Questionnaire

Low	No projections of capital spending are published beyond the budget year.
Medium	Projections of total capital spending are published over a 3- to 5-year horizon.
High	Projections of capital spending disaggregated by ministry or sector are published over a 3- to 5-year horizon.

View Table

Low	No projections of capital spending are published beyond the budget year.
Medium	Projections of total capital spending are published over a 3- to 5-year horizon.
High	Projections of capital spending disaggregated by ministry or sector are published over a 3- to 5-year horizon.

Definitions of Key Terms

Term	Definition
Projection	An anticipated amount of money available for spending in the future. It reflects policy. It is not a baseline or a mechanical calculation of a percentage of expected future revenue.
Capital spending	See the Glossary
Budget year	Fiscal year for which a budget is being prepared, and thus, typically, it has not yet started.

View Table

Term	Definition
Projection	An anticipated amount of money available for spending in the future. It reflects policy. It is not a baseline or a mechanical calculation of a percentage of expected future revenue.
Capital spending	See the Glossary
Budget year	Fiscal year for which a budget is being prepared, and thus, typically, it has not yet started.

Institutional Design

The purpose of this dimension is to see whether there is a basis for multiyear prioritization of capital spending. The existence of financing constraints for capital spending is expected to encourage priority setting among projects. A multiyear perspective on financing constraints is necessary because major projects are usually implemented over more than one year. This is similar to the purpose of establishing financial constraints when planning, as described in Dimension 2.b.

• If the budget document only provides budget allocations for one year, or if multiyear estimates do not distinguish capital spending, the institutional design should receive a low score. This will often be the case in countries with annual budgets and no systematic medium-term fiscal or budget frameworks.

• If there are medium-term projections of aggregate capital spending, the score should be medium. These projections will often be part of a comprehensive medium-term fiscal framework (MTFF), providing aggregate spending projections for all major budget components. Medium-term frameworks will usually cover at least three years, including the budget year, and may have a longer horizon. For a medium or high score, capital spending projections must be published, but not necessarily as part of the budget document itself. Production of internal estimates of future capital spending is not sufficient. It would be most useful if the projections were included in the budget documentation for information purposes. However, publication of medium-term estimates as part of a fiscal policy document or in an annually updated public sector investment program is also acceptable, as long as these are consistent with the estimates in the budget documents.

• If the medium-term projections for capital are published and presented according to ministry or sector, the design score should be high. This will generally be the case in countries with well-defined medium-term budget frameworks (MTBFs). In addition, some countries may produce medium-term projections for capital spending or for major capital projects, even in the absence of a general MTBF. If these projections provide a complete overview of capital spending, the score could be high even in the absence of a full MTBF. Capital spending may be projected by either ministry or sector. The projections should be made using the same high-level categories that appear in the budget classification. The purpose of disaggregating the overall capital funding amount is to guide units preparing detailed budget proposals on the total size of projects they can propose. Projections of capital spending in national or sectoral plans do not meet the intent of this dimension. Plans are often prepared without regard to fiscal constraints. For this dimension, projections of capital spending must be made by the Ministry of Finance (MoF) and derived from projections of overall fiscal variables, for example, total revenue, expenditure, and deficit. Similarly, ministry and sector projections must be determined in the budget process, typically as part of a top-down budget process in which ceilings are announced at the beginning of the budget preparation cycle.

It is not a requirement for the scoring under this dimension that capital spending projections are binding. As is the case of MTFFs, there are significant benefits in projecting capital spending based on the available information, even if economic and fiscal circumstances, and policies, may change in the future. Under Dimension 6.b, which covers multiyear budget ceilings, there is also a separate assessment of whether the projections are binding. This is not the case under Dimension 6.a.

Important Documents

Documents	Uses
Medium-term budget documents	Assess whether multiyear estimates for capital spending are presented and in which form
Other documents including medium-term capital projections	Assess whether multiyear estimates for capital spending are presented and in which form

View Table

Documents	Uses
Medium-term budget documents	Assess whether multiyear estimates for capital spending are presented and in which form
Other documents including medium-term capital projections	Assess whether multiyear estimates for capital spending are presented and in which form

Effectiveness

If aggregate capital spending forecasts are significantly different from actual total budget spending for capital projects, then the effectiveness is limited. In particular, if the projection is higher than the approved capital budget allocations, then the benefit of making the projection is weaker proportional to the size of the forecast error. The forecast error should be measured as a weighted average across ministries or sectors over the past three years.

If projections of capital spending are prepared but not published, then effectiveness may be higher than the design score indicates. Countries may wish to make capital spending projections but use them internally in the government without publication. Countries may choose not to publish as a political consideration to avoid creating public expectations during periods of economic, fiscal, or policy uncertainty. This is especially true for countries that do not have a formal medium-term budgeting system.

• Low effectiveness indicates that medium-term capital spending projections do not exist or that they deviate significantly from the subsequent budget allocations. This applies if budget allocations are significantly higher or lower than projections.

• Medium effectiveness indicates that capital spending projections deviate somewhat from budget allocations. This applies if aggregate capital budget allocations are somewhat higher or lower than projections as an average over the past three years. Alternatively, the comparison can be based on capital allocations by ministry or sector, weighted according to their value. The result should be the same. A medium score is also warranted if medium-term capital spending projections are made by the MoF and made available to line ministries but not published. If the MoF can document that the information is used when preparing budgets, the effectiveness score may be raised from low to medium. However, publication of projections adds to the transparency and consistency of the budget process, and it will generally not be possible to achieve a high score on effectiveness if the projections are not published.

• High effectiveness implies that projections are consistent with subsequent budget allocation. Some deviation is always possible, but medium-term projections should on average be broadly in line with budget allocations. Box 6.1 describes the capital projections in Jordan, which have been robust.

Useful Data Series

Data	Questions to Address
Compare projections of capital spending with approved budget amounts for capital spending for the same years	Did capital spending projections effectively facilitate prioritization of capital budget allocations?
If the country has a development budget and an operating budget, provide breakdown of these budgets in capital and current spending	Is analysis based on capital spending, not on development spending?

View Table

Data	Questions to Address
Compare projections of capital spending with approved budget amounts for capital spending for the same years	Did capital spending projections effectively facilitate prioritization of capital budget allocations?
If the country has a development budget and an operating budget, provide breakdown of these budgets in capital and current spending	Is analysis based on capital spending, not on development spending?

^{Box 6.1.}

Medium-Term Capital Spending Projections in Jordan

In Jordan, capital expenditure is forecasted over a three-year period, on a rolling basis. Both the General Budget Law and the Budget Laws of Government Units provide estimated capital expenditure for the budget year and indicative capital expenditure for the following two years, by ministry, program, governorate, and project. While the ceilings set for the out-years are not binding, the outturns have not exceeded the indicative ceilings set for capital expenditure since 2012. On average, the budget outturn is 11 percent below both the annual budget and the previous year’s indicative forecast (Figure 6.1). This indicates that budget allocations on average are broadly consistent with medium-term projections.

View Full Size

Medium-Term Budgeting of Capital Expenditure Versus Actual in Jordan, 2010–16

Source: Jordan PIMA 2017.

• Download Figure
• Download figure as PowerPoint slide

Questionnaire

Low	There are no multiyear ceilings on capital expenditure by ministry, sector, or program.
Medium	There are indicative multiyear ceilings on capital expenditure by ministry, sector, or program.
High	There are binding multiyear ceilings on capital expenditure by ministry, sector, or program.

View Table

Low	There are no multiyear ceilings on capital expenditure by ministry, sector, or program.
Medium	There are indicative multiyear ceilings on capital expenditure by ministry, sector, or program.
High	There are binding multiyear ceilings on capital expenditure by ministry, sector, or program.

Definitions of Key Terms

Term	Definition
Multiyear	The budget year plus two years or more after the budget year.
Ceiling	In this context, the maximum funding that can be requested during budget preparation. Given that most budget entities will request funding up to the maximum allowable amount, a ceiling often represents a preliminary budget allocation.
Ministry, sector, or program	As defined in the budget classification.
Indicative	For information purposes only and subject to change.
Binding	A decision that is final (unless fundamental factors change and there are explicit decisions and documented changes to budget ceilings).

View Table

Term	Definition
Multiyear	The budget year plus two years or more after the budget year.
Ceiling	In this context, the maximum funding that can be requested during budget preparation. Given that most budget entities will request funding up to the maximum allowable amount, a ceiling often represents a preliminary budget allocation.
Ministry, sector, or program	As defined in the budget classification.
Indicative	For information purposes only and subject to change.
Binding	A decision that is final (unless fundamental factors change and there are explicit decisions and documented changes to budget ceilings).

Institutional Design

The purpose of multiyear ceilings is to operationalize the financing constraints for capital spending that follow from the MTFF and the multiyear capital spending projections. This is a continuation of the discussion under Dimension 6.a. The difference here is that budget ceilings are more operational in nature, more detailed, and have a shorter time horizon. Projections, as used in Dimension 6.a, are analytical and informational in nature. They are anticipated amounts. Ceilings, as used in Dimension 6.b, are operational limits on budget requests, whether indicative or binding.

• A low score on this dimension implies that no multiyear ceilings on capital spending are provided to ministries before the submission of budget requests. Ceilings are intended to constrain detailed budget requests and must be issued before budget preparation is finished. Therefore, ceilings may be issued in an MTFF at the beginning of the budget preparation cycle or in annual budget instructions. If neither document exists, a ceiling must be issued before the deadline for submission to the central budget office of detailed budget requests.

• A medium score indicates that capital ceilings are issued, but that they are indicative. Ministries may submit budget requests that go beyond the indicative ceilings without these submissions being rejected. There may, however, be specific requirements for justification or documentation of submissions beyond the ceilings. It is common that the ceilings for the budget year are more binding than the out-year ceilings. For the purpose of this dimension, binding ceilings for the budget year and indicative ceilings for the out-years would still qualify as a medium score.

• A high score indicates that the capital spending ceilings are binding, for the budget year as well as the out-years. This does not mean that the ceilings might not change in next year’s budget process. However, any changes in budget ceilings would need to be explicit, explained, and documented. The MoF should provide a step-by-step explanation of any changes in budget ceilings from one budget year to the next. Box 6.2 provides an overview of the capital budgeting framework in Mali, which was assessed to be well designed and highly effective in the 2018 PIMA.

Capital spending ceilings may be issued by ministry, sector, or program. The ceilings must correspond to groupings also used in the budget classification. Otherwise, the ceilings will not have direct effects on the budget submissions.

^{Box 6.2.}

Multiyear Programming in Mali

Budget ceilings with actual and comprehensive costs have been introduced in multiyear frameworks (Table 6.2.1). Documents issued and published by the units in charge of multiyear programming are both complementary and consistent.

• The Medium-Term Economic and Fiscal Framework (DPBEP) for 2020–22 has been submitted to Parliament during the FY 2020 prebudget debate (end June 2019) and includes multiyear projections by type of expense and by ministry over 2020–22.

