Varieties of disappointment: why has decentralization not delivered on its promises in Southeast Asia?

• Author: Malesky, Edmund J.

1. Introduction

   Few policy prescriptions are as theoretically enticing as decentralization. Scholars have argued that bringing administrative, fiscal, and political decisions closer to the individuals most affected by those choices would: generate productivity gains (Oates 1972); enable better-tailored policies (Besley and Coate 2003); and reduce corruption (Fisman and Gatti 2002; Bardhan and Mookherjee 2008). Furthermore, greater experience with participatory processes would plant seeds for future democracy (Manin, Stokes and Przeworski 1999, p. 10; Fung 2004). It was also argued that decentralization would release furious competition between sub-national units that would lead to enhanced public service delivery and innovation, as local leaders sought to attract labour, capital, and votes (Tiebout 1956; Inman and Rubinfeld 1997; Weingast 1995; Montinola, Qian and Weingast 1995).

   These theoretical predictions had significant influence on policymakers and the international development community. In the 1990s, a wave of decentralization spread across the world with almost every sizeable country devolving some sort of responsibilities to sub-national units (Rodden 2006). Many of these programmes were advised or funded by multilateral and bilateral donors. Countries varied, of course, in the nature of decentralization they chose and the scale of the devolution, but around the world there was a strong faith that decentralization was an elixir that would heal many of the ills of existing political structures.

   Southeast Asian countries were caught up in the decentralization "wave" to differing degrees. On one hand, the Philippines and Indonesia embarked on far-reaching reforms to devolve revenue-raising and service delivery responsibilities to sub-national governments. In its 2005 review of decentralization in East Asia, the World Bank labelled these two countries as "fast starters". Countries such as Vietnam were termed "incrementalists" and Cambodia and Thailand labelled as "cautious reformers", due to their gradual and piecemeal approach to decentralization (White and Smoke 2005).

   With time, however, it became clear that the decentralization potion was not quite as magical as had been anticipated (Prud'homme 1995). Empirically, the excitement about the benefits of local autonomy was not met by improvements in measurable outcomes, as the World Bank demonstrated in a meta-analysis of 500 studies of decentralization (Mansuri and Rao 2013). The gap between prediction and reality has become obvious to both citizens and elites. Consequently, around the world, we are now observing a partial rollback of reforms (Eaton 2004; Dickovick 2011).

   These trends are observable in Southeast Asia as well. As the authors in this special issue document, the Indonesian government implemented a series of measures that strengthened the supervisory powers and coordination responsibilities of provincial governments over municipal and regency governments in 2004, and top politicians called for the end of district elections in the wake of the 2014 presidential election. In 2014, Thailand's interim constitution suspended the election of the heads of provincial administrative organizations and subdistricts (tambons), and foresees their replacement by appointed officials upon completion of their current term (Pawakapan 2014). For its part, Vietnam piloted the elimination of elections for local people's councils in a number of locations, in order to see if this reduces organizational duplication and inefficiency (Malesky, Nguyen and Tran 2014).

   Given these events, we decided it was time to carry out a comparative evaluation of the experiences of Southeast Asian countries with regard to decentralization on one hand, and recentralization on the other. However, at present, the literature is ill-suited to offer empirical predictions on recentralization efforts for three reasons. First, while each of the countries in Southeast Asia has a dense literature on decentralization, these have largely been confined to the borders of the specific state under investigation. Direct comparison has been hampered by: the unique constellations of decentralization choices in each country; variance in the desired outcomes of stakeholders; and case-specific terminology that complicates identifying cross-case analogues. Second, as our contributors demonstrate, the instances of recentralization are new and scattered, making it difficult to propose general theories. Third, as we noted above, the extant literature on decentralization is fiercely contested, offering few findings that withstand theoretical and empirical scrutiny. There are very few agreed-upon facts that can serve to anchor the discussions.

