11 Performance in the New French Budget System

Marc Robinson
Published Date:
October 2007
  • ShareShare
Show Summary Details
Brice Lannaud

On August 1, 2001, when parliament passed a new organic budget law—the Loi Organique relative aux Lois des Finances (LOLF)—France acquired a new system of expenditure planning and execution. The LOLF has been progressively implemented and came fully into effect on January 1, 2006.

This new system replaces the traditional budgeting system, in force since 1959, which had become unsatisfactory. In a context of high debt levels and burdensome levels of taxation and social security contributions, France had to devise a more efficient way of managing public resources. This was not possible under the traditional budgeting system.

Under the new system, parliament has acquired increased power over the budget. Line ministries and agencies (henceforth ‘Tine agencies”) have acquired increased managerial autonomy in exchange for committing themselves to improving the effectiveness and efficiency of their expenditures. The essential aim of the new system is to improve the results delivered by public expenditure and the government agencies that manage that expenditure, for any given level of resources.

Reasons for reform

The previous budgeting system instituted in 1959 was weak

The budgeting system which LOLF has replaced was established by an organic budget law dating from 1959. This law was not conducive to efficient public expenditure management and severely restricted parliament’s control over the budget. For example, the process of parliamentary approval of the annual budget under the 1959 law promoted incrementalism by requiring a two-stage budget approval process. The first stage involved making general appropriations for maintaining the expenditure necessary to ensure continuity in public services and for meeting government commitments (the “current services appropriations”—services votés). These accounted for 95 percent of the total budget. The second phase, which was more detailed, concerned measures for achieving savings or increases in appropriations for “new expenditure authorizations” (mesures nouvelles). This covered less than 5 percent of the budget.

In addition, the annual budget law was a traditional “line-item” budget which appropriated expenditure by ministry and by economic classification (remuneration, operating, capital, and so on). Appropriations were not linked to public expenditure objectives, and the appropriation process was therefore not conducive to optimal resource allocation.

In the execution of the budget, the 1959 system did not encourage the proactive management of appropriations. Indeed, approved budgets were extremely fragmented, broken down into 850 capped appropriation “envelopes” for central government managers (the “budget chapters”) and 2,000 capped appropriation envelopes for decentralized government units (that is, service delivery units of the state located in the regional) and regional or departement-level directorates (the “estimates articles”—articles de prevision). As each of these appropriation envelopes was further broken down by economic classification, whenever managers wished to redeploy appropriations from, say, operating expenditure to capital expenditure—for example, to implement public policy—they had to seek approval at a higher administrative level (central government in the case of decentralized units, and the Ministry of Finance and the Prime Minister in the case of central government units). This hampered the reallocation of appropriations.

There was, as a consequence of the appropriations regime, very limited discussion or debate at the end of the fiscal year about budget performance. In particular, the focus of discussions between the Ministry of Finance and line agencies was upon the extent to which appropriated amounts were spent, and not on the efficiency of expenditure or on the extent to which expenditure objectives had been achieved.

The 1959 system severely restricted parliament’s power to amend the budget to the extent that parliament was virtually powerless to change the appropriations proposed in the budget law proposed by the government. Article 40 of the 1958 Constitution provided that “bills and amendments introduced by Members of Parliament shall not be admissible where their adoption would have as a consequence either a diminution of public resources or the creation or increase of an item of public expenditure.” Article 42 of the 1959 organic budget law made these provisions much tougher by confining parliamentary budget amendments to those “eliminating or effectively reducing an expenditure, creating or increasing an item of revenue.” This barred parliament not only from increasing aggregate expenditure or reducing aggregate revenue relative to the levels envisaged in the government’s draft budget, but also from increasing any individual item of expenditure or reducing any individual revenue item. The result was a diminution of parliament’s power to authorize and debate the budget within the framework of “rational parliamentarism” (one law, one vote, with the assumption of responsibility for an overall legislative package), markedly modifying prior practices that were more favorable to parliament but which had led to substantial budget instability.

The time was ripe for reform

There were, over the years, 35 attempts in parliament to reform the 1959 organic budget law, all of which failed. Only in 2001, with the passage of the LOLF, was reform achieved. This was because, finally, the external context made it possible to effect reforms which had become pressing as a result of the traditional budgeting system’s inherent flaws. The key factors which came together to produce this change were:

  • A more demanding society, in a more difficult fiscal context Mounting government deficits and debt from the 1980s onward led to spirited debates on public expenditure. In a context of ever scarcer resources and increasingly burdensome levels of taxation and social security contributions, the debate naturally focused on the proper use of the taxes paid by citizens and, therefore, on the effectiveness and efficiency of expenditure. This led naturally to discussion of the improvement of government management, the introduction of new management methods, more decentralized management, the decentralization process itself, and even the outsourcing of certain government activities. In a more general way, citizens—who are both users of public services and taxpayers—began to seek a better understanding of public expenditure and to demand more transparency as well as greater efficiency and quality. With public debt exceeding 60 percent of GDP and public expenditure alone representing 54.7 percent of GDP (2003 figures), the issue of governance reform could not be ignored.

