Decentralised Governance in Indonesia’s Disadvantaged Regions: A Critique of the Underperforming Model of Local Governance in Eastern Indonesia

Abstract

This article investigates the challenges facing decentralised governance in poor and underdeveloped areas in Eastern Indonesia. The Timor Tengah Selatan (TTS) regency in West Timor in Nusa Tenggara Timur province is taken as a case study. Indonesia’s radical decentralisation programme applied a national model of decentralised governance, not taking into account the different conditions applying to disadvantaged regions (daerah tertinggal, DRs). In the TTS regency, decentralised governance is underperforming in two core areas – administration and fiscal viability – while making some progress in political decentralisation. Governance is restricted by limited social capacity, a poor resource base, and a lack of investment capital and infrastructure. The question then arises: if the uniform model of decentralisation is not performing adequately in TTS, is there a more appropriate model of local governance and central subnational relations that can better perform in DRs? While not detailing the features of a new model, this article identifies the areas requiring policy development.

Introduction

This article investigates the challenges to decentralised governance in disadvantaged regions (daerah tertinggal, DRs) in Eastern Indonesia. It reviews the particular issues confronting the governance of Indonesia’s poorest regions in the context of the radical programme of decentralisation introduced in Indonesia from 2001, following the fall of the authoritarian New Order regime. It locates the investigation in terms of the comparative discussion of decentralisation in developing, poor states and then presents a case study. The regency selected for the case study is Timor Tengah Selatan (TTS), in West Timor, north-east of the provincial capital, Kupang, in the province of Nusa Tenggara Timur (NTT) province. It is a regency that, as explained below, on all rankings is one of the most disadvantaged in Indonesia. In the extensive literature on the programme of decentralisation in Indonesia, insufficient attention has been paid to the outcomes of the programme in Indonesia’s poorest regions – a gap that this article seeks to identify.

The purpose of the research project was to identify deficiencies in the present system of regional governance in DRs in the three core areas of fiscal, administrative and political decentralisation (Schneider, 2003: 33; Scott, 2009: 5–6). In assessing the performance of TTS in providing public service outcomes, the focus was on the delivery of health and education services, finding that they are seriously deficient. The question arises: if the uniform model of decentralisation is not performing adequately in regencies such as TTS, is there a more appropriate model of local governance that can better perform in DRs?

Research was conducted in the regency and through interviews with directors general in the key ministries in Jakarta. The consultations in Jakarta were with the Ministry of Villages, Development of Disadvantaged Regions and Transmigration (Kementerian Desa, Pembangunan Daerah Tertinggal dan Transmigrasi, abbreviated “Kemendesa”); the Ministry of Home Affairs (Kementerian Dalam Negeri, abbreviated “Kemendagri”), referred to here as MoHA; and the Ministry of Social Affairs (Kementerian Sosial, abbreviated “Kemensos”). Field work in TTS included interviews with political leaders and senior administrative officers, subdistrict heads, and medical and health personnel. A list of interviews is provided as an appendix.

The socio-economic profiles of NTT province and TTS as described later suggest that the problems of significant underdevelopment explain why the uniform model of decentralised governance adopted uniformly across Indonesia is significantly underperforming in DRs. Senior ministerial officials in Jakarta indicated an awareness of the underperformance of the subnational government generally and in DRs in particular, but efforts to review the model in DRs remain incomplete.

Radical Decentralisation as Democratic Empowerment

In 1999, in a climate of democratic enthusiasm following the fall of President Suharto and his authoritarian New Order regime, a team of Western-educated policymakers in Indonesia’s MoHA moved quickly to introduce an ambitious – indeed, radical – programme of decentralised governance (Fritzen, 2009).

The democratic movement that had removed President Suharto provided the political drive for the decentralisation programme (Hofman and Kaiser, 2002: 2). Decentralisation found its champion in President Habibie, who was keen to establish his reformist credentials (Allen, 2014: 231). Law 22/1999 and Law 25/1999 constituted the basis of the reform package. The intention, in the context of reform (reformasi), was to redistribute political responsibility from the centre to the regencies.

Law 25/1999 on the fiscal balance between central government and regional governments transferred financial powers and taxation authority to the regencies and the urban municipalities (kota), bypassing the provincial governments (Kesuma Nasution, 2016). The regions received general allocation grants and the right to collect taxes and other revenues. At the same time, the central government tightly supervised regional governments through unified auditing of regency budgets (Kesuma Nasution, 2016). Auditing and regulatory requirements were seen as onerous from the local view and sometimes appeared unworkable. The misuse of funds involves serious consequences, including fines and imprisonment for individual regional officers. Misused funds have to be returned to Jakarta.

