Updated December 1997

Food and Agriculture Organization of the United Nations

United Nations Capital Development Fund

International Fund for Agricultural Development

German Agency for Technical Cooperation

Swiss Agency for Development and Cooperation

World Bank

Rome
16-18 December 1997
Technical Consultation on Decentralization
Documentation

Decentralization in Indonesia
Part 1: The Indonesian decentralization policy and the District Autonomy Pilot Programme

1. Law No. 5/1974 as basis for decentralization and local autonomy

Decentralization in Indonesia is primarily based on the law No. 5 from 1974. The law determines the status and the form of Local Governments in Indonesia and defines the basic structure of the political administrative system in the region. It covers therefore not only the autonomous functions of the Local Governments but also the deconcentrated structures and tasks.

There are two autonomous levels of government below the central level in Indonesia: the provinces (Tingkat I, Propinsi) and the districts (Tingkat II, Kabupaten/Kotamadyas). The sub-districts (Kecamatan).are part of the 2nd-level government in their deconcentrated function and have no own autonomy. The villages or communities (Desa), although having some kind of local council, an own budget and the right to regulate their own affairs, are not perceived as an autonomous level of government in the sense of law No. 5/1974. Their status and their structure are defined in the law No. 5/1979.

The Local Governments of the first and second level consist of the head of the Local Government (Gubernur and Bupati/Walikota respectively) and the DPRD, a kind of local parliament. The head of the Local Government has a double function. He is at the same time representative of the central government in the region and representative of the local population of the region. However, the head of the government is only truly accountable to the President, via the Minister of Home Affairs, and not to the local parliament. This is also reflected in the way the Gubernur and the Bupati/Walikota are elected: in executing their right to approve the list of candidates the Ministry of Home Affairs is controlling and influencing who is becoming Gubernur or Bupati/Walikota.

According to the law 5/1974 Local Governments are allowed to issue and to pass local laws and regulations. They have a certain discretionary power over their own personnel and they have their own budget. This autonomy is however already limited by intensive control and supervisory functions of the central and the provincial government towards the lower levels of government.

The law provides for three different forms of decentralization: desentralisasi, dekonsentrasi and tugas pembantuan. Desentralisasi means transferring tasks from a higher level of government to a lower autonomous level; what is often called "real decentralization" or "political decentralization" in international literature. Dekonsentrasi means transferring tasks from a central ministry to its field offices or line agencies in the region. If tasks are implemented by Local Governments on behalf of the central government, it is called tugas pembantuan (co-administration).

According to the law, the transfer of tasks in the form of desentralisasi has always to be accompanied by the transfer of personnel, money and the necessary equipment. Article 11 of Law 5/1974 determines that the focus of local autonomy has to be on the second level (Tingkat II).

Already from the law it can be concluded,

• that decentralization is always defined from the top and is seen as one form of implementing national development policy

• that political participation or democratization are not the primary goals of decentralization, but rather increasing efficiency and effectiveness of the government apparatus.

The last point is reflected in a fundamental shift in the general concept of local or regional autonomy (otonomi daerah) with the law 5/1974. While in the early times of the Republic of Indonesia, the concept otonomi yang seluas-luasnya was defined by the founding fathers of the Republic, which meant a regional autonomy that was to be as broad as possible, with the law 5/1974 the concept otonomi yang nyata, bertanggung jawab dan dinamis (real, responsible and dynamic autonomy) became binding.

Nyata means that the transfer of tasks has always to take into account the capacity and capability of the Local Governments. According to the law the capacity and capability has to be measured and analyzed in a scientific way to become the basis for the decision how many and what kind of tasks have to be transferred to which regions.

Bertanggung jawab means that decentralization policy has always to be designed in line with the development goals of the country; stability and national policy always have priority as compared to local needs and requirements. Regional autonomy is rather understood as an obligation than as a right of Local Governments.

Dinamis means that tasks which have already been transferred to the Local Governments can always be recentralized if frame conditions have changed or it is felt necessary by the central government.

