Civil Service Reform and Decentralization

Site Index
Main Page
Issues in Program Design

Contributor: World Bank
Author: Decentralization Thematic Team
Contact: Jennie Litvack

Civil Service Reform and Decentralization

Civil service reform is usually a supporting strategy for more general decentralization in government operations or service delivery. One does not decentralize the civil service as an end in itself -- one does so in order to provide services better, manage resources more efficiently, or support other general outcome goals. The civil service as a whole can be seen as one of the main instruments with which the government fulfills its obligations. In the context of decentralization, this tool must often be reshaped in order to perform a new set of duties efficiently, equitably, and effectively. Reform of the civil service, therefore, is the process of modifying rules and incentives to obtain a more efficient, dedicated and performing government labor-force in newly decentralized environment.

This note will first discuss the various civil service issues that sectoral or general decentralization strategies raise. It will then focus on various reform priorities to cope with the changes decentralization can bring.

How does Decentralization affect the Civil Service?

Civil services at all levels of government need a capable, motivated, and efficient staff in order to deliver quality services to its citizens. When civil service functions and structures are decentralized, existing bureaucratic patterns must be reorganized as roles and accountability are shifted. Decentralization thus intensifies the need for capable staff and increases the importance of capacity-building programs.

The process of decentralization:

Disperses power, both geographically and institutionally: Decentralization inevitably changes the location of power and jobs. Movement geographically or across tiers of government is often impeded by issues related to statute, prestige and poor labor mobility. In the Eastern European transition economies, for example, de-legitimation of the central state and the emergence of representative government at local and intermediate levels of government has complicated human resource allocation. Incentive programs and mechanisms for inter-post mobility, which compound the costs of decentralization, may be required in order to introduce flexibility.

Creates new responsibilities for inexperienced actors: Decentralization creates more opportunities for local autonomy and responsiveness to more specialized constituencies, but it also gives subnational governments more room to fail if specific steps are not taken to build local technical and managerial capacity.

Can disperse scale economies/expertise groups: The need for specialized personnel is related in part to the size of the territory covered by the entity. Below a certain size, it might be counterproductive or cost inefficient to have specialists or technical personnel. There are methods which can be used to address this issue, one of which is to allow in the context of the decentralization schemes the possibility of empowering local self-governments units to form associations and pool their resources in order to cover activities requiring specialized personnel.

Introduces more levels into the state: Decentralization, especially political decentralization creates a class of government workers which, based on the specific information which they receive (feedback from their constituencies) may have different preferences than workers at the next higher level. This divergence in views and convictions can create conflict within the civil service that will require mechanisms to manage effectively.

Creates a tension between local autonomy and national standards: Decentralization relaxes national control and creates the potential for more regional variation in civil service conditions. Some room for variation allows regions the flexibility to hire a civil service that matches a community's needs and budget constraints. National salary, eligibility, and performance standards can ensure consistent quality, but they can also lead to personnel expenditures (especially for locally administered education and health sectors) beyond some local capacities; grant transfer systems will need to take different financing capacities into account in these and other types of mandated expenditures. The Philippines, Indonesia, and Pakistan are examples of decentralized states with essentially uniform terms and conditions of service for government employees in different regions.

Can increase administrative costs: Creating additional layers of government is an expensive proposition, and while the central government - in the best of cases- might reduce its role and shed personnel in the context of decentralization, empirical evidence suggests that these workers are often reabsorbed by local governments. There is thus no net change in public sector employment. In the worst of cases, central government employment remains unchanged, while local government employment grows.

Civil Service Reform to Support Decentralization

The main questions in assessing the civil service reform priorities parallel those in more general decentralization policies: Under what conditions does one deconcentrate or devolve human resource management or organizational responsibilities to lower tiers of government? What requisite capacity does one need at various levels to make a system work?

The twin tasks of building local capacity and adjusting to the changes in intergovernmental coordination needs can be daunting even when budgets allow comprehensive training and all stake-holders support the reforms. The more frequent realities of budget constraints and mixed support, however, practically ensure that large-scale civil service reform will be a long drawn-out, expensive process that does not keep up with the pace of service or sector decentralization.

