Indonesia began
a rapid decentralization framework in 1999. Its intention was not only to
transform governmental structure from centralized regime to democratic one, but
also to restore relationship pattern among actors of development, i.e. to
construct and to strengthen the notion of good governance.
In many ways,
decentralization and (good) governance has been reciprocally explaining.
Decentralization is a process of transferring
power, resources and responsibilities from the central to the sub-national
levels of government, while governance is a new mode of government
characterized by heterarchy rather than by hierarchy, creating a horizontal
relationship among a multitude of actors both public and private.
Decentralization aims to promote good governance by enabling citizen
participation and democratic elections. In other words, the outcome of
decentralization should result in a more participatory government for citizens
and improved delivery of public services because of the local participation and
accountability (Green, 2005).
As Litvack,
Ahmad and Bird (1998) exposed, one of the benefits that can be expected from
decentralization is the realization of good governance. Likewise, Oyugi (2000)
reveals that for the last fifty years or so, there was an impression that
decentralization is a prerequisite for good governance. Meanwhile, good governance is crucial for successful development by
providing principles and indicators for a successful decentralization, such as
per capita income (Grindle, 2007).[1] The mutual
relationship between decentralization or regional autonomy and good governance
has, therefore, been easily accepted. Ebimaro (2006) even strongly asserts that
decentralization is synonymous with good government.
Nharnet
Team (UNDP, 2005) supports the idea on such mutual relationship by stating that
decentralization
is the logical application of the core characteristics of good governance at
the sub-national and local levels. These characteristics include
accountability, transparency, rule of law and responsiveness. This statement
infers that without decentralization, good governance seems to be
unsuccessfully implemented Similarly, the Center for Democracy and Governance
(USAID, 2000) confidently says that when effective decentralization and
democratic local governance advance in tandem, local governments – and the
communities they govern – gain the authority, resources, and skills to make
responsive choices and to act on them effectively and accountably.[2]
This paper tries
to elaborate some factors explaining and contributing to the failure of
building good governance through decentralization in the current stage of
bureaucratic reform in Indonesia. Before doing so, it would provide a rough
picture on good governance and government capacity in implementing
decentralization. At the end, it proposes expected strategy to renew the spirit
of decentralization in order to achieve more meaningful good governance in
Indonesia.
2. Preview
on Good Governance in Indonesia
Indonesia’s experience on
decentralization implies different results on good governance. Lankaster (2007)
has evaluated six indicators of good governance[3] in Indonesia after the fall of Suharto’s presidency in 1998 and
decentralization law of 1999. Basically, there has been significant improvement
since 1998, but the ratings for all indicators are in most cases below the 50th
percentile.
The above data demonstrates that the progress of governance reform in Indonesia has
been rather substantial, even though not so promising. Decentralization, for
sure, has its limits in promoting good governance. As Green (2005) insists:
“Indonesia fails to fully meet the World Bank’s criteria for successful
decentralization. Indonesian local governments lack financial resources
completely under their control to provide public services; accountability and
transparency mechanisms are not fully implemented, and the legal system is not
as effective as it should be in order to facilitate decentralization … The
decentralization being implemented in Indonesia continues to be vague and
outcomes vary because the inputs are highly dynamic and accountability is
poor.”
However, there are always many rooms for improvement. As
decentralization constitutes a strategy to achieve good governance, the failure of
good governance might be caused by unsuccessful decentralization framework. In
that sense, the failure of decentralization is caused mostly by the low
capacity of local government. From the beginning of decentralization policy,
there was wide anxiety on the capacity of local government to run the new wave
of government management. Such concern implies that government capacity
constitutes condition sine qua non for
rewarding decentralization.
3. Governance
Capacity in Implementing Decentralization and Reinforcing Good Governance
It is quite fair
to admit that government capacity to perform decentralization functions in
Indonesia is getting more observable. In April 2011, the Ministry of Home
Affairs (MOHA) has announced Ministerial Decree No. 120-276/2011 regarding the
rank and performance status of local government as alleged by Government
Regulation No. 6/2008. The publication reports that the performance of 29
provinces is deemed to be “high” and only 4 provinces are considered to achieve
“average” performance, those are, Aceh, Papua, West Sulawesi, and Central
Sulawesi. In the city/district level, 269 districts and 82 city governments
have attained high performance, 70 districts and 4 city governments are
reaching average performance, while 5 district governments are still under
performance. [4]
Simultaneously,
on March 2011 the Ministry of Finance (MoF) has passed Regulation No.
61/PMK.07/2011 concerning General Guidance and Allocation of Regional Incentive
Fund (Dana Insentif Daerah - DID)
2011. By this regulation, 5 provinces, 17 cities, and 38 district governments
were awarded with incentive fund for their excellent performance on financial
management, education, and welfare and economic development. [5]
The evaluation
conducted by both MOHA and MoF provided a broad picture on the promising
enhancement of local government capacity in dealing with decentralization
tasks. Nevertheless, such kind of evaluation was mainly using desk evaluation
method and common-sense survey, which, to some extent, doesn’t reflect the real
performance. In fact, the capacity of local government is still in need of serious
improvement, particularly in the early period of “big-bang” decentralization.