2. Implementation of decentralization policy in Indonesia

Most of the tasks which are today - at least on paper - under the discretion of the Local Governments have been decentralized in the 1950s. During that time decentralization policy was still oriented towards the concept otonomi yang seluas-luasnya (autonomy as broad as possible). With the introduction of the new concept of otonomi yang nyata dan bertanggung jawab the need arose to review the kind and the amount of tasks which have been under the discretion of the Local Governments regarding their capability and their capacity. The classification of the Districts according to their capacity and capability was considered as a precondition for a decentralization policy which is in line with the basic law 5/74.

The Ministry of Home Affairs took over the task of classifying the Districts which led to tremendous research efforts of the Research and Development Unit of the Ministry (Badan Litbang Depdagri), known as Kemampuan -study. Starting in 1979/80 and spanning more than 10 years Badan Litbang, together with the most famous universities of the country, tried to classify the Districts of Indonesia according to their capability and their capacity for managing autonomous functions or tasks. The final result, a ranking of the Districts in four groups, was however considered by all parties as insufficient and not yet applicable. The second step of the research project, namely relating these levels of capacity/capability to a specific number and to specific kinds of functions and tasks to be transferred, was never done.

Not least because of this failure and the lack of clear rules and regulations about the mechanisms and procedures of how tasks should be transferred according to the law 5/74, the way of transferring as well as the respective amount of tasks and functions which have been transferred to the Local Governments in the '70s and '80s differ considerably between the sectors. In general however it can be stated that

• deconcentration has been favored as opposed to decentralization by the sectors;

• most of the transfers happened from the central government to the province governments, so that the focus was in reality at the first not the second level of Local Governments;

• tasks which include important sources of revenues were seldom decentralized;

• only parts of the tasks were transferred; there is a tendency to decentralize only limited implementation functions while policy and strategy formulation as well as decisions about the allocation of money are mostly kept on higher levels;

• there is some confusion at local level regarding the question which of their tasks local governments are actually handling as decentralized, which as deconcentrated and which as co-administration tasks.

Only in 1992 was a government regulation passed (PP 45/92) which stipulates some mechanisms for the transfer of tasks to the District Governments (Pemda Tk. II). It was - after 18 years (!) - the first government regulation which tried to realize article 11 of law No. 5/74.

The regulation reaffirms the necessity to measure and to classify the regions according to their capacity and capability and confirms the dominant role of the Ministry of Home Affairs in this task. Beside this, it sets clear time limits for the transfer of tasks from the first to the second level-Governments. If the Provincial Governments are not further devolving tasks to the District Governments the Ministry of Home Affairs is entitled to transfer the tasks itself, without further consultation with the Provincial Government.

Regarding the kind, the number and the form of tasks which have to be transferred to the districts, the government regulation 45/92 contains, however, only very general statements and criteria. On the basis of these criteria and principles it is still impossible for the Departments as well as for the Local Governments to decide clearly which tasks have to be transferred to which level. It is therefore not surprising that the government regulation has had hardly any impact on the decentralization process in Indonesia so far.

3. The District Autonomy Pilot Programme

Starting in late 1994, the Minister for Administrative Reform, in conjunction with the Ministry of Home Affairs (MoHA), negotiated the conversion of the deconcentrated units of the central and provincial governments to units responsible directly to the district head. During these negotiations a set of tasks was defined together with the sectoral ministries for each of the involved sectors which had to be transferred to the district governments; and together with the provincial and the pilot district governments the new organizational set-up at district level was fixed.

Most Ministries and provincial Governors were supportive and cooperative, but some had deep reservations. Some government departments or agencies were even able to stay out of the initiative, arguing that they needed to maintain their presence in the district. Others sought to reduce the scope of the transfer of functions that were formerly undertaken in the departmental arms at the district level. The same is true for some of the provincial governments which tried to reduce especially the kind and number of income sources which had to be transferred together with the respective tasks.

