(Professor) Dwight Y. King, Ph.D.
Department of Political Science and  Center for Southeast Asian Studies
Northern Illinois University
DeKalb, IL 60115-2887
Tel. (815) 753-7054
FAX 815-753-6302
dking@niu.edu

Introduction

Decentralization of power was one of the main demands of the reform movement that took shape in 1998. After Suharto resigned many regions began voicing their discontent. The interim Habibie administration responded to these developments with a policy of wide ranging regional autonomy with special additional arrangements for the provinces of Aceh, Irian Jaya, the capital region of Jakarta and East Timor. Accompanying and reinforcing these decentralizing efforts were others intended to democratize the political system. In this paper I begin by briefly summarizing the effects on decentralization of new political laws passed by the House of Representatives (Dewan Perwakilan Rakyat or DPR) and promulgated in 1999. I then examine aspects of the Asecond wave of reform consisting of constitutional amendments passed by the People's Consultative Assembly (Majelis Perwakilan Rakyat or MPR) in 2000-2002 and of proposals for amendment of the Law on Regional Government (Undang-Undang Pemerintahan Daerah or UU 22/1999). Next I identify factors that help explain inter-district variation, or why political reform, decentralization and good governance have progressed further in some districts than in others. I conclude with speculations on the likely future evolution of decentralization reforms in Indonesia.

Habibie's Initiatives

Several aspects of the new political laws proposed by interim President Habibie had implications for decentralization. The Law on Political Parties (Undang-Undang Partai Politik or UU 2/1999) stipulated that, in order to contest the election, a party had to have an organization (executive committee) in at least nine provinces and in half of the districts in each of those nine provinces. To compete in the following election (2004), the law also required that a party must have gleaned at least two percent of the votes in the 1999 election. The effect of these requirements was to make it highly unlikely that a regionally based party pushing a regionalist agenda could emerge.

A government regulation negotiated simultaneously (Peraturan-Peraturan or PP 12/99) prohibits civil servants from serving concurrently as party functionaries. The intention was to keep the governmental apparatus neutral in the competitive electoral arena and to insure that civil servants render service to the public without regard to partisan sympathies or affiliations. This restriction on the political rights of civil servants was the most intractable and difficult issue in the entire package of laws on electoral reform, and for good reason. It probably contributed more than any other single factor to the defeat of Golkar in the 1999 election. Since Golkar was a highly centralized (and centralizing) organization, its defeat helped open up the political space for implementation of decentralizing policies.

The Law on the General Election (Undang-Undang Pemilihan Umum or UU 3/1999) adopted a unique version of proportional representation by province combined with some elements of a district plurality system. Every candidate for a legislature at any of the three levels had to be publicly identified with a particular territorial constituency, in addition to a party. Rather than assigning seats to particular candidates based on parties' provincial vote totals and candidate priorities stated in party lists as in previous elections, candidates were supposed to be assigned to seats on the basis of how well the parties performed at the district level. This had potential for strengthening the hand of the regional party leaders vis-a-vis the national party leaders. However, the General Election Commission's (Komisi Pemilihan Umum or KPU) decision to give central party leaders considerable discretion in filling their party's seats (as in the old system) brought about a substantial diminution of the significance of the district element in the new hybrid electoral system. It turned out that 97 of 462 elected House members (21 percent) represent districts other than those to which they were originally assigned.

At least three features of the Law pertaining to the Legislatures (Undang-Undang Susunan dan Kedudukan DPR, DPRD, MPR or UU 4/1999) were relevant to decentralization. First, reserved domains or appointed seats for the military and police in the legislatures were cut in half. Theoretically this weakened one of the main mechanisms of centralization at all three administrative levels. Second, under the previous government, 16 out of 27 (59 percent) electoral districts were over-represented, all of which were located outside the island of Java. The new law continues this mal-apportionment. How it impacts decentralization may depend on the level of development and how much the government is controlled by a dominant party. Third, the law provides for a continuation of regional representatives (utusan daerah) in the Assembly, although now they have to be selected by the provincial legislatures instead of being appointed by the executive. However, these representatives were denied their own separate faction in the organization of the Assembly until November 2001. As a result they were integrated into parties and voted with party factions rather than as a block articulating and defending regional interests. The political reforms having the greatest influence for decentralization, of course, are the Law on Regional Government (UU 22/1999) and the Law on Financial Balance between the Center and the Regions (Undang-Undang Perimbangan Keuangan Antara Pemerintah Pusat dan Daerah or UU 25/1999). The previous law (UU 5/1974), formulated by the General Suharto's New Order government, emphasized the mobilization of regions in the effort of national development. Although the law gave lip service to regional autonomy, implementing instructions were more oriented to the promotion of national stability through the establishment of an authoritarian, hierarchical structure reaching down from Jakarta to the provinces, the districts, and ultimately to the smallest villages. Thus, there was no mention of the rights of regional governments or subjects living in a particular administrative area. The experience of implementing the law

...provoked a great deal of controversy among many observers within the government calling for greater real autonomy to be given to districts or for more meaningful balance between provinces and districts. The hesitancy in pursuing decentralization aspects of the law more forcefully was caused in part by a lack of confidence that the district had the human and fiscal resources for the task, and the fear that autonomous provinces would always tend to push for independence.