The practical consequences of radical decentralisation were left largely unconsidered in 1999:

The drafting of decentralisation legislation has been characterized by many as hasty, inconsistent, radical, ‘big bang’, generally non-transparent […]. As a result of the haste of the adoption as well as its ambitious scope, the decentralisation laws suffered from many gaps and omissions that continue to dominate debate on administrative and political affairs. (Fritzen, 2009)

Many conceptual and operational problems were left unresolved; “the political factors that [drove] Indonesia’s attempts at decentralisation had overlooked technical and economic problems” (Suharyo, 2009: 76). The abrupt transfer of responsibilities was not followed up by equipping subnational governments with the administrative capacity to deliver public goods and to promote economic development (Nasution, 2017: 277).

Political decentralisation was the focus, to the neglect of policy development of the two other core dimensions of the decentralisation process: sustainable fiscal autonomy and efficacious administrative decentralisation. Fiscal decentralisation involves the transfer of the responsibility for expenditures and revenues from the national government to subnational jurisdictions. Administrative decentralisation involves the transfer of government functions and decision-making to subnational jurisdictions.

The literature distinguishes between three categories or stages of decentralisation: de-concentration, delegation, and devolution. De-concentration refers to the spatial and geographic transfer of administrative responsibilities to the field offices of central government ministries. Delegation goes further where policy responsibility is passed to semi-autonomous local governments, with central government exercising its authority through contractual arrangements with local government.

Devolution requires the central government to allow “quasi-autonomous local units of government to exercise power and control over the transferred policy” (Schneider, 2003: 38). Effective decentralisation depends upon the degree to which powers and resources are surrendered by the central state and the success or otherwise of subnational entities in deploying these powers and resources. The degree of devolution can be measured by the extent to which local government can attract and independently deploy revenues from taxes, transfers, grants and loans.

Political decentralisation involves the transfer of “mobilization, organisation, articulation, participation, contestation and aggregation of interests” to local political systems. It requires political representation directly through elections and formal public consultations to “map the multiplicity of citizens’ interests onto policy decisions” (Litvack et al., 2000 cited in Schneider, 2003: 39). Fiscal, administrative, and political dimensions operate distinctively and should be examined separately while recognising that they interact with each other (Schneider, 2003: 33–35)

This is the ideal. In practice, developing and, particularly, poor developing states face multiple challenges in achieving a successful programme of decentralisation. These challenges become particularly severe in the attempted transition from the de-concentration and delegation stages to effective devolution of decision-making and actual local control. The main finding of Scott’s (2009: 5) analytical review of the recent comparative literature on the links between decentralisation and development was:

There is a vast chasm between the benefits that proponents of decentralisation have claimed that reforms can have on service delivery, economic development and social cohesion and the reality, according to empirical research […] in fact, several authors argue that decentralisation can exacerbate long-term conflicts in fragile, socially divided contexts which indicate that a comprehensive rethinking of development policy in this area is needed.

The most commonly cited factors creating this dramatic difference between the potential benefits and the actual gains realised on the ground are the political subversion of decentralisation policies, limited administrative capacity, and financial constraints. The local political context is highly significant (Scott, 2009: 5–6). The case study of decentralisation in TTS will assess the regency’s performance against these inhibiting factors. The development of a model of local governance model suitable for DRs should address all these factors.

It is important not to generalise from a “whole-nation bias” (Snyder, 2001: 94) in considering subnational politics in Indonesia. Insofar as the regions of eastern Indonesia share the characteristics of poor less-developed states, they operate in a disadvantaged condition, unlike the generality of subnational jurisdictions nationwide. They face challenges in the uneven nature of the decentralisation process that do not confront better-endowed regencies in Java or Sumatra. Further, the mixture of developed and underdeveloped sectors in Indonesia requires different relationships between the central and the subnational levels rather than a single model of interaction that operates in Indonesia (Snyder, 2001: 98, 101).

In Indonesia, the de-concentration and delegation stages of decentralisation were bypassed in favour of the adoption of devolution, the immediate transfer of political as well as financial and managerial powers to subnational governments (Yusoff et al., 2016). Law 22/1999 on regional administrations gave broad authority to the regions in all areas of local administration. The central government, however, retains final authority over subnational governments and exercises this rigorously. Regency heads (bupati) were given the authority to issue local by-laws and local parliaments also produce regional regulations. This produced a degree of uncertainty and confusion, and where local legislation was considered in conflict with national legislation, the central government intervened (Butt, 2012).

Problems of Implementation

The experience of the TTS regency suggests that the model of decentralised governance applied there is significantly underperforming, particularly in the first two of these three core areas. This not to be unexpected in a regency with the characteristics of underdevelopment: poverty, unemployment, limited capacity, a lack of capital, a poor resource base, and a lack of infrastructure. TTS negatively experiences the regional disparity in Indonesia that seriously widened after decentralisation reforms (Talitha et al., 2019).

In terms of fiscal devolution, it will be explained that TTS lacks the capacity to independently produce its own revenues, leaving it almost completely dependent upon grants from the central government and the auditing regimes this entails. In terms of administrative devolution, the regency has its own bureaucracy and exercises decision-making powers over local policy, but the central government retains powers to overrule regency decisions. Issues of restricted administrative capacity further qualify the degree of administrative autonomy the regency commands. In terms of political devolution, however, it will be shown that this is the one core area where devolution has been more effective.

