficit has important macro-economic significance. However, it is important to emphasise that the entire deficit of public sector enterprises cannot be attributed to QFAs. Most public sector enterprises are dominantly engaged in commercial activity operating in a competitive environment with other private sector units. A part of the deficit of these enterprises is undoubtedly due to the implicit loss of revenue arising from social objectives but the bulk of the deficit is a reflection of their normal commercial operations, including especially financing of investment and it is not possible to distinguish between the two. In view of this factor we are not convinced that aggregating the central government fiscal deficit with the deficit of public sector enterprises provides a useful measure of fiscal stimulus.

Nevertheless, while noting that the public sector deficit is not a purely fiscal phenomenon, it can be argued that the size of the deficit should be made known. Since the expected deficit of the public sector enterprises can be derived from the financing plans of these enterprises for meeting plan expenditure, and this is known at the time the budget is presented, we recommend that a start should be made in providing information on the expected deficit of the public sector enterprises in the budget documents. This could be included in the Budget at a Glance.

Under this guideline, the Code contains separate guidelines for the accounting system, procurement and employment, internal audit and tax administration.

3.3.1 A comprehensive, integrated accounting system should be established. It should provide a reliable basis for assessing payment arrears.

There is a well-established accounting system based on financial rules and account Codes. However, it does not provide information on payment arrears. It also does not capture information at the commitment stage nor does it maintain records on aid-in-kind.

The process of internal audit also does not function as a mechanism for generating information on performance effectiveness. However, this is in large part due to the lack of interest in developing such a mechanism on the part of Ministries which are central, to focus on expenditure rather than effectiveness of expenditure in achieving ultimate objectives.

This is substantially met.

Regulations governing government purchases are clear. Tenders for contracts above a threshold size are conducted in a transparent manner and are open to audit.

Procedures regarding employment in the public services come up to the required standards of transparency. Vacancies are advertised and filled through competition. Recruitment for senior positions in the central government is through the Union Public Service Commission which is an independent body established under the Constitution. Similarly, recruitment of senior staff in state governments is done by the State Public Service Commissions, which are also constitutional bodies.

This guideline is substantially met.

There is a well established internal audit system in all central government departments. The system works with Internal Audit Cells under the guidance of Controller of Accounts and Financial Adviser of the ministries and departments. A detailed account of the system is given in Appendix VIII. The effectiveness of the internal audit system is examined and commented upon by the Constitutional Audit conducted by the Comptroller and Auditor General of India.

Although the internal audit system is well established, it can be argued that not enough use is made of internal audit as a critical management information tool to appraise performance.

The Manual elaborates that in order for the tax administration to be seen to be conducted in a fair and impartial manner free of political intervention, the heads of the tax administration should be appointed by law and be given some protection against removal from office and also against political direction in interpreting tax laws. The Manual also proposes that revenue collecting agencies should provide a timely annual report to the legislature on their activities during the year. Their reports should cover performance data (actual collections relative to budget) and should also provide details of actions being taken to improve compliance with tax laws.

This guideline is partially met.

The Annual Report of the Finance Ministry submitted to Parliament includes a report on the activity of the Revenue Department. The requirement that the heads of the tax administration should be appointed by law and given time protection from removal from office is partially met by the fact that some protection is normally given to all civil servants. However, there are no special provisions for this category.

Under this general guideline the Code includes detailed guidelines relating to budget and extra budgetary out-turns, final accounts and program results.

There is no provision for a mid-year report to the legislature at present. However, the Fiscal Responsibility Bill, when enacted, will provide for regular quarterly reporting to the Parliament of trends in expenditure and revenues in relation to projections in the budget.

Although there is no mid-year report in Parliament, the aggregate budget outturns of the central government covering expenditure, revenues and fiscal deficit are publicly available on a monthly basis based on the information provided by the Controller General of Accounts (CGA).

At the state government levels for most states, there is no intra-year information on their budgetary outturns. However, information on the market borrowing programme of the states is provided regularly by the RBI.

The information on extra-budgetary outturns is available only in the Annual Finance Accounts at the Central level.

This requirement is fully met.

Finance accounts are prepared by the Controller General of Accounts in the Ministry of Finance. These are audited by the Comptroller and Auditor General (CAG) and audited Accounts are presented to Parliament within a year of the end of the fiscal year. However, there is no law or rule which prescribes the timing or coverage of these accounts. While the coverage of the accounts is comprehensive, they are not presented in the same format as in the budget. In the case of the state governments, the CAG prepares both the Finance Accounts and the Appropriation Accounts.

Ministries present performance budgets, which specify the costs and results to be achieved by major programs and contain a review of recent performance in each programme. Past performance is also reported in the Annual Reports of respective Ministries presented to the Parliament. However, these performance assessments cannot be viewed as critical assessments of the effectiveness of the programmes in achieving stated objectives. They tend to focus mainly on whether expenditure levels have been met and, where some physical targets have also been laid down, whether these have been achieved. Many of these physical targets are not sufficiently indicative of effectiveness in achieving the underlying objective.

