Photo: Arnold Lee (

A combination of low wages and perverse incentives motivate civil servants to supplement their income with per diem payments connected to participation in seminars and workshops. The abuse of travel and training- related payments results in excessive expenditures and in a distortion of incentives that can frustrate development efforts. Three main factors contribute to facilitating this type of practice: insufficient controls, management (dis)incentives, and donors? role and attitudes. Strengthening controls alone is unlikely to curb this kind of abuse, the culture of "per diem hunting" needs to be changed as part of a broader reform of the civil service. Coordination among development partners can also contribute to preventing per diem abuse.

Public service provision

The provision of quality public services is a central task of government, which requires

a competent civil service. Therefore, states and donors attach great importance to enhancing the skills of their staff through training. Governments implement training programmes to build capacity or to introduce new approaches and strategies, often in collaboration with, and with funding from, development partners. As a result, significant resources are allocated to seminars and workshops for civil servants. These activities should contribute to increasing the capacity of civil servants and to enhancing bureaucratic efficiency.

However, the assumption that training leads to improved skills and better performance is open to questioning. Whether workshops and seminars achieve their objectives depends largely on the way they are organized and on the participants’ receptivity to learning. Participants are frequently paid per diems, that is, non-salary daily subsistence allowances to cover extra expenses incurred when travelling for work-related purposes. These allowances are normally set at standardized rates and are payable according to stipulated rules and regulations. Research suggests, however, that training programmes are often organized and attended for the mere purpose of obtaining extra income (Cooksey et al. 2007; Cooksey 2010; Jansen 2009a, 2009b; Vian 2009).

While per diem payments are supposed to be strictly compensatory, they can become a form of additional salary (Policy Forum 2009). In countries where salaries are generally low, these extra payments can amount to a significant proportion of civil servants’ total income. As a result, civil servants may be more interested in obtaining these allowances than in the content of the activities.[i] Indeed, the incentives inherent in travel compensation systems may prevent governments from reaching the full potential of their training programmes--or worse, the training programmes themselves may become another source of inefficiency. 

Use and abuse of travel compensation in three countries

Case studies showing widespread irregularities in Tanzania, Ethiopia and Malawi illustrate the adverse effects of per diem abuse and fraud.[ii] The public sector in all three countries has adopted standardized travel reimbursement systems, including rates and rules for determining compensation according to rank and destination; standardized claim forms for reimbursement; and rules regarding supporting documentation, such as original receipts. In all the countries, management approval is required prior to travel, and systems of internal and external audit are in place. Formally, their travel compensation systems resemble international standards. However, they work remarkably different in practice, which suggests that contextual factors are important. At one extreme, the Ethiopian system has become a disincentive to travel (see Box 1).

Conversely, the travel compensation systems of both Malawi and Tanzania provide strong incentives for making per diems a form of supplementary income. In Malawi, the government does not provide many allowances. Yet, the government’s travel related expenditure has reached unsustainable proportions in a country where the tax base is narrow (see boxes 2 and 3). 

Box 1. Ethiopia: Insufficient per diem rates

In Ethiopia, the official per diem rates are far below the real costs of meals and accommodations, which may be three to four times higher than the maximum rate (Council of Ministers 1993, 2001, 2008). As a consequence, civil servants are reluctant to travel unless reassured that the number of travel days reported will be artificially inflated to match actual expenses. Otherwise, needed official travel would not take place. The practice is hence condoned and defended by government officials. Uncertain budgets and high inflation compound the problem. The practice provides room for managerial discretion as well, which opens the door to favouritism on economic or party/political grounds, under the conditions of a repressive one-party state (Human Rights Watch 2010; Freedom House 2011).

Box 2. Tanzania: A multitude of allowances

The situation in Tanzania is the opposite of that in Ethiopia. The travel compensation system is generous both in terms of the rates established and of the rules that govern their application. Equally important, however, is the wide range of discretion in the assignment of allowances, intended to incentivize performance, including special duty allowances and sitting allowances (paid, for example, for attending meetings). Until recently, these allowances were tax exempt, which added to their appeal. Many civil servants think of travel reimbursements as just another source of extra income. In some cases, civil servants can obtain per diem payments for more days than justified or for meetings not attended, as well as reimbursements for expenses not incurred.[iii] These practices should not be considered mere cases of stretching the rules, but are tantamount to small-scale fraud.

Box 3. Malawi: One ‘clean salary’ but still over-expenditure

When reforming the staff compensation system in 2004, Malawi incorporated within a single ‘clean salary’ many of the extra allowances previously paid to civil servants. Travel -related per diems are currently the only allowances that remain at the managers’ discretion. Although this was a commendable reform, it induced some managers  to offer employees extra incentives to stretch excessively the travel compensation system. For example, a manager satisfied with a department’s performance might offer the whole department two days of travel compensation as a bonus, even when no trips or training has taken place.[iv] The travel -related allowances constitute a large proportion of civil servants’ total income, and clearly assume the character of a salary supplement rather than of reimbursement for expenses. Ultimately, these unwarranted disbursements aggravate the country’s budget deficit.