At that time, local elites have viewed decentralization as golden opportunity
to propose the creation of new local government units.
As a result of proliferation of local governments (pemekaran) euphoria, the number of cities (kota) and districts (kabupaten)
has increased by half, from 292 before decentralization to 434 in 2003 (Fitrani
et.al., 2005). [6] Since the enactment Decentralization Law No. 22/1999, the number of
autonomous local government until the year 2011 has increased by 205,
consisting of 7 provinces, 164 districts, and 34 city governments (Partnership for Governance Reform in
Indonesia, 2011). The total
number of local government until recently is 33 provinces, 399 districts, and
98 city governments (MOHA, 2011).[7] The creation
of new layers of government, in turn, may overwhelm national budgets and reduce
the capacity of local government in delivering excellent service to its
citizen.
One of the indicators showing the low
capacity of local government is fiscal dependency on transfer fund from central
government. According to Suhendra and Amir (2006) and Hirawan (2007), on average, the districts relied around 85 to 90 percent on the
intergovernmental transfers to fund their expenditures, while the provinces
relied only about 55 to 70 percent on average on the transfers. Brodjonegoro
(2004) confirms such situation by noting that many new local
governments had relatively insignificant local own revenue (PAD) and relied
heavily on transfer fund, more specifically general purpose grant (Dana Alokasi Umum).
Due to their low fiscal capacity, local
governments tend to produce massive regulation in order to generate revenue.
Preferably, local governments are supposed to have ability to formulate and implement sound policies and regulations that
permit and promote private sector development. Unfortunately, many regulations
in province and district level produce high-cost economy and make local
governments capacity even worsen. Such situation explains why central
government has canceled thousands of local regulations (Peraturan Daerah). There are 2399 Peraturan Daerah which have been canceled from 2002 until March
2011 (MOHA, 2011).[8]
Government capacity might also be best
identified from the degree of integrity shown by public sector institution,
especially in the region. In this case, Corruption Eradication Commission (Komisi Pemberantasan Korupsi - KPK) has
conducted Public Sector Integrity Survey,[9]
covering 353 service units scattered among 23 central
agencies and ministries, 6 vertical institutions, and 22 city governments. The
findings indicate that the public service in Indonesia is highly susceptible to
corruption. The vulnerability of public service to corruptive behavior can be
observed from 2010 National Integrity Index (IIN) which reached 5.4 point (of
scale 1-10), or 1.1 points lower compared to the 2009
IIN. The decrease in integrity index symbolizes the decrease in both service
delivery and service quality.
At the local level, integrity index
among service agencies is not so promising as well. Of 22 municipalities
surveyed,[10] only
2 whose score above 6, i.e. Samarinda and Surabaya, while the other 20 cities
received score below 6, i.e. Yogyakarta, Ambon,
Tanjung Pinang, Pontianak, Serang, Jakarta Barat, Jakarta Timur, Jakarta Pusat,
Mataram, Jakarta Utara, Bandung, Semarang, Jakarta Selatan, Pekanbaru, Manado,
Jayapura, Makasar, Palembang, Bandar Lampung, and Medan (KPK, 2011).[11]
Overall, there is a big challenge facing
the central and local governments to fortify their capacity in order to assure
a better implementation of good governance in the decentralization era. In
other words, the quality of good governance in Indonesia can be advanced by
accurately detecting and refurbishing some impeding factors for effective
decentralization.
4. Challenges
and Fortifying Strategies towards Decentralization
After a decade of its implementation,
decentralization in Indonesia will be entering a new phase. Decentralization
Law of 2004 is about to be reviewed, as many expectations have not materialized
during this period. Revision on the law doesn’t necessarily mean that
decentralization has been failing;[12]
it is an effort to boost the efficacy of decentralization in the future
context. In that senses, there are three conditions where decentralization is
likely to be best implemented. Firstly, decentralization must have significant
contribution to nation building processes. Secondly, decentralization needs to
be comprehensively designed and applied with bureaucratic reform programs.
Thirdly, decentralization has to be convergent with regulatory and
institutional reform. Lacking of those three requirements will lead to weak
decentralization as well as “nominal” governance.
4.1.
Decentralization and nation-state building
The basic rationale of decentralizing
state powers is mixed, from political grounds, economic motives, to
administrative objectives.[13]
Among such numerous rationales, this paper focuses on the correlation between
decentralization and nation state building. Strong state is assumed to be
cumulative result of effective decentralization. In other words,
decentralization should be appropriately seen as an ingrained strategy in
strengthening nation-state building. In Indonesian context, Matsui (2003)
suggests that decentralization as the “opportunity to learning by doing” for
the central and local governments in the process of the nation state building.[14]
Unfortunately, there are obvious tendencies
that decentralization does not convincingly contribute to the formation of
solid nation-state building. Throughout the decade, ethnocentrism and
primordialism increases along the democratic processes at the local level.