The Planning Commission has a system of Program Evaluation to evaluate plan projects which provides a more critical independent evaluation of whether programmes are achieving their objectives. Some ministries have also contracted out evaluations to be conducted by research institutions and independent consultants. The CAG has also subjected selected programmes to critical review and these reports are available to the legislature. However, it is also true that because of the time lag involved these reports receive less critical attention than they deserve.

The quality and integrity of fiscal data reported by the government is obviously critical for making any claim to fiscal transparency. The Code provides a number of guidelines relating to data quality standards and institutional mechanisms for independent audit.

Under this general guideline, the Code provides specific guidelines outlining good practices relating to budget data, the accounting basis and assurance of quality.

The Manual points out that poor budget preparation leads to many problems of implementation such as short-term cash rationing, payment arrears, frequent supplementary budgets, etc. Credible budgeting therefore requires that revenue forecasts and expenditure estimates should be of high quality. Revenue forecasts are particularly important and should be fully explained in terms of recent revenue trends, likely macro-economic developments and impact of tax policy changes.

This guideline is partially met.

Expenditure and revenue forecasting has not been as accurate as it should be. This can be attributed to the inherent difficulties in achieving high standards of 'fiscal marksmanship' in a situation where structural changes are taking place. However, transparency, according to the Manual, requires that the specific method used for forecasting should be indicated. This is not done.

The situation at the state level is much worse than at the center with substantial deviations between Budget Estimates and actuals in many areas. State budgets have typically over-estimated the likely revenue realisation and the states borrowing potential to supplement these revenues while under-estimating expenditures and liabilities on account of losses of the State Electricity Boards and other such undertakings.

On the expenditure side rigorous costing procedures, effective control measures and a critical review of existing schemes/programmes through systems like zero-based budgeting are crucial to avoid overshooting of expenditure and supplementary budgets.

Accounting policies are not stated in the budget documents, but they are based on established rules (General Financial Rules and Receipt & Payment Rules) and published accounts Codes and Manuals. Budgeting and Accounting in India is on cash basis. No attempt has been made so far to shift to accrual accounting. Information is, however, available separately on committed liabilities, revenue demands raised but not realised, payment arrears, etc.

The issue of cash versus accrual accounting is a difficult one. Appendix VII provides a brief description of accrual accounting as it is practiced in New Zealand and indicates the advantages of the system. It is our assessment that a switch to accrual accounting may not be feasible in the near future until a more substantial database is developed. However, a beginning can be made by providing certain information as part of budget documents. A list of such documentation is given in Appendix VII.

This guideline is substantially met.

Appropriation Accounts and Finance Accounts provide accurate and comprehensive accounting information to Parliament. The veracity, correctness and quality of these accounts are certified by the CAG.

India also subscribes to the SDDS which enforces its own discipline.

This part of the Code deals with good practices relating to independent audit, independent assessment of fiscal and macro-economic forecasts and the integrity of fiscal statistics.

This is fully complied with.

The Accounts of the central and state governments are examined in accordance with the requirements of Article 149 and 151 of the Constitution of India and the Comptroller and Auditor General’s (Duties, Powers and Conditions of Service) Act, 1971. The CAG is an independent constitutional authority that reports to Parliament and state legislatures. The CAG cannot be removed by the government and is also not eligible to take up any government employment on completion of his term.

The audited accounts are presented in a Report, titled Appropriation Accounts, respectively, for the centre and the states. Additionally, the Finance Accounts contain the accounts of receipts and disbursements for the respective government for the specified year, together with the financial results disclosed by the revenue and capital accounts, the accounts of public debt and the liabilities and assets as worked out from the balances recorded in the accounts.

Finance Accounts of the central government are prepared by the Controller General of Accounts and certified by the CAG of India. In the case of states, the CAG prepares both the Appropriation and Finance Accounts.

The Manual states that best practice is that institutional mechanisms should be established to provide the public with independent assurance that fiscal and macro-economic forecasts are of high quality. For example, in France, the macro-economic forecasts are reviewed by the National Economic Commission. The Commission is chaired by the Minister of Economy & Finance and includes twenty two members chosen for their economic and financial expertise. In the Czech Republic, before the budget forecasts are released, a panel of experts including individuals from the private sector scrutinise the macro-economic assumptions. In the United States, the Congressional Budget Office prepares a complete set of macro-economic and fiscal forecasts which are considered by the US Congress alongside those contained in the budget proposals submitted by the Executive Branch.