   To resolve these problems, we commissioned a set of articles from world-class scholars on the region, asking them to reflect on the decentralization/recentralization debates within their countries of study. To facilitate comparisons, however, we asked all of the authors to follow a common format and lexicon for describing elements of decentralization, allowing readers to easily draw out similarities and differences across the research settings and facilitating structured comparisons regarding: the history of subnational administrative developments; the politics behind recentralization/decentralization debates; the specific changes in sub-national institutions; and the economic and political outcomes of institutional changes. As readers will see, the articles in this special issue are excellent. While all adhere to the suggested format, they work within those confines in fascinating ways, playing to their authors' strengths and to the parameters of the debates in their respective countries. What emerges is a complex and nuanced picture of the origins, application, and outcomes of the decentralization and recentralization programmes.

   An unmistakable tone of negativity pervades the articles in the issue. Except for Myanmar, which is only now embarking on its decentralization experiment, authors express either disappointment that decentralization did not achieve its lofty goals (Indonesia and Philippines) or was never given a chance to succeed by central leaders who: were reluctant to fully devolve power; issued contradictory legislation that undermined decentralization's effectiveness; or used alternative levers to recentralize authority at the same time (Vietnam, Cambodia and Thailand).

   In this introductory article, we explore some of the unifying themes that help explain the disenchantment. First, a common thread appears to be that hand-wringing about the missing economic benefits of decentralization may be misplaced, as the primary goals for decentralization may have been non-economic including responding to previous authoritarian periods and creating space for regional identities. Second, the theoretical models on decentralization discussed above relied on restrictive assumptions for success that were not met in our Southeast Asian cases. Third, the complexities of decentralization meant that a wide array of powers needed to be devolved to sub-national governments. Within each country, different mixes of sub-national powers were chosen, leading to wide varieties in the type, scope, and jurisdiction of authorities. This has generated a number of internal contradictions that have slowed and confused implementation. Finally, the authors of the article on Thailand, Unger and Mahakanjana, make the provocative argument that the failure of decentralization reforms in that country is in large part the result of the end-users--the citizens and businesses who have failed to organize, mobilize, and take advantage of their newfound powers.

   Surprisingly, profound disappointment regarding the benefits of decentralization has not led to extensive reversals. In all of the cases above, recentralization efforts were not implemented or were abandoned midstream. All countries in our analysis therefore rest in an odd halfway house with regard to the architecture of their polities. The past forms of local-central power-sharing have not borne fruit, but the alternatives are unclear and, in the case of Philippines and Indonesia, unpopular. The next research agenda, perhaps inspired by the New Fiscal Federalism literature (Weingast 2014) will need to point the way to hybrid systems that are able to capture the benefits of devolving local authority without falling prey to its excesses.

2. Types of Decentralization

   Before delving into the details of the articles, it is helpful to develop a conceptual vocabulary that will be used throughout the issue, regarding the features of decentralization and recentralization. Scholars generally distinguish between modes of decentralization in three ways: (1) the arenas in which power has been granted to local authorities; (2) the extent of the power provided; and (3) the level or node in the government hierarchy invested with the authority (Failed 2010).

   2.1 Types of Decentralization

   Three types of responsibilities are commonly decentralized. Fiscal decentralization provides local governments with the power to: tax citizens and business; raise money through borrowing either domestically or overseas; and decide how to spend that money through the preparation and implementation of local budgets. (1) On paper, the most comprehensive fiscal decentralization occurred in Indonesia, which granted districts greater authority over revenue collection by increasing the types of taxes available, and also provided greater autonomy over expenditures. Ironically, Ostwald, Tajime and Sampantharak (henceforth, OTS), the authors of the article on Indonesia, conclude that fiscal decentralization in that country did not go as far as it could have, and even lagged behind other types, leaving most districts reliant on central transfers (p. 147). According to Unger and Mahakanjana, Thailand's 1999 reforms also involved comprehensive fiscal decentralization, which...