  • A supportive international environment. Financial governance issues were underscored by various international organizations, and some countries set up systems to improve governance. The OECD and the IMF had both developed codes of good practice on financial governance. Relevant elements of these codes include, in particular, strengthening parliamentary control over finances, greater accountability of line ministries to the Ministry of Finance, transparency of budget information, and establishment of accrual accounting similar to corporate accounting. Also, many other countries had initiated or developed governance improvement processes, including Canada (through its program review exercises), the United Kingdom (multi-year budgets, performance budgeting, accounting, and so on), and the Netherlands (performance budgeting). France closely studied these various experiments in other countries, particularly through the mission headed by the Inspector-General of Finance, Henri Guillaume, on performance management. Finally, with the Stability and Growth Pact, the Economic and Monetary Union established the requirement of multi-year fiscal programming for all the governments of the European Union.

  • Parliamentary initiative. The origin of the LOLF reform can be traced directly to a parliamentary initiative. As early as 1998, a working group on the effectiveness of public expenditure and parliamentary control had been formed in the National Assembly by its President, Laurent Fabius. A specialized parliamentary committee was then established alongside the assessment and audit mission (the MEC). A further impetus arose from a political debate begun in late 1999 on the credibility of big tax revenue surpluses claimed by the government (the “tax jackpot” debate), which led a strong push in the Senate for greater transparency in financial reporting. The political response was an announcement on March 16, 2000, by Prime Minister Lionel Jospin that the 1959 organic budget law would be reformed.

    With the way cleared by parliament, the National Assembly’s general rapporteur on the budget, Didier Migaud, tabled an initial draft of the LOLF in the National Assembly on July 11, 2000. He was supported by Alain Lambert, President of the Senate Finance Committee, who agreed to work on the drafting of the new law. Cooperation between the National Assembly, the Senate, and the government gave rise to an agreement to reform the budgetary system without touching the Constitution itself. Both the government and the Audit Office (Cours des Comptes) contributed to the drafting. The result was that when the new law was presented to the parliament, there were no amendments from the government (Laurent Fabius, formerly President of the National Assembly, was by then Minister of Finance) and the vote was nearly unanimous (with only the communists in the Senate abstaining).

The new budgeting system is aimed at moving from input-based to performance-based fiscal management in France, essentially by strengthening the managerial autonomy of government units and increasing parliamentary control of the budget.

Strengthening governmental autonomy and parliamentary control

Strengthening managerial autonomy of government units

Based on the notion that financial managers in the field are in the best position to facilitate the optimal use of available resources, the French budget reform seeks primarily to increase the autonomy of these managers. Accordingly, the appropriations structure of the budget has been reformed. The number of capped appropriation envelopes allocated to ministries and approved by Parliament was reduced from 850 to 133.1 This multiplied by a factor of six the average size of the appropriation envelopes available to each minister. On the other hand, restrictions on the movement of appropriations within the fiscal year are tighter under the LOLF than they were under the organic law of 1959 (see Box 11.1).

Crucially, these capped appropriation envelopes are now based on “programs” which are defined by purpose—in other words, by public policy objective rather than by input type.2 Within each program, appropriations can therefore in large measure be used freely for capital, operating, and personnel expenditure, as long as they contribute to the public policy involved. There is only one restriction on the free use of appropriations: within the appropriations for a program, a ceiling on personnel expenditure must be observed. Although it is possible to redeploy appropriations from personnel to operating or capital expenditure, the reverse is not allowed. This is what is known as “asymmetrical fungibility.” This control of personnel expenditure is reinforced by a separate ceiling on the numbers of civil servants (expressed in full-time equivalents) each ministry is permitted to employ. The special treatment of personnel expenditure is a reflection of the long-term rigidity that characterizes such expenditure within the framework of the employment guarantee that comes with civil servant status in France.

Box 11.1.Changing budget allocations during the fiscal year

Under the LOLF, a distinction is made between the transfer of appropriations, and the virement of appropriations.

