Unorthodox Expenditure Procedures in CEMAC and WAEMU Countries
Author:
Mr. Bruno Imbert
Search for other papers by Mr. Bruno Imbert in
Current site
Google Scholar
Close
,
Hoda Selim https://isni.org/isni/0000000404811396 International Monetary Fund

Search for other papers by Hoda Selim in
Current site
Google Scholar
Close
,
Ms. Gwenaelle Suc
Search for other papers by Ms. Gwenaelle Suc in
Current site
Google Scholar
Close
, and
Qing Zhao
Search for other papers by Qing Zhao in
Current site
Google Scholar
Close

This paper takes stock of unorthodox expenditure procedures in CEMAC and WAEMU countries and assesses their potential fiscal impact. “Unorthodox procedures” are defined as spending practices that bypass legal provisions governing public expenditure processes and circumvent regular controls or other budgetary rules, including those related to budget time limits, approved ceilings, or approved appropriations. The paper shows that despite PFM reforms, recourse to such procedures has persisted—resulting in the accumulation of arrears; inadequate fiscal reporting, including large stock-flow adjustments; and corruption vulnerabilities.

Abstract

This paper takes stock of unorthodox expenditure procedures in CEMAC and WAEMU countries and assesses their potential fiscal impact. “Unorthodox procedures” are defined as spending practices that bypass legal provisions governing public expenditure processes and circumvent regular controls or other budgetary rules, including those related to budget time limits, approved ceilings, or approved appropriations. The paper shows that despite PFM reforms, recourse to such procedures has persisted—resulting in the accumulation of arrears; inadequate fiscal reporting, including large stock-flow adjustments; and corruption vulnerabilities.

Introduction

Ex ante spending controls, put in place through legal and regulatory frameworks, aim to ensure that public spending is executed in accordance with the budget approved by Parliament, within authorized limits and following sound Public Financial Management (PFM) principles (Pattanayak 2016).1 The absence of such controls can lead to budget execution slippages and deviations between voted budget allocations and actual spending.

A well-functioning fiscal framework may allow expenditure controls to be temporarily relaxed to create flexibility during budget execution. In such cases, procedures may initially bypass the normal chain of expenditure to either help accelerate payments (exceptional procedures) or meet unforeseen spending pressures (emergency procedures). PFM rules should clearly specify the circumstances and processes for the recourse to these procedures and ensure they are strictly enforced. Spending through these procedures should also be promptly regularized; controls that were initially bypassed should eventually be imposed later so that the expenditure is cleared and properly recorded in fiscal accounts.

In practice, however, although these less stringent controls can effectively increase flexibility during budget execution, they carry the risks of being more broadly used to settle either noneligible or unauthorized (often ad hoc) expenditure. To the extent that these procedures deviate from regular controls and are not followed by ex post regularization, they become unorthodox procedures that conflict with the legal and regulatory budget rules and processes. Specifically, unorthodox procedures are defined as practices that relate to the abuse of exceptional procedures—which are allowed under the legal framework—as well as to other spending that bypasses legal provisions, such as off-budget spending or other ad hoc procedures that circumvent regular controls or other budgetary rules, including those related to budget time limits, approved ceilings, or approved appropriations.

As part of the convergence efforts and to promote well-functioning currency unions, the regional fiscal frameworks of the two Francophone monetary unions in Africa—the Central African Economic and Monetary Community (CEMAC) and the West Africa Economic and Monetary Union (WAEMU)—provide guidance to member states on shared PFM principles that are essential to support fiscal responsibility and the system of fiscal rules. Member states in both regions are expected to coordinate national fiscal policies through harmonizing budget laws and procedures including budget and accounting laws, as well as laws governing public accounting, budget transparency, and the chart of accounts. At a minimum, common PFM systems should aim to ensure that (1) proposed budgets are consistent with the fiscal rules; (2) timely and transparent fiscal information is produced to confirm the observance of the rules during budget execution and corrective actions are in place; and (3) transparent enforcement mechanisms, including specific sanctions, are present (Corbacho and Ter-Minassian 2013).

In this context, regional PFM directives governing national laws and regulations were revised in 2009 for WAEMU countries and 2011 for CEMAC countries to provide a common framework for enhanced PFM practices following international best standards. Among other things, the new regional directives aimed for more efficient expenditure controls (for example, risk-based controls) and enhanced budget execution procedures (for example, decentralization of budget authority to line ministries).

However, the effectiveness of these de jure PFM improvements in both regions remains impeded by delays in the implementation at the national level, as well as weak enforcement and ineffective sanctions mechanisms at the regional level (David, Nguyen-Duong, and Selim 2022; Lando, Rosa, and Suc 2019). To date, most studies show that the adoption of the revised regional PFM directives related to expenditure controls have lagged in many CEMAC and WAEMU countries (David, Nguyen-Duong, and Selim 2022; Lando, Rosa, and Suc 2019).2 Moreover, existing controls are generally perceived to be cumbersome—rigid, redundant, and slow—and to have contributed to the proliferation and overuse of unorthodox procedures (Bouley, Fournel, and Leruth 2002; Doe and Pattanayak 2008; Lienert 2003).3 Because these unorthodox procedures lead to the execution of expenditures outside the regular chain of expenditure, they could have detrimental macro-fiscal effects and eventually undermine overall budget integrity and credibility.

This paper takes stock of unorthodox expenditure procedures in CEMAC and WAEMU countries— procedure that bypass the legal or intended PFM framework—and assesses the potential fiscal impact of these procedures. The paper draws on assessments of PFM systems (including Public Expenditure and Financial Accountability (PEFA) assessments, Public Investment Management Assessments, and Fiscal Transparency Evaluations), as well as on findings from the IMF’s technical assistance (TA) reports, to identify unorthodox budgetary procedures. The analysis focuses on expenditure processes; it does not cover other budget components, such as revenue and debt management. At the outset, it is also important to note that the lack of comprehensive data on the amounts of spending associated with such procedures over time and across the 14 countries has limited this analysis. The paper is only able to establish cross-section correlations between the prevalence of unorthodox procedures and adverse fiscal outcomes. Nevertheless, the analysis clearly shows that these procedures are prevalent and have harmed public finances.

The paper is structured as follows: The next section describes the different types of expenditure controls and their role in regulating unorthodox procedures. It also provides insights from the literature on the potential adverse effects of recourse to unorthodox procedures. The third section takes stock of the different types of unorthodox procedures in CEMAC and WAEMU countries. The fourth section discusses the potential fiscal impact of these procedures. The final section proposes policy recommendations to reduce and better control the use of these procedures.

This paper builds and expands on previous analysis on PFM weaknesses, including the recourse to unorthodox procedures in Francophone Africa (Bouley, Fournel, and Leruth 2002; Lienert 2003; Moussa 2004).

Expenditure Controls and Unorthodox Budgetary Procedures

Expenditure controls aim to ensure that public spending is executed in accordance with the budget approved by Parliament, within authorized limits, and following sound PFM principles so that resources are used efficiently, and obligations are cleared in a timely manner (Pattanayak 2016).

Accordingly, controls are designed to ensure that expenditure is executed only with proper authorization and are applied by the relevant authorities during the stages of budget execution (Allen and Tommasi 2001 and Box 1).4 In a simplistic form, controls should ensure that (1) a commitment is only incurred when it is in line with the budget appropriation, respectively, within the expenditure ceilings and cash releases approved by the Ministry of Finance;5 and (2) no payment is made unless it is based on an approved commitment. For instance, the public accountants are expected during the payment stage to verify, based on supporting documents, that certain conditions are met for the release of funds. To this end, they need to ensure that the expenditure (1) is appropriated in the budget (appropriation control), (2) respects the commitment schedule based on the annual cash plan and for this they need to ensure that cash is eventually available to meet the underlying payment (aggregate cash control); (3) is accompanied by all necessary documents, such as the certification of delivery of goods (control of regularity); and (4) is correctly recorded in the accounts for future controls and fiscal reporting (accounting control). Also, during the verification stage, the budget officer is responsible for verifying that the goods or services are delivered and that they match the terms of the initial contract (for example, quality, quantity, and prices) and recognize the liability and the due date of the payment.