Instead of spreading powers and empowering people, decentralization produces
vortex of power among particular groups. Zulkieflimansyah (2009) declares that
the most current political system in Indonesia is characterized by politics of
dynasty, a widespread phenomenon of nominating a figure from the incumbents’
family and the old political elites as Member of Parliament or candidate of
head of local government. Dynasty is also defined as a primitive system of
reproducing power based on blood relation and descendant of particular elites.
In the kinship or dynasty system, public participation is neglected, so that it
constitutes the real enemy for democracy (Robet, 2010).[15]
Dwipayana (in Gatra, February 18, 2010)
affirms such situation as neo-patrimonial
phenomenon, i.e. political regeneration based on genealogical ties. [16]
In such situation, local democracy erected through decentralization is no
longer believed as part of revitalizing national integrity. Inevitably, not
only new local kingdoms but also new classes of oligarchy elevates during
decentralization era. According to Taslim (2007), the rise of primordialism
spirit could be seen as a sign of waning nationalism spirit. Once again,
Indonesia is facing solemn challenges with its decentralization framework.
In addition, the desire of maintaining
power might be done by any means, including occupying lower position. In the
case of Surabaya, for example, Bambang Dwi Hartono has served
as Mayor for
two periods (2000-2010), so that he
is not allowed to run for the third
period. He chose to run for vice mayor
and has been inaugurated for 2010-2015 period, while the elected Mayor, Tri
Rismaharini, was Bambang’s subordinate during his term.
The above description indicates that
political powers in grass-root level are still concentrated around specific
groups. Decentralization in Indonesia, frankly speaking, is still powerless of
spreading powers and creating checks and balances among political actors and
other stakeholders. The opportunity to gain power is formally opened
inclusively, but the access to power is empirically a very complex effort, not
to say impossible. Under such political climate, therefore, good governance can
hardly be promoted.
This condition is analogous to some West
African and South Asian countries experience. In their study, Manor and Crook
(2000, quoted by Prasojo, 2009) demonstrate a clear connection between popular
election and bad governance. They conclude that “some of the worst cases of corruption and ineffectiveness are
associated with the direct popular election mayors of chief executives ...”.
There are three reasons why direct local
election contributes to poor governance practices. Firstly, political control
from local legislative (council) over mayor is weak since they are both
directly elected by the same people. Secondly, there is tendency of local
elites in developing countries to keep resources limited to a very small elite
circle. It explains why incumbent candidate is always co-opting bureaucracy and
financial resources to win the election.[17]
No wonder that decentralization insignificantly improves economic growth and well-being of local people due to misallocation of financial
resources. Thirdly,
awareness, knowledge, and people networks in controlling the local government
is extremely weak and limited, especially in the rural and hinterland areas.
The high-cost of local democratic
processes[18] may
be indicted as inciter of the problem. It means, only those with affluent
resources, more specifically financial ones, will have greater chance to get
the power. In other words, those having huge resources may get bigger
opportunity to be new aristocrats and oligarchs in the region. Moreover, in the
country and in the environment where economic gap is remarkably visible, money
politics is more likely to take place, not only in affecting people’s voice but
also in influencing political parties’ decisions.
In order to prevent such situation
getting worse in the future, “political” decentralization itself is not enough.
It should be accompanied by greater fiscal decentralization and economic
decentralization, as decentralization would grow faster in strong and
prosperous communities. When people are strong and prosperous, they will able
to control the government, and when the government is adequately controlled,
there would be an effective, accountable, and responsive (local) government. At
last, strong and prosperous communities in the one hand, and effective,
accountable, and responsive government in the other hand, reflect the incidence
of good (local) governance. In this case, Chinese government experiences in
implementing economic decentralization might be an outstanding lesson for
Indonesia in constructing stronger and more prosperous communities.[19]
Another determinant factor affecting the
quality of good governance in the region lies in the human resource policy and
management. In fact, decentralization is not only about the transfer of
authority and budget from the center to the region, it also has to do with
recruitment of civil servant (Pegawai
Negeri Sipil – PNS). On behalf of decentralization, local governments have
managed recruitment by giving priority for local resident and limiting the
opportunity for other residents. Such practices result in disadvantages such as
reducing the prospect to attain national standard among civil servants from different
regions and different levels of government. Regions having advanced Human
Development Index, for instance, will have a better chance to recruit qualified
candidates. Quality standard of civil servant will completely be differentiated
among provinces and districts. In this case, lacking of national standard in
human resources management triggers the occurrence of common impression that
federalism is much more prominent in Indonesia than that of decentralization.
Additionally, cultural exchange, cross-regional learning, and enriching
experience, will unlikely to happen in such surroundings.
In fact, civil servant or PNS is a vital
position expected to be the glue to the unitary state of the Republic of
Indonesia (Negara Kesatuan Republik
Indonesia - NKRI).[20]
Decentralized personnel management, unfortunately, tends to weaken the role of
PNS as an adhesive to NKRI, as provided by Government Personnel Law No.