The new Law 22/1999 marks a radical departure from the structures of governance built up over the preceding quarter century. It recognizes two basic levels of governance, the central government and the autonomous district governments (kabupaten, kota) whose relationship involves a division of responsibilities and powers and is not strictly hierarchical. District governments are given both rights and duties, including the duty to support the initiatives of the citizens within their jurisdictions. They no longer double (merangkap) as a administrative area of the central government. In other words, the head of a district is responsible to the district representative assembly (DPRD-II) rather than to the President through the provincial governor as previously.

However, under the new law, provincial governments are constituted as autonomous units but at the same time as extensions and administrative regions of the central government. In other words, provincial governors continue to wear two hats, as head of a region and as representative of the central government in a province. They are responsible both to the the provincial representative assembly (DPRD) and directly to the President. Hull explains the intent of the new law as follows:

The decision to give provinces this contrasting, and in some ways contradictory set of functions was justified as a way to maintain a strong link between central and district governments in the context of the constitution of a unity state. The institutional duality is also designed to allow the provincial government to serve both central and local interests by assuming representational duties of the former and residual duties of the latter. For instant if for reasons of efficiency or effectiveness the central government cannot carry out activities directly from Jakarta [the seat of central government], the provincial government can be given such responsibilities as a part of de-concentration of administration (Article 1 (f)). On the other hand, if the regional [district] governments are unable to carry out all their duties the province has limited autonomy to take on these responsibilities.

Under the new law, the district legislatures (Dewan Perwakilan Daerah or DPRD-II) are established as political institutions separate from district government. They have the duties to legislate, monitor and supervise the executive and channel the aspirations of the citizenry, including the responsibility of choosing and dismissing the executive (bupati, walikota) without any involvement of the central government. The candidate who obtains the most votes is declared the winner and is ratified by the President, who is obliged to approve the legislature's selection. The executive is fully accountable to the legislature; although in theory coequal, this accountability feature elevates the legislature above the executive. The legislature, in turn, is accountable to the voters every five years. The devolution of power over executive selection and termination was intended to make the executive more attentive to the needs, interests and politics of his jurisdiction. The new laws do not recognize the right of recall, previously held by the party organizations over their representatives in the legislature, which was intended as a protection of free speech and, perhaps, political independence.

Unlike the old law on regional government, Law 22/1999 more clearly specifies the range of functions over which regional governments have authority. Chapter 11, verse 2 mentions public works, health, education and culture, agriculture, transportation, industry and trade, investment, environment, agrarian (land) affairs, cooperatives and employment. However useful the specification may be, the division of authority between the central government and the regions is open to a variety of interpretations and thus continues to provoke controversy.

The new law also improves the qualifications required for regional government executives. The previous law required experience in government, which resulted in priority being given to candidates from the bureaucracy and the armed forces. This requirement is erased in the new law and replaced with a weak residence requirement. Now a candidate must have lived at least one year in Indonesia (for provincial governor) or in the province (for district chiefs, bupati and walikota).

Article 8 of the new law on regional government explicitly links it with the new law on financial balance. It reads as follows: Athe authority of the government which is devolved to the region must be accompanied by the transfer of funds, means and infrastructure, together with human resources appropriate to the authority being transferred. Several provisions of the financial law are noteworthy. One is the recognition of types of revenues the regional governments are allowed to collect, including their own funds (pendapatan asli daerah or PAD) and production sharing tax (pendapatan bagi hasil or PBH). The proportion of the tax given to the regions has increased quite dramatically (see Table 1). If we focus on the column labeled Acentral government and compare percentages in the old and new columns, we see a decline or decrease in virtually every revenue item. A similar situation obtains with the column Aprovincial government, except for revenues on oil (3 %), natural gas (6%) and reforestation funds(40%). Conversely, the new column under district government registers an increase on virtually every type of revenue. The new law provides special autonomy for the provinces of Aceh and Papua, shown in the bottom half of the table. There large increases in the share of revenue go to the provinces which are given discretion over district governments' share. Another provision allows regional government to borrow money from either domestic or international sources with the agreement of the legislature. It seems clear that the law was intended to lessen the dependence of regional governments on subsidy from the center.

