Using basic mathematics to fight corruption and bad practices: the realities of day-to-day practice in CameroonBy Marcellin Djeuwo, Head of the Risk Management Unit, Cameroon Customs
One of the difficulties in reforming African public administrations derives from the absence or weakness of proper instruments to monitor public action. There is a direct link between bad practices, corruption and a lack of results. For Customs, an improvement in governance automatically brings about an increase in budgeted revenue collections and trade facilitation.
The purpose of this article is to show how a public administration like Cameroon Customs, as part of a performance measurement exercise involving Customs clearance agents, has sustainably achieved a certain level of efficiency in border controls, by combating bad practices using basic mathematical techniques, such as addition, subtraction, multiplication or division of figures extracted from its computerized Customs clearance system.
In parallel, this article explains how some Customs officials, such as frontline inspectors, whose behaviour should be overseen, attempt to bypass the system, and describes the measures taken by the performance measurement implementation team to prevent them from doing so.
Instilling a results culture
In 2010, Cameroon adopted a system of performance contracts for appraising its staff, using statistical data together with various indicators. The appraisals are based on arithmetic, with statistical data that is simple to verify and easy to understand.
Indeed, in most African countries, poor communication, among other things, means that there is a serious asymmetry of information between managers and operational services in public administrations – managers do not always have a suitable means of ascertaining whether the reforms adopted are actually being implemented in practice.
If we consider this situation from the perspective of a Customs administration, it is easy to see that a Director General cannot reasonably know, in a relatively short term, whether, for example, risk analysis guidelines are being complied with, whether checks are being carried out properly, or whether an inspector is acting efficiently to combat fraud.
In the same vein, in his communication entitled ‘L’expérience du Cameroun en matière de gestion axée sur les résultats’ [The experience of Cameroon in results-based management], the Cameroonian Minister for the Public Service acknowledged very aptly that the country’s system “has led to a paradox: that of an administration with a performance deemed to be well below average, but whose staff always have grades well above the average.” This simply reflects the fact that, at the time, the staff appraisal system was lacking in relevance.
In an environment of this kind, where effort, or simply the absence of effort, is not structurally visible, how is it possible to instil a results culture in border controls? Answering this question is the challenge that has been taken up by Cameroon Customs, an institution itself suffering from serious negative bias among the general public.
Thus, since 2010, the Cameroon Customs Administration has been using figures extracted from its ASYCUDA information system to monitor how the public Customs clearance service is being carried out at the borders. Specifically, this means measuring the behaviour of officials against carefully selected performance indicators, with the aim of seeing them develop in a positive sense.
By persuading frontline inspectors, in line with pre-set targets, to gradually reduce administrative bad practices that can be conducive to corruption, or, on the other hand, encouraging them by the same token, to adopt better practices that make the working environment less prone to certain afflictions, Cameroon Customs has been able to improve revenue collections and trade facilitation.
The procedure consists of setting quantified targets for frontline inspectors, calculated from computer data on the basis of median values determined from historical data. This median is the reference starting point for an inspector’s monthly and quarterly appraisal. The inspector is rewarded if his or her performance exceeds the median and penalized if it is below the median. The thresholds at which an official may be regarded as very good, good, average or poor are set out in a table. All this data is included in the contracts signed between the inspectors and the Director General.
The need to adapt performance indicators
When the performance measurement procedure was introduced, some inspectors tried to find ways around it. Precise analyses and impact calculations are, therefore, carried out on an ongoing basis, in order to adapt the performance indicators to changes in the environment, or the behaviour of officials. This operation is essential, as may be seen from the following illustrative examples:
- For instance, certain inspectors meticulously studied the computerized system for the random assignment of Customs declarations. Given that, in a Customs office, declarations are allocated first to those with a low workload, they proceeded systematically to deal with the declarations they received, so as to draw in more and more files for themselves. In this way, they were seeking to increase not only their chances of being regularly involved in disputed claims, and thereby to boost their pay (an inspector who uncovers an instance of fraud receives a share of the monetary penalty imposed on the offender), but also their opportunities for contact with users (risks of corruption). A simple workload calculation revealed that certain inspectors were handling five times more declarations than their colleagues, or even more. A second calculation showed that, over a whole year, some inspectors were handling more than 80% of files from the same importers or forwarding agents, betraying the existence of some collusion between them. They were penalized, and a new indicator was introduced with the aim of balancing workloads.
- Not all the Douala Customs offices were brought under contract at the same time. Hence, certain rogue operators diverted traffic normally placed with offices under contract to offices without a contract, so as to benefit from preferential treatment by certain officials serving there. Similarly, some offices under contract engaged in competition to attract traffic to themselves improperly. It is worth mentioning that in the port of Douala, the Customs offices specialize in particular types of traffic: one office for Customs clearance of goods imported in containers, another for goods imported in bulk or otherwise packaged, a third for goods in transit, a fourth for oil company imports, etc. The criteria for allocating files to the different offices are well known, and have been incorporated into ASYCUDA. But some officials found ways to manipulate the computer system to their own advantage, so that specific officials continued to deal with the files of certain forwarding agents, irrespective of the designated Customs office. Well-targeted analyses and calculations revealed this bad practice, described in the Cameroon Customs Administration as ‘transhumance’ of declarations. The culprits were penalized, and the practice ceased. A periodic monitoring exercise, again based on the analysis of figures extracted from ASYCUDA, is carried out to rule out a resurgence of this practice.