This tendency persisted even after the time of the formal announcement of the program when as a result of a "crash program" the necessary central and local government regulations had been issued and a set of responsibilities were formally devolved towards the pilot districts. Some sector departments as well as some of the provincial governments started to weaken the result of the negotiations with the Ministry of Administrative Reform and the Ministry of Home Affairs by only devolving certain functions or certain parts of a task which has been agreed to be devolved and therefore limiting the authority of the Local Governments regarding the transferred task. For those tasks that have included opportunities for revenue generation, the realization of the transfer of the tasks has been plagued by reversals, limitations and delays.

Moreover, the principle that all functions and tasks transferred to lower levels must be accompanied by attendant resources, has been grossly breached, especially in relation to the development budget. This alone has meant that the resources available to implement the functions in the 26 districts have been less than when these functions were held in the deconcentrated arms of higher levels, and less than most of their non-participating neighboring districts continue to receive from higher levels. For the clients and public touched by the services which have newly been transferred, the DAPP has probably in many cases meant a reduction rather than an improvement in services . The inability (some would say unwillingness) to transfer the requisite funding has been particularly deleterious for districts with a low level of own revenues (ex. Aileu in East Timor). These districts now must support a greater number of Dinas with their own routine funds, and they are often denied the development budget that were once attached to these functions. Only for districts with a substantial own revenue base (ex. Badung in Bali), has this problem been attenuated by an infusion from the district budget.

Based on the experiences so far, the Government of Indonesia is actually reconsidering its decentralization strategy and has delayed the extension of the programme to other districts. Especially the Ministry of Home Affairs is still very much concerned about the question in how far the decentralization measures developed in the framework of the DAPP can better be adjusted to the capabilities and the capacities of the district governments. The right sequencing and timing of decentralization in Indonesia is therefore still a demand and not yet solved satisfactory.

Decentralization in Indonesia: Part 2
Preliminary assessment of the design and the strategy of the District Autonomy Pilot Programme

Introduction

1. Significance of DAPP

In the field, the scope of DAPP is quite visible. Indeed, it quickly becomes evident that the program reflects the lead conceptor' organizational view (i.e. that of the Ministry for Administrative Reform, MENPAN). Where districts previously may have had less than ten sectoral departments of their own (Dinas), most can now boast of more than 20. For the population at large, to the extent that they are aware of a government presence, the large signs outside the transferred offices provide proof of change. This visible feature unfortunately also obscures the actual degree of change. Nevertheless, the visible change does have an effect at least on the local administration, both in terms of the number of new employees added to local rolls, and to regional officials' sense that progress is finally being made on the question of regional autonomy.

Although some departments and agencies escaped the first round of DAPP, it is worth emphasizing that nearly all sectoral departments have released certain functions and some resources to the new district departments/agencies.

Discerning the longer term significance of the DAPP is more problematic. DAPP was framed as special kind of pilot program. The choice of words was deliberately made, in the Indonesian text the initiative is explained as a "percontohan", meaning "example". The implication of the use of this term is that there will be a follow up to the pilot stage. Indeed, there have been plans afoot to expand the pilot to a combined 68 districts and, for the first time, cities. However, the "example" approach does not carry the same sense of exploration as might be implied in the term "uji coba", that is "experiment", which underpins most English's speaker's understanding of a pilot program. This shade of difference is in the end quite important in more ways than one.

For our purposes, the dominant Indonesian understanding of the pilot program is reflected in the less than rigorous monitoring and evaluation system put in place for DAPP. Focused mainly on the question in how far the transfers fixed in the legal products of DAPP have been realized, the DAPP approach itself is not made an object of a broader reflection about the best way to manage the decentralization process in a more sustainable and effective way in future. Hence the opportunities to learn from the pilot program have to some extent been limited by the framing of the pilot program concept.

