The Performance-Based State

Chapter profile

Table (1): Chapter profile, remit and boundaries.

Chapter Two concludes that an indicator is not a numerical slogan but a value with a denominator, year, definition and source that can be re-examined. This chapter takes that register and asks a different question: who has the authority to change the indicator, with what resources, who reviews their report, and what decision follows persistent deviation? Without an institutional answer, the national register becomes a sophisticated display rather than a system of governance.¹

Governing rule: the governing chain is: national outcome → legal mandate → executive owner → plan and resources → documented data → review and correction → accountability governed by law.

1. Executive summary

The problem this chapter addresses is not that ministries receive no instructions or provide no data. In a state organised around expenditure, pledges and official correspondence, many activities may be carried out without knowing whether citizens’ circumstances improved. A “performance-based state” can trace the gap between what was announced, funded, completed and actually changed, and enables a competent institution to correct policy before its losses accumulate.²

Iraq has a foundation to build on. In February 2025, the Ministry of Planning established a monitoring framework for the 2024–2028 National Development Plan, with three levels covering political and technical oversight, reporting and impact measurement. In September 2026, it announced the development of IDMS for managing and monitoring project information. In August 2026, the Ministry of Finance announced continued preparation of programme and performance budgeting to link resource allocations to objectives and outcomes. These are documented institutional steps, but they do not alone prove that every ministry has a complete national system of accountability for its service outcomes.³

The Vision proposes a system of five functions, not the creation of a fifth public body: (1) translating an indicator into an executive commitment within the mandate; (2) an outcome card containing the baseline, denominator, owner, resources and risks; (3) regular monitoring supported by relatively independent verification; (4) an annual public statement understandable to citizens and parliament; (5) administrative, fiscal and policy correction, while leaving political, disciplinary and criminal responsibility with their legally competent bodies and procedures. A performance card alone creates no powers.⁴

The most important methodological safeguard is that the state does not punish an official merely because an indicator worsens. Oil conditions, drought, epidemics, population or measurement coverage may change; a ministry may deploy the instruments it controls without achieving the aggregate impact. Decision quality, implementation, information integrity and responsiveness to correction must therefore be measured alongside outcomes, separating individual responsibility from external causes and shared mandates.⁵

Figure (2): A performance dashboard becomes accountability only when the chain of authority and decision-making is complete.

2. The central question and the chapter's limits

The chapter’s guiding question is: how can the Iraqi state link government, ministers, governors and institutions to defined public outcomes and make performance reporting part of the draft budget, annual review, parliamentary oversight and administrative practice, without conflating political oversight with employee discipline or technical evaluation with criminal judgment?

The chapter addresses executive ownership of indicators, administrative performance agreements that create no penalties, the monitoring and correction chain, annual-report design, use of data in budgeting, evaluation and oversight, and safeguards against manipulation. It sets no final numerical targets for education, health or production and does not repeat investment or sovereign-fund policy, left to Chapter Four. It does not replace “Effective Government”, V1-D02-C04, which developed general capacity; it adds management of outcomes at the level of the national Vision.

3. What is a performance-based state?

Operational definition: a performance-based state is a governance system linking approved public objectives to the mandates of specified institutions, realistic financing sources, sound indicators, and cycles of review, correction and transparency. It enables society and competent authorities to know what was achieved, why unmet goals deviated, and who must act next. It does not mean that the state is a profit-making company, that a minister is an employee under a private contract, or that every service can be reduced to a single score.

Table (2): A glossary of the performance-based state.

Every sector should use the “outcome, quality and equity” triangle. Outcome asks what changed; quality asks whether the service is adequate; equity asks whom it reached and who was left out. If average electricity provision rises while rural supply deteriorates, the average alone is insufficient for judgment. If educational outputs grow without reading proficiency, activity may have improved while outcomes remain deficient.