The transfer of appropriations refers to changes in the ministry responsible for executing the expenditure, which nevertheless leave unchanged the programmatic purpose of the expenditure. That is, the transferred appropriations must still be used to finance an action that could be funded by the program for which that expenditure was originally appropriated (for example, co-financing a joint commitment or research concerning individual ministries). Such transfers are now authorized by decree of the Prime Minister, whereas under the system established by the 1959 law they only required ministerial authorization.

The virement of appropriations refers to “changing the distribution of appropriations among the programs of a given ministry.” Virement modifies parliamentary expenditure authorizations more significantly than do transfers, inasmuch as they change the purpose of the expenditure. The LOLF provides that “the total amount, during a single year, of appropriations that have been vired may not exceed 2 percent of appropriations allocated by the annual budget law to each of the programs concerned.” Technically, the purpose of this procedure is to correct budgeting errors or cover specific management contingencies. However, in practice they will have the broader consequence of facilitating programmatic reallocations within the 2 percent limit. Under LOLF, virement modifies the purposes for which money is being spent, whereas under the 1959 system, it changed the type of input upon which money was spent (albeit within a single ministry and within the limit of 10 percent of the initial appropriation for each of the chapters concerned). Overall, the margin for shifting appropriations remains approximately the same, although slightly greater by virtue of moving from a redeployment limit divided by five (that is, 2 percent of the appropriation instead of 10 percent) to a basis divided by six, with the change from 850 chapters to 133 programs under the general budget.

A “program manager” is appointed for each program, more often than not selected from among central government directors. These program managers are empowered to allocate and reallocate appropriations within programs, and to set performance objectives for the services covered by the program. Figure 11.1, as an example, illustrates the new budget structure of appropriations for the Ministry of Justice.

Figure 11.1.New budget structure of appropriations for the Ministry of Justice

The government has been careful not to simply base programs on existing organizational units (“directorates”). Thus, of 124 non-grant programs, 56 do not correspond to a particular central government directorate. The government’s aim has, rather, been to define centers of responsibility. With respect to the relationship of programs and organizational structure, three cases may be distinguished:

  • A program corresponds to a specific organizational unit, usually a central government directorate. Such a one-to-one correspondence is to be expected insofar as there are directorates with mandates focused on specific public policy objectives which are naturally designated as programs under the new system (police, gendarmerie, judicial system, correctional system, and so on).

  • A central government directorate handles several programs because the breadth of the role of the directorate made it logical to recognize it as pursuing several distinct programmatic objectives. One consequence of this has been that there are a number of cases where the same individual serves as program manager for more than one program.3

  • A single program involves several central government directorates. Examples of this include the “heritage” program of the Ministry of Culture, which brings together six directorates, or the “central and local government tax and financial management” program of the Ministry of Finance, which brings together the two major directorates with networks (General Tax Directorate and General Directorate of Government Accounting), as well as part of Customs.4

As discussed in Chapter 20, the reform of organizational structures was not considered to be an element of the LOLF reforms at the time of their introduction. Nevertheless, this new budget structure should gradually encourage agencies to adapt their organizational structures to the new budgeting and management framework, usually through the regrouping of central government directorates.

In those cases where organizational and program structures differ, each program has been asked to develop a “management charter” specifying the manner in which the program’s budget allocation will be shared between the directorates concerned, as well as the responsibilities of each directorate in contributing to the overall performance of the program. In most cases where a number of directorates contribute to a single program, the program manager will be the head of one of the directorates concerned. He or she will, as a consequence, have no hierarchical authority over his or her colleagues. The management charter is intended as a means of specifying the rules of the game under these types of circumstances. However, this objective has not yet been fully achieved, because there are a number of cases where the management charters that have been developed for specific programs do not as yet accord to the program manager the full extent of authority that they ought to have.

Apart from affecting central government managers, the budget reform is intended to increase the managerial autonomy of the decentralized government units that manage a large proportion of government appropriations and actually deliver the services through which many public policy objectives are realized. With this in mind, each ministry has been asked to accord to its own decentralized units the flexibility afforded to the ministry as a whole under the new budget framework. To understand what this means, it needs to be borne in mind that each budgetary program is broken down within ministries into a number of “operational program budgets/’ more often than not at the regional level (which would mean that one program could be broken into 22 operational program budgets, in line with the number of regions in metropolitan France). The new system has created considerably greater flexibility at the local level, by contrast with the 1959 system under which there were a far larger number of capped appropriation envelopes which local level managers were obliged to observe. In fact, the amount of resources that the manager of an “operational program budget” can freely use has thus multiplied about twelvefold relative to the 1959 system. Unfortunately, some central government units have prevented local managers from benefiting fully from such expenditure flexibility by imposing additional disaggregated controls on how the operational program budgets may be spent. Furthermore, although the size of the appropriations allocated to the decentralized level was increased, parallel practices of “earmarking” appropriations persist through various instructions that may limit the options of the decentralized units.