2. Process of Design and Implementation of DAPP

The resistance met in some sectors and other key actors at central level may not only be based on the general unwillingness of central level agencies to share power and funds; it reflects also the crash nature and the poor design of the program. It can be argued that these agencies had 20 years to put their house in order after L. 5/74, but even so, the speed of negotiations did not favor thoughtful analysis and wide participation within the sector and with other stakeholders. The most powerful or less committed sectoral departments and agencies thus withdrew their participation, or limited their involvement to what seems to be rather token transfers of functions (see next section). Fiscal decentralization efforts developed in the Ministry of Finance and the National Planning Agency kept unlinked with the pilot program. The first level regions, unclear about the final form of autonomy to be retained at the provincial level, also generally felt uneasy about the DAPP. Here, the direct control of the Ministry of Home Affairs meant that these regions in the end had to accept the initiative, and indeed many of the most significant transfers of functions in DAPP were made from this level.

This top down and precipitous approach to decentralization resulted not only in some unsatisfied "givers" but also in some unsatisfied recipients. As will be discussed in a later section, the list of functions selected in the DAPP would be quite different from that drawn up by the districts, based on their own assessment of what was needed and their capabilities. The top down approach extended even to the setting of the organizational structures for Dinas, supposedly autonomous district agencies. Much disappointment in districts was generated by this detailed imposition by the center; some districts would not have established some Dinas, preferring to combine certain functions under one roof or integrating them in other existing Dinas. Given that routine budgets to support the Dinas come mainly from the districts, this is no small objection.

Of even greater importance than the specific complaints with organizational structures or functions received is the lost opportunity for widespread and fundamental discussion including the question of proper institution building for decentralization at central and local level that could have been engendered by DAPP. Such a discussion was never realized as a result of the pace of negotiation (and top down decision making). The focus of the discussion came to rest on the features of the DAPP itself. Even so, the degree of autonomy actually provided in DAPP remains to be fully assessed and is only now being discussed more widely.

3. Degree of Enhanced Regional Autonomy

By and large it appears that the most important functions transferred have come from the first regional level, perhaps reflecting the clout of the MENPAN/MoHA team in "negotiating" with this level (see Table 1 for a list of sectors and origin of functions involved in DAPP; those from the first regional level are drawn from the province of Bali, other provinces may differ slightly). Some Dinas at district level received some functions from both the center and first regional level, but this is the exception rather than the rule (ex. employment). A total of 165 functions and 353 tasks have been transferred to the Badung district in Bali, and some of the task have been further detailed into activities. However, this general count can be deceptive. It is difficult to compare sectors due to the terminology or categories used in each sector. Some functions in some sectors would only be tasks or activities in the categories or language employed in others sectors. There is no hierarchy of functions agreed across sectors. Hence the number of functions, tasks or activities transferred is not a good indicator of the weight or importance of the transfers.

The public works sectors (roads and irrigation), tourism and to some extent the education sectors have seen significant transfers of functions. For the roads sector, the upkeep of national roads was turned over to the districts. For irrigation, the district also became responsible for the upkeep of larger scale irrigation systems. In the tourism sector, approval and supervision over a broader range of tourism facilities was given to the district.

Others district Dinas, while accepting "new" transfers on paper in the framework of DAPP, had actually already received these functions via other legal instruments years before or had de facto been undertaking these functions (the agricultural subsectors in particular). Other Dinas received but a token increase in their functions (the Health Dinas for example only received "medicine storing/distribution").

In general, there has been a tendency to only transfer functions which are not linked to sizable sources of revenues. Many of these "dry" functions, as they are labeled in Indonesia, relate to the guidance and strict implementation tasks. The sign in front of the office may have changed but that has been no clue as to whether the functions normally carried out in these deconcentrated arms have been transferred as a whole, or whether some have been excluded from the DAPP, or whether some have been included in the DAPP but as yet not implemented in practice.

The districts have voiced their concern over the lack of substantial functions in DAPP, but some sectors and some districts have also voiced their satisfaction. Some have indicated that the functions transferred have been or will be an added burden, though this sentiment seems to be linked to the lack of transfer of requisite resources. Overall, it would seem that the DAPP represents a modest step in decentralization in terms of added functions, but the full impact of this modest initiative is less than optimal due to the failure to pass on the required financial resources passed.