4. Where does Iraq stand? An institutional baseline, not assumptions

The institutional baseline shows that plans and systems exist, but their integration into a consistent public national register is incomplete. The 2024–2028 National Development Plan contains five main programmes and monitoring matrices; IDMS supports government project data; and the Ministry of Finance is moving towards programme and performance budgeting. However, the open official sources reviewed for this chapter provide no unified national series identifying, for each ministry, the share of services completed on the first attempt, service unit costs, the proportion of corrective decisions implemented, or coverage of independent outcome audits. These are therefore “measurement gaps”, not zeros.⁶

Table (3): What actually exists and what requires new capacity.⁷

The IMF adds a fiscal dimension that cannot be ignored. Its Article IV consultation report published in 2025 recorded risks from fiscal expansion and the accumulation of implementation arrears in 2024, and called for stronger estimation of available financing and implementation capacity and better monitoring of ministries. An outcome agreement without realistic cash financing or procurement authority therefore burdens an official without enabling delivery.⁸

Governing rule: baseline decision: do not announce a “percentage of ministries subject to performance accountability” before defining the ministry, outcome, verification type and denominator of covered bodies. Establishment begins in 2027.

5. The constitutional and legal basis of accountability

The Iraqi Constitution distinguishes those who formulate executive policy and run a ministry from those who oversee it, question it and withdraw confidence under specified conditions. Article 80 governs the Council of Ministers’ management of general policy, the draft budget and oversight of ministries; Article 78 defines the prime minister’s position; and Article 61 regulates parliamentary tools of questions, discussion, questioning and withdrawal of confidence. A long-term Vision therefore cannot create an “automatic confidence vote” based on an indicator’s colour or turn a delivery unit into a substitute for the Council of Representatives or the judiciary.⁹

Meanwhile, the State and Public Sector Employees Discipline Law No. 14 of 1991, as amended, establishes employee duties, penalties, procedures and appeals; it cannot be set aside or its penalties expanded through an unspecified metric. The distinction is essential: ministers are politically accountable through the competent constitutional mechanisms; professional civil servants are subject to their statutory duties and administrative appraisal under applicable rules; financial violations undergo audit; and criminal allegations belong to investigative bodies and the courts.¹⁰

Table (4): Multiple forms of accountability without conflating mandates.¹¹

The proposed practical design begins with an executive decision requiring bodies to send accurate data to the coordinating authority and establishing a standard monitoring format within the Council of Ministers’ powers. New disclosure obligations requiring legislation, special penalties or transfers of powers between institutions must be submitted as draft legislation and constitutionally examined before taking effect. A “performance agreement” must not amend the budget law or authorise expenditure unsupported by the underlying legal basis.

6. Outcome architecture: no responsibility without an owner and data

Every outcome should have an “outcome owner”: the holder of administrative or political authority able to deploy the instruments that can change it, not merely the body collecting statistics. Roles comprise a political owner of the commitment, an executive owner of the service, a data custodian controlling definitions and sources, an auditor that does not solely validate its own figures, and a reporting body relatively independent of implementation. This does not require a new organisation for each role; existing units can be assigned these functions and their staff’s capabilities strengthened.

Table (5): Mandatory minimum contents of the outcome owner’s card.

A ministry must not be held responsible for outcomes whose instruments it does not control. The Ministry of Education is responsible for education quality within its mandate, but it does not independently control household income, student housing or the electricity network. The card must therefore distinguish controllable variables, supporting variables requiring partner institutions, and external-shock variables. This clarifies what the ministry can answer for: design, implementation, measurement and response quality, alongside the outcome’s trajectory itself.

7. Linking ministries to national indicators

Attaching ministries’ names to the 40 indicators is insufficient. The preceding chapter identifies who owns production of the figure; this chapter identifies who can act when it deteriorates. These owners may differ. Each ministry should have a short list of three to seven principal outcomes, with a more detailed service matrix, rather than placing dozens of major indicators on the minister’s desk and overwhelming administration with presentation work. This number is a design recommendation, not an enforceable regulatory limit.

Table (6): Converting a national indicator into examinable service responsibility.

The table is an example of an “initial analytical allocation”, not a binding map of powers. Federal and local mandates, contracts and private entities must be reviewed before any responsibility is fixed. Each indicator should be published with its baseline, unit, denominator, update date, spatial breakdown and reason for target changes, as established in C02.