Expenditure Controls during the Different Stages of the Expenditure Chain in Francophone PFM Systems

Five types of expenditure controls exist at different stages of the expenditure chain (Figure 1):

  • Appropriation controls by both authorizing and accounting officers ensure that expenditure is regularly appropriated in the budget and that amounts do not exceed approved ceilings.

  • Aggregate cash controls ensure that the expenditure is consistent with the annual cash plan and that it does not exceed the payment capacities of the treasury.

  • Commitment controls ensure that the committed expenditure by the spending units does not exceed the overall ceilings, as well as the allotted in-year ceilings (quarter, month, and semester), in relation with the annual cash plan.

  • Controls of regularity verify that supporting documents comply with legal and regulatory requirements and that procedures are respected (for example, authorizing officers have the authority to act as they have done).

  • Accounting controls by the public accountant before payment ensure the recording of the expenditure in the accounts that would allow subsequent control, for example, from the Supreme Audit Institution, and the production of fiscal reports.

Figure 1.
Figure 1.

Controls Applied at the Various Stages of the Expenditure Chain

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Source: Authors.

A well-functioning fiscal framework may allow expenditure controls to be relaxed temporarily to create flexibility during budget execution. For instance, exceptional procedures bypass the normal chain of expenditure to help accelerate payments or smooth the execution of recurrent expenditures with fixed-date payment terms (Pattanayak 2016). Emergency procedures also allow the rapid mobilization of resources to meet unforeseen spending pressures. PFM rules should clearly specify the circumstances and processes for recourse to these procedures and ensure they are strictly enforced. Accordingly, spending through both procedures should be promptly regularized; controls that were initially bypassed should be imposed so that the expenditure is cleared and properly recorded in fiscal accounts.

In practice, however, these less stringent procedures carry the risk of becoming the standard way to settle expenditures, including those that are not eligible under the criteria established by the rules or those that were not appropriated or formally committed. In these cases, they become “unorthodox procedures” that circumvent regular expenditure controls or other budgetary rules and procedures required under the regular expenditure process. As such, they are also neither recorded nor accounted for in the financial management information system (FMIS).

Excessive use of these procedures could have five detrimental macro-fiscal effects.

  • First, these practices could undermine fiscal discipline and budget credibility. Because payments through these procedures are often made without corresponding approved appropriations and formal commitments, they are an important cause of expenditure overruns. Accordingly, these procedures often conceal the true scale of public expenditure and underestimate the reported fiscal deficit on a commitment basis.

  • Second, the regular release of cash to meet unplanned payments through unorthodox procedures could disrupt effective cash management and contribute to the accumulation of arrears. These payments absorb cash resources that were initially allocated for approved expenditure and result in the emergence of payment arrears, which, in turn, contributes to the underestimation of expenditure and of the reported deficit (Flynn and Pessoa 2014; Pattanayak 2016). In the long term, unorthodox procedures could eventually crowd out regular expenditure.

  • Third, unorthodox procedures could undermine the quality and integrity of fiscal reporting and provide a distorted picture of public finances and fiscal data and increase fiscal risks. If exceptional procedures are subject to adequate ex post regularization, then payments are eventually reported in fiscal accounts but with some delay. For instance, advance payments are temporarily transferred to a suspense account until it is time to clear the account, and then they are correctly registered under proper classification in the budget execution system.6 In contrast, unorthodox payments circumvent normal accounting procedures and the FMIS and are not recorded in fiscal accounts. The expenditure often remains indefinitely in suspense accounts that are not subject to adequate and regular clearance. In such cases, the expenditure can no longer be tracked or verified. Unorthodox payments may thus often result in below-the-line financing that has bypassed regular controls and that may not even have had a corresponding above-the-line budget allocations. These discrepancies in fiscal accounts can be seen in differences between above-the-line fiscal balance and below-the-line net financing data, as well as in large stock-flow discrepancies between government debt and deficits (Khemani and Wiest 2016). Shortcomings in PFM—in particular, fiscal reporting and lack of fiscal transparency—could result in significant fiscal risks because they allow some operations to remain outside the headline the deficits until the risks materialize.7

  • Fourth, persistent recourse to these procedures may have a detrimental effect on private sector performance. To the extent that the private sector depends on government contracts, weak expenditure controls that result in large arrears to private contractors can adversely affect their liquidity positions, their profits, and eventually their investment levels (Checherita-Westphal, Klemm, and Viefers 2016). Overall, this situation may undermine trust in the government and its ability to honor its commitments (IMF 2019).

  • Fifth, these unauthorized and unreported expenditure may also create increased scope for corruption, including the misappropriation and misuse of government resources.8 Exceptions or overrides of regular controls may conceal potentially fraudulent and dubious activities. Increased incentives for corruption exist for politically sensitive and secretive transactions, such as military spending, that often avoid regular expenditure controls and oversight (Omitoogun and Hutchful 2006).

Unorthodox Procedures in CEMAC and WAEMU Countries

In the mid-1990s, both CEMAC and WAEMU countries adopted regional surveillance frameworks that prescribed ceilings on main macro-fiscal aggregates, including the fiscal deficit and public debt and inflation. Member countries observe convergence criteria to contribute to macroeconomic stability by ensuring the sustainability of national fiscal policies and the consistency of these policies with the common monetary policy (including the fixed exchange rate regime). The multilateral surveillance framework also monitors the compliance of member states with the convergence criteria.

The regional PFM directives that govern national PFM systems are influenced by the Francophone tradition and cover most PFM aspects. The directives in both regions also share a number of similarities. First, they include detailed rules and procedures for the different stages of the budget cycle (for example, loi de finances, mise à disposition des crédits, liquidation, engagement, certification du service fait, liquidation, ordonnancement et paiement). Second, they set specific controls (for example, credit reservation is not formally included but occurs in practice in most countries, particularly for capital expenditure) and rules for cash operations during budget execution (Box 2). Third, the verification stage represents the starting point for the reporting of arrears if payments remain overdue for more than three months.9 Finally, the directives enforce a strict distinction between the commitment officer ordering the payment (“ordonnateur”) and the treasury officer or public accountant making the payment itself (“comptable”) (Pattanayak 2016). Under this system, the public accountant assumes personal financial responsibility for compliance with the regulations and is accountable to the supreme audit institution (Cours des comptes) (Pattanayak 2016). This accountability means that the public accountant is responsible for paying any compensation related to the financial loss resulting from shortcomings and/or failures in expenditure controls. This overall set-up is designed to minimize the misuse of public funds and enhance compliance with budget laws.

The PFM Principles Set by CEMAC and WAEMU Directives

Six main rules provide the general framework for PFM operations in CEMAC and WAEMU countries and impact budget execution:

  • Annuality: The budget authorization from Parliament for both revenue and expenditure is only valid for one calendar year.

  • Specificity: The budget law specifies the purpose of the expenditure and the responsibility (line ministry, department, unit) according to the existing budget classification (for example, economic, administrative, programmatic); this specificity limits what the Government may do with approved funds.

  • Unity: All revenue and expenditure must be included and presented in the budget law.

  • Universality: All expenditure and revenue must be presented in the budget with their nominal value, without netting one with the other. This rule also implies that revenue should not be earmarked and dedicated to one single type of expenditure.

  • Sincerity: The budget law must reflect a truthful image of the government’s fiscal position and deficit: revenue must not be overestimated, and expenditure must not be underestimated.