Why are per diems abused?

Travel compensation systems and the way they are applied in practice are not trivial issues. They have a serious impact on work environment, staff motivation and work priorities, and hence on the quality of public service delivery. In the three countries studied, the irregularities and abuse of per diem payments are a significant problem. Civil servant salaries are low in these countries. For lower level civil servants, poverty is a driving force of the individuals’ pursuit of per diems, but there is no indication that the level of malpractice decreases at higher levels of government (Fjeldstad 2003). From a government perspective, it is also clearly an organizational problem. A large proportion of staff payments is allocated informally, i.e., in a discretionary way and at variance with the public interest. Thus, there is typically less control over payments, and their results are less fair than proper formal payments would be. Why does per diem abuse persist even when the problem is well known by the government? Research suggests three main factors: (1) insufficient controls, (2) management (dis)incentives, and (3) donors’ role and attitudes.

Insufficient controls

The manipulation of travel expenses should be seen a form of fraud. At first sight, it is the result of weak controls. This judgment presumes a principal-agent characterization of the problem, in which the main reason for the problem’s persistence is lack of information by the principal (a manager) about the activities of the agents (subordinate staff) (Rose-Ackerman 1975; Mookherjee and Png 1992). Whether the principal is a manager within a given institution or a separate controlling institution, such as a ministry, a principal-agent problem could, in principle, be solved by adopting better monitoring efforts and/or incentive schemes connected to staff performance. There are good reasons for considering the lack of controls an important part of the problem. In all three countries, the staff, as well as the managers, could forge receipts and produce fake seminar attendance sheets. Furthermore, in many cases it was possible for civil servants to attend training sessions just long enough to obtain the offered allowance. In fact, some managers would show up unexpectedly at seminars just to confirm how many employees were actually present.

Management (dis)incentives

Still, the lack of strong controls is not a full explanation. In many cases, managers knowingly allowed the per diem irregularities, or accepted the opportunistic organization of seminars primarily intended to maximize per diem payments. They often misused the system to offer extra payments to their staff, and inflated their unit’s budgets to create financial space for the manipulation of the travel compensation system. Managers have a strong interest in retaining discretionary authority over the travel compensation system. The managers’ authority to make such decisions allows them to demand not just improved performance but also loyalty from their subordinates. Similarly, many staff members are keen to attend seminars or travel for the extra payment. Managers often resist the reform of the travel compensation systems to establish more stringent controls on the grounds that these would reduce flexibility. By keeping the system ‘flexible’ the managers not only reinforce their capacity to reward loyalty, but also increase their opportunities to extract additional personal income. While some flexibility is needed in the interest of efficiency, controls are required to limit potential abuses.

Donors’ role and attitudes

Donor agencies play an important role in designing and financing development programmes and training events. In all the countries studied, there is a de facto opportunity for civil servants to benefit more in donor-financed programmes, partly because the rates are higher and partly because the allowances are sometimes seen as a means to attract participants to seminars. One consequence is that when similar training sessions are replicated by other agencies, there is an economic incentive for staff to participate repeatedly in them to obtain more payments instead of prioritizing other tasks.

Nevertheless, the attempts to harmonize donor rates and systems with those of recipient countries have generally failed, partly because donors often compete for participants. They outbid each other to attract the most relevant or senior staff from government. Failing to harmonize their policies, some donors have opted for using government systems, which in effect means relinquishing control. Others, however, demand detailed information and control by independent auditors. Generally, donors consider per diem abuse an insignificant problem. They fail to appreciate the distortionary effects it can have on development programmes; even worse, they fail to see the aggregate magnitude of the problem.

Policy implications and recommendations

Policy implications and recommendations

Some recommendations that can help address in the short term the issues related to insufficient controls are:

  • The information about rates, rules and the amounts spent on travel expenses should be transparent and publicly available;
  • The rates should be standardized regardless of staff status and should be compensatory only, as distortionary incentives are created by rates set at multiple levels; and
  • Training programmes and travel expenses should be subject to ex post external auditing in addition to internal control routines.

However, the increase of controls should not be taken too far (such as demanding original receipts for every expenditure item) as this would only result in huge transaction costs. Efficiency hinges on a balance between trust and control. Discretion is a prerequisite for flexibility, and complete control is not cost-effective.

