The Role of Parliament in the Budget Process

Title: The Role of Parliament in the Budget Process

Date: 10/01/1999
Language: english

THE ROLE OF PARLIAMENT IN THE BUDGET PROCESS

Warren Krafchik and Joachim Wehner

Executive Summary

Recent events have highlighted the confusion about the role of Parliament and civil society in the South African Budget process. However, the Constitution grants Parliament an active role, which is important for a variety of reasons.

The Budget is the government's most important economic policy tool and provides a comprehensive statement of the nation's priorities. As a representative of the people, Parliament is the appropriate place to ensure that the Budget best matches the nation's needs with the available resources, an ability that is especially critical considering the current challenge to reprioritise between and within departments. This exercise demands detailed engagement with the Budget - potential that is wasted in the present vacuum. Moreover, the Budget drafting phase is an intrinsically closed process. The committee stage therefore provides the most significant opportunity for the public and civil society to voice its preferences and commit itself to the choices made.

The first part of the paper reviews international experience with regard to the balance of power between the executive and the legislature. In particular we are concerned with explaining why some parliamentary systems are able to affect greater changes in the Budget than other systems. For example, in Australia and the U.K. the amendments passed are few and minor. By contrast, during the 1988 German Budget deliberations, the Budget Committee cut total expenditure by DM 4.2 billion and shifted more than DM 17 billion between votes. Our research suggests that two primary factors determine the ability of Parliament to change the Budget. We do not try to explain all variations in de facto amendment powers, but concentrate on two sets of necessary conditions that can be objectively determined and measured.

The most obvious determinant of amendment power is the actual configuration of powers vested in Parliament. Most parliamentary systems apply restrictions to the powers of the legislature in budgetary matters, often relating to increases in expenditure. Reductions in expenditure can be considered the lowest common denominator across parliamentary systems. However, it is not possible to explain the major variations in the magnitude of actual amendments across countries simply with references to differences in allocated powers.

Our research shows that de facto amendment power depends on a second set of factors relating to the effective role of committees in the Budget process. This role is determined by a combination of the following factors:

The location of amendment powers, i.e. whether committees have the power to suggest actual amendments to the deciding House. Where committees do have this power, the committee report fosters in-depth debate in the House, which is likely to lead to amendments.
Time allocated to committee debate relative to the total time available for consideration in the deciding House. Where committees do not have sufficient time to undertake adequate analysis, their power to suggest amendments is weakened.
Committee involvement, i.e. which committees are involved in the budgetary process and the relationship between them. A strong co-ordinating role of a Finance or Budget Committee, combined with other specialist input, supports Parliament's ability to affect substantial amendments. Unless there is a mechanism to co-ordinate specialist input, Parliament will not be able to reprioritise between votes.
Access to independent research capacity. The ability to affect amendments depends on detailed scrutiny, requiring analysis by specialised research staff.
Access to departmental information, i.e. timely access to executive information. Information on the implementation and impact of the current Budget and the development of next year's Budget are vital to provide a background for necessary changes.

It is the interaction of these factors that define the practical impact of amendment powers. The German system results in substantial expenditure amendments as well as reprioritisation between departments. In Germany, amendment powers are located at committee level, there is sufficient time for the committee stage, the Finance Committee has access to extensive, independent and Government research capacity, supported by a continuous process of monitoring Budget implementation.

Conversely, substantial amendments are not a feature of either the Australian or the British systems. In both countries, permanent committees are not part of the parliamentary Budget process in the deciding House. Although both of these countries provide for committee access to independent research capacity and comprehensive Budget documentation ensures strong oversight through a Public Accounts Committee, these systems fail to feed such expertise into the actual Budget process.

The second part of this paper examines how the above factors currently apply to South Africa. Even if Parliament were to immediately gain these powers, committees do not currently have the capacity to effectively utilise them. The most serious constraints are the lack of time for committee deliberations, independent research capacity, as well as access to departmental expenditure and impact analysis. The effectiveness of the South African system will further depend on such factors as the location of amendment powers, the co-ordination of input from various committees, and the relationship between the Finance and the Public Accounts Committee.

In South Africa, amendment powers and associated information and institutional requirements should be managed and sequenced as part of the broader Budget reform process. In practice, the following may form a first step:

Extending time for committee debate relative to debate in the House.
Regular publication of national and provincial expenditure reports and incidence information.
A comprehensive review of Budget documentation.
Regular meetings between departments and committees focussing on Budget implementation and development.
An analysis of the options and costs associated with enhancing independent parliamentary research capacity.
Amendment powers should be introduced and tested in the National Assembly first. A mechanism to include NCOP representatives in the debate may take the form of a joint committee stage to avoid duplicating costs.

1. Introduction

In South Africa, there is currently confusion about the role of Parliament (and civil society) in the Budget

process. This is highlighted by two recent events. First, the Congress of South African Trade Unions (Cosatu) refused to participate in the Portfolio Committee on Finance hearings on the grounds that Parliament does not have the power to change the Budget (Coetzee 1998). Second, the Minister of Public Service and Administration did not appear, nor send a representative, to the scheduled debate on the public service and the Budget in the Finance Committee.

The aim of this paper is to understand the role of Parliament in the Budget process. In particular we will focus on powers of amendment over Money Bills

and conditions for effective use of these powers.

Why is this an important debate? The Budget is the Government's

most important economic policy tool and provides a comprehensive statement of the nation's priorities. As the representative of the people, Parliament is the appropriate place to ensure that the Budget best matches the nation's needs with available resources. This ability is especially critical considering the current fiscal squeeze where the primary challenge is reprioritisation within and between departments. This exercise demands detailed engagement with the Budget, a potential that is wasted in the present vacuum. As Cosatu has pointed out, the committee stage also provides the only opportunity in the Budget process for the public and civil society to voice its preferences and commit itself to the choices made.