  • Strict separation of functions between authorizing and accounting officers: The current practice, confirmed by WAEMU directives, requires a strict distinction between the person who initiates the decision to spend (the authorizing officer, ordonnateur) and the person handling the payment itself (the accounting officer, comptable). Both functions are mutually incompatible; only accounting officers are entitled to manipulate public cash.

Source: Authors, based on CEMAC and WAEMU PFM directives.

The remainder of this section first describes the different types of exceptional procedures in CEMAC and WAEMU countries. It then highlights their difference with respect to unorthodox procedures, and then it takes stock of the different types of unorthodox procedures in both regions.

Exceptional Procedures in CEMAC and WAEMU Countries

Some progress toward shared PFM systems in CEMAC and WAEMU countries was observed in 2009 and 2011 with the revision of the regional PFM directives in both regions, respectively. The regional directives seek to provide a common framework for more effective and transparent PFM practices. They followed international best standards in harmonizing the presentation of fiscal statistics across member countries, in accordance with the IMF’s Government Finance Statistics Manuals and Guides (GFSM) 2001 standards; they strengthened the accountability of public expenditure by fostering the transition to results-based budgeting and clarified the responsibility lines; and they buttressed the internal financial controls over budgetary execution. The WAEMU Commission regularly undertakes a self-assessment of the PFM reforms required by the directives with a view to provide incentives for reform and show visible improvements in performance (David and others 2022). The CEMAC Commission, with more limited resources, has undertaken this assessment a few times with the support of IMF TA.

Meanwhile, national organic budget laws in CEMAC and WAEMU countries allow for exceptional spending procedures under specific circumstances (for example, emergency, petty cash operations and recurring sensitive expenditure, such as salaries). The rationale for these procedures is to accelerate expenditure in cases when “normal” procedures for expenditure approval and control are too cumbersome and slow to deal with particular spending cases (Bouley, Fournel, and Leruth 2002; Doe and Pattanayak 2008).10 These alternative procedures are perceived to be simpler and less stringent because they allow expenditures to initially bypass the regular required controls under the normal expenditure cycle. However, the regulations also subject this spending to ex post controls (including a rapid regularization) and other accountability mechanisms to ensure that they do not become a source of budget circumventions. Once

accounted for, exceptional expenditure should be cleared and recorded properly as a regular expenditure. Two widespread types of exceptional procedures are allowed in the budgetary legal frameworks of

CEMAC and WAEMU countries:

  • Advance decrees enable the Ministry of Finance to open supplementary credit lines during emergency situations to allow countries to swiftly respond to unforeseen spending pressures. This procedure allows the amendment of the initial budget law without prior Parliamentary approval. The legal framework requires that the Parliament is immediately notified and that the supplementary appropriations are ratified by the next budget law; this ratification usually occurs through a supplementary budget law passed during the closest session of Parliament.

  • Imprest accounts (régies d’avance) allow the upfront release of cash before goods and services are delivered but also require ex post regularization. There are two types of accounts: (1) the regular imprest, capped at low ceilings and largely used for petty cash needs and urgent minor expenditure; and (2) the special imprest, used for emergency spending purposes and allowing one-off advance payments that exceed the ceilings under the regular imprest. The imprest account is often created by a decree that specifies such items as the ceilings, the eligible spending, the period during which it will be used, and the end date. Ex post controls require that underlying transactions are settled by the end of the fiscal year and that they are adequately reported in financial reports. It is important to note that these procedures are allowed under the legal framework and are strictly regulated by PFM rules. They, therefore, differ from unorthodox procedures that are in direct conflict with legal provisions, as will be described in more detail.

Unorthodox Procedures in CEMAC and WAEMU Countries

The exceptional procedures described can become a source of unorthodox procedures. Many assessments have indicated a systematic and often unjustified recourse to exceptional and other ad hoc procedures to settle expenditure (Bouley, Fournel, and Leruth 2002; Doe and Pattanayak 2008). These assessments point to two main reasons for such recourse.

First, the provisions for the ex post regularization of exceptional expenditure were not always adequately implemented (Doe and Pattanayak 2008; Lienert 2003). In fact, unrecorded expenditures (from advance payments, for example) remained permanently registered in suspense accounts; the latter were not settled at the end of the year, as required. These expenditures were also not systematically included in financial reports and were not adequately reported in the budget (Khemani and Wiest 2016; Pattanayak 2016).

Second, vague or imprecise legislative definitions leave room for interpretation and arbitrariness in the criteria to justify the recourse to exceptional or emergency procedures. For instance, in principle, imprest accounts should only be used for minor expenditures; however, these accounts are often used to settle other noneligible spending (nonurgent and sometimes major expenditure).11 Similarly, while regulations provide that urgency is a criterion for emergency procedures, the regulations do not precisely define the term; as a result, officials use “urgency” indiscriminately to justify the application of emergency procedures to expenditures that would otherwise not qualify. In practice, the application of the emergency criterion to a particular expenditure remains subject to subjective interpretation. More generally, emergency spending procedures could also lead to unorthodox procedures if the scale of spending is large, interventions are implemented rapidly, and multiple actors are involved without adequate coordination (Khasiani, Mfombouot, and Singh 2020). In this context, it is important to note that emergency procedures undertaken during the COVID-19 crisis could potentially lead to unorthodox budgetary procedures if they are not subject to adequate ex post regularization and other PFM rules (Box 3).

Beyond unregularized exceptional procedures, unorthodox procedures also include those procedures that were associated with any executed expenditure that circumvented regular expenditure controls or that was in conflict with the regular budgetary rules and procedures required under the normal chain of expenditure.

COVID-19 Crisis: Emergency Spending in 2020 and 2021 and Associated PFM Risks

The COVID-19 pandemic has prompted unprecedented and urgent fiscal spending needs in CEMAC and WAEMU countries since 2020. Measures were developed to support the health sector and assist businesses and households experiencing economic hardship. Fiscal measures included either above-the-line measures or liquidity support (below-the-line measures, such as equity injections, or contingent liabilities, such as guarantees), or both.1

In CEMAC countries, fiscal measures between January 2020 and September 2021 are estimated at about US$2.4 billion (2.7 percent of GDP). Almost all support was in the form of revenue and expenditure measures (above-the-line); only about US$67 million (0.1 percent of GDP) was in the form of liquidity support (Box Figures 1 and 2). Measures in countries ranged between 1.1 percent of GDP (Equatorial Guinea) and 2.7 percent (Cameroon).

Box Figure 1
Box Figure 1

Fiscal Measures: Jan 2020-Sept 2021

(% of GDP)

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Box Figure 2
Box Figure 2

Liquidity Support Measures: Jan 2020-Sept 2021

(% of GDP)

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Note: The fiscal measures include resources allocated or planned in response to the CO VID-19 pandemic since January 2020, which will cover implementation in 2020, 2021, and beyond.

Fiscal support was higher in WAEMU countries over the same period; it is estimated at US$6.1 billion (about 3.8 percent of GDP). Above-the-line measures also accounted for most of the stimulus (3.3 percent of GDP), but liquidity support was more than 10 times higher than in CEMAC countries (US$0.8 billion and about 0.5 percent of GDP). About three-quarters of this support was in the form of public guarantees. Moreover, about one-half of WAEMU countries adopted liquidity support measures above 1 percent of GDP.

In many cases, countries resorted to emergency spending procedures to execute this crisis spending. Nevertheless, results from a survey carried out by the IMF’s African department showed that around one-half of CEMAC and WAEMU countries implemented crisis-related emergency spending without the adoption of supplementary budget laws. In a smaller number of countries, a share of this spending was executed outside of regular budget controls through extrabudgetary funds outside the treasury single account (TSA). In addition, several CEMAC countries did not hold public tenders for procurement or resorted to ad hoc methods for the purchase of medical materials; one country changed the threshold for simplified procurement procedures. All of these practices bear the risk of ultimately becoming unorthodox procedures.