As noted above, this enhanced managerial autonomy is combined with increased parliamentary control of the budget.

Increased parliamentary control of the budget

Parliament now has the power to redeploy appropriations across programs, subject to certain conditions described below. In the budget law, the 133 programs are grouped into 34 missions that are linked to major areas of public policy, of which the programs constitute subdivisions. Some missions may be interministerial and encompass programs falling under the authority of various ministers.

Within a single mission, parliament can change the distribution of appropriations among the various programs, thus departing from the initial breakdown proposed by the government. Interpreting the Constitution, Article 47 of the LOLF opens up the possibility of parliamentary amendments to expenditure: “within the meaning of Articles 34 and 40 of the Constitution, the responsibility is understood, in respect of amendments applicable to appropriations, to be that of the mission.” Therefore, amendments proposing increased appropriations for a program offset by an offsetting reduction in appropriations for another program within the same mission are now admissible. This interpretation of the Constitution has been approved by the Conseil Constitutionnel, France’s constitutional court. Parliament’s capacity to amend the budget still remains limited by a bar on changes in the aggregate level expenditure proposed in the government’s budget, even if financed by equivalent changes in taxations so as to leave the fiscal balance unchanged. Changes by parliament to the distribution of appropriations among missions are also prohibited.

Since missions are established by government initiative, members of parliament may not create new missions. They may, however, create or abolish programs within missions. It is for this reason that the government began consultations with the finance committees of the National Assembly and the Senate on the new budget architecture, two years before the organic law came into force. Parliament wanted to ensure that the number of programs would be large enough for it to exercise real power of amendment. It therefore became necessary to strike a balance between parliament’s wishes and the objective of making appropriations fungible in order to allow for more efficient management. The government developed a prototype program/mission classification and submitted it to parliament in January 2004. In response to the views expressed by the parliament, the government modified this classification by adopting most of the suggestions made jointly by both the National Assembly and the Senate, including the establishment of new interministerial missions such as “Security” and “Territorial Policy,” and increasing the total number of programs.

The corollary of this increased power of parliament to amend the appropriations proposed in the government’s draft budget has been the disappearance of the distinction between “current services appropriations” (services votés) and “new expenditure authorizations” (mesures nouvelles), which—as noted at the outset of this chapter—tended to promote incremental adjustments to expenditure from one year to the next. Under the new system, parliament may now question any program expenditure. This puts pressure on government units to justify the total amount allocated to each program, rather than simply any incremental changes relative to the previous year. This idea has been captured under the system by reference to the principle of justification au premier euro. This “justification from the first euro” is intended to lead to better information on the output and activity content of programs, and not only their expenditure content. Specifically, the intention is that agencies will be required to clarify the drivers behind the program appropriations requested of parliament. This will include physical drivers (number of users, volume of activity, number of staff, building floor space, and so on), financial drivers (determinants of a wage bill, unit costs of intervention mechanisms, and so on), or a combination of the two. Such information is to be included in annexes to the draft budget law, and should enable members of parliament to gain a more in-depth understanding of what each program involves. The overarching consideration here is that the fungibility of appropriations within broad programs should not result, especially for parliament, in a loss of information concerning the purpose and nature of expenditure relative to the current situation. It is also intended that the principle of “justification from the first euro” will also be applied to end-of-the financial-year reporting on program performance, in such a way as to strengthen the link between expenditure planning and execution.

These new arrangements are likely to lead to less inertia in the process of allocating budget resources.

Formal commitments by government units to enhance expenditure effectiveness

In exchange for increased managerial autonomy, agencies are now required to commit formally to the achievement of objectives for improved efficiency and effectiveness. These objectives, specified in terms of performance indicators and associated performance targets, must now (according to Articles 7 and 51 of the LOLF) be specified in “annual performance plans” (projets annuels de performances) presented as annexes to the draft budget law. At the end of the fiscal year, the results achieved must be reported and commented upon in “annual performance reports” (rapports annuels de performances) presented as annexes to budget closure law5 (Article 54 of the LOLF). Performance targets may be set, on an annual or multi-year basis, for periods of up to five years.