The international debate is framed as a tension between the roles of the executive and the legislature - between technicality and democracy. In virtually all countries, it is accepted that the executive has a mandate to prepare the Budget, since it possesses the most comprehensive information on which to base revenue and expenditure decisions. The role of the legislature is to exercise oversight and to authorise the executive to raise revenue and spend money.

Advocates of legislature power argue that effective oversight requires amendment powers. It is also argued that the legislature has the potential to add value to the Budget process by checking priorities and effective allocation of resources.

Legislatures further have a stronger contact with citizens and interest groups, contrary to the executive drafting process, which takes place behind closed doors. Advocates of weaker legislative powers argue that amendment powers could be detrimental to long-term planning and macroeconomic management if the Budget were altered on the basis of populism and shortterm point scoring. It is further argued that practical considerations limit the scope for amendments. For example, if the legislature does not have sufficient research capacity it is likely that it will have to fall back on already constrained departments for research support. Furthermore, a flood of amendments would clog the system and frustrate all role-players, thereby wasting already scarce energy and resources.

This debate highlights the opportunities and constraints of parliamentary involvement in the Budget process. Greater legislature involvement in the Budget is an opportunity for closer scrutiny, which is critical for inter and intra-departmental reprioritisation. It also represents a process encouraging greater civil society commitment to macro-plans. Nevertheless, it is important to recognise some constraints. It is generally accepted across countries that the preparation of the Budget is the prerogative of the executive, although this does not preclude contact with departments during the drafting phase. The acceptance of this principle may go some way to meeting concerns about the integrity of macro-economic policies. Second, the executive will always have a greater information base than the legislature and Parliament will enter the process at a late stage after roughly a year of departmental deliberations. However, unless legislatures are presented with sufficient information, they run the risk of not effectively utilising their powers or, even worse, utilising their powers to serve short-term partisan ends. Third, there is indeed a danger in some systems (for example, the United States) of generating an inordinate amount of suggested changes. The need to respond to each amendment will put a considerable additional burden on the executive and is likely to obscure the real issues. The fact that this does not occur in most systems implies that this possibility can be ruled out by careful design of the amendment process. Finally, amendments should not be considered votes of no confidence. In fact, the ability to amend paradoxically safeguards Government's macro-economic plans, at least to some extent. If the only option would be to vote for or against the Budget in its totality, this would indeed leave a vote of no confidence, the rejection of the Budget, as the only choice to voice dissatisfaction.

The basic dilemma between technicality and democracy has led to a compromise internationally whereby most Parliaments have restricted their powers over the Budget. In return, most were able to maintain their sovereignty by keeping a closer watch over Budget execution, thereby strengthening the capacity to make informed amendments. Every system will have to balance this tension through the design of the process and its rules.

We will survey different designs of this balance. For this purpose, the following sections provide an analysis of the broader South African context, a discussion of international experience and its specific relevance for the South African situation. We will conclude by outlining some policy implications in the light of this analysis.

2. The South African context

What is the current situation in South Africa? The new Constitution (Act 108 of 1996) grants Parliament the right of amendment over Money Bills (s 77).

To activate this right requires a separate act of Parliament, spelling out exact details of how this should work in practice. To date enabling legislation to this effect has not been tabled which implies that committees still labour under the considerably weaker powers of the past.

The presentation of the Budget on Budget Day initiates a 3 - 4 month review process in the national and provincial legislatures. During this stage, committees have two possible points of interaction with the Budget- the Portfolio Committee on Finance hearings on the overall Budget and other portfolio committee hearings on individual votes. In the National Assembly the bill is referred to the Finance Committee which has only 7 days to hold hearings and present a report to the House. Portfolio committee hearings on individual votes are optional and must be completed in time for the debate in the House (Kahn). For instance, during the 1998 Budget process, the Portfolio Committee on Labour held public hearings.

The Constitution grants Parliament strong powers to hold open hearings and to call Government officials and other experts to give evidence. But Parliament and its committees do not yet have the right to suggest changes to the Budget. In theory, a committee can recommend that the entire Budget (or specific votes) is rejected in total and this could lead to a no-confidence motion or force the Government back to the drawing board. In practice, however, this is unlikely to happen for several reasons, including the large African National Congress (ANC) majority in Parliament. Nevertheless, Parliament's own lack of clarity on this matter is reflected in the fact that the Joint Rules of Parliament do not currently deal with Money Bills. Parliamentary officials recall only one instance in the past where one of the Houses in the tri-cameral system rejected Chris Heunis' Constitutional Affairs vote. The President's Council subsequently overrode the decision.

Following the National Assembly process, the bill is referred to the National Council of Provinces (NCOP) for consideration. According to the Constitution, this process should mirror the procedure for section 75 legislation (ordinary bills not affecting the provinces). In theory the NCOP could reject the entire Budget and force the National Assembly to vote again, but this has yet to happen for any bill, let alone a Money Bill.

The introduction of amendment powers could substantially change Parliament's engagement with the Budget from a rubberstamp exercise to effective oversight. The design of the amendment system is therefore critical. Unfortunately, the South African Constitution does not prescribe the form and limits to amendment powers. As a starting point for this discussion, we consider international experience.

3. International experience

3.1. Analytical framework

3.1.1. The design of amendment powers

Our research into amendment powers indicates that while most countries allow Money Bill amendment powers, there are wide variations in the design and impact of these powers.

First, there is a significant distinction between parliamentary and presidential systems of Government.