At present, there is no domestic mechanism for subjecting fiscal forecasts and the macro-economic forecasts and assumptions on which they are based to scrutiny by independent experts. The normal Article IV Consultations of the IMF qualifies as an independent mechanism but the report is not publicly available. Part of the problem is that we do not have a complete specification of the macro-economic picture framework in which the budget is formulated which makes it difficult to evaluate internal consistency of assumptions. If the proposal to introduce a Medium-Term Fiscal Policy Statement, Fiscal Policy Strategy Statement and Macro-Economic Framework Statement along with the annual budget is implemented, it would provide the basis for independent scrutiny. Even if a formal institutional mechanism for independent review is not established, the very presentation and publication of these documents would lead to comment by academics, research institutes, rating agencies and the economic press, all of which would be of great advantage. Our recommendation that alternative macro models be developed by selected national research organisation for use by Ministry of Finance, Planning Commission and Reserve Bank of India, should also, in turn, put under examination the extent of use of the models by the official departments.

This is partially met.

At present, the Central Statistical Organisation (CSO) is the principal national statistics agency which functions under the Government of India. There is no legislation providing assurance of independence of Central Statistical Organisation. However, the entire issue of the organization of the statistical system, including possible changes in the system, is currently under the consideration of a National Statistical Commission, which is expected to submit its report shortly.

The existing system of scrutiny and audit by the CAG provides adequate assurance of the quality of fiscal data at the central and state government level. However, the CSO could be charged with integrating the available data and presenting it in a consolidated fashion which can be used consistently with the national accounts.

The overall assessment that emerges from our review is that current fiscal practices at the central government level satisfy the minimum requirements of the Code on Fiscal Transparency in many areas, though there are deficiencies in some important areas which need to be addressed. The position at the state government level is much less satisfactory with most states being well behind the standards achieved by the central government. However, since at present the Code is expected to be applied only at the national government level, non-compliance at the state level does not amount to non-compliance with the Code.

In the rest of this chapter, we present our summary assessment of the degree of compliance with the Code, highlighting the areas where current practice is deficient and indicating our recommendations for action needed to achieve full compliance.

Most of the requirements of the Code in this area are fully met. The roles and responsibilities of the central and state governments are well defined and there is a clear legal framework governing the management of the budget and extra-budgetary funds. The division of expenditure and tax powers is complex, with areas of overlapping responsibility and multiple channels for resource transfers, but the basic requirements of transparency are met.

In the interest of transparency, we recommend that the institutional table for India in the Government Finance Statistics, IMF, should be revised to make it comparably detailed with the entries for other countries. For example, various central government institutions which have their own budgets outside the central government budget could be identified e.g., the employees provident fund, central universities, the Indian Institutes of Technology, central research institutions could be shown separately.

The most important deficiency relates to the prevalence of quasi-fiscal activities (QFAs) undertaken by the banking system and by non-financial public sector enterprises which are not transparently identified and quantified.

An area where current practice could be improved relates to the prescription that the basic principles of budget management should be embodied in a general budget system law which should have constitutional or near constitutional status. Although the scope of the budget has been defined by the Constitution and clear budget procedures have evolved over time, we do not have a budget system law. We recommend that the government could consider amplifying the scope of the FRBMB to include the essential elements of a budget law.

Another area of concern relates to the issue of transparency in tax laws. Although the principle that taxation must be levied on the basis of explicit legal authority is strictly complied with, our tax laws are lacking in transparency. The complexity of the tax structure, especially the large number of exemptions, create room for uncertainty and administrative discretion. Administrative procedures are archaic and involve direct interaction between the assessee and the tax administrator which creates the possibility of case by case determination of tax liability. A major effort at simplification, and greater use of information technology, especially electronic filing, is urgently largely needed.

Government involvement in the private sector through regulation and equity ownership is exercised on the basis of clear legal authority. However, the criteria to be used in the exercise of executive authority are not always very clear. We recommend that the OECD recommendations relating to the characteristics of transparent regulations, be treated as indicators of best practices and existing rules and regulations be systematically reviewed in the light of these guidelines.

There are significant gaps in this area between current practice and the requirements of the Code.

This part of the Code seeks to make the assumptions underlying the budget and the rationale of budget policy more open and available for scrutiny by the legislature and the public. There are significant deficiencies in current practice in this area which need to be corrected though many of these will be substantially addressed once the FRBMB is enacted.

This part of the Code is substantially complied with. The institutional mechanisms for independent audit of fiscal data are very strong. The accounts of both the central government and the state governments are audited by the CAG which has developed an enviable reputation for high standards of independent and strict scrutiny.

The main deficiencies in this area are the following:

To summarise, current fiscal practices at the central government level meet most of the minimum requirements of the Code and the gaps identified above could be closed in a relatively short time. We recommend that action should be taken as outlined above to close these gaps so that India can be said to be complying fully with the Code. The FRBMB contains many of the important initiatives needed in this context.