CEMAC and WAEMU countries that received IMF emergency financing undertook PFM commitments to improve the execution of COVID-19-related spending. Such commitments ranged from conducting ex post audits of spending and enhancing ex post controls and reporting—all of which should ensure better transparency and help mitigate the risks of mismanagement of the COVID-19 funds.

1 Above-the-line spending includes additional spending or tax cuts resulting in higher budget deficits. Liquidity support to companies in financial trouble through loans or equity injections does not impact budgets directly, but it may increase debt or require additional borrowing. Guarantees do not affect deficits or debt in the near term, but they expose the government to medium- to long-term fiscal risks (IMF 2020). Source: Authors, based on the FAD Fiscal Monitor Database of Country Fiscal Measures in Response to the COVID-19 Pandemic (Fiscal Policies Database (imf.org) and findings of the AFR survey.

Turning to unorthodox practices in CEMAC and WAEMU countries, the following analysis provides a stocktaking exercise of these procedures, based on existing sources of information (IMF staff reports, TA reports, and PEFA reports). Table 1 lists several types of such procedures used in the 14 countries from 2006–18 and identifies the mandatory controls that were bypassed or violated. The Table reports four useful observations:

  • First, unorthodox practices carry different names in the different countries, but they show clear similarities in their practice.

  • Second, the procedures remain widespread among all countries in both regions.

  • Third, all countries have had recourse to unorthodox procedures at some time.

  • Fourth, unorthodox practices have evolved; some procedures were abandoned (often under pressure from donors) or replaced by others with slightly different modalities that left the overall unorthodox mechanisms fundamentally unchanged.

Table 1.

List of Reported Unorthodox Procedures in CEMAC and WAEMU Countries, 2006–18

article image
Source: IMF TA and Staff reports (2006–2018)

Expenditures executed under unorthodox procedures are difficult to identify and quantify since they do not have corresponding approved appropriations and formal commitments, circumvent regular controls, and were never registered in fiscal accounts. In some cases, the analysis in this section is able to provide some estimates of the underlying expenditure executed under these procedures based on technical assistance efforts provided by IMF staff.

Unorthodox procedures in CEMAC and WAEMU countries can be classified into three categories:

  • Any procedure leading to a cash disbursement that bypasses one or several budget execution steps and controls and that is not followed by proper ex post regularization; this category accounted for around 75 percent of practices observed in CEMAC and WAEMU countries and reported in Table 1

  • Any procedure bypassing the regular budget time limit (annual) and approved ceilings

  • Any extrabudgetary procedures circumventing budget law authorizations and unity, as well as revenue/spending ceilings.

Box 4 provides examples of the first and most common unorthodox procedure: cash payment without either supporting documentation or ex post regularization. The procedure’s precise name varies across countries, but the mechanisms remain fundamentally the same. In contravention of all the rules of budget execution, the cash release (payment) would often take place before the implementation of all required spending procedures. The public accountant would make the disbursement (for example, payment to a contractor or transfer to a bank account outside the TSA) without supporting documentation (for example, invoices, payment orders, and proof of rendered service or goods delivery) and, in most cases, without even knowing the underlying budget appropriation. Although while ex post disbursement regularization should occur promptly, in practice, it seldom happened. A similar practice in Chad is the Dépenses avant ordonnancement (DAO), which is an emergency spending procedure extensively used since 2016; it allows the regular controls to be temporarily bypassed to expedite payments. This procedure encompasses all spending that did not go through the normal spending chain, regardless of whether the spending was executed relative to an approved budget appropriation that was not regularized, or in the absence of such an appropriation. In the case of the execution of spending without a budget appropriation, regularization requires either an adjustment in a revised budget or a ministerial order to reallocate funds within the budget (IMF 2018). The DAO was a source of more than one-third of all payments in one semester in 2017 (IMF 2017). Recent IMF reviews show that this procedure intensified in 2021 in the context of COVID-19 emergency spending and accounted for about 20 percent of primary expenditure (IMF 2021)

Examples of Unregularized Cash Advances in Selected CEMAC and WAEMU Countries

The simplified procedure for the release of funds in Burkina Faso allows for the commitment of spending, the verification, and the payment order to happen simultaneously. Under this procedure, the funds are typically released for payment before the service is rendered or the goods delivered. This procedure is mostly used for (but not limited to) capital expenditure. Proof of good usage of the funds is required during later stages, to be followed by clearance; however, current practices show significant delays in ex post regularization. Moreover, the controls performed by the financial comptroller and the public accountant are not required, thereby weakening verifications to ensure the correct use of the funds. Amounts related to the execution of investment expenses through this procedure have been significant. In 2013 and 2014, nearly 40 percent of approved capital expenditure was executed by the release of funds, of which a substantial amount was pending regularization.

Unregularized cash advances in Central African Republic (CAR) intervene directly in the accounting phase. The disbursement is made by the accountant without a budgetary commitment, and no supporting documents are produced afterward. The rules specify that regular cash advances should be used only under specific and urgent circumstances; in practice, however, they are commonly used outside of these conditions. Disbursements made under cash advances are often partially regularized or are not recognized at all. Beyond the significant fiduciary risk for the government, cash advances diminish budget and accounting credibility; a significant part of the expenditure remains recorded in suspense accounts, the regularization of which could remain pending for many years.

Unregularized imprest accounts in Mali are overused. In principle, imprest accounts should be used for payments of “minor” operating expenses (equal to around CFA 10 million francs in Mali), and they are supposed to meet punctual and exceptional needs. However, in practice, the amounts paid from these imprest accounts are sometimes very high. TA reports reported 18 imprest accounts amounting to over FCFA 100 million; the special imprest account of the Ministry of Defense indicated monthly expenditure of more than FCFA 2.3 billion in FY2013. Moreover, the supporting documents are not systematically presented; this (1) reduces the quality of the general accounts of the State, as well as (2) reduces the possibility for controls. This practice could even present opportunities of mismanagement, such as the use of advances for noneligible expenditure and fictitious purchases, as reported by TA reports.

Sources: IMF TA reports (June 2016), (May 2018), and (March 2014); PEFA report (2018).

Box 5 shows examples of the second type of unorthodox procedure: circumventing the budget’s time limit (annuality principle) and approved ceilings. This practice also bypasses the regular checks and controls, because payments are not based on appropriated expenditure in the budget. It also creates a payment commitment for the budget in future years, either because budget appropriations are rolled over from one year to another and are disconnected from cash management, or because future appropriations are already committed regardless of Parliamentary approval or future spending needs or priorities.

Examples of Irregular Multiyear Expenditure in Cameroon and Senegal

Third-party accounts in Cameroon were initially introduced with the adoption of the TSA. They aimed at separating the funds from public institutions and agencies or investment projects from the rest of the deposits. Over the years, this initial purpose deviated to allow managers to do the following: (1) carry over budget authorizations (for example, credit transferred to agencies), and (2) bypass the lengthy chain of checks and controls through an accelerated procedure. This practice complicates cash management by violating the annuality of the budget, especially when cash resources are limited. The accumulated budget authorizations carried over from previous years reached about 1,000 billion FCFA (5 percent GDP) at the end of September 2017. Since then, authorities adjusted their legal framework to prohibit such practices.