In general, presidential systems grant greater powers of amendment, although this often counter-balanced by an executive or presidential veto (for a good analysis of the Philippine system, see LRC-KSK 1996). Given that South Africa is a parliamentary system, the following will concentrate on comparable countries.

The vast majority of parliamentary countries allow amendment powers. However, these powers have not had a uniform impact on these parliamentary systems. In Australia and the U.K., amendments passed are few and relatively minor. The most significant change in Australia in the past years has been a 1995 amendment, which resulted in AUS$ 250 000 reduction in capital expenditure. There were no amendments made in the most recent 1997/8 Budget (Fowler). The U.K. system is a further example of a system where significant changes are uncommon. In the House of Commons, changes in taxation rates are hardly ever made against the wishes of the Government as long as the governing party retains even a small working majority. "The only significant event of this kind in recent history" (Proctor) was in December 1994.

It can be categorically stated that amendments to expenditure proposals are never made except at the instigation of the Government.

By contrast, during the 1998 Budget deliberations, the German Budget Committee cut total expenditure by DM 4.2 billion and shifted more than DM 17 billion between votes. Over the past years, controversial capital expenditure projects, such as the construction of a high-speed train line, were effectively limited by tens of millions of DM by attaching stringent conditions to such expenditure (Levermann).

This paper will explore the practical impact of amendment powers in several parliamentary systems in order to understand why some systems have proven a greater ability to change the Budget than others. We restrict most of our discussion to a comparison of Germany, the U.K. and Australia. These countries have constituted an important resource for the design of the South African Constitution and the Budget reform programme.

3.1.2. The ability of Parliament to change the Budget

Based on our research, we argue that the ability of Parliament to change Budgets depends on two sets of necessary conditions: the extent of conferred powers and the effective role of committees in the Budget process.

Conferred powers refer to the configuration of powers vested in Parliament to change the Budget. Logically, these powers are positively correlated with the ability to affect change, but they cannot explain the full variation in Budget changes across countries referred to above. The ability to amend Budgets is also strongly influenced by the effective role of committees in the Budget process. Our research indicates that the following are good indicators of the extent of committee involvement in the parliamentary Budget process:

the "location" of amendment powers, such as whether amendment powers are vested in committees or the House itself,
the amount of time dedicated to committee debate in the relevant House,
which committees are involved, and the interaction between committees,
committee access to independent research capacity,
committee access to departmental information.

We will argue that an effective amendment system must combine conferred powers and an effective and resourced role for committees in the Budget process. In systems where Parliament is granted power of amendment without an effective role for committees in the process, Parliament's ability to change the Budget is limited in practice (e.g. Australia). In systems where Parliament is granted power of amendment, combined with an effective role for committees, Parliament's ability to change the Budget is strengthened (e.g. Germany).

In our model, Parliament's ability to significantly change the Budget is maximised when:

Parliament does have some power of amendment,
committees have the right to suggest amendments to the House,
there is sufficient committee time for Budget scrutiny,
a co-ordinating committee is able to combine finance and other specialist input,
committees involved have access to sufficient independent research capacity and
committees involved have access to detailed, timeous departmental information.

Our model contains a couple of reasonable assumptions. First, we have already observed that the existence of amendment powers in different countries does not ensure that they will be used equally. We as3.me that the absence of significant changes, especially over a number of years, is an indicator of a lack of actual ability to make changes. (It is unlikely to indicate general satisfaction with the Budget.) Our aim is to explore the reasons for the differing ability of countries with amendment powers to make significant changes. Our focus is not simply on the quantity of changes since amendment powers sometimes promote a game of symbolic demonstrations that do not filter into real shifts. For example, in India Members may move "cut motions" to express a general policy disagreement involving the sum of one Rupee. Similarly, in the U.K. reductions are often suggested as a means to voice disapproval and force the Government to furnish explanations prior to approval. In this paper, we track the magnitude of actual changes in the Budget including shifts between votes.

Second, we are aware that it is impossible to fully factor in the background institutional and political structure of the country. It is also difficult to adequately capture the de facto rules that develop in each system through the interaction of individual and political agendas with institutional structures. This unknown element may be best summarised as the dynamics of political and institutional culture. Such dynamics would include, for instance, the German tradition of co-operative governance reflected in a committee system in which opposition parties chair or co-chair essential committees, such as the Budget Committee. Similarly, a list system as opposed to a constituency-based system is likely to reduce the role of internal party debate, as may the size of the ruling majority. Traumatic events may also alter parliamentary practice. This was the case in Australia, where a deadlock between the two Houses over the Budget led to the dissolution of Parliament in 1975. A de facto convention, that the Senate would never reject the Budget in total, has since emerged to prevent a re-occurrence (Kerley).

We acknowledge that these factors will have an impact on the way in which conferred powers are designed and utilised. We do not claim to explain all variations in amendment ability. Rather, we concentrate on two essential sets of necessary conditions that can be objectively determined and measured.

3.2. The framework applied

3.2.1. Differences in conferred powers

The first possible explanation for differences in the effectiveness of committees is the extent of conferred powers. Money Bill amendment powers can be grouped into three broad categories. Unrestricted powers generally imply the ability to vary expenditure and taxation in either direction, without the consent of the Government. Most presidential systems fit this model. Restricted powers refer to the power to amend the Budget within set limits, often relating to increases in total expenditure. Balanced Budget powers refer to the ability to raise or lower expenditure and / or revenue as long as there is a counterbalancing charge to maintain the Budget balance.