43/2009.[21] In
the current personnel system, there are types of public officials, i.e. Central officials and Regional officials
(article 2). It indicates that Indonesian personnel system implements a
fragmented system instead of unified system. In the unitary state, government
functions may be transferred to local governments, but personnel system and management
needs to be centrally managed. Simultaneously, the system of employee exchanges
across regions and inter-governmental levels should be introduced.
The prevalent phenomena on augmenting
dynasty system and weak roles of civil servants in the decentralization era
indicate that good governance does not satisfyingly work or optimally achieved.
In other words, decentralization does not significantly contribute to the
development of strong state or hard state (“institutionalization” in present
term), but promoted an opposite concept of fragile state (Brinkerhoff, 2008) or soft state (Leftwich, 2000).[22]
To conclude, considering that decentralization tends to produce kinship
(dynasty) system and fails to boost civil servants’ roles in strengthening
unitary state, it is somewhat reasonable to support an idea that
decentralization in Indonesia has failed to advocate nation-state building
processes.
4.2.
Decentralization and bureaucratic reform
Decentralization and bureaucratic reform
has twofold faces. On the one hand, decentralization itself is a fundamental
reform in terms of sharing power and resources, as well as rearranging new
relationship between the central and local government. As a reform,
decentralization in Indonesia has successfully changed authoritarian regime
into a democratic one. On the other hand, decentralization would be useless
without further i.e. bureaucratic reform to deal with delegated new functions.
Bureaucratic reform is the best answer for local governments when facing
capacity problems. This makes obvious that decentralization and bureaucratic
reform constitute a “one coin with two sides”.
Unfortunately, there are some tendencies
that decentralization is not convergent to bureaucratic reform. “Big bang”
decentralization was effectively implemented since 2011, 10 years before
bureaucratic reform has been initiated by the enactment of Presidential
Regulation No. 81/2010 on the Grand Design of Bureaucratic Reform. During 10
years of its implementation, decentralization was inadequately guarded by clear
and measurable objectives and outcomes should be achieved by central and local
government. Minimum standard of services (MSS) of obligatory and basic services
were not well defined as mandated by Law No. 32/2004.
In the health sector, for instance,
SMERU (2006) shows that only 53% of districts in the sample have
met the minimum standards of service set by the central government, while only
few number of Puskesmas (health
center at sub-district level) have the resources required to meet the MSS. Consequently,
it is not startling that the performance of government especially among
autonomous local government is not so promising. Government regulation
concerning the performance evaluation of local government has been promulgated
just in 2006.
These situations explain that
decentralization has no direct connection to accountability and performance
management, so that corruption is about to expand in the era of
decentralization. In this sense, Rinaldi, Purnomo, and Damayanti’ study (2007) exposes that shift of power
relation between central and local government have
given rise to rampant ‘money politics’ – by District Heads seeking to gain and
maintain support from the legislature; and legislators exploiting their newly
acquired power over local budgets to secure financing for their political
parties. The study provides further explanation that both District Heads and
local legislators have taken the chance to embezzle funds for self-enrichment.
Opportunities for corruption have been opened up by the enactment of
inconsistent regulations governing local budgets by the national and regional
parliaments, regular ‘cooperation’ between the legislative and executive bodies
as well as low levels of public participation and control in local governance.
The modus of corruption deployed are as follows: 1) multiplying and
enlarging budget items; 2) distributing budget to fictitious
foundations/agencies; 3) manipulating budget for official travelling; 4)
procedure infringement of cash disbursement; and 5) manipulating the
procurement processes. Similarly, Dwiyanto (2011) writes that decentralization
has been followed by spreading corruptive behavior among local elites. The
augmentation of patronage and clientilism between bureaucracy and local
politicians complicates the control efforts over corruptive behavior.[23]
Under such situation, bureaucratic reform is becoming much more
essential, though, to be frank, it’s rather late. The stipulation of
Presidential Regulation No. 81/2010 on the Grand-design of Bureaucratic Reform
and Minister of Administrative Reform Regulation No. 20/2010 on the Road-map of
Bureaucratic Reform is fundamentally designated to accelerate the reform. In
order to improve the effectiveness of the reform, National Steering Committee of Bureaucratic Reform (Tim Pengarah Reformasi Birokrasi Nasional –
KPRBN) and National Bureaucratic Reform
Team (Tim Reformasi Birokrasi
Nasional – TRBN) has been established. In performing their duties, KPRBN is
assisted by an Independent Team and Quality Assurance Team that are tasked
with monitoring and evaluation and ensure the good implementation of the
reform. At the same time, TRBN is assisted by the National Bureaucracy Reform Management Unit (Unit Pengelola Reformasi Birokrasi Nasional - UPRBN).
Unfortunately, the basic design of
bureaucratic reform is also problematic. It is applying agency level approach
instead of national level approach. Under agency level approach, every ministry
and agency is supposed to compose reform proposal which is submitted to TRBN.