In the division of responsibilities transferred to the regencies, it soon became apparent that there was confusion about the distribution of power and roles between the three tiers of government: central, provincial, and regency. Law 32/2004 was promulgated as an attempt to address these problems of conflicting responsibility and practice (Hofman and Kaiser, 2002: 1; Kesuma Nasution, 2016: 444). This issue continued to reappear; the government attempted to resolve the issue by Law 23/2014 on the local government, with revisions in 2015.

The legislative review of the decentralisation policy has been iterative, introducing administrative changes, retracting them, and introducing more changes (The Asia Foundation, 2014). Our interviews with senior ministry officials in Jakarta and interviews conducted in TTS confirmed that currently significant policy review continues without a final resolution of the policy issues. The director general for regional autonomy in the MoHA confirmed that this created some uncertainty in decentralised operations (Interview, 20 September 2017). He noted that the central government has reconsidered the degree of autonomy granted to the districts, concerned about mismanagement, maladministration, and the possibility of local collusion over district resources that could lead to corruption. This raises the question of whether decentralisation in Indonesia has led to more or less corruption. That there is a connection between decentralised governance and decentralised corruption is supported by comparative studies (Fjeldstad, 2004). Recent policy changes suggest that the central government believes that this is occurring and has decided to exert firmer control. Regency subnational governments have recently had a number of responsibilities taken away from them, such as control of forests and mining, that are now shifted upwards to provincial government control. There is a concern in Jakarta that local elites will appropriate local resources for themselves (Allen, 2014; Hofman et al., 2009). The concern is that local political luminaries will act as little kings (raja kecil), presiding over networks of patronage and nepotism (Hidayat, 2017a: 68).

While some regency responsibilities have been recently retracted, Law of the Republic of Indonesia No. 6, 2014 concerning villages has provided new responsibilities for village heads (kepala desa) and village councils (déwan desa). Villages now have a degree of autonomy from the authority of higher levels of government, space for cultural diversity, and responsiveness to local aspirations (Antlöv et al., 2016). Village budget funds were now directly funded to allow them to exercise more independent fiscal control.

Village heads and village council members are not civil servants (Sir et al., 2019); they are not part of the government structure, but they act as agents of the central government. Their official status, therefore, is somewhat anomalous (Nurcholis et al., 2019). As with the regencies, there is a single model for village governance. The director general for village governance in the MoHA confirmed that there is no special model of governance for disadvantaged villages (desa tertinggal; Interview, 19 September 2017).

Law 6/2014 attempted to address problems of poor local participation at the village level and weak financial management capacity noted by a number of Indonesian observers (Wanusmawatie, 2014). Disadvantaged villages, which include the villages in the TTS regency, suffer from a very low Village Development Index (IDM), and the Kemendesa accepts that they require extra support to govern (Kemendesa, 2015, 2018). Our interviews in TTS confirmed that village chiefs and councils in TTS are overburdened with administrative responsibilities and, in fact, remain answerable to a confusing array of external authorities, an experience they now share with regency governments.

Challenges for Good Governance in Disadvantaged Regions

The most significant gap in the model of decentralised governance implemented from 2001 has been its failure to take sufficient, practical account of serious regional differences in capacity and resources (Kesuma Nasution, 2016: 111). Interviews with senior ministry staff in the Kemendesa and MoHA confirmed that despite the national government’s recognition of the problems of low capacity and limited resources facing DRs, there remained one basic policy model for all autonomous regions (otonomi daerah), from the richest and most developed to the poorest and most underdeveloped (Interview, 20 September 2017). There is some variation of the basic model: the central government has adopted special arrangements with Aceh, Papua, and West Papua to allocate a greater share of resource revenues to them through the tax sharing system (Vujanovic, 2017: 26).

The two key ministries that have overlapping but under-co-ordinated primary responsibilities for regency and village governance are the Kemendesa and the MoHA. The Ministry of Finance has further overlapping responsibilities in the villages, now making direct funding transfers to village councils. The Ministry of Social Affairs has further responsibilities. Other line ministries, such as Health and Education, also pursue their policy goals at the subnational levels. The vertical and horizontal operation of central, provincial, and district responsibilities has proved cumbersome and ambiguous (Hidayat, 2017b). Functional assignment between central and subnational governments is the essence of decentralisation, but it has proved difficult to achieve in practice (Purwanto and Pramusinto, 2018: 592).

Hill (2014) has noted that at the district level, “the richest region has a per capital income more than 50 times that of the poorest.”

Were they independent states, some parts of Indonesia would be classified as upper middle-income states, comparable to much richer Malaysia and Thailand, while other regions would be in the least developed group of extremely poor states.

NTT, considered as a separate jurisdictional entity, could be considered a fragile or “weak state” (Rotberg, 2013).