8. Shared indicators where bodies share an outcome

Accountability often falters on outcomes that cross ministerial boundaries. Reducing youth unemployment requires labour markets, education, training, investment and financing. Road safety requires design, traffic enforcement, emergency care and the judiciary. These issues cannot be resolved if each body says the indicator is not wholly within its mandate, or if the prime minister attributes every cause to one body. A “shared-issue owner” should therefore have coordinating authority and escalation schedules, while every decision, contract and expenditure remains with the legally competent body.

Table (7): Shared governance does not remove each body’s mandate.

Coordination minutes should record disagreements over causes, financing or legal requirements. Cases beyond ministerial resolution should be escalated to the Council of Ministers within its powers. Successful coordination is measured by resolving conflicts and delivering services, not the number of meetings between deputy ministers. Observations should be retained in Chapter One’s dependency register so that chains of delay can be correctly explained.

9. The administrative performance agreement and its legal limits

The chapter proposes an internal “administrative performance agreement”, signed by the competent official and their subordinate or a partner body, specifying outcomes for which they control the instruments, data and costs, and the route for escalating risks. It is an instrument for organisation, monitoring and development, not an independent civil employment contract or legislation for penalties or rewards contrary to law. It must be reviewed periodically when budgets, mandates or relevant circumstances change. A target cannot be fixed while management is prevented from requesting adjustment despite a documented shock.

Table (8): The proposed administrative performance agreement and conditions for its legality.

Outcome indicators differ according to their production cycles. Road deaths are measured annually with periodic reports; service time is an almost daily administrative measure; tax collections are monthly; and school reading is assessed through periodic standardised evaluations. The Vision does not impose a single monthly deadline on every phenomenon, because that produces poor data rather than timely data.

10. The decision cycle: from signal to correction

The indicator dashboard serves its purpose when it leads to an identifiable decision. The proposed monitoring cycle is: detect deviation → verify data accuracy → identify whether it arises from a shock, planning, implementation or measurement → present corrective options and costs → decide within the mandate → implement action → measure again and document the lesson. A meeting without an archived decision, or a decision that never reaches an implementation owner, does not close the cycle.¹²

Figure (3): The four-stage correction loop; action is not complete until its effect appears.

Table (9): Not every deviation is addressed in the same way.

11. Public annual reports

The chapter proposes a public annual performance report from each ministry and a consolidated government summary. An ordinary reader should be able to understand the objective, previous value, current target, actual achievement, expenditure, who was harmed or did not receive the service, the limits of confidence in the figure, and the planned correction. Coloured success tables or photographs of openings are insufficient. The report should include downloadable datasets where compatible with data protection and legitimate security requirements.¹³

Table (10): Proposed structure of the public annual report.

The proposed rhythm is monthly internal operational monitoring for services that permit it; quarterly review of national priorities; a short half-yearly summary; and an initial annual report within 90 days of year-end, labelling provisional values, followed by updates as final statistics and closing accounts become available. These are proposed policy deadlines, not Iraqi statutory deadlines currently in force, and they do not replace financial or legislative deadlines imposed by law.

Reporting must not breach the confidentiality of an ongoing investigation or disclose individuals’ identities or health, security or intelligence data. Security performance should be measured publicly through aggregate indicators of harm, rights and services, while operational details remain within competent audit and oversight channels. In civilian sectors, withholding information covered by a right of access should require reasons and a right of review under existing law or legislation to be introduced later.

12. Accountability for outcomes: distinctions that protect fairness

Performance accountability has three layers: public explanation (Answerability), always required where a material gap exists; administrative correction (Corrective Action), based on mandate, resources and choices; and legal penalty (Sanction), imposed only where a violation, procedure, evidence and jurisdiction exist. Failure to meet a political target may justify political accountability and questioning, but it does not alone establish a crime or disciplinary offence. An improving indicator may conceal wrongdoing if achieved by excluding beneficiaries or corrupt contracting.¹⁴

Before judging an unfavourable outcome, the ministry should apply four sequential tests: is the figure reliable? Were powers and resources sufficient? Did fundamental circumstances change beyond the decision-maker’s control? Did the official take the professionally expected action after the warning appeared? Reporting then becomes a means of learning and accountability, rather than a selective tool for settling political disputes or punishing an employee unable to change conditions.