Table 1 : Members' Rights in Budgetary Matters

Member's rights	Number of countries
1. May reduce and increase expenditure and revenue	32
2. May reduce but not increase expenditure	17
3. May reduce expenditure, but only increase it with the permission of the Government	4
4. May reduce and increase expenditure if alternative provisions are made elsewhere	13
5. Rights not specified in detail	15
6. Not applicable (Nicaragua)	1

TOTAL	82

Source: IPU 1986: Table 38A

Most parliamentary systems are clustered in the restricted power category, however, there are some differences in amendment powers within parliamentary systems. Germany, as one example of a Western European country, generally allows increases in revenue or decreases in expenditure without the consent of the Government. Proposals to decrease revenue or increase expenditure are allowed but require the consent of the Government.

In practice, the revenue powers of both Parliament and the executive are further limited to the recommendations of an independent working group on revenue estimates that includes the Ministries of Finance and Economics, L�nder (provincial) MECs for Finance, organised local government, the Federal Bank, the Federal Statistical Service, a recognised experts group, as well as major economic research institutions (BMF 1996: 17). Similarly, India allows Members to reduce expenditure and vary taxation. Increases in expenditure require the recommendation of the president. Taxation is also dealt with slightly differently to expenditure in that changes take effect immediately, but the House still has 75 days to approve.

The UK and Australia, occupy the other end of the parliamentary continuum. The House of Commons may not increase expenditure or, more surprisingly, revenue. This configuration has no logical rationale, but has historical reasons.

The only legitimate amendments are those that reduce requested expenditure or abolish a tax or duty. In the Australian system, Members may move to reduce expenditure or revenue. Only the Government can introduce or increase an expenditure or tax.

Despite these differences, most parliamentary systems allow a fairly high minimum level of amendment powers. Although there are often restrictions on increasing expenditure, reduction in expenditure can be considered the lowest common denominator for parliamentary systems. Changes to taxation are usually also possible, although restrictions are varied. It is therefore not possible to explain the large differences in the magnitude of actual amendments highlighted above simply with reference to differences in allocated powers. Observed Budget changes are greater than observed variations in amendment powers. Effective amendment power depends on a further set of factors that determine the extent to which committees are able to utilise conferred powers.

3.2.2. The role of committees in the Budget process

The role of committees in the Budget process is the second explanation for the observed variance in budgetary amendments between countries. Actual amendment power is essentially about matching formal committee powers with the capacity to utilise these powers. Our research shows that this ability depends on the five factors outlined above.

Our analysis suggests that, while each of the above factors has an important relationship to effective committee power, it is the combination of these factors in each country that is important. Any one or two factors occurring in a particular country is not likely to suffice, effective committee role in the Budget process depends on the balance of these factors to matching allocated powers.

(i) Location of amendment powers

This variable refers to the power of committees to suggest amendments to the House. In all parliamentary systems amendments have to be moved on the floor of the House. We are concerned with the distinction that in some countries (for example, Germany) the committee has primary responsibility for suggesting amendments to the deciding House. In these cases, the focus of Budget debate is in the committee, committee involvement is strong and largely determines the debate in the House. Whereas, in other models (such as India), the role of committees is restricted to providing comment to the House without suggesting actual amendments (Lok Sabha 1997). In these cases, committee involvement is weaker. The focus of the debate is on the floor of the House as it is up to individual Members to move amendments, which may or may not take committee comments into account.

The international literature (IPU 1986) and our research suggests that where committee involvement in the Budget is weak, broad criticism restricted to general debate in the House will prevail. In these circumstances substantial amendments are unlikely. In the Australian House of Representatives, there is no committee stage. While committees in the Senate deliberate on the Budget, the report to the House only mentions issues of concern. It is up to individual Senators to circulate a request for amendment. On the contrary, the German Budget Committee's report contains the suggested amendments that are debated and usually accepted on the floor of the Bundestag. Therefore, where effective powers are located in committees, it is likely that detailed scrutiny will predominate over general debate. Amendments are more likely under this scenario.

(ii) Time allocated to committee debate

This variable refers to the time allocated to committee debate relative to the total time available for parliamentary consideration. Longer time allocated to committee debate enables detailed analysis to identify and support suggested amendments. Where committees do not have sufficient time for analysis, their role to suggest amendments is weakened. This variable also has to take into account in which House the committee deliberations occur. If committees exist and make recommendations to the deciding, usually the lower House, the ability of Parliament to make changes is strengthened.

Australia allows 1 - 2 months for the legislative Budget process. However, there is no committee stage in the deciding House - the House of Representatives. The committee stage only occurs in the Senate where it lasts approximately one month. However, by convention the Senate plays largely an advisory role in the Budget process.

The U.K. does feature a rather extensive committee stage lasting several weeks of a three-month parliamentary Budget process. However, the ad hoc Standing Committee on Finance only deals with the uncontroversial parts of the Budget as agreed by the opposition and the Government. The truly controversial discussions are reserved for debate on the floor of the House.

In contrast, Germany allows 4 months for the Budget deliberations, including several weeks allocated to the Budget Committee stage in the Bundestag (lower House). In addition, the Committee on Finance in the Bundesrat (upper House) also has approximately one month to consider the Budget.

(iii) Choice of committees involved

Parliament's ability to change the Budget is influenced by which committees are involved in the legislative process and the relationship between these committees. In most countries the Finance Committee accepts responsibility for the process, alone or as a coordinating body for other relevant committees. The international trend seems to be towards broader committee consultation. One example is India. Up to 1993/4, the Budget was examined by the Committee on Estimates only. It is currently also examined by specialist sector committees. Since 1970, a similar development took place in Australia with the introduction of a committee stage in the Senate. In the U.K., a committee stage involving the Standing Committee on Finance was introduced in 1968. This broad international trend is an indication that Parliaments themselves have realised the value of committee involvement in the Budget process.