Table 1: Comparison of the Old and New Central-Local Government Revenue Sharing

The laws on regional autonomy are a main component of political reform in Indonesia. Although broad and far-reaching in their concepts, they were drafted hurriedly and in hope that both the expected and unanticipated problems to emerge would be manageable or correctable. But the new laws have provoked intense, wide-spread and surprisingly comprehensive debate.

In order for Law 22/1999 to take effect, a large number of implementing regulations had to be issued. This process proved to be both drawn out and controversial. Arguments arose relating to the lack of hierarchy between authorities, the relative authorities of the DPRD and the head of the region (kepala daerah), the powers of local authorities to introduce taxation, the validation of new regional regulations, the arrangements for cooperation between authorities, the management of natural resources and its impact on the environment, the powers relating to ports, airports and maritime questions, the recruitment of civil servants at regional level, and a wide range of other issues. The definition of minimum service standards proved to be a lengthy exercise in which suspicions arose that some central government sectoral ministries were attempting to retain decentralized powers.

In January 2002 the Ministry of Home Affairs proposed revisions to Law 22/1999 that have been strongly attacked as an attempt to re-centralize and return to authoritarianism. The latter criticism usually mentions Article 41 that gives the President the power to dissolve a regional legislature. Some observers, however, argue that the overall effect of the proposal is less re-centralization than a shift of power from the elected members of the regional legislature to the (indirectly) elected Head of Region and his/her officials (i.e. executive). If one of the problems with the original (UU22/1999) has been that some regional legislatures have misunderstood the nature of their relationship to the regional executive, the elucidations to the draft revision offer questionable solutions. They reject the concepts of legislative, executive and judiciary which are used universally in specifying the relationships among the components of democratic regional government, deny the legislature';s roles of supervision and oversight and give unclear power to the head of region to formulate regional policies. A more logical (and simpler) solution would be to clarify the actual powers of the legislature. Another example of shift of power to the executive is the contention that the legislature's funds are part of the regional government budget and therefore not independent (Article 25). But there would seem to be better ways to control budgetary irresponsibility.

A second tendency in the proposed revisions is for major and potentially sensitive areas to be left to subsequent clarification by government regulation. A case in point is the provision for withdrawal of regional powers in the event a region fails to undertake obligatory functions (Article 14). Both the criteria for judging regional failure and the process for carrying out a withdrawal are left up to a government regulation. Another example concerns the mechanism for recalling a member of the legislature (Article 21). Once the required signatures have been obtained, the process is left entirely to government regulation. AA step which has such significant implications for the status of elected representatives should be further defined in the Law itself.

In any event, on May 29, 2002 the government announced that revisions to the law had been postponed. Then on January 9, 2003 a government official reaffirmed its commitment to a partial revision of Law 22/99, indicated that discussion of revisions with the House of Representatives (DPR) was scheduled for late 2003, and that it would include the introduction of direct election of heads of regional governments (i.e. governors, regents and mayors).

Constitutional Amendments and Other Legislation

Despite the concessions by the central government as represented in the laws on regional autonomy, many regions remained dissatisfied because they believe that their autonomy was based only in laws that could be rescinded at any time by a decision of the central legislature and the President. Hence pressure mounted for the decentralization of power to be enshrined in the Constitution, making it harder to reverse in the future. When the sovereign Assembly (MPR) met in its annual session in August 2000, it amended Chapter VI of the Constitution on regional authorities in ways which captured the spirit of the laws on regional government and financial balance.

Known as the Second Amendment, several aspects of the amendment are of particular interest here. One is the strongly regionalist character conveyed by

... the principle that regions may act on any subject that is not reserved by law to the central government. There is a constitutional provision for special legislation and/or special status for particular provinces. There is a requirement for justice and equity and regard to local distinctiveness and diversity in the financial arrangements for regions.

The amendment also stipulates that regional executives be democratically elected, with the precise method (direct or indirect) to be determined by law. The alternative of enshrining universal direct election of regional executives was not accepted, although the special legislation for Aceh passed by the House provides for direct election of these positions.

The Assembly also passed MPR decree IV/2000, entitled Policy Recommendations in Implementing Regional Autonomy. This decree criticized the central government for failing Ato view the implementation of regional autonomy as a constitutional mandate so that the decentralization process has tended to become bogged down. It noted wide discrepancies exist between the center and the regions and among the regions themselves in respect to the control of natural resources, cultural resources, economic infrastructure and the quality of human resources and that the interests of various parties obstruct the implementation of regional autonomy.