This procedure is based on the application of a number of key principles which were agreed to by parliament, the Audit Office, and the government, and which are embodied in a methodological guide entitled The Performance-Based Approach (June 2004). Prior to the LOLF coming into full force, these principles were tested by all ministries through the publication of pilot annual performance plans in the fall of 2004. The further refinement of these principles has been facilitated by the review of these pilot plans by the finance committees of the National Assembly and the Senate, as well as by the Audit Office and the Interministerial Program Audit Committee. In addition, several pilots carried out at the decentralized level by certain ministries during 2003, 2004, and 2005 have also provided an opportunity to investigate the practical effects of applying these principles.

Objectives aimed at enhancing expenditure effectiveness: “spending better”

The objective of the performance-based approach is to “spend better.” This approach encourages those who design and manage public policies and the personnel who implement them to improve the results achieved by public services and the quality of those services, at least cost and within the framework of the resources allocated to programs.

Accordingly, agencies are expected to articulate three types of objective in their annual performance plans:

  • Outcome objectives, reflecting citizen expectations. These outcome objectives should spell out the expected benefits for citizens or society which each program will deliver, in terms of changes brought about to economic, social, ecological, health, or cultural conditions. They are intended to indicate the socioeconomic impact which the government’s actions are intended to have, rather than the services (outputs) delivered. For example, the socioeconomic effectiveness objective of the “public elementary education” program is “to result in all pupils mastering essential basic skills by the time they complete primary education.”

  • Service quality objectives, reflecting expectations of the users of public services. These objectives define the expected quality of services delivered to users, that is, the suitability of the service to meet the needs of the recipient, whether the recipient is a willing customer of the service or an unwilling subject (prison inmate, for example). The user may be an external user (user of a public service) or, in the case of programs based on internal support services (such as senior management and resource management), an internal user (that is, staff of the supported programs). For example, a service quality objective in the “development, urban planning, and public engineering” program is “to provide users and local governments with quality service in the delivery of urban planning permits.”

  • Management efficiency objectives, reflecting taxpayer expectations. These objectives relate to technical efficiency—that is, the maximization of the outputs or activities delivered given the level of resources. For example, the efficiency objective of the “central government and local public sector tax and financial management” program is “to reduce the cost of tax management.”

Objectives related to the reallocation of inputs or activities within programs to achieve new priorities are also recognized as management efficiency objectives.

It is expected that each of these three types of objectives will, to the extent possible, be specified for each program. The improvement of public expenditure should be pursued in each of these three dimensions simultaneously, all the more because the expectations of citizens, users, and taxpayers do not always coincide. Experience has shown that citizen expectations for socioeconomic outcomes often tend to receive more consideration than user expectations in terms of service quality, which in turn tend to receive more attention than taxpayer expectations with respect to efficiency. Accordingly, of all the objectives contained in the pilot annual performance plans published in 2005, 39 percent related to outcomes, 18 percent to service quality, and 25 percent to efficiency (with 18 percent not falling into any of these categories). Parliament therefore requested that the objectives related to quality and especially to efficiency be strengthened, and that the management efficiency objectives be specified in respect to the bulk of program appropriations. In the annual performance plans published with the draft 2006 budget law, outcome objectives account for 54 percent of the total, service quality objectives 18 percent, and management efficiency objectives 28 percent.

Activity, input, and output quantity objectives are not covered in the performance-based approach, as they are attained mainly as a result of increases in the volume of resources and therefore are not evidence of improvements in spending effectiveness and efficiency. The specification of such objectives would tend to undermine the objective of providing increased managerial autonomy because it would in many cases, in effect, reduce the flexibility provided by the new budget framework with respect to managerial choices in the use of allocated inputs. Although 13 percent of the objectives in the 2004 pilot annual performance plans related to inputs or activities, such objectives had virtually disappeared from the 2006 annual performance plans.

It has been deemed essential that these objectives be assigned to a specific program manager, and that the program manager and his/her minister should be accountable to parliament for the achievement of their assigned objectives. An objective can be assigned to a program if:

  1. its achievement depends on the activities, appropriations, tax expenditure, or personnel of that specific program

  2. it is feasible for the program manager and his/her minister to commit to a target for performance within a timeframe not exceeding five years.

In the pilot annual performance plans of 2004, approximately 5 percent of objectives could not be assigned to programs because they did not meet these criteria.

In some cases, programs have wide-ranging objectives which go beyond the field of competence of a single program manager. In these cases, responsibility for performance in respect to that objective has been assigned to a minister appointed to take the leading role in respect to the cross-cutting policy objective concerned. This is the case, for instance, with certain interministerial public policy objectives of considerable budgetary significance, identified by the Prime Minister, which related to several programs falling under the authority of various ministries and not necessarily belonging to the same mission. The lead minister appointed by the Prime Minister is responsible for coordinating the activities of the various programs involved, with a view to achieving the common socioeconomic results shared by those programs. Because of the coordination powers vested in the lead minister, the program can have cross-sectoral objectives reflecting the combined impacts of the activities being carried out under several programs.