UPRBN will assess the document and conduct field verification, and the result
will be re-submitted to TRBN. TRBN will then convey the assessment result to
the Ministry of Finance for the calculation of budget need to be allocated for
additional remuneration (performance incentives). The new remuneration scheme
will be discussed and approved in the KPRBN, and will subsequently be presented
to the Parliament. The consent by the Parliament will be employed to determine
the grant of remuneration for approved ministry and agency.
The way how
bureaucratic reform works in Indonesia as described above sparks much criticism
that bureaucratic reform is merely about getting additional financial benefits,
while the genuine transformation of governance never materializes. In
Indonesian terms, reformasi (reform)
is misleadingly perceived as remunerasi
(remuneration). Since remuneration is a logical consequence of reform, state
budget burden increases dramatically to funding the reform.[24]
Moreover, bureaucratic reform doesn’t work as expected since document of reform
proposals from ministries and agencies are oftentimes prepared by consultant,
not by the corresponding agencies, so that they do not experience any learning
process, mindset changes, and real improvements. As a result, the ongoing
bureaucratic reform is actually representing a nominal (vague) reform, not a
substantial reform.
Another weakness
of current bureaucratic reform is that it only copes with business process
enhancement instead of building culture of outcomes or building public
trust. According to Effendi (2011), the
ultimate objective of bureaucratic reform is rebuilding and bringing back trust
from the people. It may be achieved by providing and delivering excellent
service to the public. To do so, cultural change is unambiguously required.
Those three dimensions of reform, unluckily, are not getting enough attention
in the recent process of reform in Indonesia. In fact, only organizational
dimensions that received adequate attention, such as efficiency and
simplification of work procedures, human resources management, elimination of
overlapping regulations, and improvement in transparency and accountability. In
the near future, organizational dimensions should be combined with three
reinforcing dimensions to assure the best deed of bureaucratic reform.
It is widely acknowledged that
Indonesian bureaucracy is facing lots of cultural problems. In the system level, distrust between
government and people is widespread, as distrust among and within government
bodies are rampant. In the institutional
level, efficiency, to some extent, is justified as incapability of budget
absorption, and therefore, judging as low performance. Culture of
accountability is halfheartedly developed, just opposite of culture of
reporting. Culture of output is overwhelming culture of outcomes. In the individual level, indisciplinary actions
are frequently encountered, while disobedience to the rules and against the
superior is also commonplace.
It is truly unfortunate that
bureaucratic reform in Indonesia is lagging of cultural touch to solve such
problems. Because of cultural deficiency, bureaucratic reform in Indonesia
needs to be redesigned by incorporating cultural dimensions. In this case,
organizational culture is an aggregate function consists of three components
namely (superb) leadership, legal obedience and enforcement, and fulfillment of
basic needs. Only when all three components are met, then organizational culture
will flourish, and, in turn, will contribute to the efficacy of bureaucratic
reform.
Leadership is a
very essential factor to make government institution more dynamic. At least,
there are two roles of leadership in ensuring the sustainable reform for the
organization.
In the macro
level, good leaders are
those who are able to endorse new ideas, fresh perceptions, continual
upgrading, quick actions, flexible adaptations and creative innovations.
Simultaneously, they are expected to fully promote continuous learning, fast
and effective execution, and unending change. Such roles are becoming much more
imperative when government institution is typically regarded as a slow, stodgy
bureaucracy that consistently and, sometimes, mindlessly enforces outdated
rules and sticks to procedures without any care or concern for individuals or
businesses (Neo and Chen, 2010). In order to realize dynamic governance, Neo
and Chen (ibid) propose three critical governance capabilities:
“1) thinking ahead — the ability to perceive early signals of future
developments that may affect a nation in order to remain relevant to the world;
2) thinking again — the ability
and willingness to rethink and remake currently functioning policies so that
they perform better; and 3) thinking
across — the ability and openness to cross boundaries to learn from the
experience of others so that new ideas and concepts may be introduced into an
institution.”
In a micro level, leadership can only
be effective whenever it is able to provide with outstanding behavior and
concrete examples. By practicing “walk the talk” principle, a leader is more
likely to develop
dynamic and effective followership. As Latour and Rast (2004) points out, without effective followership, a leader
at any level will fail to produce effective institutions. Valuing followers and
their development, therefore, is the first step toward cultivating effective
transformational leaders.
From the
influence model point of view, leaders are model for their subordinates. People
tend to change their mindset and behavior if they see their leaders behaving
differently. In this sense, Barsh and
Cranston (2009) recommends five new roles of a leader:
· Framing: looking at problems in new ways to find better solutions;
· Connecting: actively shaping networks to heighten one’s sense of belonging, ability
to influence change, and personal growth;
· Engaging: taking personal accountability for one’s life experience and setting
aside fears to step up to opportunities;
· Energizing: actively managing experiences to achieve maximum “flow” in the work day;
and
· Meaning: finding an inspiring purpose that is built on strengths, and using it to
generate hope and action.