Fiscal practices at the state level are generally behind the standards achieved at the central government level and there are many gaps in comparison with the requirements of the Code. Since the Code is not being applied at present below the national level, non-compliance at the state level does not constitute non-compliance with the Code. However, from a substantive point of view, it is obviously important that fiscal transparency should be extended to state government levels also, since the scale of fiscal activity at the state and local levels is very large.

In January 2000, a Core Group on ‘Voluntary Disclosure Norms for State Governments’ was constituted consisting of the Finance Secretaries of Gujarat, Kerala, Meghalaya and West Bengal and representatives of the Reserve Bank of India, Planning Commission and Ministry of Finance. The objectives of the Core Group were to study transparency practices in state governments budgets, and to make recommendations for improvement.

As an initial step, the Core Group suggested a common format for all the states along the lines of page 1 of Budget at a Glance (BAG) of the central government to be provided in the state budgets to be presented in 2000-01. As a follow up, the Governments of Tamil Nadu, Madhya Pradesh, Gujarat, Maharashtra, Punjab, Meghalaya, NCT Delhi, Orissa, Haryana and Uttar Pradesh have already presented critical summary indicators in the budgets for 2000-01. Further, Bihar, Goa, Himachal Pradesh, Kerala, Rajasthan, Tripura, West Bengal, Chattisgarh and Jharkhand have also introduced these indicators for the budgets of 2001-02.

The Core Group has further recommended as follows:

In the absence of full fiscal transparency at the level of states, it is obviously difficult to evolve a consistent approach to fiscal policy at the state level. We therefore recommend that as part of item (iv) above, the Finance Secretaries Forum could review the Report of the Advisory Group on Fiscal Transparency and determine a set of minimum standards on transparency which all state governments should achieve within a three year period. In particular, we would recommend that state governments should be encouraged to increase the extent of reporting on contingent liabilities and at least major tax expenditures and QFAs (especially losses of SEBs).

Appendices

I. Composition of Advisory Group on Fiscal Transparency
II. Extra Budgetary Funds
III. Oil Coordination Committee (OCC) Oil Pool Account
IV. Budget Related Documents at Central and State Levels
V. Tax Expenditure Reporting in OECD Countries
VI. Classification of Expenditures: Plan vs. Non-Plan
VII. Accrual Accounting in Government
VIII. Internal Audit and Internal Control System in the Central Government

The Group benefited immensely from valuable interactions/inputs from Dr.Y.V.Reddy, Deputy Governor, Shri S.S.Tarapore, Former Deputy Governor, Reserve Bank of India, and Shri A. Premchand, Fellow, National Institute of Public Finance and Policy, New Delhi, as well as officials of the Comptroller and Auditor General of India and state Finance Secretaries.

The Group also acknowledges the valuable assistance rendered by Dr. R Kannan, formerly Adviser with the Reserve Bank of India who was initially the Convenor of the Group.

The following are the main extra-budgetary funds currently in existence. As pointed out in Box 2, all revenue receipts into these funds and expenditure from the funds are accounted for in the budget.

1. NATIONAL SMALL SAVINGS FUND: This fund was established on April 1, 1999 and reflects a new method of accounting for small savings collections. All small savings collections (including Public Provident Fund) are credited to this fund. Similarly, all withdrawals under small savings schemes by the depositors are made out of the accumulations in this fund. The balance in the fund is invested in the central and state government securities. The investment pattern is as per norms decided from time to time by the Government of India. From 2000-01, 80% and 20% of the net collections (gross collections minus withdrawals by depositors) is being invested by National Small Savings Fund in special securities issued by the state and central government respectively. The debt servicing of these government securities is an income of the fund while interest paid to subscribers and cost of management of small savings schemes is an expenditure of the Fund.

2. CENTRAL ROAD FUND: The proceeds of the cess levied on motor spirit and on high speed diesel are credited to this fund. It is utilised for financing constructions and maintenance of approved schemes of road works in the states and union territories and also for financing construction of Railway overbridges/Railway safety works at unmanned railway crossings.

3. SUGAR DEVELOPMENT FUND: It is being used for development of sugar industry and the matter connected therewith and also making loans, grants and other expenditure relating to development of sugar industry. The fund is financed from levy of cess of Rs. 5 per quintal from 1^st June, 1982 and Rs. 14 per quintal from 1^st November, 1982 on production of sugar.

4. STEEL DEVELOPMENT FUND: In June 1978, a development surcharge of Rs. 100 per tonne was levied on certain categories of steel for the purpose of modernisation, rehabilitation and development of the steel industry. The Joint Plant Committee (set up under the iron and steel control Order, 1956) was authorised to add an 'element' to the ex-works prices of the main producers for the purpose of the Steel Development Fund (SDF).