• The three-year investment plan (PTI) for 2020–22 prepared by the National Development Planning Directorate (DNDP) provides the three-year rolling investment plan and is appended to the 2020 Budget Law as Annex K (Etat K).

• The Special Budget for Investment (BSI), which is the annual component of the three-year investment plan, is incorporated into the budget and presented in the 2020 Budget Law. The PTI reflects all project costs as well as annual expenditure for each project for a three-year period.

• All ministries prepare sectoral Medium-Term Expenditure Frameworks (CDMTs).

^{Table 6.2.1.}

Medium-Term Capital Budget in Mali, 2020–22

(Millions of African financial community francs)

Sources: Mali PIMA 2018; Mali Finance Law 2020.

^{Table 6.2.1.}

Medium-Term Capital Budget in Mali, 2020–22

(Millions of African financial community francs)

Three-year investment plan, 2020–22	2020	2021	2022	Total
Total	643,104	646,448	412,229	1,701,781
Including internal financing	293,104	316,491	251,376	860,971
Including external financing	350,000	329,957	160,853	840,810

Sources: Mali PIMA 2018; Mali Finance Law 2020.

View Table

^{Table 6.2.1.}

Medium-Term Capital Budget in Mali, 2020–22

(Millions of African financial community francs)

Three-year investment plan, 2020–22	2020	2021	2022	Total
Total	643,104	646,448	412,229	1,701,781
Including internal financing	293,104	316,491	251,376	860,971
Including external financing	350,000	329,957	160,853	840,810

Sources: Mali PIMA 2018; Mali Finance Law 2020.

Mali’s multiyear capital expenditure programming system is largely consistent. The multiyear budgeting mechanism secures the allocation of resources for public investments. The programming tools (DPEBP and PTI) are consistent with the annual budget. The mechanism to carry funds forward is used as follows: systematically for externally financed expenditure or subject to a government’s decision for domestically financed expenditure. Progress made in multiyear budget management has been further consolidated by the gradual implementation of commitment control through commitment authorizations and payment allocations, starting in 2018.

Ceilings on total spending by ministry, sector, or program do not qualify as a capital spending ceiling under this dimension. The ceiling should be specifically for capital spending. If the ministries are only provided with ceilings on total spending, including current as well as capital, the score should be low.

There is no requirement that capital spending ceilings be published. Budget ceilings are primarily a tool for internal management of the budget preparation process and are confidential in many cases. Although Dimension 6.a indicates that it is useful to publish projections for aggregate capital spending, there is no similar requirement for the specific budget ceilings.

Important Documents

Documents	Uses
Annual budget call	Analyze whether budget call provides multiyear ceilings and in which form
Other document specifying capital budget ceilings	Analyze whether other documents provide multiyear ceilings and in which form

View Table

Documents	Uses
Annual budget call	Analyze whether budget call provides multiyear ceilings and in which form
Other document specifying capital budget ceilings	Analyze whether other documents provide multiyear ceilings and in which form

Effectiveness

If ministry or sector capital spending ceilings are issued but the approved budget amounts are significantly different, then the ceilings do not have the desired effect. This applies for both binding ceilings (design score high) and indicative ceilings (design score medium). The deviation should be measured as the deviation between total capital ceilings and approved capital budget or as a weighted average across ministries or sectors. The result should be the same. It should be measured over the past three years, if data are available. If data for ceilings in previous years are missing, the assessment can be made on the basis of the ceilings for the last approved budget.

• Low effectiveness indicates that approved capital budgets are significantly higher than the ceiling.

• Medium effectiveness implies that approved capital budgets are somewhat higher than the ceiling.

• High effectiveness indicates that approved capital budgets are broadly in line with the ceiling. Governments may use other methods than formal budget ceilings to constrain capital spending requests. For example, ministries may be informed that they cannot request more than a certain percentage of the capital budget they received in the current year, or that they must include in their request certain projects but not others. If there is written guidance from a central authority within the executive (for example, MoF, Ministry of Economy, or cabinet) that clearly constrains a ministry’s capital budget proposal, this may be considered as an implicit budget ceiling. The effectiveness of such constraints should be assessed by comparing the implicit ceiling imposed by the constraint with actual capital spending, in the same way as when there are formal budget ceilings. If the constraint on capital spending can be formulated as an implicit budget ceiling, and this implicit ceiling has proven to be effective, then this can justify changing the effectiveness score on this dimension as compared with the design score.

Useful Data Series

Data	Questions to Address
Compare ministry or sector ceilings for capital spending with capital spending amounts in the approved budget for 3 years	What is the effectiveness of the capital ceilings?

View Table

Data	Questions to Address
Compare ministry or sector ceilings for capital spending with capital spending amounts in the approved budget for 3 years	What is the effectiveness of the capital ceilings?

Questionnaire

Low	Projections of the total construction cost of major capital projects are not published.
Medium	Projections of the total construction cost of major capital projects are published.
High	Projections of the total construction cost of major capital projects are published, together with the annual breakdown of these costs over a 3- to 5-year horizon.

View Table

Low	Projections of the total construction cost of major capital projects are not published.
Medium	Projections of the total construction cost of major capital projects are published.
High	Projections of the total construction cost of major capital projects are published, together with the annual breakdown of these costs over a 3- to 5-year horizon.

Definitions of Key Terms

Term	Definition
Total construction cost	The total amount required to plan, prepare, design, construct, and hand over a public investment project. Equivalent to total project cost (see Dimension 8.a).
Publish	See the Glossary.
Annual breakdown	The amount of spending expected each year over the total period of construction.

View Table

Term	Definition
Total construction cost	The total amount required to plan, prepare, design, construct, and hand over a public investment project. Equivalent to total project cost (see Dimension 8.a).
Publish	See the Glossary.
Annual breakdown	The amount of spending expected each year over the total period of construction.

Institutional Design

Total cost estimates are necessary to ensure funding and to assess the performance of major public investment projects. Because major projects are implemented over more than one year whereas appropriations typically are made annually, publication of the total project cost is necessary to identify financing needed beyond the budget year, including adjustments during project implementation (see Dimension 14.b) and cost overruns. The total project or construction cost for a project equals the sum of actual expenditures through the prior year, planned spending in the current year, and planned spending beyond the current year required to complete the construction. It should include the cost of feasibility studies and detailed design that may have been supported by separate appropriations. An alternative way to present the total costs of a project would be to publish both the current and the remaining expenditures for the project.

• A low score indicates that there are no published estimates of total construction costs for individual major projects. In these cases, it is not possible to assess whether the funds allocated to a project in the budget are sufficient to ensure efficient project implementation or how long it will take to complete the project. It will not be possible to see how total cost estimates for a project evolve over time and whether there are cost overruns.

• A medium score indicates that the total costs for each major project are estimated and published, but that there is no indication of the distribution of these costs over time. When cost estimates for major projects are published it is possible to assess how these estimates develop and whether estimates increase over time, but it is still difficult to see whether annual project implementation is in line with plans.

• For a high score, the total cost estimates for each major project must be supplemented by indications of how these costs are distributed over time, at least for a three- to five-year horizon. With this information, it will be possible to compare annual and total project costs with available resources over the medium term. It will also be possible to assess annual project implementation compared with plans. However, the information will still be incomplete for projects with an implementation period longer than three–five years.

Total costs of each major project should be published as part of or together with a medium-term expenditure framework (MTEF) or budget documents. This can be done in the MTEF document itself, an annually updated public investment public investment program (PIP), or another suitable document. The information should be available for the purpose of medium-term budgeting. This is different from Dimension 8.a, which asks if information is available to the legislature at the time of budget appropriation.

Important Documents

Documents	Uses
Budget and PIP preparation regulations	Are there regulatory requirements for publication of total projects costs?
Budget documents	Analyze whether total project costs are published in budget documents
Public investment program	Analyze whether total project costs are published in public investment programs

View Table

Documents	Uses
Budget and PIP preparation regulations	Are there regulatory requirements for publication of total projects costs?
Budget documents	Analyze whether total project costs are published in budget documents
Public investment program	Analyze whether total project costs are published in public investment programs

Effectiveness

If cost overruns and other changes in budget estimates are not explicitly identified and explained, the effectiveness of cost estimate publication is reduced. Budget documents should explain when and why cost estimates are changed and provide a reconciliation of such changes over time. This should include the initial cost estimate, the changes over time, and the most recent cost estimate along with short explanations.

• Low effectiveness indicates that changes in total construction costs are not identified and explained. It may be because there are no published estimates of total construction costs, consistent with a low score on institutional design. In addition, it may be that total construction costs are published at some point, for instance, when a project is approved, but that there is no information about subsequent changes to this estimate.

• Medium effectiveness implies that total construction costs are published, and changes in estimates are recorded and explained. These changes may be explained in a published document, such as the annual MTEF or budget documents or a PIP, or in internal government documents.

• High effectiveness implies that total construction costs and the annual breakdown of costs on a three- to five-year horizon are published, and changes from one budget to the next are recorded and explained. These changes should be explained in a published document, such as the annual MTEF or budget documents, or a PIP. Box 6.3 describes the Kiribati Development Fund budget, which provides extensive information about total costs of each major project and discloses any revisions.

^{Box 6.3.}

Capital Spending Development Fund in Kiribati

In Kiribati, Development Fund reports provide useful information on the cost of capital projects over time (Table 6.3.1). The detailed Development Fund table in the budget shows project listings organized by ministry. The total cost of each project is approved at its inception. Full funding of projects allows contracting for multiyear projects, prevents end of year funding pressure, and prevents the need to find funding for ongoing projects. The Development Fund includes both recurrent and capital spending organized by ministry. Recurrent spending in many cases is for discrete noncapital activities that will not continue once they have been completed; projects that would not be included in baseline recurrent expenses.

^{Table 6.3.1.}

Project Information in Development Fund Reports in Kiribati

Source: Government of Kiribati 2018.

^{Table 6.3.1.}

Project Information in Development Fund Reports in Kiribati

Details	Costs and Expenditure Before Current Year	Current Year Information	Budget and Forward Year Information
Implementing ministry	Total Approved Cost of Project	2017 Budget	2018 Budget
Project name	2016 actual	2017 Supplemental Budget	2019 Estimate
Donor (Including the government of Kiribati)	Remaining balance as of January 1, 2017	2017 Revised Budget	2020 Estimate
		2017 Estimated Expenditure	2021 Estimate

Source: Government of Kiribati 2018.

View Table

^{Table 6.3.1.}

Project Information in Development Fund Reports in Kiribati

Details	Costs and Expenditure Before Current Year	Current Year Information	Budget and Forward Year Information
Implementing ministry	Total Approved Cost of Project	2017 Budget	2018 Budget
Project name	2016 actual	2017 Supplemental Budget	2019 Estimate
Donor (Including the government of Kiribati)	Remaining balance as of January 1, 2017	2017 Revised Budget	2020 Estimate
		2017 Estimated Expenditure	2021 Estimate

Source: Government of Kiribati 2018.

Total cost estimates that are revised frequently can still be effective. Looking forward, they give an updated view of the total cost of a project. If information on all amendments to project costs are clearly documented and explained (starting with initial and showing each revision), then cost overruns and delays can easily be seen (see also Dimensions 13a and 14b).