It can be underlined that the most
basic role of a leader is empowering the collective efforts of the organization
toward meaningful goals. Under such new role, there will be a growing sentiment
among people in organization that they are an integrated part of the
organization. This kind of feeling will, in turn, lead to willingness to comply
with rules and consensus. It implies that qualified leadership might be
appropriately functioned as leverage to draw people’s obedience over the law as
well as to enforce the law.
One more important thing needed to
ensure the best running of cultural reform is discharging basic needs of the
people. Corruption cases, in many ways, are not caused by greed and
opportunity, but more by needs. Despite
recent efforts to increase participation, accountability and transparency,
basic needs fulfillment remains an opaque process to many government employees.
Certainly, bureaucratic reform is not mere a policy of providing remuneration
but minimum living standard is indeed a crucial dimension of the reform. When
people are no longer troubled by daily basic needs, they may focus more on
their duties as well as having a higher level of law obedience. At the same
time, they are also more resistant to the temptations for corruption acts.
Again, only when the aspects of leadership, obedience to law, and fulfillment
of basic-needs are well-constructed, there will be strong cultural foundations
for a perfect bureaucratic reform.[25]
4.3.
Decentralization and regulatory and
institutional reform
It is assumed that decentralization
will bring about the reduction of central government institution number due to
transfer of major government functions to local government. In turn, it is
expected to increase the amount of budget transferred to local level, and
therefore, improving local development as well as enhancing public services.
Similarly, decentralization is also supposed to reduce the tendency of over
regulating local government.[26]
In fact, decentralization policy in
Indonesia has no significant impact on both institutional and regulatory
reform. The number of central agencies and local autonomous governments has steadily
increased. In the central level, the number of existing ministries is 34, a
maximum number allowed by Law No. 39/2008 on State Ministries.[27]
Additionally, there are 25 non-ministerial government institutions (Lembaga Pemerintah Non-Kementerian –
LPNK).[28]
Indonesian post-reform public
administration is also characterized by formation of new commissions (komisi), councils (dewan), board (badan), committee
(komite), team (tim), etc. They are all called non-structural institutions (Lembaga Non-Struktural – LNS). The
number of LNS in 2005 was 42, rising to 85 in 2010, or an increase of more than
100 percent within the last five years. The uncontrolled growth of LNS has
created effectiveness problem of state administration, particularly concerning
potential overlapping of duties across institutions. Furthermore, the
establishment of LNS is also considered as a burden to state budget (State
Secretary GoI, 2010).
Then, an interesting question occurs:
if the establishment of LNSs is engendering task ineffectiveness and budget
inefficiency, why are there still desires to create new LNSs? One possible
answer is that formation on new LNSs reflects a growing public distrust over
existing state agencies. For example, the presence of Ombudsman (Komisi Ombudsman Negara – KON) is a sort
of public distrust over service-providing institutions. Likewise, deviant
behavior of judges is encouraging the formation of Judicial Commission (Komisi Yudisial – KY). In other words,
government institutions both in central and local level are lacking of
appropriate capacity so that they fail to demonstrate the best performance.
It is a weird thing that the problem of
weak capacity and low performance of government institution is addressed by
establishment of new institutions. Reforming the twisted institutions might be
more appropriate than forming the new one. Unfortunately, enlarging the size of
government is much more preferred than fortifying organization functions and
its capacity to execute. It is truly regrettable, therefore, that institutional reforms tend to be
neglected in the spirit of bureaucratic reform, so that decentralization and
institutional reform is going to opposite direction.
Similar
situation can be observed in the case of regulatory reform. 10 years of
decentralization is not marked by the wider space of expressing aspiration or
better climate to run business, but it is flawed by new regulations that tend
to restrict and burden the citizen. The number of local regulation (Peraturan Daerah – Perda) increased
dramatically, causing deceleration of local economic engines and escalation of
high-cost economy in the region. That’s why the central government (i.e. Ministry
of Home Affairs – MOHA) needs to strictly evaluate every single Perdas,
especially relating to local taxes and levies. Until March 2011, there are
2,399 Perdas which have been canceled by MOHA.
The reasons of
canceling Perdas are diverse, such as incompatible with higher regulation,
resulting in double payment/charging, no basis of authority, and so forth. In
short, Perdas do not function as a decentralization instrument to improve the
welfare of the people; it is misleadingly deployed as income generation tool.
Regulatory reforms are abandoned, so that decentralization and regulatory
reform is going to opposite direction.
Concluding
remarks
Decentralized governance has always
consequences. Transferring government functions and its financial and human
resources to local government necessitates an alteration of government
structure. Local government structure tends to slightly enlarge, whereas
structure in the central level is supposed to shrunken. In such structure, the
role of central government is simultaneously altered from “steering rather than
rowing” to “serving rather than steering”, and from “serving rather than
steering” to “enabling rather than serving”. As an enabler factor, the central
government plays crucial roles of providing sound policies and nation-wide
standardization, and inviting more active participation from private sectors
and the communities.