4 Contribution of DAPP toward decentralization goals

One important consideration in making a plausible link between the specific measures of the DAPP and the broad goals of increased public participation, improved services and increased government efficiency lies in the degree of effort made by the district apparatus in implementing the functions in a better way. Certainly the Dinas are exhorted by many district heads to do just that, and the climate produced by the initiative is in general supportive, although the frustration with the lack of follow through on resources is undermining the initial build up of enthusiasm. In general, our field observations lead us to conclude that the district level apparatus is not giving sufficient attention as to how the district can do better with these functions than when they were held by higher levels. Specific arguments or proposals indicating efficiencies and improvements in coverage, speed, quality and other relevant indicators are few and far between in the districts we have visited. Moreover, the subdistrict level staff of the district Dinas have by and large not been fully apprised of the intent and mechanisms of DAPP or the subdistrict's possible role in improving the implementation of the transferred functions. It is difficult to see how participation or service delivery will appreciably improve directly as a result of the DAPP.

The fate of the new functions is not entirely in the hands of the district. If the promise held in DAPP is to be realized, the central and first level sectoral departments must continue to support the district Dinas beyond the initial transfer (assuming that in itself is concluded properly). It is the higher level technical departments that must provide technical specifications, training and other support to make the implementation of the functions at the district level a success. At this point, some departments do not even have the necessary indicators ready to assess whether indeed the functions are being properly implemented. Effort on this front is essential and should follow the transfers in a timely manner.

The achievement of the broad goals pursued by the Indonesian government calls for significant country wide initiatives in transfers of functions and requisite resources. But this alone is not enough. As a base for cooperation in the decentralization effort, a range of critical issues must be addressed to have an overall significant impact in terms of these broad societal goals. The district apparatus must become more accountable toward the district legislative branch, and the legislature itself must be more accountable to the district constituency. In other words there is a clear need for a political decentralization as a complement to the more administrative decentralization of the DAPP. Additionally, strengthened subdistricts and a revitalized and autonomous village are essential in assisting the districts to shoulder the burden of decentralization.

5. Prospects for further decentralization

However, the perception regarding the form and the content of local autonomy still differs considerably between the most important institutions on central level as well as within the responsible Departments; there is no concept of local/regional autonomy which is jointly agreed on by all stakeholders, no clear criteria which could be used to determine the functions which should be implemented autonomously by the respective local governments and no concept on how to deal with local/regional variations; there is even no concept regarding the process which could lead to a commonly agreed upon concept of local autonomy which could then be the basis for an integrated decentralization policy.

Some of the issues brought to the fore in DAPP relate to the resistance met in the sectors and the first regional level. The approach to communication and negotiation needs to be reformulated if further steps are to be more successfully implemented in the field. There is a clear need to clarify and to further elaborate the concept of local autonomy in Indonesia including the issue of the autonomous role of the provincial level to give some certainty or vision for what it means to emphasize the second regional level. Of great importance, as already alluded to earlier, will be the linking of administrative decentralization to ongoing efforts in fiscal decentralization and to complement the administrative with political decentralization measures.

Decentralization is usually initiated by the center, and only when the center feels safe, or compelled, to allow for it. The prospect for meaningful decentralization in Indonesia is limited due to the resistance to be found in the bureaucracy. Even when Presidential force is brought to bear the bureaucracy finds ways of delaying or diluting the initiative. Indeed, the communication pattern of the bureaucracy in Indonesia works against meaningful dialogue; the tendency of the bureaucracy is to become self-referential and isolated from external claimants. As difficult as the task may be, it appears that pressure on the bureaucracy must come principally from the bottom. In this respect, forward looking Indonesian officials, and their supporters (in non- government organizations and the donor community for instance) must seek to organize the lower levels of government to more effectively and forcefully channel their aspirations to the political/administrative apparatus. Beyond that there is a need for strengthening and empowering elements of a civil society in raising their claims towards the bureaucracy.