Table (11): An accountability matrix that separates failure from violation.

13. Programme and performance budgeting

Programme and performance budgeting does not mean reducing every ministry’s appropriation simply because it fails. Deteriorating services often require additional reform or maintenance funding, while programmes that meet some objectives may cost more than alternatives. The Vision uses “Performance-Informed Budgeting”: attaching evidence to fiscal decisions while retaining appropriation and accounting under Federal Financial Management Law. The OECD’s 2023 survey showed that 28 of 33 responding countries used some form of performance budgeting, but approaches differ; this does not make any single model obligatory to copy.¹⁵

In August 2026, the Ministry of Finance announced work on this transition, linking resources to programmes, objectives and outcomes. The Vision proposes beginning in 2027 by adding performance cards to pilot programmes while retaining existing legal accounting classifications, then progressing after testing data quality, commitment accounting and lifecycle costs. Spending capacity, not allocation value alone, must be examined. Operating and maintenance costs must be included to avoid projects that produce an opening ceremony rather than a continuing service.¹⁶

Table (12): Outcome evidence informs fiscal decisions; it does not replace them.

14. Data quality and independent audit

As incentives based on numbers expand, so does the risk of changing the figure instead of the service. Work centres on a national system for preserving data series: the indicator definition and version, source identity, input records, automated and logical validation, explanations of outliers, approval by the record owner and an independent review sample. A central system must not collect all personal data unnecessarily; it should use purpose-limited data exchange, access permissions and an audit trail.¹⁷

Table (13): How does an outcome-oriented state prevent measurement from becoming manipulation?

This chapter’s measurement system distinguishes facts, estimates, assumptions and targets. Every reporting dashboard should therefore include a revision log: who changed a value or its definition, when, on what evidence, and whether the new version has documentation enabling reproduction. A report that corrects and discloses an earlier error deserves more trust than one that hides the change to appear consistent.

15. Measuring services and citizens’ experience

Citizens do not experience government through the number of projects listed in the budget. They experience a service journey: could they access it? How long did they wait? Did they have to return because a file was incomplete? Was the outcome correct and safe? Could they appeal? “Citizens’ experience” measures how the service works, rather than relying solely on a satisfaction survey that may vary with expectations. Indicators should be disaggregated by service location and beneficiary type without compromising privacy.

The Vision adopts the “complete journey” rule: application → document receipt → decision or service → payment, if any → notification → appeal and correction. Digitisation cannot be declared successful because a form is online, nor can service time be reduced by closing difficult requests or requiring beneficiaries to visit five offices excluded from the official time measure.

Table (14): Service-journey indicators that complement national outcomes.

16. Federal governance, governorates and spatial differences

The performance-based state principle extends to governorates, but it does not turn decentralisation into complete administrative subordination to a central ministry. In federal Iraq, functions are distributed among federal authorities, the region and governorates under the Constitution and applicable laws. Each outcome card must therefore identify the competent authority, data owner and service financing at every level before assigning responsibility to a governor or minister. Indicators should be presented on maps and by population group, not only as a national average.¹⁸

Where a service is implemented locally but funded or standardised federally, a three-party coordination agreement should apply: the federal body approves the standard and safeguards the fiscal transfer under the law; the governorate manages implementation within its powers; and the statistical/oversight body verifies comparability. Disputes should be escalated through the competent mechanism. Genuine comparison between areas must consider population, terrain, costs, security and water conditions and poverty distribution, rather than ranking governorates by absolute levels without explanation.

Table (15): Managing spatial performance within the federal system.

17. Monitoring is not impact evaluation

Monitoring asks: are we proceeding as planned? Evaluation asks: was our choice appropriate, and did the intervention actually produce the outcome compared with feasible alternatives? A ministry’s data may show improved reading after new schools opened, but timing alone does not establish the cause. The test, cohort size or attendance patterns may have changed. Some programmes therefore require independent evaluation using a valid comparison method, not more monthly indicators.¹⁹

The Vision proposes selecting major or particularly consequential programmes for in-depth evaluation, starting from a defined Theory of Change. Where ethics and data permit, methods may include before-and-after comparison with a similar group, a controlled experimental design, or cost-effectiveness analysis. Where these are unsuitable, implementation-process evaluation and case studies should be used with the limits of inference disclosed. A ministry cannot claim to have “created” every job generated during a year merely because it launched a training programme that year.