It seems that strong co-ordination from the Finance Committee, combined with specialist sector input, supports strong committee involvement in the Budget process and therefore Parliament's ability to change the Budget.

Moreover, and from the perspective of South Africa very important, unless there is a process or structure to co-ordinate committee or sector inputs, Parliament will not be able to reprioritise between votes. The lack of a mechanism to shift funds between votes, also acts as a strong general disincentive for expenditure cuts. Parliamentarians are less likely to enact cuts if they are not able to shift funds elsewhere.

The relationship between the Finance / Budget Committee and Public Accounts Committee is also crucial for amendment ability. In many countries, the Finance Committee establishes a special relationship with the Public Accounts Committee. This interaction is very important as the latter usually monitors the implementation of the Budget, including the Auditor-General. Where the two committees work closely together, effective oversight strengthens the ability to change the Budget. Where the work of the two committees is not well integrated, the system may achieve strong oversight without being able to translate this capacity into effective amendment.

The German Budget Committee is able to draw on the expertise of designated reporting Members who are each assigned to monitor the implementation of the Budget in a single department. Importantly, Members retain the same portfolio for several years building up significant sector expertise (Levermann). This enables detailed changes within departmental votes, as well as substantial shifts between departments. As mentioned above, in the 1998 Budget, the committee decided to shift DM17.1 billion between votes. In Germany, the important Public Accounts Committee is a sub-committee of the Budget Committee. This means that expenditure authorisation and monitoring decisions are uniquely integrated in a single committee. Detailed knowledge of spending and incidence patterns gathered through this subcommittee helps to ensure the accuracy and relevance of the amendments suggested by the broader Budget Committee.

In both India and Australia, the committees involved are able to scrutinise departmental Budgets in detail. However, there is no co-ordinating mechanism or report to support shifts between votes.

In the U.K., the process dealing with revenue proposals is much more formalised- involving a temporary Standing Committee on Finance

- than that dealing with expenditure. The permanent Treasury Select Committee usually inquires into the Finance Bill between the Budget Statement (outlining its provisions) and the Second Reading debate (on the principle of the Bill). There is, however, no requirement on it to do so or for the Government to wait for its report before holding the Second Reading. On the expenditure side, there is no obligation for the Government to wait for reports from any committee. Proposals for expenditure are sometimes examined by the appropriate select committees. However, committees hardly ever propose to have expenditure reduced (Patrick).

In both the U.K and Australia, the link between public accounts and the Budget process is broken. In both countries, the Public Accounts Committees are strong and resourced but not able to feed this expertise in to the Budget process itself. In both countries, the Public Accounts Committee is not part of the actual Budget process. They have no direct input in the debate nor is there a strong relationship with the committees that do have input.

(iv) Independent research

Sufficient time for committee debate is unlikely to lead to budgetary change, unless committees have access to independent research expertise. The ability to change Budgets deptids on detailed scrutiny that is only possible with detailed analysis. Committees are likely to depend on departmental expertise during the initial stages of a new system. However, this is neither sufficient nor satisfactory in the longer-term from either the department or from the committee's point of view. In the short-term, it is also possible for committees to share research expertise. In the longer-term, given the size and technical nature of the Budget, effective research services are likely to depend on dedicated, specialised personnel. In general, the research capacity of committees often varies in proportion to effective committee power, but even weaker committee systems make provision for dedicated Finance or Public Accounts research capacity.

The German Budget Committee has a secretariat that consists of 5-6 people. The committee can also draw on an independent scientific service, only available to Members, that has a dedicated finance unit consisting of 5 persons and an overall staff of 80.

In Australia each of the 8 Senate committees that consider the Budget have a secretariat, consisting of the secretary, an executive assistant and one or more research officers. In addition, committees have access to an independent parliamentary research service with a permanent staff of 80. The powerful Joint Committee on Public Accounts and Audit, that examines the Auditor-General's reports, has a secretary and 10 dedicated staff Members.

In the U.K. the Standing Committee on Finance has no advisors of its own. The Select Public Accounts Committee, that is not directly involved in the Budget process, relies on the National Audit Office's research and is supported by one clerk only. The Treasury Select Committee is relatively well resourced with two clerks, two specialists and two administrative staff. It can also employ advisors if necessary. The library of the House of Commons further has a large research section with a permanent staff of approximately 75 people (Long). Strictly speaking it is only the Standing Committee on Finance that is actually involved in the Budget process itself, but this committee has the fewest independent research resources available.

(v) Interaction with departments

Given the executives preferential access to information, access to independent analysis is unlikely to be sufficient to inform effective committee engagement. In most models, committees utilise a mix of independent sources of analysis together with access to the executive and other public agencies. This means that committees are reliant on the breadth and depth of the information supplementing the Budget speech. In addition, committees require information on the implementation and impact of the current Budget and development of the next year's Budget. Access to this information is critical - greater capacity to monitor Budget implementation is often the trade-off that legislatures accept in return for restrictions on amendment powers (see above). In general, the greater the quality and timeliness of information available to the appropriate committees, the stronger will be their input and therefore Parliament's ability to change Budgets.

In Germany, the supplementary information on the Budget is extensive and includes the relation between inputs, outputs and outcomes. Budget Committee interaction with departments during the year is based on the principle of continuous control introduced in 1985. Interaction is largely the responsibility of the designated reporting Members for each department, selected from governing and opposition parties, and includes regular departmental briefings. The Committee on Public Accounts, a subcommittee of the Budget Committee, oversees frequent expenditure reports during the year. In addition, it receives Government reports on extra-budgetary expenditure 4 times a year, as well as comments by the Auditor-General.