To remedy these weaknesses, seven recommendations were addressed to the government and the House. One of them was the instruction to issue all remaining implementing regulations to the laws on regional autonomy by the end of December 2000. If this was not done, then regions that were fully capable of implementing autonomy were free to produce their own regulations. Also recommended was justice in financial equalization among regions and the possibility of using the local profits of state-owned enterprises in regions where natural resources are limited. The decree also recommended the development of regional autonomy master plans in each region to define the process of transition, the establishment of coordination teams in each region to smooth the implementation of autonomy and a requirement for a fundamental (medium term) review of the two laws on regional autonomy to bring them into line with the amended Constitution, especially in relation to the hierarchy between different levels of regional government, something that had specifically rejected in Law 22/1999.

The Assembly convened for its next annual session in November 2001 and enacted the Third Amendment to the Constitution which addressed and provided answers to a large number of the questions relating to the structure of the Indonesian state. Especially relevant here was the decision to establish a second, regionally-based chamber of the national legislature, the Regional Representatives Council (Dewan Perwakilan Daerah or DPD). Its members will be elected as individuals, not openly linked with any political party, from every province through the general election. The total membership of the Council may not exceed one-third of the House (currently 500), and each province will have the same number of representatives. The Council may propose to the House bills related to regional autonomy, center-regional relationship and financial balance, and management of natural resources. It also has the duty to provide consultation to the House over bills on the state budget and on draft laws relating to tax, education and religion. It has no role in foreign policy, defense or security. It is, thus, a minor chamber and not analogous to the U.S. Senate. Further statutory definition of this body is underway and is expected to be completed in time for the 2004 general election.

The Third Amendment clarified that the state will be structured as a conventional presidential system, which has implications for decentralization. By providing for a direct presidential election, the amendment creates political pressure to elect governors, regents and mayors directly as well. Recall the announcement, mentioned earlier, that the government will propose revisions of Law 22/1999 to the House in late 2003, including direct election of regional executives.

Another article in the Third Amendment with implications for decentralization states, all taxes and other levies for the needs of the state of a compulsory nature shall be regulated by law (No. 23A). This article arose out of concern that a brake was needed on regions becoming too liberal in levying charges and retributions.

In the internal organization of the Assembly (MPR), as was previously noted, regional representatives (utusan daerah) have always been denied their own faction. But in 1999 during the first post-Suharto meeting of the Assembly, regional representatives began to press for their own faction as a way of increasing their influence. They finally prevailed in the 2001 annual session of the Assembly on condition that regional representatives relinquish their connections with political parties and agree that the regional representatives faction would not be entitled to its own deputy speakership of the Assembly.

At its 2002 annual session the Assembly ratified another round of changes, collectively known as the Fourth Amendment. Perhaps most noteworthy was the decision to abolish all appointed seats in the legislatures including, nota bene, those of police and armed forces after the next general election in 2004. Further, the Assembly reached agreement on the role and composition of the Assembly which henceforth will consist entirely of the two chambers, the House of Representatives and the Regional Representatives' Council. A final change with implications for the regions pertained to rules governing the election of the president and vice president. The winning pair must have 51 percent of the votes cast as well as at least 20 percent of the vote in half the provinces.

Since the passage of the laws on regional autonomy in 1999, an issue that has periodically occupied the attention of central government has been the establishment of new regions (propinsi, kabupaten, kota). Five new provinces and hundreds of districts (kabupaten and kota) have been created from existing provinces and kabupaten. Most of the bills creating these new regions were initiated by the House to accommodate political aspirations of local elites and new power linkages for national political actors.

Explaining Inter-district Variation

Why have political reform, decentralization and good governance reforms progressed further in some districts than in others? What is the source of demand for decentralization? Are the sources mainly political or economic? These questions intrigue us but answers are elusive and tend to be highly localized and idiosyncratic. Both policy makers and social scientists are in need of more generalizable knowledge that holds with high probability across districts (nation-wide). But lacking appropriate and comprehensive information that can be systematically analyzed, we tend to fall back on anecdotes or (deductive) theorizing, neither of which are very reliable. In the remainder of this paper I want to suggest some tentative answers to these questions that are suggested by recent empirical research, including my own analysis of available, district-level quantitative indicators as well as observations from field work in West Sumatra, Riau and East Java carried out in August 2002.

A. Administrative Decentralization (Deconcentration) and Democratic Decentralization (Devolution)

In 1995, the New Order government undertook the Pilot Regional Autonomy Project (Proyek Otonomi Daerah Percontohan, PODP), considered by some to have been its most significant attempt at administrative decentralization. Under this project, one district from each of the twenty-six provinces was selected for participation on the basis of logistical convenience (for monitoring purposes), their representing a reasonable mix of local situations (e.g. poor and rich) and of advice of governors. What are the administrative, economic and political legacies of this project?

One method of answering this question is to divide all districts into two groups (PODP districts (N=26) and non-PODP or other districts (N= >265) and then compare them on several indicators. (See Table 2.)

Table 2. Comparison of PODP and Other Districts on Selected Indicators
(averages)