18. Less rhetoric and more measurement

“Less rhetoric and more measurement” is not a ban on political speech; it means subjecting public promises to information and outcomes. In review meetings, government should replace lengthy activity reports with a one-page decision brief: the previous position, current value, what changed and what did not, three possible causes, the decision required, cost, timing and implementation owner. Every press release announcing delivery of a major service should link to a public results page distinguishing beneficiary numbers, operating capacity and distribution from expenditure and buildings.

Figure (4): Turning government communication from media activity into verifiable transparency.

Success claims should be reviewed regularly through three modest but decisive questions: did the programme produce an outcome beyond the working paper? Do the data demonstrate that impact for those who need it? Were costs, rights and risks acceptable? Government should publish its most important failures and corrections in the same report as its successes. Ignoring failure or postponing its explanation weakens trust more than acknowledging it and providing a professional route to change.

19. What can international experience teach us?

The OECD’s 2024 report on the centre of government explains that monitoring public priorities can be led by small units close to political leadership, but these do not implement in place of ministries and must build trust, analytical capacity and learning. The OECD’s 2025 paper on reporting practices highlights combining outcomes and finances and publishing them in time to enable accountability. For Iraq, the lesson is a limited, professional central monitoring unit with coordination and escalation powers, not a shadow ministry or a laboratory for punishments.²⁰

In the United Kingdom, the planning and performance framework updated in November 2025 explains the links among priorities, plans, resource allocation, progress measurement and parliamentary and public accountability, with an accounting officer inside the executive. In Canada, experience with the Results and Delivery Unit highlights three recurring questions: what outcome is required? How will it be achieved? Has it been achieved, and if not, what should change? These are mechanisms to select from, not instructions to copy another country’s constitutional structure or civil service.²¹

Table (16): Adapt a suitable mechanism; do not replicate another state.²²

20. The starting point: the first hundred days

The first hundred days begin after a decision by the competent political and organisational authority, not by building an expensive platform or declaring the Vision complete. Government starts with limited priorities whose dependencies were ordered in V4-D10-C01 and uses the headline indicators proposed in V4-D10-C02, subject to quality and mandate verification. The Vision cannot impose these steps on a government before adoption, but it defines the minimum starting point for any government that adopts it.

Table (17): The first-hundred-days model as an accountability system, not a media campaign.

The governing condition is that government must not choose outcomes merely because they are easy. The portfolio should include a service visible to citizens, an indicator of fiscal and project discipline, a human-capital indicator, a production indicator and a rights-protection indicator. The centre does not determine who delivers every service locally; it identifies when resolving a bottleneck requires a higher-level decision, coordination or procedural reform.

21. Transformation phases for 2027–2045

Table (18): Phases of the performance-based state without repeating C01’s details.

Years are review markers, not automatic guarantees. One institution may mature slowly while another advances rapidly, but no body may change a definition to appear to have reached the next phase. Every gate requires a verification report, assumptions register and reasoned decision; its results should be published to the extent legally permitted.

22. The government performance dashboard and proposed indicators

Assessing the performance system itself requires more than repeating the forty national indicators. An internal dashboard of twelve administrative indicators is proposed, each with a denominator, definition and owner before a figure is published. The targets below are “proposed” organisational commitments for the initial period, not facts about Iraq in 2026. Meeting them does not mean reform is complete if services have deteriorated.

Table (19): A performance-system management dashboard; it is not automatically combined into a national ranking score.

PM01–PM12 must not become a secret score for ranking ministers and publishing personal judgments. The dashboard tracks government’s capacity to manage information and correction and must always be accompanied by professional assessment of actual outcomes and citizens’ rights. Reaching an internal target must not be described as achieving a national outcome among C02’s indicators.