Useful Data Series

Data	Questions to Address
Compare initial cost estimates and final expenditures (aggregate and by year) for selected major projects	What is the change in total project cost during project implementation?
Identify changes in total project costs from one year to the next (that is, rolling updates and the gross amount and percentage of annual change)	How have changes in total project costs occurred over time?

View Table

Data	Questions to Address
Compare initial cost estimates and final expenditures (aggregate and by year) for selected major projects	What is the change in total project cost during project implementation?
Identify changes in total project costs from one year to the next (that is, rolling updates and the gross amount and percentage of annual change)	How have changes in total project costs occurred over time?

Questionnaire

Low	Significant capital spending is undertaken by extrabudgetary entities with no legislative authorization or disclosure in the budget documentation.
Medium	Significant capital spending is undertaken by extrabudgetary entities, but with legislative authorization and disclosure in the budget documentation.
High	Little or no capital spending is undertaken by extrabudgetary entities.

View Table

Low	Significant capital spending is undertaken by extrabudgetary entities with no legislative authorization or disclosure in the budget documentation.
Medium	Significant capital spending is undertaken by extrabudgetary entities, but with legislative authorization and disclosure in the budget documentation.
High	Little or no capital spending is undertaken by extrabudgetary entities.

Definitions of Key Terms

Term	Definition
Significant capital spending	See the box on recommended scoring criteria.
Extrabudgetary entity (EBE)	See GFSM 2014, paragraph 2.41—”government entities with a separate legal identity and substantial autonomy, including discretion over the volume and composition of their expenditures and a direct source of revenue, such as earmarked taxes. Such entities are often established to carry out specific functions, such as road construction or the nonmarket production of health or education services.”
Legislative authorization	Prior approval by the legislature.
Disclosure	Information is included in budget documents to inform the legislature and the public of ongoing activities, for instance, investments by EBEs, but is not subject to approval by the legislature.
Budget documentation	See the Glossary. Note that budget documentation under this dimension refers to the government budget documentation, not EBE budget documentation.

View Table

Term	Definition
Significant capital spending	See the box on recommended scoring criteria.
Extrabudgetary entity (EBE)	See GFSM 2014, paragraph 2.41—”government entities with a separate legal identity and substantial autonomy, including discretion over the volume and composition of their expenditures and a direct source of revenue, such as earmarked taxes. Such entities are often established to carry out specific functions, such as road construction or the nonmarket production of health or education services.”
Legislative authorization	Prior approval by the legislature.
Disclosure	Information is included in budget documents to inform the legislature and the public of ongoing activities, for instance, investments by EBEs, but is not subject to approval by the legislature.
Budget documentation	See the Glossary. Note that budget documentation under this dimension refers to the government budget documentation, not EBE budget documentation.

Institutional Design

The aim of this dimension is to assess whether selection of projects is made with knowledge of the pool of all possible projects, and therefore if the right projects are selected. Dimension 7.a only applies to investments by EBEs. Public corporations (PCs) and special purpose vehicles for public-private partnerships (PPPs) are usually not defined as EBEs and projects funded by PCs or through PPPs should not be considered here. They are considered in Dimension 7.b. As discussed under Dimension 5.c, there may be uncertainty about which state-owned enterprises (SOEs) are PCs and which are EBEs or government units. If this can be ascertained, the SOEs that are EBEs should be discussed under this dimension. Otherwise, the whole SOE sector should be covered under Dimension 7.b. The assessment should in any case clarify the institutional coverage.

• A low score implies that the legal and regulatory framework allows for significant extrabudgetary public investment and there is no formal requirement for legislative authorization or disclosure. This is the case in many countries where key parts of public infrastructure have been moved out of the central government, for instance, road funds. Significant capital spending by EBEs means that it on average constitutes more than 10 percent of capital spending approved in the budget.

• A medium score also implies that the legal and regulatory framework allows for significant extra-budgetary public investment, but this investment is required to be disclosed in budget documentation and subject to some form of legislative authorization. This authorization can take different forms. It can be a detailed approval of each major project, or it can be a more general endorsement of the EBEs’ investment plans. Disclosure in the budget documentation exists, for the purpose of this institution, only if individual projects are named and their cost indicated, or if projects of similar purpose are presented as a group. Excessive grouping means that there is not enough information to coordinate EBE-funded projects with projects appearing in the central government budget.

• A high score applies when the legal and regulatory framework requires that the volume of extra-budgetary spending is small compared with the regular budget and is authorized or disclosed in the budget. In these cases, there are restrictive rules for the establishment of EBEs or for investment activities by such entities. This dimension may be concerned with organizations other than regular EBEs. Some countries use extrabudgetary financial institutions, such as development banks or special funds, to finance projects that could be included in the government capital budget. Such institutions and their public investment activities should be included in the assessment of this dimension as long as they fall under the GFSM 2014 definition of an EBE.

Important Documents

Documents	Uses
Legal framework for EBEs	Assess definitions of EBEs and requirements for budget approval and reporting
Consolidated overview of EBE activities, including in budget documentation	Assess size of EBE investment activities
Annual reports for major EBEs (If no consolidated overview)	Assess size of EBE investment activities

View Table

Documents	Uses
Legal framework for EBEs	Assess definitions of EBEs and requirements for budget approval and reporting
Consolidated overview of EBE activities, including in budget documentation	Assess size of EBE investment activities
Annual reports for major EBEs (If no consolidated overview)	Assess size of EBE investment activities

Effectiveness

The effectiveness assessment of this dimension should reflect whether the formal rules for capital budget approval and disclosure are applied in practice. If the regulatory framework for EBEs requires that their investments be authorized by, endorsed by, or disclosed to the legislature but this is not done in practice, the effectiveness score could be lower than the design score. Alternatively, if there is a well-established practice to disclose EBE capital projects in budget documents although this is not legally required, effectiveness could be higher than the design score indicates.

• Low effectiveness indicates that there is significant extrabudgetary public investment without legislative authorization or disclosure in the budget. This might be because there are no formal requirements, as indicated by a low score on institutional design. Or there may be formal requirements that are not applied or followed in practice.

• Medium effectiveness implies that even if extra-budgetary capital spending is significant, actual budget authorization or disclosure covers most EBE capital spending. This score may be higher than or the same as the score on institutional design.

• High effectiveness indicates that there is little investment spending by EBEs and that mos t of this spending is authorized in the budget or disclosed in budget documents. This would generally coincide with a high score on institutional design. Box 6.4 describes the high level of capital budget comprehensiveness in Armenia, including little extrabudgetary spending.

Useful Data Series

Data	Questions to Address
Number and size of EBEs	What is the relative importance of EBEs in the public sector?
Amount of EBE capital spending on economic infrastructure	What is the role of EBEs in the provision of public infrastructure?

View Table

Data	Questions to Address
Number and size of EBEs	What is the relative importance of EBEs in the public sector?
Amount of EBE capital spending on economic infrastructure	What is the role of EBEs in the provision of public infrastructure?

Questionnaire

Low	Capital projects are not comprehensively presented in the budget documentation, including PPPs and externally financed and PCs’ projects.
Medium	Most capital projects are included in the budget documentation, but either PPPs, externally financed or PC projects, are missing.
High	All capital projects, regardless of financing sources, are included in the budget documentation.

View Table

Low	Capital projects are not comprehensively presented in the budget documentation, including PPPs and externally financed and PCs’ projects.
Medium	Most capital projects are included in the budget documentation, but either PPPs, externally financed or PC projects, are missing.
High	All capital projects, regardless of financing sources, are included in the budget documentation.

Definitions of Key Terms

Term	Definition
Capital project	See the Glossary
Not comprehensively presented	Fewer than 75 percent of capital projects in each category (by value) are included in the budget documentation.
Budget documentation	See the Glossary
Externally financed	See the Glossary
Most capital projects	More than 75 percent of capital projects in each category (by value).
Financing source	See the Glossary
All capital projects	More than 90 percent of capital projects in each category (by value).

View Table

Term	Definition
Capital project	See the Glossary
Not comprehensively presented	Fewer than 75 percent of capital projects in each category (by value) are included in the budget documentation.
Budget documentation	See the Glossary
Externally financed	See the Glossary
Most capital projects	More than 75 percent of capital projects in each category (by value).
Financing source	See the Glossary
All capital projects	More than 90 percent of capital projects in each category (by value).

Institutional Design

The aim of this dimension is similar to Dimension 7.a: to assess whether projects are selected with the knowledge of all possible projects, and therefore if the right projects are selected. The difference is that Dimension 7.b focuses on financing sources rather than EBEs. The term financing source is used here in a broad sense. It includes PPPs and PC infrastructure investments, as well as government investment financed by external sources. The effects of EBEs on the comprehensiveness of the capital budget are fully addressed in Dimension 7.a. As discussed under Dimension 5.c, there may be uncertainty about which SOEs are PCs and which are EBEs or government units. If this can be ascertained, the SOEs that are EBEs should be discussed under Dimension 7.a. Otherwise, the whole SOE sector should be covered Dimension 7.b. The assessment should in any case clarify the institutional coverage.

• A low score indicates that the legal and regulatory framework requires no budget disclosure of projects financed by any of the three major nonbudgetary financing sources (PCs, PPPs, and external sources). PCs include fully owned as well as partially owned entities. Legal forms may vary—PCs can be statutory bodies as well as joint stock companies. PPPs may also take many different forms (see the discussion under Dimension 5.b). External financing may come from international financial institutions, as well as multilateral and bilateral development partners, and comprise grants, concessional loans, and nonconcessional loans.

• A medium score indicates that most projects are required to be comprehensively presented, but that projects for one of the three financing channels are missing in budget documentation. If no projects are financed by a specific source, that financing source should not be considered in meeting the scoring criteria. For example, if there are no PPPs in a country, then a budget must include either externally financed or PC projects to achieve a medium score.

• A high score indicates that all projects undertaken through the three channels mentioned are required to be included in budget documents:

  • For PPPs, all financing should be shown in the budget, to indicate the size of the project. The budget will usually authorize and control spending for direct expenditures of public funds, for instance, accessibility payments. The budget should show for information purposes private investments related to the PPPs or investments by an independent agency, bank, or fund capitalized with public funds. The budget document should also disclose guarantees and other contingent liabilities related to PPPs, but the assessment of this disclosure is done under dimension 3.c and not here.

  • For PCs, the budget should disclose investment projects related to major infrastructure markets. In most cases, this information should be provided in the budget for information purposes only. The exception would be for countries and PCs for which there are legal requirements that the legislature approve their capital budgets. This is common in many Latin American countries.

  • For externally financed projects, all project expenditures should be shown and authorized in the budget on a gross basis. The inflow of external funds should be shown as revenue (for grants) or as financing (for loans) in the budget documents. If the budget only includes government cofunding of externally financed projects, this does not meet the requirements under this dimension.