The fund was set up initially for the development and rehabilitation of the steel industry. The scope was subsequently expanded and the fund was utilised mainly for:

1. Financing schemes, projects and capital expenditure for modernisation, research and development, rehabilitation, diversification, etc.
2. Payment of rebates to the Small Scale Industries Corporations (SSICs) on supplies made by the main producers.
3. Meeting the expenditure of the Economic Research Unit (ERU) of the Joint Plant Committee (JPC) set up on 1.9.1983.

The development surcharge which generated revenues for the Fund was discontinued in 1994. However, the corpus of the Fund continues.

5. RAILWAY RESERVE FUNDS: These funds comprise of (i) Railway Pension Fund, which is intended to meet the pensionary charges of railway employees. Suitable amounts are transferred annually to the fund by debit to revenue and capital expenditure heads. (ii) Railway Depreciation Reserve Fund, which is to meet the requirement for replacement and renewal of assests including the improvement element and is fed annually by debit to revenue. (iii) Railway Development Fund, which is to meet the cost of all works for providing amenities to passengers and other railway users, the cost of labour welfare works and the entire expenditure on un-remunerative improvement works costing more than a prescribed limit. The fund is financed by appropriation from the portion of railway surplus. (iv) Railway Capital Fund, which was created in 1992-93 to enable the railways to utilise the part of the internally generated resources for building up the infrastructure of the railways. This is also fed from railway surplus. (v) Railway Safety Fund, which is created from 1.4.2001 for financing safety works relating to the conversion of unmanned level crossing and construction of railway over/underbriges at busy level crossings. The fund is financed from railway surplus and also from Central Road Fund.

6. TELECOMMUNICATIONS RESERVE FUNDS: Since the Department of Telecom Services has been corporatised with effect from 1.10.2000, operation of Telecommunications Reserve Funds have been dispensed with.

In the context of the international oil price crisis in 1973, government undertook two major initiatives: one to step up efforts to enhance domestic crude oil production, which resulted in the establishment of Oil Industry Development Board (OIDB) and the levy of a cess under the OIDB Act for funding OIDB activities. The second was the setting up of the Oil Coordination Committee (OCC) to provide for an institutional mechanism for ensuring consumer price stability of POL products. Prior to 1974, import parity was the basis on which the prices of petroleum products were determined, which was abolished in 1974; with the pooling of prices of indigenous and imported crude oil for the purpose of fixing the prices of the petroleum products (started from March, 1974). The main reason for the abolition of import parity was that the imports then formed less than 10 per cent of the total demand of the country and that the additional requirement of petroleum products in the future years was expected to be met by refining additional indigenous crude oil in India.

The oil refineries obtain crude oil from domestic sources as well as from imports. The petroleum products are both manufactured as well as imported. In order to ensure that there is no variation in price for the end use customer depending on whether the product item has been imported or manufactured from domestic crude or manufactured from imported crude, the refineries are treated as the primary pricing points with a uniform ex-refinery price alongwith a scheme for freight equalisation. The refineries get reimbursed on the basis of cost of inputs and operating costs (on a normative basis) plus a fixed rate of return. This requires operation of a series of accounts – Crude Oil Price Equalisation (COPE) account, Cost and Freight Adjustment (C&FA) account, Product Price Adjustment (PPA) account etc. These pool accounts are in the nature of price equalisation funds and are administered by the Oil Coordination Committee, which was constituted in July 1975 by a Government of India Resolution. Functions of OCC are controlled by government and the expenditure is met out of C&F Account. Under this scheme, surpluses of oil companies over and above their retention prices, which make allowance for 12 per cent post tax return on net worth or capital employed, whichever is less, are to be surrendered to OCC.

In 1977-78, Ministry of Petroleum & Natural Gas (MoPNG) decided that, in place of the then practice of keeping these as deposits in the commercial banks, surplus funds of OCC were to be kept in a deposit account in the Public Account of India. Upto 1987, government allowed interest to be levied on the OCC Pool deposits. Then it was decided that interest need not be paid as the government has no liability to repay or pay this amount to any external agency. On this basis, even the interest credited earlier was got refunded.

The Oil Pool Account is maintained to provide uniform and stable prices to consumers and reasonable retention margins for oil companies. The account is expected to be self-balancing in a longer time frame. Inflows to the account are from collection of surcharges on sale of petroleum products while the outflow is for meeting variations in the cost. The difference between inflows and outflows represent the surplus/deficit position of the Oil Pool Account.

The mounting outstandings of the oil companies from the Oil Pool Account impacted adversely on the financial position of oil companies and resulted in their facing severe cash crunch necessitating huge borrowings. The borrowings which were around Rs.17,500 crore as on December 31, 1996 were estimated to rise to a staggering level of Rs.23,000 crore by March 31, 1997. On 1^st September, 1997, the Government took major decisions to deal with the accumulated deficit in the oil pool accounts and kick-started a process of gradual dismantling of Administered Price Mechanism.