23. Implementation programmes and the responsibility package

The Vision proposes seven interconnected operating programmes, each implemented within an existing structure: (a) the outcome and owner register; (b) integration of project, financial and output data; (c) open performance reporting; (d) beneficiary and service assessment; (e) cause review and correction; (f) independent impact evaluations; and (g) capabilities, training and protection against manipulation. These programmes must not become seven new bodies; most develop functions that already exist in fragmented form.

Table (20): The results-based management programme package.

The programmes overlap in time. A reporting template can be introduced before technical integration is complete, but an Iraq-wide implementation rate must not be announced before service and project denominators are standardised. Every software project should undergo a necessity, lifecycle-cost and integration test against existing platforms, particularly since IDMS and performance-budgeting processes are already being developed.

24. The implementation and financing matrix

Table (21): An implementation and financing matrix without inventing a national budget.

The chapter does not estimate a national dinar cost for building this system, because no published inventory of current systems, infrastructure, personnel and contracts permits a reliable calculation. Costing a new platform on top of available systems before that inventory would encourage duplication. The correct fiscal starting point is to assess what Iraq already has, then cost the gap, integration, security, support and training for each programme. Finance and Planning can then submit approved estimates within the budget law.²³

25. Risks and safeguards

Table (22): The initial register of risks and safeguards.

Government performance must not improve at the expense of rights and freedoms. Security and judicial reports should use an aggregation level balancing transparency, safety and individual rights; justice must not be reduced to the number of judgments, nor security to the number of arrests. An internal appeal mechanism should protect officials to whom inaccurate data are attributed, and an external review mechanism should serve citizens who establish that a report excluded their experience or community.

26. An applied test: from a school and service to an outcome

An illustrative example, not actual results: a ministry announces that it added 30 schools in a year, while third-grade reading proficiency remains unchanged. An activity system counts construction as achievement. A performance system examines school distribution, class density, teacher attendance, standardised testing and the costs of operations, books and facilities. Data may show that the reading problem lies in early education and teaching rather than buildings; government can then lawfully redirect part of its response towards quality training and learning assessment. This does not mean school construction was useless; it means that a building project alone cannot demonstrate achievement of an educational objective.

A second example: an institution announces that licence issuance time has fallen to two days. If it counts only applications accepted by an employee after preliminary visits, citizens may actually have waited 18 days. A performance system records the first submission point, the number of returns, the time distribution for easy and difficult applications, appealed decisions and compliance with conditions. This reveals genuine achievement or manipulation of definitions before the institution is rewarded or its procedures replicated.

Table (23): From counting to managing outcomes—hypothetical analytical examples.

27. The success criterion and conclusion

By 2045, institutionalised measurement should not depend on the name of a government or minister. Success means that parliament, citizens and oversight bodies know what ministries achieved, in which year, with what resources and impact, and that the next government can use the same register without erasing earlier decisions. Transparency becomes a management method rather than an exception; equity becomes an evaluation dimension; correction becomes a required administrative decision process; and penalties beyond that process occur only under law.

The Vision does not achieve its purpose simply by issuing annual reports, however attractive. The stronger tests are: was a resource allocation changed because evaluation demonstrated greater value? Was a service changed after a documented complaint? Was a failed programme stopped without losing beneficiaries’ rights? Did a record of error and learning pass from one administration to another? Do citizens know how to challenge an inaccurate achievement claim? When these conditions exist widely, an outcome-oriented state becomes capable of continuity and renewal.

Governing rule: a strong government is not one that can always announce success, but one that can prove it, correct failure and retain institutional memory when officials change.

28. A register of knowledge gaps and open decisions

• Actual service completion times and first-time completion indicators across ministries and governorates: no established unified national value; collect records and samples in 2027–2029.

• Programme and project coverage in IDMS and the relationship of each record to budget classifications and completed services: an official inventory and reconciliation are required before establishing a national percentage.

• Independent verification, service unit costs, complaint-resolution rates and corrective decisions: establish the series before any public ranking.

• The legal decision: proposed performance agreements are administrative instruments under existing law; political accountability and employment and judicial penalties remain with competent bodies.