Comfort letters in Senegal are a practice initially used in connection with multiyear contracts prior to the adoption of multiyear commitment authorization.12 The Minister of Finance typically guaranteed (by means of a formal letter) to the Director of the Central Procurement Directorate that budget resources to pay contractors will be appropriated as needed in future budget laws. However, this practice evolved over the years, with the Minister of Finance providing similar assurances (also through formal letters) to directors of commercial banks and requesting financing for a given operation to be repaid out of future budget appropriations. This procedure is problematic because committing public expenditure that has not yet been appropriated in a budget to be repaid from future revenue narrows the available fiscal space for future regular budget appropriations. This procedure bypasses all controls because (1) committed public expenditure is not yet approved in the budget, and (2) amounts indicated in comfort letters are not recorded in IT systems or public books, although they could be significant. For instance, total commitments in connection with one comfort letter amounted to 104 billion CFA (1 percent GDP) in 2018.

Source: Fouad M., and others, “Fiscal Transparency Evaluation for Senegal (July 2018) and TA report (February 2018).

Box 6 provides examples of the third unorthodox category: extrabudgetary spending. This type of spending takes place outside of the budget and totally avoids controls, reporting, and audit. In fact, neither the revenue nor the expenditure related to extrabudgetary transactions is included in the annual budget law; consequently, neither are the amounts related to the actual execution of the underlying transactions. The latter occur completely outside the scope of authorized officers and the regular expenditure chain and IT systems, thereby circumventing regular controls and PFM rules. They also circumvent budget unity and expenditure ceilings. Extrabudgetary operations are difficult to capture—not only due to the lack of reporting and traceability but also due to their very variety of arrangements (for example, extrabudgetary units, extrabudgetary funds and entities, extrabudgetary accounts).

Examples of Extrabudgetary Procedures in Guinea Bissau and Gabon

Extrabudgetary expenditure based on specific revenues in Guinea-Bissau is a practice used by some ministries (education and health) to execute current operating expenses financed by their own collected revenues, in contravention of existing PFM rules. Furthermore, part of the internal revenue collected by these ministries is deposited in accounts with private banks, which are outside of the control of the Ministry of Finance. This procedure, often resulting from cash management inefficiencies in the country, affects budget unity; the corresponding revenues and expenses are not subject to fiscal reporting and control. As a result, there is no guarantee that the funds are effectively use for the planned purposes.

Special account in Gabon is a generic term for financial operations that benefit some public entities (including the central government) in which revenues were collected and allocated without being either authorized in the annual budget law or recorded in financial statements. Part of this revenue was collected by the beneficiary administration (for example, audit products, fines, and tax disputes) and deposited into “special accounts” opened by the treasury. Another part of this revenue was passed on directly to relevant entities to finance personnel or current operating expenses. Furthermore, these operations contributed to reducing available fiscal space without ensuring the regularity or effectiveness of the related expenditure.

Source: IMF TA reports (January 2014 and October 2014).

Potential Fiscal Impact of Unorthodox Procedures

This section examines whether the recourse to unorthodox procedures in CEMAC and WAEMU countries was associated with potential adverse macro-fiscal outcomes, including the accumulation of arrears, deficient fiscal reporting, and corruption vulnerabilities.

The analysis relies on some indicators from the PEFA.13 The analysis uses the following PEFA indicators:

  • Compliance with payment control rules and procedures. Compliance This indicator is used as a proxy to capture the quality of expenditure controls, on the assumption that inadequate expenditure controls are associated with unorthodox procedures.14 This indicator assesses to the extent to which payments are compliant with regular payment procedures and whether exceptions are properly authorized in advance and justified. Lower scores indicate a high rate of dependence on unorthodox procedures.

  • Arrears. In the absence of a reliable long-term time series on the stock of arrears for this sample of countries, this analysis uses the PEFA indicator on the stock of expenditure arrears to measure the extent to which there is a stock of arrears and how large it is.15

  • Fiscal reporting. This analysis uses the PEFA indicator on financial data integrity, which assesses the extent to which treasury bank accounts, suspense accounts, and advance accounts are regularly reconciled and how the processes in place support the integrity of the financial data.16 Large unregularized spending amounts from unorthodox procedures often lead to significant outstanding balances of suspense accounts, indicating inadequate reconciliation of accounts.

In addition, the analysis also uses a control of corruption measure to capture perceptions of the extent to which public power is exercised for private gain, including both petty and grand forms of corruption, as well as “capture” of the state by elites and private interests.17

The analysis presents associations based on correlations between proxies for unorthodox procedures and the indicators gauging their fiscal impact. The main findings are:

First, the use of unorthodox procedures as a result of bypassing the regular expenditure control procedures was associated with a larger stock of arrears in both CEMAC and WAEMU countries (Figure 2). This observation confirms previous findings from other analyses; these analyses find that arrears in both regions occur as invoices and payment orders due on regular and approved expenditure presented during the payment stage, and that they remain pending because the available cash had been used to meet unplanned unorthodox payments. This finding confirms those of other studies that also associated PFM weaknesses in WAEMU countries with the inadequate recording of arrears (David, Nguyen-Duong, and Selim 2022).

Figure 2.
Figure 2.

Dependence on Unorthodox Procedures and Arrears in CEMAC and WAEMU Countries

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Source: Authors, based on PEFA scores for Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Republic of Congo, Côte d’Ivoire, Gabon, Guinea Bissau, Mali, Niger, Senegal, and Togo.Note: r is the correlation coefficient; the p-value indicates that the correlation is significant.

Second, recourse to unorthodox procedures is associated with poor fiscal reporting and weak integrity of fiscal data in CEMAC and WAEMU countries (Figure 3). In fact, a longstanding track record of using unorthodox procedures was associated with large amounts of transactions recorded in suspense accounts pending regularization, pointing to inadequate monitoring and irregular clearance of these accounts. Bouley, Fournel and Leruth (2002) generally find that the overdevelopment of suspense accounts has also played a crucial role in weakening the accounting recording in both regions and relate this to unorthodox procedures.

Figure 3.
Figure 3.

Correlation Between Unorthodox Procedures and Fiscal Reporting Integrity in CEMAC and WAEMU Countries

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Source: IMF staff, based on PEFA scores of Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Congo Republic, Côte d’Ivoire, Gabon, Guinea Bissau, Mali, Niger, Senegal, and Togo.Note: r is the correlation coefficient; the p-value indicates that the correlation is significant.

Another key reconciliation exercise to ensure the accuracy and reliability of fiscal reporting is the reconciliation of the stock and flows of government debt. Debt decomposition exercises show the flow of net incurrence of debt liabilities of the change in the stock of government debt. The difference between both is referred to as “stock-flow adjustments” (SFAs).

The accounting decomposition shows that SFA averaged about 1.1 percent of GDP per year between 2013 and 2019 in CEMAC (Figures 4a and 4c) countries and about 1.5 percent in WAEMU countries (Figures 4b and 4d).18 This finding means that the average increase in the debt-to-GDP ratio by this amount every year was explained by factors other than the recorded fiscal deficit (excluding the effects of economic cycle and exchange rate dynamics). This also suggests that these discrepancies have contributed to debt accumulation in CEMAC and WAEMU countries between 2012 and 2019. Although large budget deficits still constituted the main contribution to debt accumulation (70 percent), other factors accounted for the remaining debt accumulation of approximately 30 percent (Figure 4).

Figure 4.
Figure 4.
Figure 4.

Public Debt and Below-the-line Operations in CEMAC and WAEMU Countries, 2012–19

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Source: Authors estimates based on authorities’ data.

This disconnect between deficit and debt is partly due to SFA.19 In some cases, SFA operations could be identified for some countries; they were found to result from one-off benign adjustments that are consistent with good PFM practices, such as clearance of outstanding arrears in Côte d’Ivoire in 2018 and the adoption of a broader debt perimeter in Senegal in 2017. Overall, in WAEMU countries, they also reflected a change in public guarantees that are included in the stock of debt but are not accounted for in the budget. The change in guarantees amounted about 0.2 percent of GDP, suggesting that the SFA, excluding the change in guarantees, would be about 1.3 percent of GDP (instead of 1.5 percent).