There are districts elsewhere in Indonesia where a combination of competent leadership and favourable local conditions produces more satisfactory governance outcomes. Franklin identifies such a district in Lamongan in East Java, which benefits from its ability to attract business investment because of its location close to Surabaya, Indonesia’s second largest city (Franklin, 2014: 157–159). However, the percentage of locally generated district finance “varies dramatically between provinces” (Fritzen, 2009). More developed regions enjoy a significant capacity to raise their own revenues; disadvantaged regions lack this capacity (Suharyo, 2009). In regencies such as TTS, domestic and international investment is negligible, discouraged by limitations of scale (Firman, 2009). In a comparative study of decentralisation in Latin America, Africa, and Asia, Robinson (2007) concluded that decentralisation has a negative impact on regions that are disadvantaged.

The Indonesian government is aware of the urgent need to reduce the gap in human welfare between DRs and the more developed regions. President Joko Widodo and his predecessor, Susilo Bambang Yudhoyono, made poverty reduction and capacity-building in Eastern Indonesia a major development goal. Law 17/2007 concerning Indonesia’s National Development Plan mandated the need for priority in the development of the welfare of the people living in DRs. Indonesian Government Regulation no. 78/2014 addressed the need to accelerate the development of “backward areas.” The Bali–NTT corridor has been identified as a major component of the Economic Master Plan (MP3EI). President Widodo visited Kupang and the regencies of Belu and Malaka in West Timor in December 2014 and December 2016, respectively, to announce major infrastructure projects.

In 2014 the government moved to adopt a National Strategy for Accelerated Development of Disadvantaged Regions (STRANAS-PPDT). Government Regulation No. 78/2014 recognised that “efforts to promote the general welfare and educating the nation have not fully materialised because there is a gap between developed regions with disadvantaged regions, in particular, in planned, systematic and sustainable [development].”

STRANAS-PPDT directed that subcriteria needed to be defined by the Ministerial Regulation in executing reforms for DRs. These regulations appeared not to have been put in place by 2016 (Hidayat, 2016).

Indonesia adopted a concept of good governance for its decentralisation programme that the international community had encouraged through such agencies as the United Nations Development Programme (UNDP) and the World Bank. The guiding elements of good governance included “open, transparent, accountable, efficient, effective and responsive administration” (Curmi, 2009: 46). Government came to be seen as only one, albeit a critically important, contributor to good governance (Cheema and Rondinelli, 2007a, 2007b: 1–2).

The Indonesian experience of decentralisation soon demonstrated that the political ideal of good governance was not easily realised in practice. The model proceeds from major normative assumptions that express assumed political and cultural interpretations. While promoted for the developing world, the concept originates in a Western understanding of the social preconditions required for its operation. In traditionalist social systems such as that in TTS, these preconditions may be absent; customary societies operate on different understandings of authority and responsibility.

The expectation was that democratic participation by local communities would require local representatives to be accountable and respond to community needs (Treisman, 2002). Local elections of heads of government (pilkada) and for the district legislature (pileg) would enable voters to participate in the development of policy as well as to choose their representatives. In practice, local elections in Indonesia may involve problems concerning corruption, money politics, and patronage (Hadiz, 2010; Mietzner, 2010). Ufen (2011) argues that pilkada elections encourage a fragmentation between political parties and candidates at the local level. Buehler and Tan (2007) reported weak political institutionalisation of the party systems in the local government. Hidayat (2017b) confirmed these claims in his study of the Jember regency in eastern Java.

A Socio-Economic Profile of TTS

TTS is the most densely populated area in NTT, with a population of 459,310 in 2016 (Badan Pusat Statistik, 2010, 2016). The regency is made up of thirty-two subdistricts (kecamatan), 266 villages (desa), and 12 village administration areas (kelurahan). Socio-economic profiles of the province of NTT and of the TTS regency identify the extent of the challenge facing local government reform in DRs. NTT is one of the poorest provinces in Indonesia. Hill (2014) lists NTT as having the lowest level of gross regional product (GRP) in all of Indonesia’s provinces. Its per capita incomes are less than half the national average. NTT had the highest poverty rate of all provinces in 2011 and included seven out of ten of Indonesia’s poorest districts (Hill, 2014: 85).

Around 80 per cent of budget revenue for TTS comes from Jakarta as part of the Public Allocated Funds (Dana Alokasi Umum, DAU) and some Special Allocated Funds (Dana Alokasi Khusus, DAK). The province also provides some financial assistance. Local income generated from taxes and service levies is minimal. The biggest expenditure is on government employees. Almost half of the domestic income was generated from agriculture, forestry, and fisheries (Badan Pusat Statistik, 2016: 246 and 286).