Important Documents

Documents	Uses
Legal frameworks for investments by PPPs, externally financed investments, and PC investments	Clarify formal requirements for authorization and disclosure
Summary documents for PPPs, externally financed investments, and PC investments, including in budget documentation	Estimate the value of these investments compared with budget-funded investments
Annual reports for major PCs	Estimate the value of these investments compared with budget-funded investments (if there are no summary PC repor ts)

View Table

Documents	Uses
Legal frameworks for investments by PPPs, externally financed investments, and PC investments	Clarify formal requirements for authorization and disclosure
Summary documents for PPPs, externally financed investments, and PC investments, including in budget documentation	Estimate the value of these investments compared with budget-funded investments
Annual reports for major PCs	Estimate the value of these investments compared with budget-funded investments (if there are no summary PC repor ts)

Effectiveness

The effectiveness of this dimension is measured by the compliance with formal disclosure requirements. If a category of projects is only partially presented in budget documentation, the effectiveness score may be lower than the design score. This assessment requires estimating the total value of projects financed from nonbudgetary sources. This will often have to be done in the field and may require compilation of data from different sources. The estimates should be reconciled with the authorities.

• Low effectiveness means that few projects implemented by PCs, as PPPs or through external financing, are comprehensively presented. This may be because there are no formal requirements for such disclosure, as indicated by a low score on institutional design, or because the requirements are not complied with.

• Medium effectiveness means that most capital projects are disclosed in budget documentation. Projects from one category may be missing, as indicated by a medium design score.

• High effectiveness implies that disclosure is full and comprehensive. Most projects are included from each of the three funding categories. Box 6.5 describes how Timor-Leste budget documents provide a complete picture of public investments regardless of financing source.

If countries meet the disclosure criteria in a formal sense, but the information provided is not transparent or is difficult to interpret, the effectiveness could be lower than the design score. This could be the case if information is highly aggregated and does not provide any information on the major projects financed by the different sources. If the information provided on PPPs is limited, the effectiveness of this disclosure may also be lower.

If the number and value of projects in an omitted category is small, the effectiveness score may be higher than the design score. For instance, if externally financed projects and PC investments are fully presented but there is a single, small PPP in a country that is not included, the effectiveness score may still be high. This would require that the value of the projects in the omitted category be negligible.

^{Box 6.5.}

All Capital Projects Shown in Budget Documents in Timor-Leste

The budget presents a comprehensive picture of capital investments (Table 6.5.1). Budget Book 3A contains detailed information on the projects approved by the Infrastructure Fund, which is an agency under the consolidated fund (CFTL) and reviews all investment projects of $5 million or more. The information is presented by project and by program undertaken by respective line ministry. Budget Book 3A also contains information on loan-financed components of projects. Budget Book 3B covers projects implemented at the subnational government levels. Budget Book 3C on the special economic zones of Oecusse and Atauro presents information on their planned capital investment. Development partner grants are provided in detail in Budget Book 5, although whether they are fully or partially capital in nature is not easily identifiable.

^{Table 6.5.1.}

Budget Expenditures in Timor-Leste, by Sources of Funds, 2018–24

(Millions of US dollars)

Source: Government of Timor-Leste 2020a.

^{Table 6.5.1.}

Budget Expenditures in Timor-Leste, by Sources of Funds, 2018–24

(Millions of US dollars)

	2018 Actual	2019 Budget	2020 Budget	2021 Projected	2022 Projected	2023 Projected	2024 Projected
Combined source budget	1,343.9	1,681.2	2,123.5	2,574.5	2,496.4	2,259.1	2,117.7
Government expenditures, by fund	1,172.7	1,482.0	1,950.0	2,491.6	2,461.1	2,245.4	2,104.0
Consolidated fund (excluding loans)	1,119.0	1,375.0	1,853.1	2,386.5	238.3	2,172.6	2,042.2
Human Capital Development Fund	14.5	20.0	23.9	24.8	25.8	26.9	27.9
Borrowing/loans (disbursements)	39.2	87.0	73.0	80.2	53.9	45.9	33.8
Development partner commitments	171.2	199.2	173.5	82.9	35.3	13.7	13.7

Source: Government of Timor-Leste 2020a.

View Table

^{Table 6.5.1.}

Budget Expenditures in Timor-Leste, by Sources of Funds, 2018–24

(Millions of US dollars)

	2018 Actual	2019 Budget	2020 Budget	2021 Projected	2022 Projected	2023 Projected	2024 Projected
Combined source budget	1,343.9	1,681.2	2,123.5	2,574.5	2,496.4	2,259.1	2,117.7
Government expenditures, by fund	1,172.7	1,482.0	1,950.0	2,491.6	2,461.1	2,245.4	2,104.0
Consolidated fund (excluding loans)	1,119.0	1,375.0	1,853.1	2,386.5	238.3	2,172.6	2,042.2
Human Capital Development Fund	14.5	20.0	23.9	24.8	25.8	26.9	27.9
Borrowing/loans (disbursements)	39.2	87.0	73.0	80.2	53.9	45.9	33.8
Development partner commitments	171.2	199.2	173.5	82.9	35.3	13.7	13.7

Source: Government of Timor-Leste 2020a.

Information on PPP transactions is fully integrated in the budget documentation. Budget Book 1 contains information on the underlying rationale for PPPs and a detailed description of approved and planned projects.

Useful Data Series

Data	Questions to Address
Total financing by source of capital spending within the central budget	Are externally funded projects, PPPs, and PC projects included in the budget?
Total financing by source of capital spending outside the central budget	What is the volume of externally funded projects, PPPs, and PC projects outside the budget? How much of this spending is disclosed in budget documents?

View Table

Data	Questions to Address
Total financing by source of capital spending within the central budget	Are externally funded projects, PPPs, and PC projects included in the budget?
Total financing by source of capital spending outside the central budget	What is the volume of externally funded projects, PPPs, and PC projects outside the budget? How much of this spending is disclosed in budget documents?

Questionnaire

Low	There is no standard methodology for determining the needs for routine maintenance.
Medium	There is a standard methodology for determining the needs for routine maintenance and its cost.
High	There is a standard methodology for determining the needs for routine maintenance and its cost, and the appropriate amounts are generally allocated in the budget.

View Table

Low	There is no standard methodology for determining the needs for routine maintenance.
Medium	There is a standard methodology for determining the needs for routine maintenance and its cost.
High	There is a standard methodology for determining the needs for routine maintenance and its cost, and the appropriate amounts are generally allocated in the budget.

Definition of Key Terms

Term	Definition
Standard methodology	Methods, rules, or procedures that consistently guide the actions of persons or organizational units. Must be written and issued or designated by an appropriate government body.
Routine maintenance	Typically, cosmetic and preventive maintenance that is performed regularly to keep an asset operating according to its design. Funding needs are usually calculated using the age and number of assets in an asset class, not project by project. In other words, budgeting for routine maintenance can be estimated with a formula, rather than on the basis of asset-by-asset inspection. Maintenance funded through the operating budget is usually routine maintenance. See GFSM 2014 paragraph 8.25–8.27.
Generally allocated	Allocated within an acceptable range over time.

View Table

Term	Definition
Standard methodology	Methods, rules, or procedures that consistently guide the actions of persons or organizational units. Must be written and issued or designated by an appropriate government body.
Routine maintenance	Typically, cosmetic and preventive maintenance that is performed regularly to keep an asset operating according to its design. Funding needs are usually calculated using the age and number of assets in an asset class, not project by project. In other words, budgeting for routine maintenance can be estimated with a formula, rather than on the basis of asset-by-asset inspection. Maintenance funded through the operating budget is usually routine maintenance. See GFSM 2014 paragraph 8.25–8.27.
Generally allocated	Allocated within an acceptable range over time.

Institutional Design

The aim of this dimension is to determine if there are methodologies to assure that the funding needs for routine maintenance are known. This is essential information to know if infrastructure asset values and service levels are to be retained over time.

• A low score on this dimension implies that there are no standardized methodologies for assessing current maintenance needs. Maintenance planning and allocations will then of ten be based on mechanical approaches, such as continuing the funding levels from last year’s budget. In some cases, maintenance allocations might be residual, reflecting what remains within the overall budget ceiling after, higher-priority programs have received their allocations.

• A medium score on institutional design indicates that there are standard methodologies used for assessing maintenance needs. Standard methodologies will often vary based on type of asset to be maintained, and standard methodologies may not be developed for small asset classes. The methodology should describe the asset class to which it should be applied. For the purpose of this dimension, standard methodologies for routine maintenance exist if they address at least buildings and roads and are consistently applied to those asset classes.

• A high score on design implies that standard methodologies exist and that there is a formal requirement that budget submissions are based on these methodologies. The responsible ministries and agencies are required to provide an assessment of maintenance needs based on the relevant methodologies when submitting their budget proposals to the MoF. On the other hand, the MoF will be expected to prioritize the funding of these maintenance needs within the available fiscal envelopes.

Important Documents

Documents	Uses
Regulations regarding maintenance of infrastructure assets	Assess overall requirements for current maintenance
Maintenance guidelines— general and for main asset classes	Assess detailed methodologies for current maintenance

View Table

Documents	Uses
Regulations regarding maintenance of infrastructure assets	Assess overall requirements for current maintenance
Maintenance guidelines— general and for main asset classes	Assess detailed methodologies for current maintenance

Effectiveness

It is difficult to know the appropriate Amount of funding without a standard methodology. Countries often use ad hoc approaches and rules of thumb for assessing maintenance needs and allocating resources to maintenance. Without a consistent methodology, it is not possible to determine whether the priorities and funding needs estimated through these ad hoc approaches are fully appropriate and adequate.

However, maintenance funding levels compared with asset replacement values do provide a rough indicator for the effectiveness of ad hoc maintenance approaches. Annual maintenance cost needs vary with assets classes, but a minimum level across several different asset classes is at least in the range of 2–3 percent of replacement values, and often more. If there are data available for actual maintenance levels compared with asset replacement values for main asset classes, effectiveness can to some extent be assessed, even in the absence of systematic methodologies. In the absence of comprehensive data, the assessment should be based on maintenance of buildings and roads, which usually are important government assets in all countries.

In the absence of comprehensive asset registers and balance sheets, asset replacement values can be estimated from standard construction costs. Table 6.1 provides an overview of average costs for road construction in different countries, taken from an extensive international study. These can be used as a basis for discussion with relevant country authorities about asset replacement values and maintenance requirements in their country. Most countries will have some information on the volume of assets that can be used for this discussion (for example, square meters of building, kilometers of roads).

• Low effectiveness indicates that maintenance levels are clearly inadequate to retain asset values. If standard methodologies are systematically used, actual maintenance allocations that only provide a small share of estimated needs would imply a low score on effectiveness.

• Medium effectiveness indicates that maintenance levels are better matched to the funding needs to retain asset values. If standard methodologies are systematically used, actual maintenance allocations of a significant share of estimated needs would imply a medium score on effectiveness.

• High effectiveness indicates that maintenance levels are clearly adequate to retain asset values. If standard methodologies are systematically used, actual maintenance allocations should be in line with estimated needs. This requires that there is a standard methodology for assessing maintenance needs, as indicated by a medium or high score on institutional design. Box 6.9 provides an example of maintenance practices in Estonia. Because maintenance is based on standard methodologies and funding is broadly in line with the assessed needs, Estonia had a high score on both institutional design and effectiveness for this dimension in the 2019 PIMA .