Government decided, while approving revision of prices of POL products on 1^st September 1997, that the outstanding liabilities of the OCC to the oil companies as on 30^th June 1997 (estimated to be about Rs.18200 crore) would be taken over by the government and bonds of equivalent amount would be issued to the oil companies. The Oil Companies were required simultaneously to invest these amounts in Special Government Bonds issued by the Reserve Bank. Under this expedient '10.5% Oil Companies (Non-transferable) Special Government Bonds 2005' totaling Rs.12,984 crore were issued by the Government to the Oil Companies towards the outstanding receivables of the Oil companies from the Oil Pool Account as on March 2, 1998. To the extent of this involvement of issue of government bonds and their redemption from the OCC surpluses, the OCC accounts stand partially fiscalised.

The document and publications from the central government and the state governments relating to the budget and analysis of related data are summarized below.

Tax incentives and exemptions are often used to achieve goals that could be achieved equally through public spending. Nevertheless, such benefits are often given through the tax system rather than through expenditure channels. But the accounting for costs and benefits of tax measures is often less rigorous than for direct expenditure. Treating these incentives as 'tax expenditures' and requiring such expenditure to be reported in Budget documents is a step towards transparency and greater accountability.

Tax expenditures include exemptions from the tax base, allowances deducted from gross income, tax credit amounts deducted from tax liability, rate reliefs as a reduced rate of tax applied to a class of taxpayer or activities, and tax deferrals in the form of delay in paying tax. Conceptually, tax expenditures are difficult to calculate as it requires the classification of a tax system into a benchmark or normative level so as to arrive at the tax expenditure, in terms of the deviation from the norm. Nevertheless, a number of OECD countries regularly estimate and publish tax expenditures, either as part of a legal requirement of the budgetary process or as a source of information for parliamentarians, government officials and others.

There are three principal methods to calculate the ‘cost’ of tax expenditures:

1. Revenue foregone, which is an ex post measure to arrive at the amount by which revenues are reduced by a particular provision, for example in Australia, Austria, Belgium, Canada, Finland, Germany, Ireland, Italy, Netherlands, Portugal, Spain, United Kingdom and the United States.
2. Revenue gain, which is an ex ante measure of the expected increase in revenues if the concession is abolished.
3. Outlay equivalent approach, which measures the cost of providing the same monetary benefit as the tax expenditure through direct spending. For instance in United States, [which uses both (i) and (iii)], this measure allows a comparison of the cost of the tax expenditure with that of a direct Federal outlay.

Variation in tax expenditure calculations may also arise based on whether the accounting practice is cash or accrual based. Belgium, Canada, Finland, France, Spain and the United States use the former, Austria adopts a mixture of the two, and other countries use accrual basis.

There are also differences in the presentation of tax expenditures accounts. Ireland, the Netherlands and United Kingdom classify tax expenditures by type of tax involved while others provide further classification to enhance comparability with direct spending accounts.

The classification of expenditures as plan and non-plan gained currency with the inception of planning in India. Plan expenditure was supposed to indicate new developmental and capital formation activities, which are expected to increase the productive capacity of the economy. The classification is valid only over a period of five years in the sense that plan expenditure is the expenditure on new schemes undertaken in the five-year plan period. Continued expenditure on these schemes after completion becomes non-plan in the next plan period.

A major criticism of the dichotomy between plan and non-plan expenditures relates to the implicit priority accorded to Plan expenditure which leads to inadequate provisions for crucial expenditures like the maintenance of existing assets which are regarded as non-plan expenditure. As a consequence essential maintenance is neglected and existing assets including schools and hospitals are starved of funds while resources are allocated to new schemes. A large part of non-plan expenditure is developmental in nature and should enjoy the same priority as new plan schemes if not higher.

Since both revenue and capital components of plan schemes are clubbed under plan outlay, it is difficult to categorise Plan expenditure as investment expenditure. In fact over the years the increasing proportion of Plan expenditure consists of revenue expenditure. On the financing side this has the consequence that borrowed funds used to finance Plan expenditure are actually being used to finance revenue expenditure.

The Working Group on Expenditure Policy felt that the distinction between plan and non-plan is historical and the data may be useful to monitor the actual expenditure on capital projects and revenue schemes under Plan. In this context, the Finance Minister in presenting the Union Budget for 1998-99 had proposed constituting a Task Force to examine these issues and recommend ways for a functionally viable and focused presentation of Government expenditure in the budget. The Eleventh Finance Commission has endorsed this approach.

Government Accounts in India are kept on cash basis both at the centre and the states. Except for goods received on credit by Public Works Departments, there is no accounting entry made for credit purchases.

Several governments in other countries have moved to, or are exploring the possibility of adopting accrual measures in government accounts. According to the international Federation of Accountants, accrual accounting is used by Governments in USA, Australia, and New Zealand and by central government agencies in many other countries. Canada is reported to be using modified accrual accounting and is moving towards a full accrual system. Other countries like France, Chile etc. are seriously considering a shift to a modified or full accrual based accounts system.