Notwithstanding these benign operations, some components of SFA were due to “other” debt-creating flows, most likely reflecting some below-the-line operations that bypassed regular expenditure controls and other dysfunctions in PFM systems (Féler and Simard 2019; Versailles 2018). For example, these operations in some WAEMU countries were associated with expenditure that did have a corresponding budget allocation and took the form of extrabudgetary spending including prefinancing schemes (Benin and Togo), quasi-fiscal operations related to SOE-related fiscal transactions and roll over of past budgetary appropriations in Senegal (David, Nguyen-Duong, and Selim 2022; Versailles 2018). An analysis of country-specific factors confirms that between 2013 and 2019, below-the-line operations were particularly large in Gabon (close to 6 percent of GDP), Togo (2.3 percent of GDP on average), Senegal (2.2 percent of GDP), and to a lesser extent, Benin and Côte d’Ivoire (1.4 percent of GDP), Central African Republic and Cameroon (1.2 percent of GDP) (Figures 4e and 4f).

Third, unorthodox procedures may have potentially increased vulnerabilities to corruption in CEMAC and WAEMU countries. Figure 5 reports a significant correlation between the use of unorthodox procedures and the control of corruption. Vulnerabilities to corruption increase as a result of unauthorized spending that circumvented regular controls. Incentives for corruption are often more prevalent in defense and security spending, which have recently increased significantly in some CEMAC and WAEMU countries severely affected by the violence and conflict in the Sahel region.20 Concerns for confidentiality are often used to subject related expenditures to less transparency, to channel them outside regular expenditure controls and oversight, or to more generally execute them while bypassing regular budget procedures. For the same reasons, military spending may also benefit from derogatory procurement procedures with limited transparency and competition requirements.21

Figure 5.
Figure 5.

Correlation Between Unorthodox Procedures and Control of Corruption in CEMAC and WAEMU Countries

Citation: IMF Working Papers 2022, 148; 10.5089/9798400215681.001.A001

Source: IMF Staff, based on PEFA scores and control of corruption index of Benin, Burkina Faso, Cameroon, Central African Republic, Chad, Congo Republic, Côte d’Ivoire, Gabon, Guinea Bissau, Mali, Niger, Senegal, Togo.Note: r is the correlation coefficient; the p-value indicates that the correlation is significant.

Policy Recommendations

There is a need to reduce recourse to these unorthodox procedures and contain their macro-fiscal damage in CEMAC and WEAMU countries. At least five measures can be taken to reduce recourse to unorthodox procedures and ensure that exceptional procedures are used according to their intended purpose.

First, establish clear legal frameworks that restrict the use of exceptional and other simplified procedures to very limited and well-identified cases. National legal and regulatory frameworks should do the following: (1) clearly specify the exact rules applicable to such procedures; (2) present a comprehensive listing of all existing simplified and exceptional procedures; (3) list the conditions and circumstances (including the type of emergencies) for their use; (4) present the powers of the various actors and authorities (for example, the Minister of Finance); (5) provide details on the ex post regularization, as well as all accounting rules required for these procedures correlated to the risk emanating from their use; and (6) list the responsibilities and sanctions applicable in the event of noncompliance with procedures. Moreover, the legal framework should require annual budget laws to set maximum appropriation amounts that can be executed through exceptional procedures, as well as more frequent reporting on budget execution, for example, though quarterly reports.

Second, strengthen internal and external audits. Audits are crucial to prevent and detect the use of unorthodox procedures, and, most importantly, to apply corrective actions. Internal audits aim to ensure that internal controls comply with procedures and regulations and identify potential failures and propose remedies. If an excessive use of exceptional procedures is detected, the underlying causes of this practice must be understood and eliminated. Treasuries could publish a list of unregularized expenditure in their financial reports, for instance, in the special appendix, as provided for in the regional directives.22 In parallel, the external audit system can impose sanctions when the use of exceptional procedures does not comply with the legal framework. In the Francophone external audit system, Supreme Audit Institutions (SAIs) have the jurisdictional authority to impose penalties to public accountants when they undertake irregular expenditure operations. To be able to perform this task, the SAI needs, however, to be an independent institution with adequate financial resources and technical capacities and to receive timely fiscal reports from the Government. In several CEMAC and WAEMU countries, the SAI is still part of the Supreme Court, as per the constitution. The PFM directives in both regions require member states to grant full autonomy to their SAIs. However, progress toward this objective has been slow and uneven across countries. The independence of SAIs in both regions remains weak, as shown in the last World Bank’s Supreme Audit Institutions Independence (SAI) Index (2021).23

Similarly, the implementation of internal controls should be more broadly supported by independent and accountable authorities. In this regard, codes of conduct and/or ethic rules ensuring the accountability of staff are critical. In addition, procedures manuals aiming to operationalize the legal provisions of laws and setting concrete tasks and responsibilities for relevant actors should be produced and disseminated.

Third, leverage the use of digital solutions. Stronger and more effective internal controls are necessary to ensure the adequate use and ex post regularization of exceptional procedures. Several suggestions can be relevant in this regard. A better use of IFMIS and digital solutions can prevent deviations from appropriations or bypassing of any controls during the different phases of the expenditure chain. In Gabon, the requirement that suppliers must present a purchase order issued by the FMIS to be paid has proven to be an effective control at the commitment stage (Box 7). Other system-embedded controls could include an automatic alert to signal when a ceiling on authorized exceptional spending is reached or when the timeframe to regularize an exceptional spending has lapsed.

Moving to Regular Spending Procedures in Gabon

Gabon used a simplified procedure for all expenditures of goods and services with no limits between 2013 and 2017. Under this practice, commitment and verification stages were merged, and commitment controls were temporarily bypassed to be applied after service delivery. However, without strict ex post regularization, transactions were not appropriately recorded in fiscal accounts. As a result, commitments were made in excess of the government’s payment capacity, and arrears accumulated from 2013–16. These arrears—identified through an audit launched in 2016—were estimated at about 3 percent of GDP.

The Gabonese Government implemented the following measures to restore sound spending procedures and prevent the creation of new domestic arrears:

  • Removing the simplified procedure to strictly limit the use of exceptional procedures and to make it mandatory to produce a purchase order to issue payment for the goods supplied or services delivered (decree issued in May 2017)

  • Upgrading the FMIS to allow the systematic issuance of purchase orders at the commitment stage of the expenditure chain (July 2017)

  • Communicating (through a press release) to inform all suppliers of the new procedure

  • Training of the relevant directors (financial affairs and budget controllers)

Source: Gabon Staff and TA Reports.

Fourth, streamline budget execution processes and improve budget credibility. Doing this would be instrumental in reducing the recourse to certain types of unorthodox procedures by creating quick disbursement procedures for urgent budgeted spending. Regarding budget execution processes, the devolution of budget execution authority to line ministries, as prescribed by the regional directives, would contribute to smoother processes and strengthened controls. Under the current practice, all payment mandates or requests for payments are routed to the Ministry of Finance, which does not always have the capacity to expedite execution at the same time for all departments. This excessive centralization or concentration of financial authority explains the cumbersome nature of budget execution in the countries. In such a context, urgency inevitably becomes the standard, which encourages the recourse to exceptional and then-to-become unorthodox procedures. Paradoxically, this situation is also detrimental to the implementation of sound control procedures. The concentration of execution and control authorities in the hands of one single entity creates the conditions for noncompliant practices. As the departments in charge of executing the budget and implementing the controls (for example, financial control, public contracts authority, budget, treasury, and accounting) are part of the same entity, under the Ministry of Finance, hierarchical authority, the exercise of independent control is difficult, if not impossible, due to political and hierarchical interferences. In practice, controls from public accountants or financial comptrollers are often rushed or simply abandoned based on direct orders from the ministers and/or their direct supervisors. Devolution of budget authority to line ministries would smooth budget execution and refocus the Ministry of Finance assignment on impartial control. Regarding budget credibility, the recourse to in-year exceptional procedure also results in weak budget preparation. Largely incremental, most budget laws in CEMAC and WAEMU countries include unrealistic or ill-designed spending envelopes leading to in-year changes and providing incentives to bypass those constraints.

