Follow-the-Money
The Underbelly of Public Finance Systems: Corrupt Budgeting
In this part of the follow-the-money corruption cycle we expose the underbelly of public finance systems by discussing how budget systems work in the three nexus areas: 1) facilitating corruption; 2) striving for efficiency; and 3) political manoeuvring. We also assess the scope for different types of corruption based on the Ang framework that splits corruption between elites and non-elites, and corruption that delivers some form of exchange for the corrupt action, or if it’s just a form of theft. The Ang framework is also adapted to include whether the corrupt activity is a form of petty or grand corruption.
Facilitating Corruption
To put it bluntly, corrupt budgeting means that people basically pay to get their budgets or their preferred policies. When the budget is like an auction[1], resources are allocated based on a willingness to pay. Budgets and policies get handed out to the highest bidder. For example, budget authorizations are provided for particular projects and policies that will benefit particular vested interests that paid for the “right to win” the project or have the policy approved.
Essentially, bribes are paid for promised policies or project allocations. Project-based appropriations or a program of subsidies or grants for special interest groups is a common mechanism. Under this form of grand corruption, the promise for the project is made in the budget process and can occur irrespective of how strong a procurement process is. A World Bank Procurement Benchmarking Study of 2017 for example had many notionally highly corrupt countries receiving outstanding procurement benchmark scores. There are many ways to provide inside information to ensure a promised project wins a competitive procurement process.
Most policies that aren’t linked to a project or program still have fiscal implications. The classic example is a tax break (i.e. a tax expenditure) for a particular group paid for by someone in that group. These would normally all need to go through a budget process. To allow a tax break to go through even though there is limited evidence that it is efficiency enhancing can be done through rudimentary or sophisticated means. Rudimentary means is just a budget decision without reason or evidence. Sophisticated mechanisms can use for example paid for analysis that shows that the policy is justified through biased analysis, even though it isn’t efficient, effective or warranted. It’s worth noting that most independent assessments are not really independent, with obvious conflicts of interest.
These people, for example, can be members of parliament, ministers, contractors, officials, and their agents. The size of bribes here can be massive, representing a significant proportion of the original appropriation. There are various red flags of widespread corruption in the budget debating process within the legislature, one of which is the provision of resources to projects that have not followed due processes, like independently verified economic evaluation, fiscal impact analysis and value for money assessments.
When resources are allocated this way, rather than through an evidence-based assessment of the effectiveness of fiscal policy options, corruption is likely to be flourishing throughout the parliamentary and budget formulation system.
[1] See a Development Practice Note on “The Blight of Auction-Based Budgeting: What is it and how can we deal with it?”, slide deck on “Corrupt Budgeting” and “Corruption in the Budget Process”.
It is very hard to get agreement on a consistent methodology and therefore easy for vested interests to undermine efforts to try. Politically, corruption is very sensitive as it can undermine the confidence of the public in their governments. It is why autocratic models of government and corruption often go together. This makes it hard to have an honest and open discussion about the true costs of inefficiency and corruption.
AFI directors spent years in Afghanistan trying to raise these issues internally, with government officials and donor partners putting billions of dollars into governance. To no avail, there was little appetite to challenge powerful vested interests who were stealing the money. Everyone accepted that corruption was bad, but with no specific way to measure the cost, no serious change was possible. We believe this significantly contributed to the eventual outcome.
We have concluded that developing a robust and consistent approach to costing inefficiency and corruption and making it public is a necessary step towards a more constructive discussion about what needs to be done.
In 2020, we published a paper on something we call the “Follow-the-Money Corruption Cycle: Revealing National Accountability Failures”[1]. It was very frank about how corruption can work in any public finance system.
It showed how the very systems that are designed to control corruption and promote government efficiency can end up facilitating corruption.
In the extreme, public finance systems can be used as the foundation for formidable cultures of corruption within the very institutions entrusted by the public to help deliver good governance.
By being clear on where and how corruption works in the public finance system we believe that corruption can be better targeted where it matters most to strengthen public accountability and improve government efficiency and effectiveness.
The figure below provides an overview of the “follow the money corruption cycle”.
[1] See How Corruption Works in the Public Sector—One Easy Lesson, IMF PFM Blog, March 2020.
Focus First on the
Follow-The-Money Corruption Cycle
The Underbelly
of Public Finance Systems
The “follow-the-money corruption cycle” shows how all parts of the budget cycle work and where the risks are for corrupt activities. The scale of the problem becomes clearer as each corruption risk point is highlighted throughout the cycle. To keep things simple the following describes the basic corrupt objective at each one of the eleven (11) points of the “follow-the-money corruption cycle”.
The model uses a risk-based approach to quantify the costs of corruption and efficiency losses caused by weak public finance systems. The methodology adopted is the same we have used in various Fiduciary Risk Assessments (FRAs) but is adapted to enable assessment of risks and losses at all points in the “follow-the-money corruption cycle”. The methodology also builds on methodologies used elsewhere including on estimating costs of corruption in procurement.
Being Frank About How Corruption Can Work
Within Public Finance Systems
- Starting at the budget you pay to get your budget, including under an auction like arrangement, rather than having the budget allocate resources based on policy / evidence[1].