^{Table 6.1.}

Standard Construction and Maintenance Costs for Roads, Various Countries (Thousands of US dollars per kilometer)

Source: Collier, Kirchberger, and Söderbom 2015. Note: Updated information is available from the Road Cost Knowledge System at https://www.doingbusiness.org/en/reports/ thematic-reports/road-costs-knowledge-system.

^{Table 6.1.}

Standard Construction and Maintenance Costs for Roads, Various Countries (Thousands of US dollars per kilometer)

		N	Mean	Minimum	Maximum
Construction
	New 6-lane expressway	1	5,571	5,571	5,571
	New 4-lane expressway	65	2,839	937	7,810
	New 4-lane highway	11	2,196	660	4,561
	New 6-lane highway	2	1,990	1,289	2,691
	Widening and reconstruction	108	874	178	6,533
	Widening	138	843	9	5,786
	New 2-lane highway	68	750	22	1,986
	Partial widening and reconstruction	117	261	8	682
	Upgrading	360	250	4	941
	Partial widening	12	138	67	168
	New 1-lane road	7	92	58	168
	Total (Average)	889	678	4	7,810
Preservation
	Concrete pavement restoration	4	539	69	788
	Reconstruction	745	220	2	2,616
	Strengthening	422	139	27	554
	Asphalt mix resurfacing	458	65	12	211
	Surface treatment resurfacing	230	25	3	17
	Gravel resurfacing	275	18	2	113
	Bituminous pavement preventive treatment	47	7	1	31
	Unsealed preventive treatment	101	4	2	8
	Routine maintenance	119	2	0.3	9
	Grading	23	0.5	0.1	3
	Total (average)	2,424	110	0.1	2,616

Source: Collier, Kirchberger, and Söderbom 2015. Note: Updated information is available from the Road Cost Knowledge System at https://www.doingbusiness.org/en/reports/ thematic-reports/road-costs-knowledge-system.

View Table

^{Table 6.1.}

Standard Construction and Maintenance Costs for Roads, Various Countries (Thousands of US dollars per kilometer)

		N	Mean	Minimum	Maximum
Construction
	New 6-lane expressway	1	5,571	5,571	5,571
	New 4-lane expressway	65	2,839	937	7,810
	New 4-lane highway	11	2,196	660	4,561
	New 6-lane highway	2	1,990	1,289	2,691
	Widening and reconstruction	108	874	178	6,533
	Widening	138	843	9	5,786
	New 2-lane highway	68	750	22	1,986
	Partial widening and reconstruction	117	261	8	682
	Upgrading	360	250	4	941
	Partial widening	12	138	67	168
	New 1-lane road	7	92	58	168
	Total (Average)	889	678	4	7,810
Preservation
	Concrete pavement restoration	4	539	69	788
	Reconstruction	745	220	2	2,616
	Strengthening	422	139	27	554
	Asphalt mix resurfacing	458	65	12	211
	Surface treatment resurfacing	230	25	3	17
	Gravel resurfacing	275	18	2	113
	Bituminous pavement preventive treatment	47	7	1	31
	Unsealed preventive treatment	101	4	2	8
	Routine maintenance	119	2	0.3	9
	Grading	23	0.5	0.1	3
	Total (average)	2,424	110	0.1	2,616

Source: Collier, Kirchberger, and Söderbom 2015. Note: Updated information is available from the Road Cost Knowledge System at https://www.doingbusiness.org/en/reports/ thematic-reports/road-costs-knowledge-system.

^{Box 6.9.}

Maintenance Practices in Estonia

In Estonia, routine and capital maintenance needs are determined on the basis of sector-appropriate methodologies and systematic physical monitoring of the infrastructure (Table 6.9.1). For example, the Road Administration maintains a database of 16,600 km of national roads, and physical road condition inspections as well as electronic testing methods are conducted at regular intervals to determine maintenance requirements. Planned service levels have been set, and each road has been categorized accordingly. Estimates are comprehensive and include reconstruction and maintenance of road surfaces, lighting, pedestrian walkways and bridges, reconstruction of hazardous areas, and other items, such as road furniture and road markings. All costs for routine and capital maintenance are calculated per item as per object type in a standardized template. Maintenance of public assets is seen as a high priority and prioritized over new construction. Capital project allocations will be reduced, if the capital maintenance needs cannot be met. Seventeen road maintenance contracts have been awarded to conduct routine maintenance on the road network to maintain service levels. For Estonian Railways, five out of seven major projects are dedicated to maintenance and renovations, to preserve the network and to modernize the network control system.

^{Table 6.9.1.}

Budget Allocation for Road Maintenance in Estonia

(Thousands of euros)

^{Table 6.9.1.}

Budget Allocation for Road Maintenance in Estonia

(Thousands of euros)

Funding	2018	2019	2020	2021	2022
State funding	242,784	253,259	249,260	209,260	209,260
External funding	45,821	46,750	65,610	50,000	10,000
Total funding	288,605	300,009	314,870	259,260	219,260
Road network preservation	159,451	143,349	148,364	140,248	163,059
Development of road network	101,949	128,956	138,152	90,131	26,782
Administration	27,205	27,704	28,354	28,882	29,419
Road management works	288,605	300,009	314,870	259,260	219,260
Road maintenance cost as percentage of total budget	55.2	47.8	47.1	54.1	74.4

View Table

^{Table 6.9.1.}

Budget Allocation for Road Maintenance in Estonia

(Thousands of euros)

Funding	2018	2019	2020	2021	2022
State funding	242,784	253,259	249,260	209,260	209,260
External funding	45,821	46,750	65,610	50,000	10,000
Total funding	288,605	300,009	314,870	259,260	219,260
Road network preservation	159,451	143,349	148,364	140,248	163,059
Development of road network	101,949	128,956	138,152	90,131	26,782
Administration	27,205	27,704	28,354	28,882	29,419
Road management works	288,605	300,009	314,870	259,260	219,260
Road maintenance cost as percentage of total budget	55.2	47.8	47.1	54.1	74.4

Source: Estonia PIMA 2019.

Useful Data Series

Data	Questions to Address
Estimated funding need for routine maintenance, based on standard methodology Funding for routine maintenance included in budgets	How do funding needs compare with actual allocations?
Estimates for asset replacement values for main asset classes	What is the ratio of maintenance funding levels compared with asset replacement values (simplified indicator of effectiveness)?

View Table

Data	Questions to Address
Estimated funding need for routine maintenance, based on standard methodology Funding for routine maintenance included in budgets	How do funding needs compare with actual allocations?
Estimates for asset replacement values for main asset classes	What is the ratio of maintenance funding levels compared with asset replacement values (simplified indicator of effectiveness)?

Questionnaire

Low	There is no standard methodology for determining major improvements, and they are not included in national or sectoral plans.
Medium	There is a standard methodology for determining major improvements, but they are not included in national or sectoral plans.
High	There is a standard methodology for determining major improvements, and they are included in national or sectoral plans.

View Table

Low	There is no standard methodology for determining major improvements, and they are not included in national or sectoral plans.
Medium	There is a standard methodology for determining major improvements, but they are not included in national or sectoral plans.
High	There is a standard methodology for determining major improvements, and they are included in national or sectoral plans.

Definitions of Key Terms

Term	Definition
Standard methodology	See definition in Dimension 9.a.
Major improvement	Capital maintenance (renovations, reconstructions, and enlargements) to existing assets that increase their productive capacity, extend their service lives, or both. Maintenance funded through the capital budget will generally be considered capital maintenance. See GFSM 2014 paragraph 8.25–8.27.
National or sectoral plans	See Dimension 2.a.

View Table

Term	Definition
Standard methodology	See definition in Dimension 9.a.
Major improvement	Capital maintenance (renovations, reconstructions, and enlargements) to existing assets that increase their productive capacity, extend their service lives, or both. Maintenance funded through the capital budget will generally be considered capital maintenance. See GFSM 2014 paragraph 8.25–8.27.
National or sectoral plans	See Dimension 2.a.

Institutional Design

The aim of this dimension is to determine whether the funding needs for capital maintenance are assessed and funding for capital maintenance is evaluated against other project proposals in national or sectoral plans. Capital maintenance is often necessary to enable an asset to reach its planned life. Without periodic rehabilitation, for instance, every 10 years, the asset may have to be phased out earlier than planned. Capital maintenance also includes activities to enable an asset to extend its life and to increase its capacity. This is often referred to as reconstruction. Rehabilitation and reconstruction activities typically are funded through the capital budget and are collectively referred to as capital maintenance.

• A low score implies that there is no systematic use of standardized methodologies for assessing capital maintenance needs. There may be some ad hoc assessment of rehabilitation and reconstruction needs, but often there is no analysis at all. When the needs become sufficiently pressing, rehabilitation or reconstruction of key assets, such as major roads, may be defined as separate, new projects and funded as any other new project.

• A medium score indicates that there is a standard methodology that is used for assessing capital maintenance needs, but these assessments are not reflected in national or sectoral plans. Methodologies will differ across sectors and may not cover all asset classes. For the purpose of this score, there should be standard methodologies for capital maintenance at least for roads and buildings.

• A high score implies that standard methodologies exist and that the capital maintenance needs are fully reflected in national and sectoral plans. It is preferable that capital maintenance projects are included in national or sector plans and are subject to appraisal.

Important Documents

Documents	Uses
Regulations regarding maintenance of infrastructure assets	Assess overall requirements for capital maintenance
Maintenance guidelines—general and for main asset classes	Determine which methodologies are used to assess capital maintenance needs
National and sectoral plans	Identify how capital maintenance projects are reflected in national and sectoral plans

View Table

Documents	Uses
Regulations regarding maintenance of infrastructure assets	Assess overall requirements for capital maintenance
Maintenance guidelines—general and for main asset classes	Determine which methodologies are used to assess capital maintenance needs
National and sectoral plans	Identify how capital maintenance projects are reflected in national and sectoral plans

Effectiveness

The effectiveness of this dimension depends on whether the assessed needs for capital maintenance are reflected in budget allocations. The needs for capital maintenance should be reflected in national and sectoral plans, but the actual prioritization will be determined during budget selection. This assessment is difficult when there are no standardized methodologies or there is a lack of data on capital stocks. Average annual allocations for capital maintenance vary with asset classes, but a minimum level across several different asset classes are at least in the range of 2–3 percent of replacement values, and often more (see Box 6.10 for South Africa’s guidelines for adequate maintenance of public infrastructure). See the discussion under Dimension 9.a for suggestions on how to make rough estimates for capital replacement values.

• Low effectiveness indicates that capital maintenance levels are clearly inadequate to retain asset values. If actual capital maintenance allocations only provide a small share of estimated needs, effectiveness is low.

• Medium effectiveness indicates that capital maintenance levels are better matched to the funding needs to retain asset values. If actual allocations for capital maintenance provide a significant share of estimated needs, this would imply a medium score on effectiveness.

