Institutions, Outputs and Outcomes Two Decades of Decentralization and State Capacity in Indonesia.

• Author: Nugroho, Yanuar

1. Introduction

   The two biggest hallmarks of the 1998 Reformasi process in Indonesia are democratization and decentralization, which have transformed the country, not just in terms of politics, but more importantly in terms of administration. Decentralization, both political and administrative, is indeed the other side of the democratization coin (Crook and Manor 1998; Gopal 2008; Grindle 2007; Rondinelli and Cheema 1983).

   By giving subnational governments greater discretion in pursuing their development objectives, decentralization was aimed at fostering regional development from below (Rondinelli and Cheema 1983). This includes building administrative capacity at the local level. As subnational governments provide key services to citizens, improvements in capacity and service delivery in turn are key to the achievement of developmental goals. However, despite the rhetoric and theory of decentralization promising better capacity of local administrations, the practice over more than two decades suggests that decentralization can also result in unfulfilled expectations and the emergence of unanticipated problems (Cheema and Rondinelli 2007; Gopal 2008; Grindle 2007; Malesky and Hutchinson 2016).

   With regard to Southeast Asia, Malesky and Hutchinson (2016) set out how and why decentralization has not delivered on its promises in the region. Either it does not achieve its goals, or it has not been given a chance to succeed by central government leaders who are reluctant to devolve their power. In the latter case, national leaders issue contradictory legislations that undermine the effectiveness of decentralization reforms or use alternative methods to recentralize authority.

   On Indonesia, Ostwald, Tajima and Samphantharak (2016) note that decentralization has created a new class of regional political elites, who have pushed to shift significant power to the subnational level. While the political motivations of mitigating centrifugal pressers appear to be vindicated, the anticipated gains in service provision and downstream economic impacts have not uniformly materialized, as efficiency gains in some regions have been offset by the widespread emergence of clientelist practices and fiscal inefficiencies. The tension between the central and local governments also remains under decentralization. Von Luebke (2009) finds government leadership is an important, but often underestimated, policy determinant that can compensate for (or aggravate) weak societal checks in transitioning economies such as Indonesia.

   Reflecting on two decades of decentralization in Indonesia, this study raises an overarching question: has decentralization delivered on its promises, particularly as it pertains to state capacity? We investigate this in two areas, namely how and to what extent has decentralization led to measurable improvements in: the structure of public administration; and the provision of local public services.

   There have been several studies focusing on the relationship of decentralization reforms and subnational development in Indonesia, including: Vujanovic (2017); Firman (2009); Thufail (2016); Hill and Vidyattama (2016); Hofman and Kaiser (2004); Kimura (2010); Resosudarmo and Vidyattama (2006); and Talitha, Firman and Hudalah (2020). However, most of those studies focus either on decentralization and regional or spatial economic development. This paper contributes to the analysis of the relationship between state capacity and decentralization outcomes in Indonesia. To this end, we base our investigation on existing research regarding the importance of institutions as key determinants of the decentralization process. The existing literature focusses on local governance and local government leaders (Von Luebke 2009), local government proliferation (Lewis 2017), district-level implementation of particular services such as health services (Fossati 2016b) and local political institutions and politics (Ostwald, Tajima and Samphantharak 2016). This paper, for its part, focuses on state capacity as a lens to understand Indonesia's decentralization reforms and their implications for effectively achieving development outputs and outcomes.

   Based on theoretical arguments on state capacity by Tilly and Besteman (1985), Fukuyama (2014) and Berwick and Christia (2018), we propose a cross-cutting framework to analyse the linkage between decentralization reforms and regional development. To this end, the framework of "state capacity" is used to assess whether decentralization has managed to achieve the following outcomes: first, improve the capacity of local governments to create regulatory frameworks, establish local institutions, provide accountability mechanisms, and deliver development outputs and outcomes; and second, improve the capability of the central government to manage decentralization through effective coordination across ministries and subnational governments.