- At the commitment phase you pay to get access to the allotment (or approval to spend the budget), rather than using the commitment system to manage cash and ensure appropriations (legal authorities) are not breached.
- At the procurement stage you pay to win contracts, rather than using it to support competition to keep costs down and quality up.
- At the contract management stage you pay to change contracts to make them more profitable, rather than ensure contract terms are complied with.
- At the verification and payment stage you pay to falsely verify that goods and services were delivered on time and to specification, and then pay again to get paid, rather than ensuring payments are made in accordance with contracts.
- At the audit stage you pay for audit irregularities to be cleared, whether those irregularities are real or otherwise (i.e., made-up to get more bribes or to punish), rather than audit providing assurances that financial information is accurate and believable and auditing being a reliable mechanism to support institutional learning.
- At the personnel and payroll stage you pay for positions and pay to retain positions through family connections or outright bribes or cashing in old debts and favours, rather than promoting merit-based recruitment and retention.
- At the accounting stage you pay to facilitate licit and illicit movements of funds through the accounting systems and to cover tracks, rather than ensuring accounting, classification and reporting rules are followed.
- At the revenue collection stage you pay to facilitate favourable treatments of what you owe or what you are paying for (including for illegitimate/illegal goods or services) or speed up intentional or unintentional slow administrative processing, including in service delivery areas, rather than ensuring revenue is assessed and collected in accordance with the rules.
- During balance sheet management phases you pay to secure favourable treatments of assets and liabilities – e.g. have debts written off and acquire assets cheaply, rather than ensuring assets and debts are managed well.
- If then, at any point in the cycle evidence of malfeasance emerges and results in referrals to any or all the accountability institutions (like the police, anticorruption commission, prosecutors, inspectors-general, and auditors) you just pay again to clear the allegation, investigation, prosecution, finding or sentence, rather than ensuring checks and balances are working as intended and the rule of law is respected.
In a country with systemic corruption throughout the “follow the money corruption cycle”, it can clearly be seen that there is very little public money left to finance good quality fiscal policy aimed at improving the efficiency and effectiveness of core public services like health, education, public works and defence.
The approach takes the “follow the money corruption cycle” structure of public finance sub-systems to provide clarity of where and how corruption works in government systems and ultimately to estimate the costs of corruption and efficiency losses at the sub-system level.
[1] See also Laing, 2017, “The Blight of Auction-Based Budgeting: What is it and how can we deal with it?” http://effectivestates.org/wp-content/uploads/2017/09/Auction-Based-Budgeting-WEB.pdf
A simplified version of the formula used to quantify financial impact of weak PFM systems … or in other words … the costs of corruption and inefficiency is above. There are three parameters used in the costing algorithm:
- Quantified development or fiduciary risks (r) representing a good enough proxy for likelihood of risks materialising, including inefficiency and leakage (the current model uses development risk);
- Impact factor (i) is based on assessed importance of the public finance sub-system to efficiency and leakages; and
- Financial flows through the financial sub-system (f). The financial flows and pecuniary significance rates for every country were drawn from Government Finance Statistics (GFS) data on components of fiscal flows and stocks for every country. Likelihood (or risk) multiplied by impact factor multiplied by financial flows in the sub-system gives an approximate financial estimate for inefficiencies and corruption in the sub-system. The formula is effectively a simple linear equation where r and i form the slope parameter with variables f and c.
The Three Steps to Estimate the Losses
Step 1
Quantify Risk
Quantify risk of inefficiency, corruption, and system weakness – We use a modified index of indexes approach for Development Risk – the risk of not achieving development goals due to weak public finance systems. Our methodology is here.
Step 2
Apply Impact Factors
Apply impact factors for the relevant part of the public finance system. These are drawn from field work and confidential interviews
Step 3
Measure Amount of Funds Flowing through the Systems
Use GFS Data and economic classification categories for fund flows then apply costing algorithm
The results can be shown in various ways. The simplest way is just to report the losses in terms of dollar value. These can be in nominal or real terms (i.e. adjusting for the effects of inflation). To be able to compare losses between countries, the dollar value is not that helpful as it doesn’t account for differences in size of the country or economy. Other ways to show losses include representing the dollar value of losses as percent of: i) total government expenditure; ii) total government revenue; and iii) of GDP, or it can be shown on per capita terms. While there is plenty of fiscal and risk data to calculate losses, standard annual results would normally be limited to having sufficient risk and fiscal data available in the same year. Consequently, period averages are used to deal with this problem. The annual and period average results can be shown in league tables that rank countries.
The Costs of Inefficiency and Systemic Corruption (COSIC) Index is the first version of such a league table. AFI are currently piloting it to assess its usefulness to reformers and other stakeholders. An interactive version of the COSIC Index is provided below. Users can choose how to rank countries by the different loss measures (by clicking on the relevant heading), and can also drill down on certain groups of countries and years and can also choose what level of government to draw the fiscal data from. There are different views that can be scrolled through (using the arrow buttons at the bottom). The first view is the standard league table of 6 different measures of losses plus the average development risk for the country and period. The second provides four frequency distribution charts of development risk, showing distribution by income status, region, year and fragility. The third provides heat tables of the seven measures summarised by income group, region and fragility. The fourth is an alternative layout of the first.