• High effectiveness indicates that capital maintenance levels are clearly adequate to retain asset values. Actual maintenance allocations should be in line with estimated needs. This requires that there is a standard methodology for assessing maintenance needs, as indicated by a medium or high score on institutional design.

^{Box 6.10.}

Maintenance Guidelines for Public Infrastructure in South Africa

Many countries have established guidelines and standards for maintenance of public infrastructure, to ensure that maintenance levels are sufficient to avoid deterioration of public asset values. In South Africa, the cabinet approved the National Infrastructure Maintenance Strategy in 2007. The Department of Public Works and Infrastructure has a lead role in implementing the strategy, which includes indicative budget requirements for maintenance of different types of assets (Table 6.10.1).

^{Table 6.10.1.}

South Africa Maintenance Budget Guidelines

^{Table 6.10.1.}

South Africa Maintenance Budget Guidelines

Type of Infrastructure	Average Annual Maintenance Budget as a Percentage of Replacement Cost	Replacement or Major Rehabilitation Over and Above the Annual Maintenance Budget Requiring Specific Capital Budget
		Every 20–30 years	Every 30–50 years
Bulk water storage	4–8		×
Water treatment works	4–8	×
Water reservoirs	2–3	×
Water reticulation	4–8	×
Sewage treatment works	4–8	×
Sewer reticulation	4–8	×
Roads and storm water	5–10	×
Electricity reticulation	10–15	×
Public buildings	4–6		×
Hospitals	5–8	×
Schools	4–6		×
Electricity generation	5–8		×
Electricity reticulation	10–15	×

View Table

^{Table 6.10.1.}

South Africa Maintenance Budget Guidelines

Type of Infrastructure	Average Annual Maintenance Budget as a Percentage of Replacement Cost	Replacement or Major Rehabilitation Over and Above the Annual Maintenance Budget Requiring Specific Capital Budget
		Every 20–30 years	Every 30–50 years
Bulk water storage	4–8		×
Water treatment works	4–8	×
Water reservoirs	2–3	×
Water reticulation	4–8	×
Sewage treatment works	4–8	×
Sewer reticulation	4–8	×
Roads and storm water	5–10	×
Electricity reticulation	10–15	×
Public buildings	4–6		×
Hospitals	5–8	×
Schools	4–6		×
Electricity generation	5–8		×
Electricity reticulation	10–15	×

Source: Construction Industry Development Board 2021.

Useful Data Series

Data	Questions to Address
Estimated funding need for capital maintenance, based on standard methodology Funding estimates for capital maintenance in national or sectoral plans	How do funding needs based on standard methodologies compare with estimates included in national and sectoral plans?
Funding for capital maintenance in budgets	How do capital maintenance funding needs compare with actual allocations?

View Table

Data	Questions to Address
Estimated funding need for capital maintenance, based on standard methodology Funding estimates for capital maintenance in national or sectoral plans	How do funding needs based on standard methodologies compare with estimates included in national and sectoral plans?
Funding for capital maintenance in budgets	How do capital maintenance funding needs compare with actual allocations?

Questionnaire

Low	Routine maintenance and major improvements are not systematically identified in the budget.
Medium	Routine maintenance and major improvements are systematically identified in the budget.
High	Routine maintenance and major improvements are systematically identified in the budget and are reported.

View Table

Low	Routine maintenance and major improvements are not systematically identified in the budget.
Medium	Routine maintenance and major improvements are systematically identified in the budget.
High	Routine maintenance and major improvements are systematically identified in the budget and are reported.

Definitions of Key Terms

Term	Definition
Routine maintenance	See Dimension 9.a.
Major improvement	See Dimension 9.b.
Reported	Approved budget allocations and actual spending are included in standard reports as part of the budget documents.

View Table

Term	Definition
Routine maintenance	See Dimension 9.a.
Major improvement	See Dimension 9.b.
Reported	Approved budget allocations and actual spending are included in standard reports as part of the budget documents.

Institutional Design

The aim of this dimension is to determine whether budget allocations and spending on maintenance are systematically disclosed in budget documents in a consistent manner over time. This information is necessary for the legislature to have a view on the medium- and long-term adequacy of maintenance allocations.

• A low score indicates that it is not possible to identify either routine maintenance or capital maintenance in the budget documents. In some cases, maintenance is combined with other spending items and does not appear as a separate category in the budget classification. In other cases, maintenance might be identified under some ministries or programs but not under others, because there is no common budget classification for this type of expenditure. Maintenance might appear at different levels of the economic or program classification.

• A medium score implies that maintenance funding can be identified using either the budget classification or the analytical information regularly provided in budget documentation. The program or activity classification, if available, is best suited to identifying routine maintenance. If at least some routine maintenance is carried out by government employees, the economic classification should not be used because such maintenance requires spending on wages, equipment, and material. Grouping routine maintenance expenditures into one economic classification item distorts reporting on other economic classification items. However, capital maintenance can be identified using the economic classification. Capital projects are sometimes considered activities, if that classification exists.

• For a high score, both routine maintenance and capital maintenance must be systematically identified and regularly reported. At a minimum, there must be standard reports that show the approved budget allocations and the actual spending for routine and capital maintenance by ministry or agency responsible for it. Preferably, reports should aggregate such spending to show total spending on current maintenance and capital maintenance in the budget. Additional detail, such as maintenance funding by program or by asset class, is desirable. Budget documentation frequently includes analysis of the capital budget by type and geographic location of projects, and this could also include capital maintenance.

Important Documents

Documents	Uses
Budget classification	Analyze how maintenance is reflected in the budget classification
Budget documents	Assess disclosure and reporting of maintenance budgets and spending

View Table

Documents	Uses
Budget classification	Analyze how maintenance is reflected in the budget classification
Budget documents	Assess disclosure and reporting of maintenance budgets and spending

Effectiveness

The effectiveness of this dimension should reflect what share of maintenance spending is transparently disclosed and reported in budget documents. In many cases, maintenance spending is identified in the budget but not systematically, and there is no explicit reporting. To get a picture of overall maintenance spending it may be necessary to compile data from different parts of the detailed budget through manual, ad hoc methods. There may also be uncertainty about whether all maintenance data are fully reflected. In these cases, effectiveness may be low even if the design score is medium. The assessment could focus on maintenance of buildings and roads, which presumably would constitute a significant share.

The effectiveness also depends on whether maintenance data are used systematically for analysis and for decision-making. If maintenance reporting is merely a technical routine exercise with no analysis and no follow-up, effectiveness may be only medium even if the design score is high. If external audits regularly address needs and spending on maintenance, this does not meet the intent of this dimension. The auditor general is usually independent, and the nature and subject of audits can change at any time. Thus, audits do not systematically identify maintenance budgets or expenditures.

• Low effectiveness indicates that maintenance data are not transparent and not actively used for analysis or decisions. Only part of estimated maintenance funding is identified in the budget. There are no specific examples that the data are used during planning or budgeting.

• Medium effectiveness indicates that maintenance data are reasonably transparent and there are some examples of analysis or decisions based on these data. Most estimated maintenance funding is identified in the budget. Budget documents could include analysis that has led to adjustments in maintenance funding.

• High effectiveness indicates that maintenance data are transparent and used actively and systematically for analysis and decision-making. Most estimated maintenance funding is identified in the budget and there are regular published reports of budget allocations compared with actual spending, by ministry. Budget documents may also provide an overview of actual maintenance spending compared with target levels and outline how the government will bridge this gap. Box 6.11 illustrates the transparent reporting of maintenance expenditures in Armenia’s budget documents.

^{Box 6.11.}

Budget Visibility of Maintenance Spending in Armenia

Expenditures related to routine maintenance and major works are visible in the state budget and consistent with GFSM 2014 definitions on page 124, paragraph 6.45 (“goods and services consumed for the ordinary maintenance and repair of fixed assets constitute use of goods and services. However, major renovations, reconstructions, or enlargements of exiting fixed assets are recorded as an acquisition of fixed assets.”). There is a budget line for current repairs and maintenance under the goods and services category and a separate line for the capital repairs of buildings and construction under nonfinancial assets. As part of the MTEF process, line ministries submit project details that specify allocations to these budget lines. It is possible to assess execution of these lines in the quarterly and annual budget execution reports. The same report also contains an annex on the execution for routine road maintenance.

Maintenance spending has been well protected during budget execution. Figure 6.11.1 illustrates that all three major components of the maintenance budget have maintained significant levels of execution, suggesting these budget lines are not vulnerable to in-year cuts. The execution rate for capital maintenance (major works) has been the most volatile, with an execution rate of 86 to 89 percent for three of the past five years but an execution rate well over 100 percent in one year. Routine road maintenance averaged 99 percent and routine maintenance for all other sectors averaged an execution rate of 94 percent over the five-year period.

View Full Size

Budget Execution Rates for Maintenance in Armenia, 2013–17 (Percent)

Sources: Armenia budget implementation reports; and IMF staff estimates.

• Download Figure
• Download figure as PowerPoint slide

Useful Data Series

Data	Questions to Address
Summary table for maintenance data reported in budget documents	How transparently are maintenance data reported?
Examples of analysis and subsequent decisions on revised maintenance approaches or funding	How are maintenance data used for analysis and decision-making?

View Table

Data	Questions to Address
Summary table for maintenance data reported in budget documents	How transparently are maintenance data reported?
Examples of analysis and subsequent decisions on revised maintenance approaches or funding	How are maintenance data used for analysis and decision-making?

Questionnaire

Low	Major projects (including those funded by donors or PPPs) are not reviewed by a central ministry before inclusion in the budget.
Medium	Major projects (including those funded by donors or PPPs) are reviewed by a central ministry before inclusion in the budget.
High	All major projects (including those funded by donors or PPPs) are scrutinized by a central ministry, with input from an independent agency or experts before inclusion in the budget.

View Table

Low	Major projects (including those funded by donors or PPPs) are not reviewed by a central ministry before inclusion in the budget.
Medium	Major projects (including those funded by donors or PPPs) are reviewed by a central ministry before inclusion in the budget.
High	All major projects (including those funded by donors or PPPs) are scrutinized by a central ministry, with input from an independent agency or experts before inclusion in the budget.

Definitions of Key Terms

Term	Definition
Major projects	See the Glossary.
Central ministry	Carries out a function that potentially affects many or substantially all ministries. Typical examples are a ministry of finance, a ministry of economy, and a planning ministry or agency. Some countries establish specialized agencies for public investment oversight and support.
Scrutinize	Same meaning as “review.”
Input	Submission of data, comments, analysis, conclusions, or recommendations. This does not imply approval.
Independent	See the Glossary.

View Table

Term	Definition
Major projects	See the Glossary.
Central ministry	Carries out a function that potentially affects many or substantially all ministries. Typical examples are a ministry of finance, a ministry of economy, and a planning ministry or agency. Some countries establish specialized agencies for public investment oversight and support.
Scrutinize	Same meaning as “review.”
Input	Submission of data, comments, analysis, conclusions, or recommendations. This does not imply approval.
Independent	See the Glossary.