In addition, the lack of multiyear perspectives as a basis for budget preparation limits the capacity to implement and control budget execution over several budget exercises, encouraging circumvention strategies, especially for investment. Many of the unorthodox procedures listed in Table 1 (such as comfort letters in Senegal or prefinancing in Togo) are the direct effects of these limitations. The systematic use of multiyear budget preparation tools—such as Multiyear fiscal framework (MTFF) or Multiyear budget framework (MTBF), associated with the implementation of multiyear commitment authorizations as prescribed in the regional directives—as well as the use of use of in-year programming instruments such as procurement plans, commitment plans, and cash flow plans, are critical to the capacity to address these issues and improve budget credibility.

Fifth, develop tools to mitigate the impact of exceptional procedures on cash resources. The TSA is a pivotal tool that can help manage cash pressures and prevent paying irregular expenditures. By centralizing all liquidities on a single account at the Central Bank, the TSA provides the treasury with an instant and comprehensive view of all available deposits. It also enables centralized oversight of cash management and contributes to strengthening internal controls. The adoption of a TSA at the national level is a requirement by both the CEMAC and WAEMU PFM directives. All countries are currently rolling out and/or consolidating a TSA, albeit at uneven levels of progress. Cash flow plans linked to commitments and procurement planning, as illustrated by recent improvement in cash management in Niger (Box 7), are another tool for better monitoring the use of exceptional procedures initially not included in the forecasts. These plans also help prioritize payments of regular expenditures instead of unorthodox spending.

Improving Cash Management in Niger

Niger initiated the implementation of a TSA in 2016. The number of public entities’ accounts open with commercial banks has been divided by close to 10, from 1,568 initially to 924 at end-May 2018, and finally to 207 by end-December 2018 (including 20 for departmental agencies that benefit from a temporary derogation from the TSA); the total balance has been reduced to 1.8 billion FCFA (compared to 68.5 billion at end-December 2016). New templates for enhanced commitment and procurement plans have been designed to ensure a better alignment between spending needs and available liquidities, as well as to facilitate the shift to active cash management, thereby preventing the absorption of available liquidities at the detriment of regular expenditures.

Source: Niger TA Reports.

References

  • Alesina, A., and R. Perotti. 1999. “Budget Deficits and Budget Institutions.” In Fiscal Institutions and Fiscal Performance, edited by J. Poterba and J. von Hagen. Chicago: University of Chicago Press and National Bureau of Economic Research.

    • Search Google Scholar
    • Export Citation
  • Allen, R. 2013. “Challenges of Reforming Budgetary Institutions in Developing Countries.” In Public Financial Management and Its Emerging Architecture, edited by M. Cangiano, T. Curristine, and M. Lazare. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Allen, R., and D. Tommasi, eds. 2001. Managing Public Expenditure: A Reference Book for Transition Countries. Paris: Organisation for Economic Co-operation and Development.

    • Search Google Scholar
    • Export Citation
  • Bouley, D., J. Fournel, and L. Leruth. 2002. “How Do Treasury Systems Operate in Sub-Saharan Francophone Africa?IMF Working Paper 02/58, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Cangiano, M., T. Curristine, and M. Lazare. 2013. Public Financial Management and Its Emerging Architecture. Washington, DC: International Monetary Fund.

    • Search Google Scholar
    • Export Citation
  • Cebotari, A., J. Davis, L. Lusinyan, and others. 2009. “Fiscal Risks: Sources, Disclosure, and Management.” Fiscal Affairs Departmental Papers 0901, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Central African Economic and Monetary Community (CEMAC). 2011. Regional Directive 01/11/UEAC-190-CM-22 on Budget Laws (and related manual).

    • Search Google Scholar
    • Export Citation
  • Checherita-Westphal, C., A. Klemm, and P. Viefers. 2016. “Governments’ Payment Discipline: The Macroeconomic Impact of Public Payment Delays and Arrears.” Journal of Macroeconomics 47: 14765.

    • Search Google Scholar
    • Export Citation
  • Corbacho, A., and T. Ter-Minassian. 2013. “Public Financial Management Requirements for Effective Implementation of Fiscal Rules.” In The International Handbook of Public Financial Management, edited by Richard Allen, Richard Hemming, and Richard Potter. Palgrave Macmillan.

    • Search Google Scholar
    • Export Citation
  • Dabla-Norris, E.; R. Allen, L.F. Zanna, T. Prakash, E. Kvintradze, V. Lledo, I. Yackovlev, and S. Gollwitzer. 2010. “Budget Institutions and Fiscal Performance in Low-Income Countries.” IMF Working Paper 10/80, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • David, A., A. Nguyen-Duong, and H. Selim. 2022. “Strengthening the Regional Fiscal Framework in the WAEMU.” IMF Working Paper WP/22/49, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Doe, L., and S. Pattanayak. 2008. “Financial Control in African Countries.” PFM Technical Guidelines Note No. 5, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Féler, A., and D. Simard. 2019. “The WAEMU Surveillance Framework: Reforms to Foster Public Debt Sustainability.” Selected Issues Paper, IMF Country Report 19/91, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Flynn, S., and M. Pessoa. 2014. “Prevention and Management of Government Expenditure Arrears.” IMF Technical Notes and Manuals 14/01, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Fouad, M., Nour Chamseddine, Fabien Gonguet, Bruno Imbert, Philippe Samborski, and Adrien Tenne. 2018. “Fiscal Transparency Evaluation for Senegal.” IMF Technical Assistance Report, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • IMF. 2021. “Chad: Request for a Three-Year Arrangement under the Extended Credit Facility—Staff Report.” IMF Country Report No. 21/267, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • IMF. 2020. “Policies to Support the People During the COVID-19 Pandemic.” Fiscal Monitor Report. Washington, DC: International Monetary Fund, April.

    • Search Google Scholar
    • Export Citation
  • IMF. 2019. “Domestic Arrears in Sub-Saharan Africa: Causes, Symptoms, and Cures.” Regional Economic Outlook: Sub-Saharan Africa. Washington, DC, October.

    • Search Google Scholar
    • Export Citation
  • IMF. 2017. Tchad: Améliorer l’exécution de la dépense pour limiter le recours aux procédures dérogatoires. IMF TA report. International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • IMF. 2015–2020. Staff reports for CEMAC and WAEMU countries. www.imf.org.

  • IMF. 2011–2020. Technical Assistance reports for CEMAC and WAEMU countries. Unpublished.

  • Khasiani, K. Y. K., A. Mfombouot, and A. Singh. 2020. “Budget Execution Controls to Mitigate Corruption Risk in Pandemic Spending.” Fiscal Affairs Department Special Series on COVID-19, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Khemani, P., and B. Wiest. 2016. “How to Check Integrity of Fiscal Data.” How to Notes 4, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Lando, S., S. Rosa, and G. Suc. 2019. “A Regional Approach to Enhancing Governance and Reducing the Potential for Corruption.” CEMAC Selected Issues Paper, IMF Country Report No. 19/2, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Lienert, I. 2003. “A Comparison between Two Public Expenditure Management Systems in Africa.” IMF Working Paper 03/2, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Moussa, Y. 2004. “Public Expenditure Management in Francophone Africa: A Cross-Country Analysis.” IMF Working Paper 04/42, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Omitoogun, W., and E. Hutchful, eds. 2006. Budgeting for the Military Sector in Africa: The Processes and Mechanisms of Control. Stockholm International Peace Research Institute and Oxford University Press.