Poverty continues to be a chronic problem (World Food Programme, 2013). This is reflected in the low Human Development Index for NTT, which is ranked third-last of the country’s thirty-four provinces. TTS is ranked at the level of a least developed country (UNDP Factsheet, 2019). Livelihoods are overwhelmingly dependent upon general subsistence agriculture, with rural incomes around six times lower than urban incomes. Subsistence-based rain-fed crop production leaves rural communities highly dependent upon unreliable climate conditions. Food security is a major issue (United Nations Development Programme (UNDP) Fact Sheet, 2019). Half of Indonesia’s subdistricts in the highest category of food insecurity are located in the TTS regency. Chronic malnutrition affects over half of children under the age of five. Only one in three households have access to clean drinking water (World Food Programme, 2013).

Rates of underemployment are high. The largest employment category is agriculture, forestry, hunting, and fisheries, employing 74 per cent of the workforce. Of the workforce, over 27,000 never attended school, around 46,700 did not finish primary school, and around 70,000 completed primary school. Consequently, illiteracy levels are high. Manufacturing is limited, with only 11 medium or large businesses having 100 or more employees. There is a growing mining industry but mining demands heavy investment in infrastructure, including power, roads, ports, and other support and service industries; lack of infrastructure deters mining investment. There are only a few modest hotels in SoE, serving only 657 recorded international tourists in 2015. Infrastructure is basic. Less than half of the 1688 kilometres of roads are classified as “good”; 234 kilometres are classified as “damaged” and 387 kilometres as “badly damaged” (Badan Pusat Statistik, 2016: 231).

The difficulties of providing basic public services such as health and education operate within the particular sociocultural context of TTS. As mentioned, the concept of good governance proceeds from an introduced understanding of civil society. In traditionalist, least developed regions such as TTS, “civil society” is constituted in quite a different manner. TTS continues to consist of largely customary (adat) local communities. Regency governments need to successfully manage a dual system of local governance.

The central government encourages the inclusion of customary practice. Law of the Republic of Indonesia No. 23, 2014 on the local government states (Article 31:2 1/f ) that in the implementation of regional decentralisation, it is intended to “maintain the uniqueness of customs, traditions and culture of the region.” Law 6/2014 concerning villages similarly recognises and strongly supports the viability of customary villages. Village heads and councils, however, derive their authority from an introduced political system; they are locally elected but their position is not held according to customary law. It has been claimed that they actually act in contradiction to adat (Nurcholis et al., 2019: 391).

The national indigenous people’s organisation Aliansi Masyarakat Adat Nusantara (AMAN), established in 1999, also in the context of reformasi, defines a customary community (masyarakat adat) as:

a population group that lives on the basis of their hereditary ancestral origins in a specific geographical region; that possesses value system, society and culture of their own; that possesses sovereignty over their land and natural resources; and that orders and manages their social life through customary law and institutions. (Avonius, 2009)

The communities in the rural population of the TTS regency fit this definition. The duality of local governance in TTS, as in neighbouring rural East Timor, operates on “a complex melding of customary and state-based institutions, drawing on the world-view of both” (Cummins, 2015: 13). If the liberal-democratic system of governance is overly predominant, it threatens the survival of customary identity. At the same time, if the state-based legal-rational system fails to establish itself, decentralisation cannot provide the type of good governance expected in the approved development model. A hybrid traditional-modern system of political governance can also provide opportunities for those with local power to subvert and manipulate local resources. The chiefs (luirai) of East Timor, for instance, have exercised a parallel political authority in local communities alongside the state-based authority and this operates in TTS (Shoesmith, 2010; Interview, Mella, 2 January 2016).

The View from Jakarta

The situation determining the success or otherwise of district governance in TTS is an outcome both of local conditions and of government policy in Jakarta. The two key ministries that have overlapping but under-co-ordinated primary responsibilities for regency and village governance are the Kemendesa and the MoHA. The Ministry of Finance has further overlapping responsibilities in the villages, now making direct funding transfers to village councils. The Ministry of Social Affairs has further responsibilities. Other line ministries, such as Health and Education, also pursue their policy goals at the subnational levels. The vertical and horizontal operation of central, provincial, and district responsibilities has proved cumbersome and ambiguous (Hidayat, 2017b). Functional assignment between central and subnational governments is the essence of decentralisation, but it has proved difficult to achieve in practice (Purwanto and Pramusinto, 2018: 592).

During visits to the ministries, it appeared that senior staff tended to operate within their own separate ministerial brief rather than work collaboratively with cognate ministries. There is limited consultation at the ministerial level in Jakarta on the subnational government. This was acknowledged by the head of planning and director general designate of the Kemendesa (Interview, 19 September 2017). When asked how closely the Kemendesa worked with the MoHA, the director general for otonomi daerah in the MoHA replied that bilateral inter-ministerial collaboration was quite limited and that co-ordination tended to occur at cabinet level rather than between ministries (Interview, 20 September 2017).