The above description plainly explains
the positive correlation between decentralization and good governance. Decentralization contains paradigm and notion of good governance
such as participation, people empowerment,
transparency,
accountability, and rule of law. In other words, good
(local) governance is one of the benefits that may occur when decentralization
is effectively implemented and appropriately managed.
However, it is true that carrying out
the mandate of decentralization is not an easy way. There is a complex
situation encountered by Indonesian government.
On the one hand, the government is
facing lots of internal limitations in terms of human resources capacity,
ineffective institution and inefficient business processes. Given the limited resources, knowledge, as
well as human and organizational capacities, “good enough governance”[29] might be actually
the most realistic target of decentralization policy.
On the other hand, people’s demand and
pressure is getting stronger, while decentralization has not been able to
exhibit its function as unifying factor for diversified Indonesia and
fortifying strategy to the nation-state building. In dealing with such
complexity, bureaucratic reform is convincingly perceived as the best approach
to restore the present governance in Indonesia. The objective of bureaucratic
reform is to fortify the decentralization framework, while decentralization is
intended to boost the realization of good governance. Reaching of good
government, in turn, will lead to the betterment of public service.
Finally, we come to the conclusion that decentralization
is, undoubtedly, a reform, i.e. a process of getting intergovernmental relation
more balanced and more democratic. Nevertheless, decentralization itself is not
enough; it is not panacea for all diseases. It should be equally accompanied by
systemic enhancement in nation-state building, bureaucratic reform,
institutional arrangement, as well as regulatory reform. Otherwise, both
decentralization and good governance remains a dream or theory; they would
never take place in the real world.
References:
Bardhan, Pranab and Dilip Mookherjee (ed.), 2006, Decentralization
and Local Governance in Developing Countries: A Comparative Perspective.
MIT Press: Cambridge.
Barrios, Cristina and Said-Abass Ahamed, no year, Democracy
promotion vs. state-building: decentralization in the Democratic Republic of
Congo. First draft paper.
Barsh,
Joanna and Susie Cranston, ‘How remarkable women lead’, Crown
publishing, New York, 2009.
Basuki,
Tobias, 2006, Decentralization in Indonesia and China: An answer to Legitimacy Crisis?.
Paper presented at the annual meeting of The
Midwest Political Science Association, Palmer House Hilton, Chicago, Illinois,
April.
Brinkerhoff, Derick W., 2008, Building Resilience in Fragile States: The Role of Governance,
Paper presented at Nation-State Building 101 CNA and Peacekeeping and Stability
Operations Institute, October 28.
Brodjonegoro, Bambang, 2004, “Three Years of Fiscal
Decentralization in Indonesia: Its Impacts on Regional Economic Development and
Fiscal Sustainability”, paper presented in International
Symposium on Fiscal Decentralization in Asia Revisited, Hitotsubashi
University, Japan, February, 20-21.
Center
for Democracy and Governance, U.S. Agency for International Development, 2000,
Decentralization and Democratic Local Governance Programming Handbook,
Technical Publication Series.
Cheema,
G. Shabbir and Dennis A. Rondinelli, 2007, Decentralizing Governance: Emerging Concepts and Practices, Brookings
Institution Press
Choi, Jong Chan, 2010, Korean Government Reform during 1998-2007,
presentation in the Capacity Building for Public Officials for Government
Innovation in Indonesia (Second Phase), Seoul, December.
Corruption Eradication
Commission (KPK), 2010, Indonesia Public
Sector Integrity 2009: Corruption
Facts in Public Services, Public Sector Integrity Survey 2009.
Dwiyanto, Agus, 2011, Mengembalikan Kepercayaan Publik Melalui
Reformasi Birokrasi (Bringing Public Trust Back through Bureaucratic Reform),
Jakarta; Gramedia.
Ebimaro, Sampson, 2006, Why Performance-Oriented Political
Decentralization Matters in Good Governance and Sustainable Development:
Comparison between Japan and Nigeria.
Effendi,
Sofian, 2011, Diklat Kepemimpinan
Reformasi (Reform Leadership Training), presented at the FGD on Special
Training Program to Strengthen Bureaucratic Reform, Jakarta: National Institute
of Public Administration, May 19.
Eisenstadt,
S.N. and L. Roniger, 1984, Patrons,
clients and friends; Interpersonal relations and the structure of trust in
society, Cambridge University Press.
Eslava, Luis,
2009, "Decentralization of Development and Nation-Building Today: Reconstructing Colombia from the Margins of
Bogotá," The Law and Development Review: Vol. 2: No. 1, Article 11.
Fitrani, Fitria, Hofman, Bert and Kaiser, Kai,
2005, “Unity in diversity? The creation of new local governments in a
decentralizing Indonesia”, in Bulletin of
Indonesian Economic Studies, April.
Ford,
James, 1999, “Rationale for Decentralization”, in World Bank Institute (ed.), Decentralization
Briefing Notes, WBI Working Papers.
Green, Keith, 2005, Decentralization and Good Governance: The Case of Indonesia, MPRA
Paper No. 18097, Munich Personal RePEc Archive.