    • Search Google Scholar
    • Export Citation
  • Pattanayak, S., 2016. “Expenditure Control: Key Features, Stages, and Actors.” IMF Technical Notes and Manuals 16/02, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • PEFA. 2011–2019. Reports for CEMAC and WAEMU countries. www.PEFA.org.

  • Radev, D., and P. Khemani. 2007. “Commitment Controls.” Fiscal Affairs Department Public Financial Management Technical Guidance Note, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Versailles, B. 2018. “Sources of Fiscal Risks in WAEMU Countries.” Selected Issues Paper of IMF Country Report 18/107, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Wang F., R., T. C. Irwin, and L. K. Murara. 2015. “Trends in Fiscal Transparency: Evidence from a New Database of the Coverage of Fiscal Reporting.” IMF Working Paper 15/188, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • West African Economic and Monetary Union (WAEMU). 2009. Regional Directives 01/2009/CM/UEMOA on Budget Laws (and related manual).

Annex I. Monitoring Indicators to Control Unorthodox Procedures in IMF-Supported Programs of CEMAC and WAEMU Countries

article image
article image
1

PFM is concerned with how governments manage the budget in its established phases—formulation, approval, and execution. It deals with the set of processes and procedures that cover all aspects of expenditure management in government (Cangiano and others 2013).

2

More broadly, inadequate controls are a prevalent feature of budget execution in many low-income countries. This prevalence is due to inadequate sanctions related to the noncompliance with expenditure controls, as well as poor overall policy coordination, inefficient planning, and weak capacity in human resources and information systems (Allen 2013).

3

In contrast, excessive recourse to unorthodox procedures in other countries occurs when expenditure controls are absent and ineffective (Dabla-Norris and others 2010).

4

Budget stages include authorization of expenditure, commitment to a future obligation to pay, verification of purchase or of goods and services, payment authorization, and, finally, payment.

5

A “commitment” is defined as an obligation to effect a future payment, subject to the fulfillment of certain conditions (contractual or otherwise) (Radev and Khemani 2007).

6

A suspense account (compte d’imputation/d’affectation provisoire) temporarily records transactions until comprehensive information becomes available for proper records (Khameni and Wiest 2016). They are useful for transactions for which there is uncertainty about the correct account in which to record it.

7

Fiscal risks refer to the possibility of deviations in fiscal variables from what was expected at the time of the budget or other forecast (Cebotari and others 2009). >

8

In a context of poor oversight and accountability for the budget, government officials may be able to manipulate the rules for personal gain which could harm budget credibility and fiscal outcomes (Alesina and Perotti 1999).

9

Articles 23 and 14 of the respective CEMAC and WAEMU directives on the Government Financial Operations Table (TOFE).

10

For instance, the control of regularity is redundant since it is applied both by financial controllers and public accountants (Pattanayak 2016). Normal procedures were also perceived to be slow to permit payment for goods and services (Lienert 2003).

11

In the specific context of COVID-19, IMF staff released a series of notes to emphasize the legal grounds for the use of these exceptional procedures and provide guidance for their recourse to ensure transparency, as well as an appropriate level of control. Those guidelines can be found on the websites of the IMF regional technical assistance centers for central and western Africa (AFRITAC Central and AFRITAC West).

12

Multiyear commitment authorization (autorisation d’engagement) is a new procedure introduced under CEMAC and WAEMU regional directives allowing authorizing officers to commit funds for capital expenditure, over several years, under a multiyear ceiling granted by the budget law. Ceilings and the use of commitment authorizations are tracked in the IT system and reported in fiscal reports.

13

PEFA assessments provide a framework for assessing and reporting on the strengths and weaknesses of PFM to measure performance. Typically, PEFA indicators (and underlying scores) help assess the extent to which PFM systems, processes, and institutions contribute to the achievement of desirable budget outcomes: aggregate fiscal discipline, strategic allocation of resources, and efficient service delivery.

14

PEFA scores rely on a four-point ordinal scale: A, B, C, or D. A score of A reflects the most advanced standards or a greater compliance with PFM practices, a score of B reflects a level of performance ranging from good to medium, C reflects a level of performance ranging from medium to poor, and a score of D indicates either that a process or procedure does not exist at all or that is not functioning effectively. The analysis is based on PEFA scores of CEMAC and WAEMU countries from 2011–16. PEFA scores have been converted to numerical values (A = 4, B = 3, C = 2, and D = 1).

15

The PEFA indicator on arrears assesses the extent to which there is a stock of arrears (a score of A reflects a stock of expenditure arrears that is no more than 2 percent of total expenditure in at least two of the past three completed fiscal years; a score of B reflects a stock that is no more than 6 percent; a score of C reflects a stock that is no more than 10 percent, and a score of D reflects a performance that is less than the C score.

16

The PEFA indicator on fiscal reporting integrity includes four dimensions that measure the following: (1) the frequency of bank reconciliation for all active central government bank accounts at aggregate and detailed levels, (2) the frequency of reconciliation of suspense accounts and that these suspense accounts are cleared in a timely way—no later than the end of the fiscal year unless duly justified, (3) the frequency of reconciliation of advance accounts and that all such accounts are cleared in a timely way, (4) whether access and changes to records are restricted and recorded, and whether they result in an audit trail; and whether there is an operational body, unit, or team in charge of verifying the integrity of the financial data.

17

The control of corruption index is obtained from the Worldwide Governance Indicators in 2020 (http://info.worldbank.org/governance/WGI/).

18

In cumulative terms, SFA amounted to 8 percent of GDP in CEMAC countries and 10 percent of GDP in WAEMU countries over the whole period.

19

The SFA is the outcome of an accounting debt decomposition exercise; it captures the increase in public debt that cannot be explained by the fiscal deficit, adjusted for exchange rate valuations and GDP growth.

20

Countries affected include Mali, Niger and Burkina Faso from Chad in CEMAC and the WAEMU countries. All four countries experienced significant security spending between 2010 and 2018.

22

Articles 27 and 29, respectively, of WAEMU and CEMAC directives on the chart of accounts.

24

Dépenses avant ordonnancement (DAO) are defined as all expenditures that do not go through the normal spending procedure and relate to cash disbursement bypassing controls without ex post regularization as described in section II-B. A standard procedure entails a chain that includes the commitment (“engagement”), the validation (“liquidation”), the authorization of payment order (“ordonnance”), and the cash payment. There are two categories of DAOs. The first category consists of DAOs that are made relative to a credit line in the budget. These DAOs can be regularized (that is, recorded in the correspondent line of the budget) without difficulties. The second category consists of DAOs that are made regardless of the existence of a credit line in the budget. Their regularization requires either an adjustment in the revised budget, namely, Amended Financial Law (LFR), or a ministerial order to transfer an existing credit allocation.

  • Collapse
  • Expand
Unorthodox Expenditure Procedures in CEMAC and WAEMU Countries
Author:
Mr. Bruno Imbert
,
Hoda Selim
,
Ms. Gwenaelle Suc
, and
Qing Zhao
  • Figure 1.

    Controls Applied at the Various Stages of the Expenditure Chain

  • Box Figure 1

    Fiscal Measures: Jan 2020-Sept 2021

    (% of GDP)

  • Box Figure 2

    Liquidity Support Measures: Jan 2020-Sept 2021

    (% of GDP)

  • Figure 2.

    Dependence on Unorthodox Procedures and Arrears in CEMAC and WAEMU Countries

  • Figure 3.

    Correlation Between Unorthodox Procedures and Fiscal Reporting Integrity in CEMAC and WAEMU Countries

  • Figure 4.

    Public Debt and Below-the-line Operations in CEMAC and WAEMU Countries, 2012–19

  • Figure 5.

    Correlation Between Unorthodox Procedures and Control of Corruption in CEMAC and WAEMU Countries