Building Local Capacity

Local (or at least sub-national) capacity is one of the most important factors creating a well-functioning decentralized civil service. In countries where local institutions already exist, the challenge will be to reinforce them institutionally and legally as well as to strengthen their personnel management capacities. In places where local government institutions are embryonic or exist only at an informal level, the institutional and legal framework will have to be created before any type of reform of the administration is undertaken.

The degree of local capacity determines the kind of human-resource management strategies that will be feasible and desirable. Decentralization of human resource management is more likely to succeed in cases where lower-level authorities have the financial and managerial ability to set competitive compensation packages and salary levels that will attract local talent. In these cases, the flexibility advantages of allowing local governments the to set hiring levels might outweigh the risk of increasing inter-regional inequalities. Where talent and skills are lacking at the local level, a unitary hiring system might be preferred to ensure that the necessary skills are present locally in all regions. In these cases where the center retains more control over human resources, caution should be paid to ensure that the management options of local stake-holders are not curtailed.

Adjusting to Decentralization: General Guidelines for Country-Specific Strategies

  • The legal framework should clearly define responsibilities and standards. The creation of a strong legal framework- to address issues related to financing and reporting, to determine the type of control mechanisms (especially financial) that are necessary and who is accountable for them, to evaluate hiring practices and compensation schemes as well as address issues related to the procurement of public works - must be a priority in any reform effort to ensure sound utilization of public resources and minimize corruption.
  • Consistency and transparency gain support. On matters of staffing, compensation or oversight of local administration, and most importantly in the delivery of services, it is very important to ensure that there is transparency and that changes in the administration (and therefore the civil service) are not seen as an instrument to disenfranchise some groups or favor another.
  • Reporting mechanisms need to be clear and precise. Clear reporting procedures will need to be put in place vis--vis higher levels of government (central government, in the case of regional administrations, for example) and horizontally, vis--vis other government agencies at the same level. In the medium and longer-term, audit courts can be a useful regulatory mechanism. Transitions from the existing system to new systems have to be carefully planned to avoid conflict between new reporting arrangements and enduring mechanisms.
  • Channels for citizen-civil servant communication need to be created. By including more citizens in the process of monitoring civil service performance, decentralization creates more opportunities for friction between civil servants and citizens. Harassment by private interest groups can prevent honest and dedicated civil servants from performing their duties, while civil servants can use their positions to threaten citizens. These tensions can be avoided by relatively quick and inexpensive methods and structures for redressing grievances, whether these come from civil servants or from the citizens.

  • Training should contribute to the formation of new working relationships. In addition to building local capacity, training can be a tool for creating personal networks among various levels of government, regions, or types of government workers. One recommendation, for example, might be to train career civil servants and local politicians together to insure that they better understand what is expected of them and what they can expect from each other.
  • All levels of government should be encouraged to define and plan for the types of workers they will need in order to carry out new responsibilities. In the short term, these sorts of rough plans substitute for the computerized establishment management capacity and human resource management staff that so many countries lack and can help eliminate duplicate workers, unnecessary hires, and other expensive mistakes. At the very least, they can be an exercise in longer-term planning and role definition.


Decentralization can be a way of improving access to services, tailoring government actions to private needs, and increasing the opportunities for state-society interactions. Subnational governments, however, will only be effective when they have access to the necessary human and financial resources to undertake the services they have been conferred.

Civil service reform -- both capacity building and adjusting to decentralization -- addresses the first of these requirements. There is fairly widespread agreement that capacity-building at all government levels is an essential component of decentralization. The sequencing and priority levels of training -- whether to train local or central governments first, for example -- depends on the country itself, although the subnational governments have generally been the first to be trained to accept their new responsibilities. There is less agreement over how to deliver the appropriate human resources package to the appropriate levels of government and how to coordinate human resource management across and between levels of government. The decision to decentralize or retain central control over human resource management -- recruiting, hiring, salary-setting, etc. -- depends heavily on the existing degree of subnational capacity. The suggestions above outline some general coordination mechanisms, but the specific institutional arrangements for ensuring a consistent, efficient civil service must react to the kinds of institutional changes that decentralization has brought.