As regards the cross-sectoral policy on road safety, for example, the road safety program, as one covered by the lead minister, has a cross-sectoral objective—namely reducing the number of deaths on the roads—that is clearly beyond the scope of this program alone. Other programs which support the achievement of this objective have more specific objectives which fall within their specific fields of competence but which contribute to the higher-level common objective. Thus the national police and national gendarmerie programs will seek to increase the effectiveness of traffic-related controls in their respective areas of competence, while the national road network program will seek to reduce the dangerous aspects of the road infrastructure.

In addition to its annual performance plans, the lead ministry is also responsible for producing a cross-cutting policy document (document de politique transversale—DPT), annexed to the draft budget law, which spells out its coordination role. The DPT describes the overall strategy of the cross-sectoral policy and contains a structured presentation of the strategic objectives included in the various programs covering aspects of that policy. An objective presented in the DPT must also be shown in an annual performance plan, as only the latter document is required by the LOLF. The purpose of the DPT is to facilitate coordination of the objectives of various programs, rather than to institute a parallel, competing management mechanism. (See Table 11.1.)

Table 11.1 lists cross-cutting policies covered by a DPT and the respective lead ministers.

Table 11.1.Cross-cutting policies covered by a DPT
Cross-sectoral policiesLead minister
Civil securityInterior
Overseas territoriesOverseas Territories
Road safetyCommunity Facilities
Action outside the countryForeign Affairs
Official development assistanceForeign Affairs
Urban affairsUrban Affairs
Higher educationNational Education
Combating exclusionSocial Affairs

A limited number of objectives reflecting priorities and based on a decisive approach

It has been decided that the number of objectives specified for programs should be limited, to ensure the overall readability of the budget law and prevent a dispersal of efforts that would lead to the pursuit of too many objectives at once. Ideally, there should be no more than about six objectives per program. This means that the objectives specified for each program are not necessarily comprehensive, but rather reflect the most important program objectives.

In the annual performance plans for 2006, each program contains five objectives on average,6 with an average of two performance indicators for each of these objectives.7 Each objective relies, moreover, on the implementation of specific courses of action (levier d’action) or action plans. These courses of action may be defined as any change in the way in which inputs are used, or in the activities used to deliver services, which can assist in maximizing outcomes, quality, and efficiency Thus performance with respect to program objectives, within given resource constraints, may be enhanced through:

  • redesign of the policy; and/or

  • changes in the way in which services are delivered. For example, to improve service quality with given resources, the following intervention strategies might be used: adoption of a service quality approach, training of staff, retooling of processes, a change in opening hours, and so on.

Operational objectives

The performance objectives set out in the annual performance plans give agencies considerable autonomy in deciding how those objectives should be achieved, as they do not specify in detail which activities should be carried out, the appropria tions to be used, or the intervention strategies to be employed. It is up to agencies to make the best possible use of the new flexibility provided by the LOLF, through the fungibility of appropriations, to decide on the most appropriate course of action to achieve their objectives.

Agencies are therefore asked to put in place a performance-based steering mechanism to focus the overall action of their staff on the attainment of objectives. To this end, the strategic objectives presented to parliament are translated by agencies into operational objectives related to each of the “operational program budgets” (see above) of which the program is comprised. Indicators and targets are then specified for these operational objectives.

Objectives defined within the framework of the budget process

One of the key phases of the annual budget discussion is the discussion and joint determination of objectives, indicators, and targets.

At the beginning of the year, the ministries first discuss with the Ministry of Finance, under the authority of the Prime Minister, the amount of each ministry’s appropriation envelope for the following year and the structural reforms likely to generate savings. Program managers then discuss, in the context of the envelopes allocated by their minister, the results obtained during the previous fiscal year, and the targets appropriate for the future. New objectives are defined as needed. The same Ministry of Finance staff are involved throughout this dialogue.

Results achieved and objectives for the coming year are not used to determine the allocation of expenditure between programs, because this has been determined prior to the definition of objectives. Around June, after these discussions are over, a provisional list of objectives and indicators under consideration for the next draft budget law, without targets, is submitted to parliament. Although parliament has no power of amendment of these objectives and indicators, it can make its views known. In doing so, it can make use of the annual performance reports published with the budget closure law. The government can decide to change certain objectives or indicators in response to the views expressed by the Parliament.

Performance targets in respect of the chosen indicators may be adjusted further during the summer months, in the event that program appropriation envelopes change as a result of new trade-off discussions. The finalized annual performance plans, including objectives, indicators and targets, are placed before parliament on the first Tuesday of October.