In Australia, there is a permanent and strong Joint Committee on Public Accounts. This committee receives and reviews approximately 40 reports from the auditor-general during the year that focus on financial and performance audits. The committee holds public hearings for each report. The Public Accounts committee and Members are also able to access current expenditure information on-line. The committee has access to the Auditor- General and other relevant ministries.

In the U.K., the permanent Select Committee on Public Accounts draws upon the work of the National Audit Office, which makes for a substantial information base. It receives roughly 50 reports on expenditure and impact during the year. Other select committees are charged with examining the expenditure, administration and policy of the various departments. However, they lack the information base of the Public Accounts Committee. In addition, the Osmotherly rules (so named after the civil servant who drafted them), governing what civil servants can and cannot say before a select committee, prevent revelations on ministerial-bureaucratic relationships. The Osmotherly rules are subject to a specific exception in relation to the Committee of Public Accounts only: the head civil servant in each department is appointed Accounting Officer in respect of that department's vote (parliamentary grant of expenditure). He is responsible to the Public Accounts Committee for how he carries out this function.

In the U.K., Australia and Germany, access to information during the financial year is extensive. The difference seems to be that in Australia and the U.K. the relevant information is not accessed by the committees involved in the Budget process itself. While the Public Accounts Committee also receives most of the relevant information in Germany, this committee is part of the Budget Committee and therefore integrated into the Budget process.

3.3. Summary of country studies

Although many countries allow legislative amendment powers, these powers have significantly different impacts. We have shown that two sets of factors are useful for explaining this observed difference in impact. First, each country decides on a particular configuration of amendment powers ty t directly frames what is and what is not possible. Given the relatively common expenditure reducing powers across countries, conferred powers are a necessary but not sufficient condition for effective use of these powers. Our investigation into four parliamentary systems suggests the differing ability of Parliaments to change the Budget will also depend on a second set of factors. This set of factors relates to the role of parliamentary committees in the Budget process.

The five factors detailed above each have a unique relationship to committee capacity, but it is the interaction of these factors that defines the practical impact of powers. In Germany, the benchmark for an effective committee system, amendment powers are located at committee level. Committee debate is allocated a substantial share of the time available for the Budget process in the deciding House. To compliment this bias, committees have access to extensive dedicated, independent and Government research capacity, supported by a continuous process of monitoring Budget implementation, and an integrated Public Accounts Committees. The German system results in substantial expenditure reductions as well as re-prioritisation between departments.

In Australia and the U.K., permanent committees are strictly speaking not part of the parliamentary Budget process in the lower House, while amendment powers in the Australian Senate are vested in the House rather than in committees. Committee deliberations in the Australian Senate can deepen debate but cannot easily filter into effective amendments in either House. On the other hand, in both Australia and the U.K. access to independent and dedicated research capacity and comprehensive Budget documentation ensures strong oversight through the Public Accounts Committee. However, both systems fail to feed this detailed and intimate knowledge into the Budget process. Consequently, significant amendments are not a feature of either the U.K. or Australian system.

Second Chamber involvement in the committee stage can be observed in four distinct forms, i.e. a committee stage only in the second Chamber (Australia), no committee stage in the second Chamber (U.K.), a joint committee process (India), or a parallel committee process (Germany).

4. How does this compare to SA?

South Africa is on the brink of discussing the amendment powers of Parliament. At this important juncture, it may be useful to compare the existing situation in the country with our model.

In terms of conferred powers, South Africa is presently well below the minimum in parliamentary systems. Even if Parliament were to gain amendment powers immediately, however, committees do not currently have the capacity to effectively utilise them. Consequently, the ability of Parliament to change the Budget would be low.

The decision on where to vest amendment powers has not yet been taken. The Constitution, like most constitutions, concentrates on parliamentary rather than committee powers. Our analysis suggests that giving a coordinating committee the power to suggest amendments to the Budget is important. This would then be congruent with the powers committees have over other legislation.

The Portfolio Committee on Finance is currently allocated a minimal seven days to submit a report on the Budget to the National Assembly. This comprises less than one tenth of the four months allocated to the entire legislative Budget process in the deciding House. In the NCOP, committee involvement is optional. Even well resourced organisations are hard-pressed to deliver detailed, substantive comment on their immediate issues of interest, let alone the entire Budget. Given the poor weighting to committee deliberations currently in South Africa, most of the process is consumed by general debate on the floor of the House, often unrelated to the Budget.

The current Budget process prioritises the role of the Finance Committee, but there are insufficient mechanisms to bring together informed input from other portfolio committees with specific sector knowledge. Although the Public Accounts Committee is separate from the Finance Committee, there is a substantial overlap in membership. The work of the Public Accounts Committee does not formally feed into the Budget process. This will be an important issue to address in the amendment powers debate.

In terms of independent research capacity, the South African Parliament fares poorly in comparison to the countries considered. The national Portfolio Committee on Finance currently has access to only 1 researcher. The Standing Committee on Public Accounts has recently secured outside funding for 2 researchers who have yet to arrive. There is no independent parliamentary research service available, nor is there a tradition of researchers available to individual Members. In addition, each party has proportional to a limited pool of researcher funds. In practice, this means that the ANC study group in the Finance Committee has access to a single researcher without specialised knowledge.