Institutional Design

The aim of this dimension is to determine whether there is a required process for reviewing major projects objectively before their selection for inclusion in the budget proposal. As mentioned earlier, the selection process is separate from the appraisal process. Institution 4 focuses on filtering out projects with poor or negative value, whereas institution 10 focuses on prioritizing projects among those previously appraised. Still, there are close links between the two and the assessments of institutions 4 and 10 must be closely coordinated, to ensure consistency.

• A low score on this dimension means that there is no formally required central review process for major capital investment projects before they are considered for inclusion in the budget. This will often mean that projects are selected and submitted by the sector ministry without systematic involvement from other parts of the government. The sector ministry will generally have strong incentives to give as positive a picture of the project as possible, and projects that have not been subject to independent review will often have optimistic assumptions regarding costs, benefits, and timetables.

• A medium score indicates that there is a formally required central review process for major projects. This review will often be undertaken by a specialized department in the MoF or the Ministry of Planning, or by a specialized agency. These entities will be more objective than the ministry promoting the project and have more realistic expectations for the project. There should be a thorough review of the key features of the project, including strategic alignment, project concept, costs, benefits, and implementation plans. The review should be conducted before budget consideration—the general budget process will usually not include a project review of this type.

• A high score implies that there is a formal requirement that major projects are reviewed and that this review include inputs from an independent agency or independent experts. Some countries have set specialized agencies for this purpose, while others rely on independent expertise from technical universities or from private consultants. The use of independent experts adds another layer of objective scrutiny and further promotes the realism of the project proposals.

The review process will often lead to projects being returned to the promoting ministry for further development. The reviewers may raise specific questions regarding the strategic alignment of the project with government priorities, whether the specific concept that has been proposed is best suited to meet project objectives, and whether project costs, benefits, and implementation plans are realistic and viable. These questions must be addressed before the project can be resubmitted for central review.

The operation of a special mega-project unit, separate from the customary budget office, will usually not constitute central review and scrutiny. This dimension assumes that line ministries propose projects that are reviewed centrally. Central review, then, suggests that a disinterested party conducts the review, and that the review of a single project is made in light of all projects being proposed. If a mega-project unit develops mega-projects and reviews only mega-projects, then it does not review projects centrally for the purposes of this dimension.

Important Documents

Documents	Uses
Regulations for project selection and approval	Verify the existence of required central review of project proposals
Guidelines for project review	Assess the stringency and consistency of methodologies for central project review
Project review documents	Assess the rigor of a representative sample of project reviews

View Table

Documents	Uses
Regulations for project selection and approval	Verify the existence of required central review of project proposals
Guidelines for project review	Assess the stringency and consistency of methodologies for central project review
Project review documents	Assess the rigor of a representative sample of project reviews

Effectiveness

The assessment of effectiveness of this dimension should focus on how well formal review requirements are complied with in practice and the effects of these reviews. The assessment should be based on a representative sample of documents from the major project review process. These should include examples of the initial project proposals, the documents produced by the central ministry conducting the review, inputs from independent experts, and e adjustments to initial project proposals. Project selection effectiveness requires a review process that translates into recommendations for improvement, with some projects returned for further development. Consistent and systematic central review of major projects should foster a culture of rigor in sector ministries, and therefore the number of rejected projects would be expected to be reduced over time.

A high degree of external financing of capital projects will usually not add to the effectiveness of this dimension. The focus is on the government’s review process. Development partners and international financial institutions cannot be considered as independent in the context of this institution. They will typically be involved in promoting specific projects that are aligned with their country or business strategies and will often have interests similar to the ministries promoting the projects.

Some international financial institutions have rigorous internal processes that include independent scrutiny, but this is not always the case. Many development partners and international financial institutions do not have procedures for independent scrutiny of projects.

• Low effectiveness implies that there is no central review of major project proposals or that it is formalistic and cursory, and few projects are rejected or returned for further development. Documentation of the review process is missing or limited. The number of projects rejected or returned is low.

• Medium effectiveness implies that central review has a reasonable level of rigor, and some major projects are rejected and returned. The review process is documented and some projects are rejected or returned.

• High effectiveness implies that there is rigorous central review of major project proposals and proposals are regularly rejected or returned for further development. In this case, the rigor of the review process should be clearly documented in project review documents and at least 10 percent of the reviews should include independent inputs. It would be expected that several project proposals are rejected or returned for further development. Box 6.12 describes the central review process in Ireland.

Useful Data Series

Data	Questions to Address
Number and value of projects submitted for central review Number and value of projects rejected or returned for resubmission (with examples of central review reports recommending further development)	What is the share of project proposals that are rejected or returned? Do central review reports include relevant recommendations for further development of the project?
Number of project reviews that involve external independent inputs	What is the share of project proposals that benefit from external independent inputs?

View Table

Data	Questions to Address
Number and value of projects submitted for central review Number and value of projects rejected or returned for resubmission (with examples of central review reports recommending further development)	What is the share of project proposals that are rejected or returned? Do central review reports include relevant recommendations for further development of the project?
Number of project reviews that involve external independent inputs	What is the share of project proposals that benefit from external independent inputs?

^{Box 6.12.}

Central Review of Project Appraisals in Ireland

The appraisals of all major projects are subject to review by the Irish Department of Public Expenditure and Reform (DPER) (Figure 6.12.1), with input from external experts on an “as-needed” basis. The Public Spending Code requires the sponsoring agency to seek the views of the DPER before the sanctioning authority makes its official “decision in principle” to proceed with the project. The sponsoring agency sends the appraisal to the relevant vote section in the DPER, who then confers with the in-house Irish Government Economic and Evaluation Service (IGEES) team, with whom the department ‘s project analysis capability is situated. Depending on the nature of the project, the DPER will also draw on expert advice from the National Development Finance Agency (NDFA) and New Economy and Recovery Authority (NewERA), when necessary.

View Full Size

Appraisal and Selection Process in Ireland

Source: Ireland PIMA 2017.Note: CBA = cost-benefit analysis.

• Download Figure
• Download figure as PowerPoint slide

There is no formal obligation for the sponsoring agency to take account of the DPER’s views, but it is implicitly understood that not doing so would be detrimental to future requests for funding. As a kind of sanction, the Public Spending Code allows for the DPER to publish its review on the department website, although this lever seems to be rarely used, if at all. The review function is reinforced by the secondment of IGEES staff to departments acting as sanctioning authorities in investment-heavy sectors. These teams—the Economic and Financial Evaluation Unit in the Department of Transport, Tourism and Sport (DTTaS) being the best example—review appraisals using the same criteria as central IGEES staff, thus ensuring consistency of approach and greater objectivity in the sanctioning authorities’ decision making.

Questionnaire

Low	There are no published criteria or a required process for project selection.
Medium	There are published criteria for project selection, but projects can be selected without going through the required process.
High	There are published criteria for project selection, and generally projects are selected through the required process.

View Table

Low	There are no published criteria or a required process for project selection.
Medium	There are published criteria for project selection, but projects can be selected without going through the required process.
High	There are published criteria for project selection, and generally projects are selected through the required process.

Definition of Key Terms

Term	Definition
Publish	See glossary
Required process	A process defined in law, regulation, or instructions governing the budget process. “Process” implies that there are tasks assigned to organization units, with specified outputs, that must be carried out in a specific sequence.
Project selection	The government’s decision to implement a specific investment project.

View Table

Term	Definition
Publish	See glossary
Required process	A process defined in law, regulation, or instructions governing the budget process. “Process” implies that there are tasks assigned to organization units, with specified outputs, that must be carried out in a specific sequence.
Project selection	The government’s decision to implement a specific investment project.

Institutional Design

The aim of this dimension is to verify that there are specific criteria and a well-defined selection process to ensure that projects are selected in an objective and comprehensive manner. Project selection may be done before the budget decision, for instance, through a separate cabinet decision or as part of the budget decision process. The institution focuses on the government’s decision regarding the project, although this may be subject to subsequent endorsement by the legislature.

• With a low score, there are no published, specific criteria for project selection and the project selection process is not explicitly defined in law, regulation, or instructions. In this case, projects are often selected through ad hoc methods and approaches. Project discussions may include references to strategic priorities and goals, but there are no detailed criteria for assessing whether the project contributes to these. The actual selection is often done implicitly through the budget process, without any attention to project benefits and the realism of implementation plans.

• A medium score implies that there are published selection criteria, but that these are general and do not provide clear guidance on which projects should be selected or not. It is common that countries require projects to be consistent with national plans and priorities. However, these documents are often general and do not necessarily provide clear guidance. If all or most projects are deemed to be consistent with the selection criteria, these are not precise. The legal framework may include some exemptions from the standard process and criteria, for instance, for priority projects or emergency projects.

• A high score indicates a stringent selection process defined in law or regulation: there are published selection criteria, and these provide clear guidance on which projects are to be selected. There can be more than one required process. For example, a separate process may be required for mega-projects. The main issue is that there is a defined process for selecting all projects for the budget and that the same selection criteria are applied.

Important Documents

Documents	Uses
Regulations for project selection and approval	Verify the existence and stringency of project selection process and criteria
Project review documents	Assess the application of project selection criteria in practice

View Table

Documents	Uses
Regulations for project selection and approval	Verify the existence and stringency of project selection process and criteria
Project review documents	Assess the application of project selection criteria in practice

Effectiveness

The effectiveness assessment should include a discussion of the actual stringency of the selection process and criteria. If projects are selected outside the required process and without applying the defined selection criteria, effectiveness is limited.

• Low effectiveness implies that actual project selection is not significantly affected by a required selection process and defined selection criteria. These may be missing completely. Alternatively, if there is a required process and criteria but many projects are selected outside the formally required process, effectiveness is low. In some countries, it is not uncommon that some projects are presented and approved at cabinet meetings, without any prior analysis or review by the administration. Unfortunately, these may be large and complex mega-projects, where the needs for stringent project preparation, appraisal, and selection are particularly high.

• Medium effectiveness would indicate that the majority of projects (by value) are selected in accordance with the prescribed process and criteria. In many countries, there are dual processes, whereby some projects go through the prescribed procedures and others are selected through ad hoc approaches.

• High effectiveness requires that the entire capital budget is selected in accordance with the specified criteria and through the required process. Box 6.13 describes the well-regulated project selection process in Mexico.

Useful Data Series

Data	Questions to Address
Number and value of projects that were selected outside published criteria and outside the prescribed selection process	Which share of projects is selected in accordance with prescribed criteria?
Number and value of projects that have been rejected because they did not meet the defined selection criteria	How stringently is project selection criteria applied? Total number and value of projects that have been analyzed using the defined selection criteria/ needed to determine the share of projects selected outside process or rejected.
Total number and value of projects that have been analyzed using the defined selection criteria.	Needed to determine the share of projects selected outside process or rejected.

View Table

Data	Questions to Address
Number and value of projects that were selected outside published criteria and outside the prescribed selection process	Which share of projects is selected in accordance with prescribed criteria?
Number and value of projects that have been rejected because they did not meet the defined selection criteria	How stringently is project selection criteria applied? Total number and value of projects that have been analyzed using the defined selection criteria/ needed to determine the share of projects selected outside process or rejected.
Total number and value of projects that have been analyzed using the defined selection criteria.	Needed to determine the share of projects selected outside process or rejected.