   The main findings show that decentralization in Indonesia has, in many areas, managed to achieve partial improvements. Yet, at the same time, it has failed to deliver in others. The capacity of subnational governments to deliver developmental outcomes has increased, but disparities and gaps remain. These are linked to varying levels of local government capacity to deliver public goods, particularly between local governments in Java-Bali on one hand, and those in Papua, Maluku and Nusa Tenggara Timur on the other.

   We discovered weak local capacity in creating supportive regulation for: improving access to and quality of public services; establishing public administrative reform; and providing accountability mechanisms. Our study highlights weak policy coordination across ministries and between the central government and subnational governments. The ensuing redundancies, ambiguities, and confusion lead to low capacity at the local level to deliver developmental outcomes. Consequently, the national government still needs to improve its capacity to coordinate and manage decentralization, particularly reconciling tensions between central and subnational interests. Elite capture and political intervention at the national and subnational levels remains high, and bureaucracy at the national and subnational levels continue to serve as an extension of elite interests.

   Our findings are in accord with studies indicating that the anticipated gains in service provision and downstream economic effects from Indonesia's decentralization reform have not uniformly materialized (Von Luebke 2009; Lewis 2010, 2017; Fossati 2016a; Ostwald, Tajima and Samphantharak 2016). While local governments are responsive to the needs of local people to some extent (Ostwald, Tajima and Samphantharak 2016; Lewis 2010, 2017), there are disappointing outcomes due to: weak state capacity to create supportive regulation (Malley 2003); corruptible local accountability mechanisms (Fossati 2018); and thwarted public administration reform (Purwanto and Pramusinto 2018).

   The next section discusses relevant theoretical work before laying out the framework used to analyse the diverse outcomes of decentralization reform in Indonesia. We follow with a discussion of the relevant decentralization regulations. Next, we compare the achievement of developmental goals before and after the decentralization. Subsequently, we elaborate how state capacity in local government can be measured. (1) The last two sections discuss our main findings and then chart a way forward.

2. Decentralization and State Capacity

   Broadly, scholars agree that decentralization is a process that unfolds over time and is neither linear nor one that necessarily results in a single outcome. Hence, they propose shifting the debate from weighing the promises or perils of decentralization to examining the causal explanations of the diverse outcomes that decentralization reforms engender (Jutting et al. 2005; Grindle 2007; Hansjorg and Junghun 2016).

   In the Indonesian case, Fossati (2016b) seeks to identify what has and has not worked. In his research on multilevel politics, coordination, and local health insurance schemes, he finds that regions where cooperation between provincial and district authorities has emerged display systematically higher levels of health insurance coverage. He holds that the positive effect of cooperation between levels of government does not depend exclusively on patronage networks. Patunru, McCulloch and von Luebke (2012) show that relationship-based, rather than rule-based, cooperation between government leaders and local firms can provide an effective mechanism to boost investment and improve local investment climates. By taking a detailed look into the political economy of the Javanese city of Solo, Von Luebke, McCulloch and Patunru (2009) find that informal, relation-based cooperation can provide a constructive platform for policy reform.

   Fossati (2016b) also examines the impact of local democratic institutions on decentralization. He found that local governments are, to some extent, responsive to the needs of the most vulnerable. In years when local elections (pilkada) are held, low-income households are targeted more accurately, suggesting that electoral incentives for local elites may increase access to social services among the poor. However, this positive effect of direct local elections is only limited to districts with electorally competitive politics. Lewis and Hendrawan (2019) find that when large multiparty coalitions are newly elected, there is a shift in local government spending towards health sector activities. Yet, this positive effect is short-lived and disappears after a year or two. It is suspected that funds are then diverted to prepare for the subsequent electoral cycle.

   Fossati's work on the resurgence of ideology in Indonesia, political Islam, aliran (ideological and partisan groups) and political behaviour highlights the influence of religious political institutions following...