The Kemendesa has overlapping responsibilities in the district government with the MoHA. The Kemendesa is responsible for projects down to the village level. The MoHA is responsible for governance issues at the regency level; development issues are the responsibility of the Kemendesa. The MoHA has the role of overseeing economic and political decentralisation and providing overall guidance to the districts with regard to spatial and economic development planning (Interview, Sumarsono, 20 September 2017). The Ministry’s brief covers the formulation of policies on decentralisation, human resources development, development of regional administration, including village administration, local government affairs, local politics and public administration, and supervision of regional administration.

The MoHA oversees the performance of the heads of districts: the director general for otonomi daerah in the MoHA supervises the bupati. He applies a “carrot and stick approach”: if they perform well they are rewarded; if do not, they must attend training programmes and there may be penalties for unsatisfactory management (Interview, 20 September 2017). The director general spoke frankly concerning his Ministry’s view that there were problems of some magnitude in district governance. For instance, forestry governance, including logging, needed central co-ordination to save forests because at the district level there was connivance with the loggers. Some functions had been relocated from the district to the provincial level, because there was less temptation for self-interested or corrupt practices. Higher education had been transferred to the provinces as well as forest management and energy and mining (these changes were introduced by Law 23/2014).

The interviews with senior staff in the ministries tended to present a fairly unpromising opinion of the capacity of regencies in NTT districts to meet basic standards of competent government. TTS experiences problems in attempting to meet the requirements for attracting central funding directed to assist DRs. An official in the Ministry of Social Affairs responsible for poverty alleviation observed that the districts in NTT had been unsuccessful in attracting funding because they failed to do the work necessary to lodge a submission. In 2016, there was a submission from TTS for Rp 750 million ($US 58,000) to restore fifty dwellings to make them fit for habitation. This was approved, and TTS received the money, but they failed to use it and the funds had to be returned (Interview, 20 September 2017).

The View from the Regency

The capital, SoE, is the location of government offices and the district parliament (the Dewan Perwakilan Raykat Daerah, DPRD). As of 2014–2019, the DPRD was made up of forty elected members from nine political parties, with none holding an outright majority (Pemerintah Kabupaten, 2019; Susunan Anggota Legislatif, 2019). The National Party of Democracy (Partai Nasional Demokrasi, or “Nasdem”) was the largest, with six seats. This fragmentation of political parties may weaken the role of the DPRD but it probably means that “rent opportunities” for exploiting state resources in the district are dispersed rather than concentrated (Allen, 2014).

The head of the regency, the bupati, at the time of the study was Paulus Victor Rolland Mella, who was elected twice as district head, first for 2009–2014 and again for 2014–2019. His district at the time of the study was administered by 8,173 civil servants (Badan Pusat Statistik, 2016: 20). In 2019, as his second term neared its end, he announced that he and the vice-regent, Obed Naitboho, were resigning their positions in TTS. Mella ran with the Golkar party in the 17 April 2019 election for the national parliament, the DPR, while Naitboho ran with Nasdem for the NTT provincial DPR. Mella failed to win a seat, but Naitboho was successful.

In terms of the expansion of democratic space through political decentralisation, the programme introduced from 2001 and the adoption of direct local elections for district leaders and district councils from 2005 seem to have largely achieved that objective. The bupati and the subdistrict heads we interviewed agreed that the decentralisation programme had brought important democratic benefits and a new ability to exercise local decisions on policy and finance (Interview, 6 January 2016). Bupati Mella praised the reforms as providing a real opportunity for people to participate, opening a larger space for democratic politics and local autonomy.

There is a continuous process of community consultations for planned development, known as Musrenbang (Musyawarah Perencanaan Pembangunan). Independent candidates can contest local elections although only wealthy independent candidates can expect to find the necessary funds for a successful campaign as Simandjuntak found in North Sumatra (2012: 123). The DPRD leaders we interviewed were members of the TTS regency elite. While local government reform has been welcomed, the bupati, subdistrict heads, senior civil servants in planning, health, and education, and members of the district parliament all expressed serious concerns regarding the lack of resources while having to comply with demanding ministries in Jakarta.

A downside of democratic decentralisation, according to critics, is the expansion of the space for rent-seeking and patron–client politics (Allen, 2014). Locals acted as citizens (voters) and clients interchangeably, a system that could be called Patronage Democracy. Law 30/2002 allowed for the establishment of an independent Corruption Eradication Commission in 2003 (Erb, 2011). Law 32/2004 was enacted to replace Law 22/1999 in an attempt to address the problem of corruption, “which had become rampant upon the devolution of power to the local parliaments” (Simandjuntak, 2012: 105). This law strengthened the position of the district head, now directly elected by the people of the district and not selected by the local parliament. This may have had the unintended effect of shifting some of the opportunities for rent-seeking to the district chiefs (Erb, 2011: 185).