- It has also become clear that certain indicators very quickly become inoperative, because the people under contract have turned them to their advantage. For instance, an indicator had been provided to measure the processing time for declarations. Certain inspectors had adopted the habit of directing their declarations into the data circuit known as the ‘green channel.’ When this type of message was transmitted to ASYCUDA, it signified that the declaration had been released, although in actual fact the user had not received the release warrant. In parallel, this kind of operation relieved pressure on the file processing time, and the officials concerned, then had sufficient time to ‘negotiate’ with the importer. When they were appraised in the light of the computer data, these inspectors appeared to be very efficient, although in reality they were not. All that was needed to unmask the deceit was to calculate, on the basis of onsite information, the rate of declarations placed in the ‘awaiting green channel’ status by inspectors. A new indicator was added to the contracts, and the practice gradually disappeared in favour of the number of declarations actually assessed.
The improvement of individual performance makes sense only if it has a positive effect on the performance of the service as a whole. Thus, on a quarterly basis, in parallel to the individual staff appraisals, an assessment is carried out on the impact of contracts on the overall performance of the Customs offices concerned, to ensure that the individual results and the results for the service as a whole are in line. Correlation analyses have often raised the alarm and indicated avenues for rooting out bad practices.
The need for dialogue to identify bad practices
The supervision procedure presented above came to be seen as necessary, not only to detect and correct abuses, but also to anticipate more serious problems that could have discredited the performance measurement process. It thus became clear that it was essential to bring together all the stakeholders in the procedure from time to time, so as to examine progress and the difficulties encountered in implementing performance measurement: those who monitor the contracts from day to day, inspectors under contract and their managers, inspectors who have had experience of contracts, and people outside the administration, including Customs’ regular partners, such as the WCO. With this in mind, a workshop was held from 15 to 19 February 2016 in Cameroon, in order to pursue this issue through the sharing of contributions from all those involved.
One of the major lessons learned on this occasion was that some inspectors take advantage of the shortcomings in the wording of certain indicators to develop bad practices. For instance, inspectors may reroute declarations from the facilitation channel to the control channel, but an indicator was introduced to calculate the percentage of these declarations giving rise to an adjustment. This precaution was designed to protect users against pressure that might be exerted on them by an inspector, using the authority to reroute declarations to his or her advantage. However, data analysis showed that certain officials made do with token adjustments, particularly as no minimum amount was required for the adjustment to be deemed valid. This practice made the precaution ineffective. To deal with the problem, a new indicator was introduced, setting a minimum acceptable adjustment level.
Another indicator obliges inspectors to attain a certain number of adjustments as a percentage of declarations received. To maintain plenty of room for manoeuvre, allowing them to circumvent this requirement, some inspectors knowingly made adjustments to certain kinds of declarations that were not to be adjusted, since they related to importers with special status, such as diplomatic missions or the government. They knew that the computer system would credit them with this kind of adjustment, and that they would therefore be regarded as high-performing, whereas in fact they were burdening with charges people who were not required to pay them. Again from onsite feedback, a study backed up by figures brought the deceit to light, and the system was corrected.
The team responsible for monitoring the indicators also reviewed all the current indicators to check, by means of small calculations, that they still constituted relevant responses to the problems identified. This exercise made it possible to modify certain indicators, in particular those relating to enforcement against fraud, which, through their wording, should prioritize adjustments resulting in actual payment. New indicators were introduced (such as one on average return per declaration adjusted), and existing ones were corrected.
Almost 10 years ago, Cameroon Customs introduced a ‘figures culture’: figures are used as a performance measurement unit for people and organizational structures. This step made it possible to set up a results-based policy in a public administration. It met with internal and external resistance, and still does, but it also provides the means to overcome such resistance.
In the final analysis, the procedure continues to be a strategic steering tool for Cameroon. Almost 90% of declarations are assessed on the day on which they are lodged. Cameroon Customs is no longer, as it used to be, one of the 10 most corrupt administrations in the country according to a Transparency International ranking. Despite an exponential increase in the budget target assigned to the administration (approximately 114% in 12 years), the Directorate General of Customs has regularly achieved the results set.
The performance contract mechanism is still work in progress. A highly motivated team dedicated full-time to this task, responsible for the ‘research and development’ segment, is absolutely essential to anticipate problems that could hamper the implementation of this mechanism in the long term. It is all about constructing an environment conducive to the implementation of the performance measurement policy, and for that endeavour it is, of course, essential that political will should exist.
 Centre Africain de Formation et de Recherche Administratives pour le Développement [African Centre for Administrative Training and Development Research], Morocco, January 2013.
 Op. cit.