The Government of New Zealand had introduced accrual accounting in the early 1990s. It published its first set of accounts on accrual basis for the year ending June 1992. Since the New Zealand government has had a fairly long experience of accrual based system of accounts, it may be worthwhile to spell out the salient features of accounting policies of the Government of New Zealand.

Salient features of accounting policies of Government of New Zealand are:

• Welfare expenses are recognised when an application for a benefit has been accepted and the eligibility criterion met.
• Discretionary grants and subsidies are recognised when payment is made.
• Depreciation is on the straight-line method.

• Lands and buildings are recorded at the net current value
• State highways are recorded at depreciated replacement cost based on the estimated present cost of constructing the asset.
• Intangible assets, which cannot be sold, are not recognised.
• Foreign monetary liabilities are valued at the closing exchange rate. Exchange gain or loss is included in the operating statement.

The important statements published by the Government of New Zealand are:

The introduction of accrual accounting in government will have some definite advantages. It will greatly add to the depth and transparency of government accounts. It will also greatly improve the quality of fiscal decision making by bringing into sharp focus the claims and commitments, otherwise hidden from view, which are so vital for taking a holistic view. We quote below some relevant parts from the New Zealand government's financial statements.

It is through full and transparent financial reporting that taxpayers are able to see how their taxes are spent and what the future obligations of the government are likely to be.

The financial statements have another important role: they encourage responsible fiscal decision-making. Good fiscal control is essential if current economic growth projections are to be realised.

The kind of statements mentioned above cannot be derived from the accounts in India as these are based on cash basis, and not on accrual system. It is felt that it may not be feasible to introduce accrual accounting until a strong base of information system is first built up. This should not, however, deter us from making a beginning by compiling and publishing the following information. It is suggested that a beginning be made to compile and publish it as part of the budget papers and accounts as a supplementary information.

A large amount of direct and indirect taxes remains unrealized after the demands are raised. Some of these are in dispute in the Departments, in courts, tribunals, or is in the process of recovery for rather long periods. It is time that a statement is published showing:

• Amount in dispute pending before administrative authorities and the period of pendency.
• Amounts in dispute pending before Tribunals and their year-wise break-up.
• Amounts in dispute pending before High Courts and their year-wise break up.
• Amount in dispute pending before the Supreme Court and their year-wise break-up.
• Amount not in dispute but not realised - their year-wise break up.

Details regarding arrears of revenues (both tax and non-tax revenues, e.g. interest receipts, by broad categories) may be given in a separate statement to be appended with the Receipt Budget.

A statement of commitments on the lines of the Statement prepared by the Government of New Zealand be published to show the existing commitment. However, the statements may be confined to civil and commercial departments of the Government of India.

The position of undischarged/carried over liabilities on preceding 31^st March may be indicated in a separate Annexure to Expenditure Budget Volume-I.

A statement of the incomplete works with information on the initial estimates, the estimates updated for the year of the budget, the original time-schedule, the break-up of expenditure in each of the remaining years of the revised time schedule may be published. This may be restricted to schemes, projects costing more than Rs.5 crores each. This statement will show the commitments for completion of schemes and the financial requirements. The informed debate on competing schemes may ensure acceptable time-schedule for completion of projects. It may also restrict spreading of resources too thin on far too many schemes.

A statement showing schemes/programmes/projects in hand and cumulative expenditure with reference to sanctioned costs (for close-ended ones) may be appended in the Detailed Demand of the concerned ministry.

Almost the entire investment in public enterprises in the last few decades has been from government borrowings. Governments keep on borrowing at high rates of interest. A total picture of government's own cost of investments and the profits or dividend received by it as well as retained by undertakings should be published. The subsidy implicitly paid by the government to the public sector enterprises, which is ultimately borne by the taxpayer should also be brought out.

A statement of pension liabilities on the basis of actuarial valuation may be published every year. This statement would help in subsequent switch over to self-sustaining pension funds, relieving the exchequer of pension commitments.

Government department provide different kind of tax or non-tax concessions. Tax incentives are to encourage either development of industry and agriculture or capital formation. It is felt that a beginning be made to disclose approximate amount of tax concessions to different sectors. The list may be extended in a phased manner. This will enable both the Parliament and the public to know the total value of tax concessions.

Similarly, whenever non-tax concessions are granted like allotment of government property at a concessional rate or rent, then the annual value of the concession needs public disclosure. One advantage may be that the concessions to individuals will not continue for indefinite periods. In course of time, approval of Parliament may be obtained as in the case of demands for grants for expenditure.