NTT has been identified not only as the poorest region in Indonesia, but by some observers as the most corrupt (Erb, 2011: 171, 191). Corruption, collusion and nepotism (Korupsi, Kolusi, Nepotisme, KKN) have been a particular issue in TTS. The delegation of power to a district parliament in 1999 was marked by a number of charges of corruption against the first, appointed representatives to the DPRD. Jean Neonufa, speaker of the DPRD, reported that as many as thirty-five members of the DPRD had been investigated and a number convicted of corruption over the 1999–2004 period (Interview, 8 January 2016). The situation appears to have improved with the introduction of elected representatives. People have an understanding of their rights grounded in strong, surviving traditional identities. At the same time, the district political executive, the members of the district parliament, and the senior administrative staff at the centre of district government inevitably occupy a different and privileged domain from that of rural communities.

Allegations of misuse of district funds and property in TTS regency have appeared in the press. In September 2014, Fajar Timor published allegations that fictitious employees had been listed in the claims for salary and allowances payments for district civil servants in sixty-one of the TTS Regional Work Units (SKPDs). Although the allegations referred to reports in 2008, before Paul Mella was in place as regent, there was an attempt to somehow implicate him (Lerek, 2014). There was an ongoing hearing in 2017 by the Kupang Corruption Eradication Commission on alleged improper use of the TTS regency consumption funds (Teropong, 2017; Timor Express, 2017). The secretary of TTS, Salmun Tabun, was named a suspect in the hearing after Bupati Mella had been examined for six hours as a witness. The claim was that Mella had issued a decree on taking office to draw on regency funds to cover the costs of his inauguration. There was also a dispute between the regent and more than seventy contractors who claimed he had not paid them arrears of Rp 88 billion ($US 6.8 million) and threatened to sue (Victory News, 2019). After the vice-regent, Obed Naitboho, resigned to contest the national parliamentary elections, the incoming regent, Epy Tahun, complained to reporters that Naitboho had not returned his official car and had attempted to keep state property (Kupang, 2019). It must be noted that there were no findings by the Kupang Corruption Eradication Commission or any other investigative agencies against Regent Mella.

Relations between the regent’s office and the members of the district parliament could be acrimonious (there was a physical confrontation between a DPRD member and a senior civil servant around the time of our first field trip). DPRD members of the faction led by representative Nasdem Hendrikus Babys demanded the regent’s resignation in May 2016 on claims he had failed to follow through mining permits and the expenditure of district funds (EXPO NT, 2016). He was also criticised for the poor performance of the heads of the districts’ SKPDs.

The Delivery of Health and Education Services

Channa and Faguet (2016) note that decentralisation is probably the single most advocated measure for improving the provision of health and education in the developing world. They observe in their review of the literature, however, that the impact of decentralisation on service delivery in health and education is weak and incomplete. Assessments of the performance of the district health system in Indonesia have shown an overall low level of performance with wide variations between districts (Heywood and Choi, 2010). The central government tends to retain control over the health and education sectors while transferring responsibility to the regency. A recent study has reported that the allocation of function “does not suggest that district-level government [is] the basis of decentralisation”; the central government remains dominant (Purwanto and Pramusinto, 2018).

Health services in TTS are seriously below national standards. The only hospital in TTS is rated Category C (A is the highest category). The general hospital has four medical specialists to service the entire province, a ratio six times below the national standard. Every subdistrict has a health clinic, but most are under-resourced and many still do not have running water. The infant mortality rate is high. TTS regency has the highest infant mortality rate in East Nusa Tenggara Province (Littik et al., 2017a, 2017b).

The regency struggles to manage the resources it has. We heard in interviews that a sum of Rp 700 million ($US 54,000) had been provided to TTS for its health services at the hospital. The money was received, but the regency department responsible for using the funds was unable to have all the necessary documentation ready or to abide by the regulations for the use of the funds. Consequently, the money was not used and had to be returned to Jakarta.

Both Bupati Mella and the head of the regency health services, Ibu Fosiani, identified severely limited resources in the health sector as a major problem (Interviews, 1 and 5 January 2016). The central government requires nurses to have completed a Certificate 4 in medical training; in the past, local nurses only had a Certificate 1. After some effort, a number of nurses were brought up to Certificate 3, still below the national requirement (Interview, 2 January 2016). The Bupati said that in an attempt to attract more doctors to TTS, the administration decided to offer increased monthly salaries. This was disallowed by the Ministry of Health. Despite a recruitment campaign, the regency was unable to hire any new doctors.

The education system in 2016 had around 8600 teachers and 125,000 students at primary, middle, and senior high school (Badan Pusat Statistik, 2016: 59). On average, people attended school for 6.3 years in TTS, compared to the seven-year provincial average, itself low by national standards. A subdistrict chief in TTS said that children preferred to work and earn money than go to school because they saw that even with an education there were few job opportunities (Interview, 7 January 2016).