Compared to the countries considered, departmental interaction and information in South Africa is still insufficient, despite some recent improvements. The 1998/9 Budget, together with a move towards medium-term expenditure planning, considerably improves the breadth and depth of information available to Parliament and committees. However, the current situation is unlikely to meet the minimum requirements for effective intervention. The primary source of information - the Estimates of Expenditure - does not provide sufficient detail on programme expenditures, supplementary documentation is inadequate and there is no satisfactory provision for department aims and the outputs associated with the proposed expenditures. It is also vital that committees have access to regular reports on the implementation of the previous Budget. However, there is no provision at national level for regular public expenditure reports, nor is there capacity in the Government to produce regular reports on the expenditure incidence.

Furthermore, several aspects pertaining to the role of the NCOP in the Budget process require clarification. The Budget does impact on the provinces in many ways. For instance, the package of Budget bills includes the Division of Revenue Bill (classified as a s 76 bill affecting the provinces) that directly determines provincial shares of nationally collected revenue. In addition, many of the assumptions and decisions in the Budget itself (a Money Bill according to s 77), such as the allocation for improvements in the conditions of service, restrict provincial flexibility. The current approach to classification already recognises that the distinction between Money Bills and other legislation remains artificial and is often blurred in practice. The distinction was imported from the British system without considering whether mere adoption of such classification made sense in the South African context.

There are two substantial obstacles to enabling NCOP amendment powers. First, the potential financial costs to the system are high. One effective mechanism to reduce these costs is to establish a structure or process to involve NCOP Members in National Assembly committee hearings, whether this is through a joint committee sitting or enabling participatory or observer attendance. Second, the institutional capacity of the NCOP is not yet geared to utilise amendment powers. There is no independent research service, nor are committees staffed with dedicated, skilled researchers. It should also be added that the NCOP is a sophisticated and new institution whose systems and responsibilities are in many ways still unclear. Until these issues have been addressed, the value that the NCOP can add to the Budget debate is limited. However, as the substantial involvement of regional Chambers in other countries demonstrates, the NCOP has the potential to add value to the process by building consensus and a common understanding of budgetary issues between national and provincial spheres.

In addition, the potential role of some unique South African institutions should be clarified. For example, some have suggested that the National Economic Development and Labour Council (Nedlac) should be drawn into the Budget process (Trollip quoted in Coetzee 1998). Also, the full potential of the Financial and Fiscal Commission (FFC) may not have been fully explored yet. The research capacity of Parliament could be enhanced by drawing on the expertise of the FFC. After all, s 220(1) lists explicitly as the purpose of the commission to make recommendations to Parliament and provincial legislatures. We have, for now, kept the development of the Medium-Term Expenditure Framework and its impact on the Budget process out of the discussion. In other countries, the development of multi-year budgeting has tended to decrease Parliament's flexibility. However, for instance in Germany, this has often been balanced by increasing control and oversight. In South Africa, one open question is where Parliament should fit into the process of drawing up multi-year expenditure plans.

In sum, even if the South African Parliament were to immediately receive powers of amendment, it is unlikely to have the ability to make significant changes. The most serious constraints are the lack of time, research capacity and departmental information. It is therefore important that the debate on amendment powers is linked to the broader debate on parliamentary research and analysis capacity, including its interaction with departments. Demands for fuller information and participation are not new to South African debate. They form part of a broader programme of Budget reform that is currently being driven by the Department of Finance. Amendment powers and associated information requirements should be managed and sequenced as part of this broader reform process.

5 Conclusion

We suggest that the following steps may form the next stage in the Budget reform process.

The presentation of the Budget should be moved from March to January to allow for sufficient time for analysis prior to parliamentary hearings and the beginning of the financial year. We suggest that the time for House debate is decreased, while the time for committee analysis and debate is lengthened.

The supplementary Budget currently interrupts the Budget debate in the House and further drains stretched resources. In effect, the House is required to assent to a supplementary Budget before it has even passed the main Budget. This may be an overhang of a slower printing process in the past.17 In any event, we suggest the two processes be disentangled.
In order to strengthen committees access to information, we suggest the following:
An analysis is required urgently, considering the options and costs associated with creating and funding an independent parliamentary research service and dedicated committee research support.
This study should also estimate the additional burden on the executive, in the shortterm and long-term, of granting Parliament powers of amendment.
Amendment powers should be introduced and tested in the National Assembly first. A mechanism to involve NCOP representatives in the debate may take the form of a joint committee stage. Such representation, at least initially, may not include voting powers, but should enable participation. A joint committee stage would avoid duplicating costs, which is one disadvantage of a parallel committee process in both Houses.

References

Survey responses

1. Australia: Peter Fowler, Chamber Research Office, Australian House of Representatives, Email:

Chamber.Research.Reps@aph.gov.au

2. Germany: Herr Levermann, Wissenschatliche Dienste (Fachbereich IV: Haushalt und Finanzen), Deutscher Bundestag, Fax: +49 228 1626971
3. United Kingdom: Simon J. Patrick, Clerk of the Treasury Select Committee, House of Commons, Email:

patricksj@Parliament.uk

; Bill Proctor, Clerk of the Public Bill Office, House of Commons, Email:

proctorw@Parliament.uk

Interviews

1. South Africa: Shahid Kahn, Clerk of the Portfolio Committee on Finance, National Assembly (10 March 1998), Tel.: +27 21 4033743
2. Australia: Margot Kerley, Secretary of the Joint Committee on Public Accounts and Audit, Parliament (26 March 1998), Tel.: +61 2 62774615
3. Germany: Herr Levermann, Wissenschatliche Dienste (Fachbereich IV: Haushalt und Finanzen), Deutscher Bundestag (12 and 25 March 1998), Tel.: +49 228 1622855
4. United Kingdom: Jennifer Long, Assistant Clerk of the Treasury Select Committee, House of Commons (27 March 1998), Tel.: +44 171 2193285