The present system of internal audit and internal control is based on the scheme of Integrated Financial Advisers introduced in various departments of the Government of India in 1975. Soon thereafter, the central government also decided to separate accounting functions from auditing in all central civil ministries and departments (the Defence and the Railways already had their separate accounting organisations). After separation of account from Audit in 1976, Controllers of Accounts have been appointed for each ministry or department. The internal audit functions have been entrusted to the Controllers of Accounts who works under the overall supervision of the Integrated Financial Advisers.

2. Effectiveness of internal audit and control systems depends upon the nature and scope of internal auditing and the conceptual framework used for this purpose. It may be worthwhile to mention briefly about the nature and scope of internal audit both traditional and modern.

3. The concept of 'internal auditing' traditionally developed as an audit on behalf of the management to ensure that the existing internal controls are adequate and effective, (b) the financial and other records and reports show the results of actual operations accurately and promptly, and (c) each unit of the organisation follows the policies and procedures as laid down by the top management. Thus, the traditional concept of internal auditing was primarily concerned with questions such as whether the assets of the organisation were adequately safeguarded and properly accounted for, whether the accounting and other allied records were reliable and whether the organisational procedures and policies were complied with. With a significant emphasis on the detection of frauds, and accuracy of financial records, the internal auditor was perceived as a status quo oriented auditor of financial record and somewhat unfortunate brother of the statutory auditor. As such, he was mostly engaged in conducting routine checking of statutory records, purchase invoices, cash, sales bills and other similar record and documents.

4. Modern concept of internal auditing goes far beyond this traditional approach. Internal auditing is no longer considered as a mere routine review of financial and other records by specially assigned staff. As the Institute of Internal Auditors of USA, defines it, "Internal auditing is an independent appraisal function established within an organisation to examine and evaluate its activities as a service to the organisation. The objective of internal auditing is to assist members of organisation in the effective discharge of their responsibilities. To this end internal auditing furnishes them with analyses, appraisals, recommendations counsel and information concerning the activities reviewed".

5. The Institute of Internal Auditors, defines the scope of internal auditing as "The examination and evaluation of the adequacy and effectiveness of the organisation's system of internal control and the quality of performance in carrying out assigned responsibilities". According to the Institute, internal auditing and control system must comply with the following parameters:

Reliability and Integrity of Information: Internal auditors should review the reliability and integrity of financial and operating information and the means used to identify, measures, classify and report such information.

Compliance with Policies, Plans, Procedures, Laws and Regulations: Internal auditors should review the system established to ensure compliance with those policies, plans, procedures, laws and regulations which could have a significant impact on operations and reports and should determine whether the organisation is in compliance thereof or not.

Safeguarding of Assets: Internal auditors should review the means of safeguarding assets and, as appropriate, verify the existence of such assets.

Economical and Efficient Use of Resources: Internal auditors should appraise the economy and efficiency with which resources are employed.

Accomplishment of Established Objectives and Goals for Operations or Programmes: Internal auditors should review the operations or programmes to ascertain whether results are consistent with established objectives and goals and whether the operations or programmes are being carried out as planned.

6. Internal audit and control systems can be only as effective as the management wants them to be. However, if an internal auditor has to effectively perform his duties, he should be independent of the activities that he handles. This can be achieved only if he is given the requisite organisational status and if he possess the desired degree of objectivity.

7. It might be useful to mention the relationship between the internal auditor and the statutory auditor or constitutional auditor. There are some similarities in the areas of operation of the internal auditor and the statutory auditor. However, there are basic differences in the status, responsibilities, approach and the scope of operations of an internal auditor and statutory or constitutional auditor. The internal auditor is the representative of the management. The nature and scope of his operations are determined by the management or the government and therefore may differ from one organisation to another according to the needs and perception of different managements. Rights and duties of the statutory auditor on the other hand are defined by the Institute, e.g., in case Corporate Sector, Section 227 of the Companies Act lays down detailed provisions in this regard. In the case of constitutional audit, the provisions are available in the Constitution and in the CAG's (DPCS) Act 1971. (In case of public sector enterprises, the scope of audit by constitutional auditors is laid down in sections 619 sub-section 3 and 4 of the Companies Act.)

8. The system of internal audit and controls in the case of India is very well defined and has been in operation for a number of decades; only the agency performing that function might have changed. There is a fair degree of independence enjoyed by the internal auditor of the Ministries. They have full freedom to review the expenditure, to appraise various programmes and projects. Their Reports are submitted to the Administrative Secretary through the integrated financial advisers. Normally the reports form useful instruments of management information.

9. However, a Committee which went into the functioning of the Civil Accounts Department, observed that more than 50 per cent of the staff and the officers employed in the work of internal audit and internal control functions did not have the requisite background, training or experience either of finance or of accounts. In its opinion, the work was also accorded a lower priority than what it deserved. These observations suggest that the system of internal audit and internal control, which has been in place for a long time, needs to be renewed to make it more effective and meaningful as an instrument of internal management control.