Regulations periodically issued from Jakarta demand total compliance. Teachers must have a degree and a certificate. The district administration says it has no other option than to ignore these requirements; the national standard simply does not work in TTS (Interview with Bupati, 2 January 2016). The authority to employ teachers sits with the ministry in Jakarta. In fact, many are appointed locally. Most are not adequately qualified, lacking university teaching degrees, and certificates, and many have just senior high school certificates. School children may have to walk more than two or three kilometres to school and back, some much longer distances. Of the thirty-two subdistricts, seven still do not have a high school or a vocational school, so students may have to travel up to thirteen kilometres to reach the nearest school facility (Mella, 2018). A further issue is the difficulty the regency experiences in complying with strict, detailed requirements for school construction when designated building materials are unavailable. Failure to comply can involve fines and individual penalties.

Conclusion

The project of radical, democratic decentralisation pursued in Indonesia since 1999 has produced notable achievements accompanied by some significant drawbacks. The key political achievement has been the expansion of democratic space for the local community. At the same time, this has opened new opportunities for local élites to appropriate state resources for their own benefit.

The experience of decentralisation has demonstrated that the ideal of good governance has not been easy to realise in political terms or in terms of the provision of adequate and accessible public services in Eastern Indonesia, confirming Scott’s (2009) review of contemporary judgements of the gap between the expectations and the actual outcomes of decentralisation in under-developed areas. The central government has struggled to address issues of ambiguity and inadequate coordination between central, provincial, district, and village authorities. The legislative review of decentralisation policy has been iterative, introducing policy changes, retracting them, and introducing more changes.

The Eastern Indonesian case is particularly instructive of the need for a rethinking of development policy in regions marked by the political confusion of the management of decentralisation and the consequences of limited administrative capacity and financial constraints (Scott, 2009: 5–6). The most obvious gap in the model implemented since 2001 is the failure to take into account adequately the serious regional differences in terms of the problems confronting DRs. The NTT province and its regencies start from the situation that they operate under the conditions of “extremely poor states,” now carrying autonomous responsibilities but effectively almost totally financially dependent on the central state.

Interviews with the key ministries in Jakarta confirmed that central policymakers recognise that in terms of successful local government reform, one size does not fit all. There have been recent attempts to address this problem for DRs, represented by Regulation No. 78 2014 on accelerating the development of backward areas. The principles set out in that regulation had not been translated into practice at the time of the case study.

The view of those interviewed in the regency was that the situation in TTS was better than under the New Order regime. Financial autonomy enabled the regency government to make decisions that responded to local needs more closely than total central control from Jakarta would allow. However, on the core dimension of fiscal autonomy, TTS is unable to generate independent income and consequently is heavily dependent upon grants from Jakarta. This involves complying with detailed and imposed requirements that the regency administration struggles to satisfy. The degree of fiscal autonomy is consequently quite low. The ongoing problems concern the difficult relationship between the regency government and the central ministries where national standards in health and education services, for instance, simply cannot be reached. Central government demands for compliance with national regulations were seen as onerous and often unsupportable. A second set of problems concerned the very limited human and natural resources available. This again was clearly the problem in the health and education sectors. The model of good governance in TTS needs to simultaneously operate in a state-based and customary context, producing a hybrid system of local government that presents its own challenges. In terms of the core dimension of administrative autonomy, TTS has made limited progress. According to local informants, real progress has only been achieved in one of the three core dimensions: political autonomy.

In summary, the policy challenge is to achieve inclusive, representative, transparent, and efficacious decentralised good governance in DRs in Indonesia, a sustainable and specific model that works in DR regencies, a model that ensures democratic engagement of the local community, respecting and affirming local adat, while ensuring that it delivers optimum public services and operates successfully within the state system.

In order to develop a successful model, there is a need to clarify the alignment of the respective roles and responsibilities of the four levels of government: central, provincial, regency, and village, as these operate in DRs. There appears to be a need to improve the co-operation and co-ordination between the key Jakarta ministries, most importantly between the Kemendesa and the MoHA to overcome their currently ambiguous and overlapping responsibilities. The MoHA has the primary role of overseeing the politics and public administration of the regencies and villages. The Indonesian Cabinet could consider improving the overlapping roles of central government agencies in local government by assigning the MoHA a position as a co-ordinating lead ministry in overseeing subnational governance. At least, the lines of communication between ministries and their directors general with local government responsibilities could be improved.

The Indonesian government accepts the need to develop a better model of central support for DR regencies. This should include a rethink of the model of funding, probably the provision of more necessary human resources from outside, such as health and education professionals, and perhaps the deployment of more suitably trained central administrative staff into DRs. It may be necessary to narrow the scope of complete devolution in DRs and to partly reverse the model to something closer to delegation in terms of administrative and fiscal decentralisation, where semi-autonomous district governments are not directly controlled by the centre but are yet more directly managed by it (Yusoff et al., 2016). Provincial governments have already resumed some areas of responsibility previously allocated to regencies; a review of the role of provincial government in disadvantaged areas would be useful. The issue is how to rework the roles of the centre, the province, and the regency in DRs while preserving the real gains in representative political autonomy at the regency level. The larger, long-term challenge, of course, is to address the conditions of severe underdevelopment that face DRs in eastern Indonesia to the point where the conditions for successful decentralised governance are present locally.