^{Box 6.13.}

Project Selection Process and Criteria in Mexico

The process for selection of public investment projects for the federal budget is governed by Article 34 of the Federal Budget and Fiscal Responsibility Law. The law is supplemented by project selection guidelines. Following the criteria set out in the law, selection is carried out in two phases; the first-level prioritization is for what is called “irreductible investment” (for example, pluri-annual projects, ongoing projects, and maintenance for productive infrastructure, followed by administrative acquisitions and maintenance, and then new projects).

The second level of prioritization (primarily for new or reformulated projects) uses the following criteria: (1) progress on feasibility studies; (2) net present value; (3) regional effects; (4) extensiveness of beneficiaries; and (5) support to the Green Budget initiative. Each project is ranked following a valuation of all projects based on weights assigned to each of these five criteria. The project ranking relative to the total expenditure budget ceiling generates the list of selected projects. This list of projects is reviewed and formally approved by the Inter-ministerial Commission for Public Expenditure, Financing and Disincorporation (Comisión Intersecretarial de Gasto Público Financiamiento y Desincorporación).

The selection of projects for inclusion in the budget follows a clear, criteria-based, and consistent process (Figure 6.13.1). This process is applied to all ministry submitted projects approved by the Ministry of Finance Investment Unit (IU) and registered in the project portfolio (cartera). Additional projects of up to 10 percent may be proposed by Congress for inclusion in the budget. These projects must be registered in the cartera and thus go through the same IU review and selection process required for ministry-submitted projects, including socioeconomic and financial analyses. In principle, the IU can reject poor projects proposed by Congress. Active project selection by the IU is limited to projects funded by the federal budget; other projects funded by nonbudgetary sources (for example, EPE public corporations such as Pemex or CFE, or extrabudgetary sources and from some trust funds are not part of the selection process.

View Full Size

Types of Projects Covered by the Mexican Government’s Project Selection Process

Source: Mexico PIMA 2019.Note: CFE = Federal Electricity Commission; EPE = Empresas Productivas del Estado (category of public corporation, such as PEMEX and CFE); FONADIN = Fondo Nacional de Infraestructura (federal infrastructure fund); IU = Ministry of Finance Investment Unit; Pemex = Mexican Petroleum.

• Download Figure
• Download figure as PowerPoint slide

Questionnaire

Low	The government does not maintain a pipeline of appraised investment projects.
Medium	The government maintains a pipeline of appraised investment projects, but other projects may be selected for financing through the annual budget.
High	The government maintains a comprehensive pipeline of appraised investment projects, which is used to select projects for inclusion in the annual budget, and over the medium term.

View Table

Low	The government does not maintain a pipeline of appraised investment projects.
Medium	The government maintains a pipeline of appraised investment projects, but other projects may be selected for financing through the annual budget.
High	The government maintains a comprehensive pipeline of appraised investment projects, which is used to select projects for inclusion in the annual budget, and over the medium term.

Definitions of Key Terms

Term	Definition
Pipeline of appraised investment projects	A group of projects that have been appraised and through the appraisal have been judged to be eligible for selection for the budget. The pipeline can be an unsequenced list of projects or it can be a list of projects sequenced by year.
Comprehensive pipeline	A pipeline is comprehensive if it includes enough projects, across all types of projects and all financing sources, to fully absorb estimated aggregate funding for the capital budget. Such a pipeline is continuously built and updated, not only to feed the coming budgetary space, but also to prepare for subsequent budgets.
Medium-term	See the Glossary.

View Table

Term	Definition
Pipeline of appraised investment projects	A group of projects that have been appraised and through the appraisal have been judged to be eligible for selection for the budget. The pipeline can be an unsequenced list of projects or it can be a list of projects sequenced by year.
Comprehensive pipeline	A pipeline is comprehensive if it includes enough projects, across all types of projects and all financing sources, to fully absorb estimated aggregate funding for the capital budget. Such a pipeline is continuously built and updated, not only to feed the coming budgetary space, but also to prepare for subsequent budgets.
Medium-term	See the Glossary.

Institutional Design

The aim of this dimension is to determine if there is a pool of appraised projects, to which the selection criteria and process are applied. This pipeline or pool is important to facilitate efficient project selection.

• A low score indicates that there is no formal requirement for a pipeline of appraised investment projects. This may be because there is no pipeline at all, so that investment projects are not identified before the line ministries propose them, often during budget deliberations. Alternatively, there may be a pipeline of projects, but the projects in this pipeline have not been subject to appraisal. In both cases, there is a clear risk that projects are selected for implementation without having gone through the necessary preparation and appraisal.

• A medium score implies that there is a pipeline of appraised projects, but no formal requirement that projects be selected only from this pipeline. The pipeline exists and there is a mechanism in place to facilitate efficient selection from this pipeline, but there is no requirement that this mechanism is always applied.

• For a high score, there must be a comprehensive pipeline in place, and there is a formal requirement that this pipeline must be used to select previously appraised projects in the annual budget and in the medium term. The pipeline should encompass all funding modalities, including externally financed projects and PPPs. Several practices may meet the intent of the condition that the pipeline must be used to select projects in the medium term:

  • There is an MTEF and specific projects are shown for forward years, even on an indicative basis, or

  • There is an annual budget process and budget documents show for information projects expected to be funded for future years, or

  • There is a multiyear PIP wherein projects are scheduled by year and the total cost of projects for each year is constrained by forward estimates of aggregate funding available for the capital budget.

For the purpose of this dimension, projects in the pipeline must be appraised before selection. It is not sufficient if projects are appraised afterward. If projects are added by cabinet or the legislature and were not previously appraised, it is unlikely that an appraisal would be rigorous or the supplemental selection process objective and comprehensive.

The pipeline could include projects that have been subject to different levels of appraisal. It is common that appraisal guidelines differentiate between small, routine projects that may be subject to simplified appraisal procedures, and large, complex projects that are subject to comprehensive appraisal procedures. See institution 4 for a discussion on this.

Important Documents

Documents	Uses
Regulation for preparation of investment project pipeline (or similar)	Assess institutional design
Project pipeline document	Assess comprehensiveness and coverage of projects in the pipeline

View Table

Documents	Uses
Regulation for preparation of investment project pipeline (or similar)	Assess institutional design
Project pipeline document	Assess comprehensiveness and coverage of projects in the pipeline

Effectiveness

The effectiveness assessment should reflect how consistently projects are selected from the pipeline. If the design score is medium, then there may be projects selected from outside the pipeline. If this is common, then effectiveness is low. If it rarely or never happens, although it is allowed, then effectiveness could be assessed as being high:

• Low effectiveness indicates that many projects are selected from outside the pipeline.

• Medium effectiveness indicates that the majority of projects are selected from the pipeline.

• High effectiveness indicates that all projects are selected from the pipeline. Box 6.14 describes Chile’s integrated project pipeline, which was probably the first fully operational system of this type in the world during the 1980s.

Useful Data Series

Data	Questions to Address
Number and value of projects that have been appraised and are viewed as eligible for selection (i.e., a pipeline) Number and value of projects that were appraised and rejected as being inefficient and thus were not included in the project pipeline	What is the rigor and objectivity of the selection process? Number and value of projects selected for implementation from the pipeline in past 5 years
List of projects selected for implementation from outside the pipeline in past 5 years	What is the effectiveness of this dimension?

View Table

Data	Questions to Address
Number and value of projects that have been appraised and are viewed as eligible for selection (i.e., a pipeline) Number and value of projects that were appraised and rejected as being inefficient and thus were not included in the project pipeline	What is the rigor and objectivity of the selection process? Number and value of projects selected for implementation from the pipeline in past 5 years
List of projects selected for implementation from outside the pipeline in past 5 years	What is the effectiveness of this dimension?

^{Box 6.14.}

Integrated Project Pipeline in Chile

Chile is a pioneer in the development of public investment management systems. The Chilean System of National Investment was developed during the 1980s and has been a model for similar systems in many countries. It includes a framework for identification and development of project proposals, with a database of projects that are available for funding decisions (Integrated Project Bank-BIP). The BIP is available online for all public institutions, and the public can access summary information about specific projects and programs.

Projects are subject to extensive analysis before being allowed to enter the BIP. Table 6.14.1 shows the value of projects in different sectors that were included in the project pipeline for each of the years 2008–11, as well as the percentage of proposals that were allowed to enter the pipeline each year (bottom line).

^{Table 6.14.1.}

Cost of Projects in Chile’s Integrated Project Database, by Sector, 2008–11

(Million US dollars)

Source: Gómez-Lobo 2012.

^{Table 6.14.1.}

Cost of Projects in Chile’s Integrated Project Database, by Sector, 2008–11

(Million US dollars)

Sector	2008	2009	2010	2011
Water supply and sewerage	24.5	35.5	13.9	11.8
Communications	1.2	12.6	0.0	0.0
Defense and security	45.8	70.4	63.6	96.0
Sports	7.7	42.9	11.8	38.3
Education and culture	734.0	705.1	582.2	426.8
Energy	0.0	0.6	0.7	0.2
Industry, commerce, finance, and tourism	27.3	20.4	14.3	19.0
Justice	136.9	125.0	94.7	111.8
Mining	1.3	1.2	0.7	0.0
Multisectoral	150.8	140.4	256.6	466.7
Fishing	5.8	5.0	36.2	58.3
Health	95.7	196.3	201.8	205.7
Agricultural and forestry	38.2	29.4	36.9	16.1
Transport	164.9	199.3	182.0	462.1
Housing	85.6	74.9	22.1	7.0
Total	1,519.7	1,659.0	1,517.5	1,919.8
Percentage of approved initiatives	76.8%	74.6%	68.8%	65.4%

Source: Gómez-Lobo 2012.

View Table

^{Table 6.14.1.}

Cost of Projects in Chile’s Integrated Project Database, by Sector, 2008–11

(Million US dollars)

Sector	2008	2009	2010	2011
Water supply and sewerage	24.5	35.5	13.9	11.8
Communications	1.2	12.6	0.0	0.0
Defense and security	45.8	70.4	63.6	96.0
Sports	7.7	42.9	11.8	38.3
Education and culture	734.0	705.1	582.2	426.8
Energy	0.0	0.6	0.7	0.2
Industry, commerce, finance, and tourism	27.3	20.4	14.3	19.0
Justice	136.9	125.0	94.7	111.8
Mining	1.3	1.2	0.7	0.0
Multisectoral	150.8	140.4	256.6	466.7
Fishing	5.8	5.0	36.2	58.3
Health	95.7	196.3	201.8	205.7
Agricultural and forestry	38.2	29.4	36.9	16.1
Transport	164.9	199.3	182.0	462.1
Housing	85.6	74.9	22.1	7.0
Total	1,519.7	1,659.0	1,517.5	1,919.8
Percentage of approved initiatives	76.8%	74.6%	68.8%	65.4%

Source: Gómez-Lobo 2012.