Literature

1. Bundesministerium der Finanzen (BMF) 1996: Das Haushaltssystem der Bundesrepublik Deutschland, Bonn (Bundesministerium der Finanzen).
2. Calland, Richard 1996: All dressed up with nowhere to go? The rapid transition of the South African parliamentary committee system, June 1996, Cape Town (Idasa: PIMS).
3. Calland, Richard and Taylor, Mandy 1997: Parliament and the socio-economic imperative- What is the role of the national legislature?, Cape Town (Idasa: PIMS).
4. Coetzee, Johan: "Begroting moet 'demokratiseer'"; in: Finansies & Tegniek, 20 Maart 1998.
5. Inter-Parliamentary Union (IPU) 1986: Parliaments of the World. A Comparative Reference Compendium (2nd Edition), Aldershot (Gower).
6. Legal Rights and natural Resources Center, Inc.- Kasama sa Kalikasan (LRC-KSK/ Friends of the Earth- Philippines) 1996: Budget for Nature, Quezon City (LRC-KSK/ Friends of the Earth- Philippines).
7. Leonardy, Uwe 1996: "Money Bills" in South African and German Constitutional Law, notes prepared as constitutional advisor, 29 November 1996.

Internet

1. House of Commons Public Information Office (January 1998): How does the Finance Bill proceed?,

http://www.parliament.uk/commons/lib/fin-bill.htm

2. House of Representatives (April 1998): The Budget and Financial Legislation. House of Representatives Factsheet No. 10,

http://www.aph.gov.au/viewFrame.asp?Item=9&x=0.1783716

3. Lok Sabha (January 1997): Budget in Parliament,

http://alfa.nic.in/intro/p23.htm

4. University of W�rzburg: International Constitutional Law Index,

http://www.uni-wuerzburg.de/law/

Appendix

Table 2: Variables of the Parliamentary Budget Process in Four Countries

	Australia	Germany	South Africa	United Kingdom
Scope of amendments	1997/8 Budget: no amendments.	1998 Budget: total expenditure was cut by DM 4.2 billion and more than DM 17 billion were shifted between votes.	1998/9 Budget: no amendments possible.	Amendments only at the instigation of the government.
Members rights to move amendments.	Members may move to reduce expenditure or revenue.	Members may move to reduce expenditure and increases revenue. Increases in expenditure or decreases in revenue require the consent of the government.	No amendment rights are currently specified.	Members may move amendments to reduce revenue or expenditure.
Location of amendment powers	Floor of the House.	Committee.	Not applicable.	Committee and Floor of the House.
Duration of parliamentary Budget process and committee stage in deciding House	1-2 months, no committee stage in the House of Representatives.	4 months, committee stage in the Bundestag lasts several weeks.	3 months, Finance Committee stage in the National Assembly lasts 7 days.	3 months, committee stage in the House of Commons can last several weeks.
Committee(s) involved and its (their) role	House of Representatives: no committee stage. Senate: departmental budgets referred to eight Legislation Committees for examination and report. These engage in public hearings, usually attended by the relevant minister.	Bundestag: Budget Committee (Haushaltsausschu�) assigns _�ch vote to selected reporting Members (Berichterstatter) of the committee, from the opposition as well as the ruling party. The recommendations to the House are almost exclusively based of the departmental reports to the committee of the reporting Members. Bundesrat: Finance Committee.	National Assembly: Finance Committee, with other committee reports optional. Report currently provides comment only. National Council of Provinces: No committee stage required.	House of Commons: Standing Committee on Finance considers the less controversial parts of the revenue side of the Budget. Proposals for expenditure are sometimes examined by the appropriate select committees.
Research capacity (e.g. staff, qualifications, data access etc.) of committees	House of Representatives: Secretariats consist of 5-6 staff members (Joint Committee on Public Accounts and Audit: staff of 10). Senate: Each pair of committees has a shared secretariat, consisting of the secretary, an executive assistant and one or more research officers. The Parliamentary Research Service, available to Members, consists of 80 staff members.	Parliamentary committees, including the Budget Committee, have a secretariat, which consists of between 5 and 6 public servants. The Budget Committee's reporting Members can draw on assistance from the auditor-general as well as officials from the respective ministries. The scientific service of the Bundestag (Wissenschaftlicher Dienst) has a designated Budget and finance unit consisting of 5 members, and 80-85 members overall. It offers assistance to all committees upon request.	National Assembly: Eight researchers shared between all committees.	The Standing Committee on Finance has no researchers. The Treasury Select Committee has two clerks, two specialists and two administrative staff, as well as the right to employ advisors if necessary. The Select Committee on Public Accounts relies on the national Audit Office's research, and is supported by one clerk only. The Library of the House of Commons has a large research section available to Members, consisting of a staff of about 75.
Interaction between departments and committees during financial year	The JCPAA receives and reviews approximately 40 reports from the auditor-general during the year that focus on financial and performance audits. The committee holds public hearings for each report. The Public Accounts committee and Members are also able to access current expenditure information on-line. The committee has access to the Auditor-General and other relevant ministries.	Committee control during the year is exercised through the designated reporting Members (Principle of continuous control, mitlaufende Kontrolle, introduced in 1985). In addition, four times a year, the Budget Committee scrutinises governmental reports on extra-budgetary expenditure. The Committee on Public Accounts, a subcommittee of the Budget Committee, oversees expenditure.	Varies; however, no up to date expenditure reports are received during the financial year.	The Select committee on Public Accounts receives roughly 50 reports on expenditure and impact from the National Audit Office during the year. Select committees, charged with examining departmental expenditure, lack the information base of the Public Accounts Committee.