6. Summary

As a beginning, the DAPP cannot be faulted; even if it is rather modest in its scope and likely impact, it is after all fairly significant in symbolic terms and also significant in practice for certain sectors. A more thorough assessment needs to be made to determine whether it has been successful in terms of various evaluation categories and success indicators. If this assessment is indeed made with the Indonesian government's broad goals in mind, then it is likely that the DAPP will fall short of being a success. But if DAPP is viewed as one step toward meaningful regional autonomy, if it leads to a richer and broader dialogue, and if it generates a more pro-active stance from the regions themselves in voicing their views and needs, then DAPP will auger well for the future.

7. General conclusions and open questions

1. There is obviously a need for more comprehensive approaches to decentralization. Pure administrative decentralization measures for example fall short if they are not complemented with the transfer of corresponding funds and with measures to increase the accountability of local governments towards the population. The devolution of more tasks, to give another example, will not automatically lead to more autonomy on local level, if other relationships between local and central governments remain unchanged: Control, planning and financing mechanisms as well as organizational development and personnel management can limit the room to maneuver of local governments to the extent that there is no real autonomy left.

   However, the question remains regarding the right degree of comprehensiveness: concepts which are too complicated and try to cover all aspects of decentralization will not only need too much time to be elaborated, they will also be too difficult to communicate to all relevant parties and will therefore fail in raising sufficient enthusiasm and support on the side of the implementing agencies of such a reform programme.

2. There seems to be a need for broader and more intensive intersectoral discussions on the contents and the strategy of a decentralization initiative and for an involvement of local government representatives in these discussions. On the other hand there is a need for a clear and strong leadership in decentralization in order to push through necessary reforms which run against strong own interests of central level stakeholders, and in order to keep coordination and communication to a feasible degree.

   The question is therefore who has to be involved in which form in the design and implementation of decentralization initiatives. Who has to take part in decision making about fundamental issues of decentralization and local autonomy and who has just to be consulted?

3. Incrementalism seems to be necessary if local conditions regarding the capacity and capability of local governments vary considerably. On the other hand there is a danger that certain parties will use the argument that some local governments are not yet be ready to receive certain tasks or functions to create unlimited delays in the implementation of decentralization. Additionally, allowing too great variation among jurisdictions of similar status could lead to an unhealthy fragmentation of responsibilities between different levels of government and among local governments.

   This raises the question if a basic set of tasks or functions could be defined which would have to be devolved to all local governments of the same level or type of jurisdiction while others could be defined as additional tasks which would have to be devolved according to capabilities and capacities. How to assess these capacities and capabilities properly would be the next question. Experiences in Indonesia show at least that there is a need for involving local governments in such kind of assessments.

4. Support from higher level government institutions in form of technical advice, guidance, training etc. seem to be crucial in order to enable local governments to fulfill their autonomous tasks properly. The danger is however that support functions are being implemented in a way which limits or even counteracts local autonomy.

   The question is therefore how to find the right mixture between guidance and support on the one hand and the application of local autonomy principles on the other hand.

5. A learning approach in form of experimentation and piloting in order to develop an appropriate decentralization process and local autonomy concept seems to be convincing. For that the concept has to be kept open for changes and adjustments; a too detailed concept may hinder the necessary openness. However, Indonesian experiences show, that the basic concept needs clarity to enable proper links to ongoing activities in the broad field of decentralization and to gain the support of key decision makers.

   Much more important in this context is, that a learning approach requires the ability and willingness of key stakeholders for self-reflection and criticism. An open and radical reflection of the experiences made during the testing including reflecting on the own role and failures is of the essence in a learning approach. However, the ability and willingness for self-reflection cannot be assumed in political-administrative systems. The opposite is true as long as there is no pressure from outside of the system itself. How this can be brought about is difficult to answer, but certainly the role of the civil society is key. Whether technical cooperation approaches can be useful in this context is another question.