Strengthening controls by itself is, in any case, unlikely to curb per diem abuse in the long term. Governments and donors should promote steps to make more fundamental reforms that change the embedded culture of “per diem hunting” and contain the misuse of per diem pay by managers. This problem needs to be addressed as part of a broader civil service reform, addressing pay and compensation structures as well as performance management. Ultimately, when justified, the additional monetary per diem benefits should be incorporated into formal wages. While states whose revenue base is narrow may not be able to raise salary levels appreciably in the short run, reducing per diem abuse would allow for a more efficient allocation of scarce resources. Efforts should be made to change the widespread acceptance of per diem allowances as a source of extra pay. Civil servants at all levels should understand that travel compensation systems are only for the reimbursement of travel expenses--and they need to behave correspondingly. Managers should be instructed not to use per diem allowances as a device to incentivize performance or for buying loyalty from subordinates.

Development partners should also coordinate to prevent irregularities. In particular, they should not foster the inflation of daily allowances offered for the purpose of attracting participants to training activities, as this practice leads to replication and waste. Where the context allows, donors may also choose to align their per diem rates with government rates to avoid civil servants chasing higher benefits from donor- funded programmes. Finally, donors should also ensure that training programmes are subjected to independent auditing, at least until the perception of the purpose of these payments has changed and the systems are reformed.



[i] For a discussion of how payments can distort intrinsic motivation, see Frey 1997.

[ii] For additional details, see Søreide, Tostensen and Skage 2012. We are grateful to local consultants, Mekonnen Gebeyehu in Ethiopia, Tom Kavinya in Malawi, and Samuel Wangwe in Tanzania, as well as to Michael Davies and Odd-Helge Fjeldstad for useful advice and comments.

[iii] For similar findings, see Vian et al. 2011.

[iv] In an extreme case that was investigated for fraud, one single person claimed per diem allowances for 1,057 days in one calendar year. The case was part of a larger fraudulent scheme in the Ministry of Agriculture and Food Security, which involved 20 civil servants.


Cooksey, B., L. Anthony, J. Egoe, K. Forrester, G. Kajembe, B. Mbano, I. von Oertzen, and S. Riedmiller. 2007. Management of Natural Resources Programme, Tanzania TAN-0092 Final Evaluation. Oslo: Norwegian Agency for Development Cooperation (Norad Collected Reviews 1/2007).

Cooksey, B. 2010. “Allowances in Tanzania”. A study conducted for Norad. Mimeo.

Council of Ministers. 1993. Council of Ministers’ Directive on the Daily Allowance of Government Employees. Addis Ababa: Council of Ministers.

Council of Ministers. 2001. Revised Council of Ministers’ Directive on the Daily Allowance of Government Employees. Addis Ababa: Council of Ministers.

Council of Ministers. 2008. Council of Ministers’ Directive on the Determination of the Daily Allowances for Foreign Business Trips. Addis Ababa: Council of Ministers.

Fjeldstad, O.-H. 2003. “Fighting fiscal corruption: lessons from the Tanzania

Revenue Authority”. Public Administration and Development 23 (2): 165–75.

Freedom House. 2011. Country Report. Ethiopia. Accessed 2 October 2011.

Frey, B. S. 1997. Not just for the money: An economic theory of motivation. Cheltenham: Edward Elgar.

Human Rights Watch. 2010. Development without freedom: how aid underwrites repression in Ethiopia. New York: Human Rights Watch.

Jansen, E. 2009a. Monitoring aid: lessons from a natural resources programme in Tanzania. Bergen: Chr. Michelsen Institute (U4 Practice Insight 2009:1).

Jansen, E. 2009b. Does aid work? Reflections on a natural resources programme in Tanzania. Bergen: Chr. Michelsen Institute (U4 Issue 2009:2).

Mookherjee, D. and I. P. L Png. 1992. “Monitoring vis-á-vis investigation in enforcement of law”. American Economic Review 82 (3): 556–65.

Policy Forum (in collaboration with Twaweza). 2009. Reforming allowances: A win–win approach to improved service delivery, higher salaries for civil servants and saving money. Dar es Salaam: Policy Forum (Policy Brief No. 9).

Rose-Ackerman, S. 1975. “The economics of corruption”. Journal of Public Economics 4: 187–203.

Søreide, T., A. Tostensen and I. A. Skage. 2012. Hunting for per diem: The uses and abuses of travel compensation in three developing countries. Oslo: Norwegian Agency for Development Cooperation (Report 2/2012).

Vian, T. 2009. Benefits and drawbacks of per diems: do allowances distort good governance in the health sector? Bergen: Chr. Michelsen Institute (U4 Brief 2009:29).

Vian, T., C. Miller, Z. Themba and P. Bukuluki. 2011. Perceptions of per diems in the health sector: evidence and implications. Bergen: Chr. Michelsen Institute (U4 Issue 2011:6).