Audited indicators

The inspection unit of each ministry, the Interministerial Program Audit Committee, the Audit Office, and parliament audit the accuracy of the results reported by ministries.

The Interministerial Program Audit Committee

The Interministerial Program Audit Committee (CIAP) is a body composed of representatives of the various audit and inspection units. It is accorded by the government with responsibility for auditing the quality of the data presented in annual performance plans and reports. It assesses the consistency of objectives and related indicators, the reliability of the information systems that produce the indicators, the pertinence of the action plans adopted to achieve the objectives, the quality of the system for managing the networks of decentralized units in the management control system, the accuracy of reported results, and the objectivity and completeness of the analysis of variances between targets and actual results.

The Audit Office

The Audit Office functions independently of the government and is specifically responsible for auditing the proper use of public funds. In its annual work auditing the execution of the budget, the Audit Office analyzes the consistency and reliability of the data presented in the budget documents, especially with regard to performance, as well as any major variances between targeted and actual performance. In addition, more in-depth audits, which may be undertaken at the request of parliament, are conducted to ensure that government actions recorded in the budget laws are performed in accordance with established rules.


The special rapporteurs,8 who are members of the parliamentary Finance Committees, may use their powers to conduct off-site and on-site audits, particularly with a view to testing the reliability of the results announced. The Finance Committees are authorized, on the occasion of assessment and audit missions, to verify—if necessary with the support of specialized bodies or the Audit Office—the methods used in compiling and reporting performance data. The work of the members of parliament involved also consists of determining, in the field, the reliability, quality, continuity, and use made of the performance indicators associated with each program.

Improving results within the framework of the allocated inputs

The link between budgeting and performance

In France’s performance budgeting system, the link between appropriations and performance is not automatic. On the contrary, an improvement in outcomes or service quality is sought with given inputs, based, for example, on program redesign or changes in the input mix.

In addition, performance commitments are made for given levels of appropriated resources, after the allocation of expenditure to programs has already been determined based on the political priority given to certain areas. The challenge is to assure parliament that government units will seek the best possible use of the funds granted to them, for a given policy.

Nor do the results achieved automatically affect the amount of appropriations for the following year. The review of results achieved should, first of all, raise questions about the courses of action used—that is, the design of policies and the nature of the activities and input mix used—prior to any discussion of the level of program funding. Nevertheless, the performance-based approach may lead indirectly to decisions concerning the level of resources provided. Indeed, the availability of information about the results achieved by programs provides members of parliament with a valuable tool for analyzing both the quality of program management and the pertinence of the underlying objectives. This information may when necessary and along with other factors, lead members of parliament to propose amendments aimed at reallocating appropriations between programs grouped under the same mission.

Use of performance information in the budgeting and management process

The Ministry of Finance, parliament, the ministry head offices, and the decentralized units all use performance information at various stages of the budget process. As a result, the performance-based approach changes the budget dialogue, by encouraging reviews or reforms of ineffective interventions and by setting objectives when program appropriations increase in order to ensure that additional funding is used optimally. During and at the end of the fiscal year, performance is used to guide the agency management toward the achievement of priority objectives, to define corrective measures, and to identify and share best practices.

In practice, when the budget is being compiled:

  • the Ministry of Finance looks for objectives that reflect reforms deemed necessary, making it possible to reduce appropriations or improve their use in real terms; it also looks for objectives that will impose performance pressure on the program

  • when a decision has been made to increase the appropriations for a program, the Ministry of Finance looks for objectives that will allow for future verification that the appropriations were used in the best way possible, especially in cases where it may have preferred another, less costly approach to achieving the objectives concerned

  • once the objectives have been formulated, the Ministry of Finance tries to ensure that performance targets are set which are consistent with the best possible use of the appropriations in question

  • program managers individually seek to define objectives that will enable them to energize and mobilize decentralized units, by assigning them a number of clear priorities to achieve during the fiscal year.

During the fiscal year:

  • program managers monitor performance measurements, so as to take corrective steps when necessary: speeding up a reform, reallocating resources, and so on

  • program managers can immediately redeploy, within a program, any savings generated by performance improvements.

At the end of the fiscal year:

  • program managers compare the results achieved by the various decentralized units contributing to the program. From this comparison, they determine which are the best practices or the most effective actions for achieving the best results with similar resources. They then disseminate these good practices, the adoption of which they recommend to all decentralized units. Over time, this exercise leads to overall improvements in program performance

  • when the desired program results have not been achieved, the program manager analyzes the reasons for this and defines the remedial measures needed to improve future results

  • where applicable, the results achieved and the analysis of the reasons for them can, along with other factors, lead to a decision to change the amount of program appropriations, in the event that there is a concomitant major revision of the procedures for implementing the policy in question.

Future paths

Experience to date has led to the development of a more substantial link, within programs, between appropriations and performance and between compensation and performance. These arrangements are aimed at rewarding the most successful units or staff members within programs, rather than providing rewards to programs which perform better than other programs. For example:

  • within a number of government units, such as the General Directorate of Taxes, the achievement of targeted results by decentralized units results in their being allocated additional funding for specific purposes such as improving working conditions for staff

  • the national police directorate has introduced a performance-based compensation system under which staff in better-performing units receive individual bonuses.

In each of these cases, the fact that performance measures have been used by management for some years has put the agencies in a good position to identify appropriate indicators on which to base their systems of rewards. Most other government units have not yet reached this stage of maturity in performance measurement. At the same time, experiments carried out in other decentralized units suggest that the autonomy granted through the fungibility of appropriations, together with the setting of explicit performance objectives, are capable of significantly increasing the motivation to perform of both management and staff.

Finally, the Ministry of Finance has entered into discussions with certain volunteer ministries with a view to concluding multi-year performance contracts.9 These contracts will provide the ministries in question with a multi-year perspective on the appropriations they receive, in exchange for the attainment of ambitious performance objectives based on major reforms. The multi-year framework is intended to promote the accomplishment of major reforms, as such reforms often require considerable time for formulation and implementation, and their effects can only be felt in the medium term. The multi-year perspective on the amount of appropriations allotted helps to create a stable framework conducive to the success of the reforms. The contract also allows for determining, in advance, the portion of potential savings that can be reallocated to the program in exchange for the efforts made.

* * *

In conclusion, the new French budget reform seeks to make the allocation of resources more transparent and more flexible, based on the new budget architecture which is defined by purpose and on increased parliamentary control of the budget. It also seeks to make the use of allotted resources more effective by giving greater managerial autonomy to central and decentralized government units, in exchange for commitments on the results to be achieved.


In the 2006 budget law.

There is an exception to this principle: certain resource management functions (human resources, property management, and so on), senior management functions (ministerial office, inspections), and certain joint expenses (investment in a joint information system for the ministry as a whole) are still classified under support programs because they cannot be distributed among public policy programs, or could only be so distributed to the detriment of their effectiveness. To compensate for this exception, the documents annexed to the draft budget law contain, for information purposes, an analysis of the costs of public policies, which consists in breaking down the support appropriations, using allocative keys, across the public policy programs, in order to reconstruct the full cost.

There are, in fact, nine senior servants—including the Deputy Director of Employment, the General Director of the Treasury and the General Director of Local Government—who are program managers for three or more programs.

This example is discussed further in Chapter 20.

The law which, after the closure of the accounts for the financial year, serves as the means for parliament to note and discuss the financial results for the year.

Among programs with objectives, programs without performance objectives.

A total of 629 objectives for the general budget, consisting of 133 programs, 118 of which have objectives; 1,284 indicators for the general budget.

The general rapporteur on the budget, appointed by the finance committee of each assembly, is the person chiefly responsible for reviewing the budget law. Special rapporteurs are appointed to analyze the budget of each mission. In early 2005, in preparation for the LOLF, the two assemblies adapted the committees’ organization and work methods by appointing rapporteurs, no longer by ministerial budget document, but by mission. Within each finance committee in the National Assembly and in the Senate, the general rapporteur analyzes the draft budget laws and proposes amendments. He lays out his findings before the committee and submits his proposals for amendments, which are then debated. The commission votes on each amendment and each article of the draft law. At the end of the debates, the general rapporteur drafts a report setting forth the opinion of the committee and submits the amendments agreed upon by the majority of the committee. The other committees to which the draft budget law was submitted for comment appoint a special rapporteur to analyze and submit a report for comment on the relevant missions. In addition to the amendments selected by the finance committee, members of parliament may enter other amendments on which the finance committee has expressed an opinion. Once the draft reports of the special rapporteurs of the other committees have been collected and the report of the general finance committee has been adopted, the draft budget law is discussed in open session.

Multi-year performance contracts covering the period 2006-08 have been concluded with the General Directorates of Taxes and Public Accounting (tax and financial management program of the central government and the local public sector); the General Directorates of Competition, Consumption, Suppression of Fraud, and Customs and Indirect Duties (program to regulate and protect trade in goods and services); and the Ministry of Foreign Affairs.

    Other Resources Citing This Publication