Grindle, Merilee S., 2007, “Good Enough Governance
Revisited”, Development Policy Review 25 (5).
Hirawan, Susiyati
Bambang, 2007, Desentralisasi Fiskal
Sebagai Suatu Upaya Meningkatkan Penyediaan Layanan Publik Bagi Orang Miskin di
Indonesia (Fiscal Decentralization as an Effort to Improve the Provision of
Public Service for the Poor in Indonesia), Inauguration speech as Professor in
Economics, University of Indonesia.
Hutchcroft,
Paul D., 2001, “Centralization and Decentralization in Administration and
Politics: Assessing Territorial Dimensions of Authority and Power”, in Governance:
An International Journal of Policy and Administration, Vol. 14, No. 1,
January. Blackwell Publishers.
Johnson, Ronald W. and
Henry P. Minis, Jr., no year, Toward
Democratic Decentralization: Approaches to Promoting Good Governance,
Research Triangle Institute.
Kim, Pan Suk, 2000, “Administrative Reform in the Korean Central Government: A Case Study of
the Dae Jung Kim’s Administration”, in Asian Review of Public
Administration Journal, Vol. XII
No. 2 (July-December).
KPPOD, 2010, Doing Business in Indonesia 2010,
Jakarta. Available at http://www.doingbusiness.org/reports/subnational-reports/indonesia
Lankaster, Sir Tim, 2007, Reform of Indonesia’s Governance: Myth or Reality? Oxford: Corpus Christi College, Institute of
Policy Studies.
Larok, Arthur, 2008, Deepening Decentralization or Regime
Consolidation, What explains the unprecedented creation of new districts in Uganda?
National NGO Forum issues paper No. 1.
Latour, Sharon M., and Vicki J. Rast,
2004, “Dynamic
Followership: The
Prerequisite for Effective Leadership”, in Air & Space Power Journal.
Lee,
Hyung-U, 2007, Characteristics
of Korean Bureaucracy, Dongguk University.
Leftwich,
Adrian, 2000, States of Development: On the Primacy of Politics in
Development, Cambridge: Polity Press.
Litvack,
Jennie, Junaid Ahmad, Richard Bird, 1998, Rethinking Decentralization in Developing
Countries, Washington DC: The World Bank.
Matsui,
Kazuhisa, 2003, “Decentralization in Nation State Building of Indonesia”, IDE Research Paper No. 2, Institute of
Developing Economies, JETRO.
Neo, Boon Siong and Geraldine Chen, 2010, “Chapter 1: Framework for
Dynamic Governance: Institutionalizing Culture, Capabilities and Change”, in
Dynamic Governance - Embedding
Culture, Capabilities and Change in Singapore, © World Scientific
Publishing Co. Pte. Ltd.
Nharnet
Team, 2005, Decentralized Governance, a Global Sampling of Experiences,
UNDP Monograph on Decentralization.
Oyugi, Walter O., 2000, “Decentralization
for Good Governance and Development, the Unending Debate”, in Regional Development Dialogue Vol. 21
No. 1, Spring. UNCRD.
Partnership for
Governance Reform in Indonesia, 2011, Desain Besar Penataan Daerah (Grand
Design of Regional Arrangements in Indonesia), Partnership Policy Paper No 1/2011.
Prasojo, Eko, 2008, Menghapus Pilkada Langsung (Eliminating
Direct Local Election), in Kompas,
February 5.
__________, 2009, Bad Governance Salah Satu Ekses Pilkada (Bad Governance, One of Negative Effects of Popular
Local Election).
Rinaldi, Taufik, Marini
Purnomo, and Dewi Damayanti, 2007, Memerangi
Korupsi di Indonesia yang Terdesentralisasi: Studi Kasus Penanganan
Korupsi Pemerintahan Daerah (Combating Corruption in the Decentralized
Indonesia: Case Study of Handling Corruption in Local Government), World Bank:
Justice for the Poor Project.
Robet, Robertus, 2010, Bahaya Politik Dinasti (The Danger of
Dynasty Politics), in Kompas, April 23. Available at http://cetak.kompas.com/read/xml/2010/04/23/03395434/bahaya.politik.dinasti
Rondinelli,
Dennis and G. Shabbir Cheema, 1983, Decentralization and Development, Policy
Implementation in Developing Countries, California, Sage Publication in
cooperation with UNCRD.
Simatupang, Rentanida Renata, 2009, "Evaluation
of Decentralization Outcomes in Indonesia: Analysis of Health and Education
Sectors", Economics Dissertations. Paper 58.
SMERU
Research Institute, 2006, The State of
Local Governance and Public Service in the Decemtralized Indonesia in 2006:
Findings from the Governance and Decentralization Survey 2 (GDS2), Research
Report.
Smith,
Brian C., 1985, Decentralization: The
Territorial Dimension of the State, London: Unpim.
State Secretary Government of Indonesia, 2010, Bunga Rampai Pemikiran Penataan Lembaga Non
Struktural (Potpourri of Thinking: Arrangement of Non-Structural
Institution).
Widyanti, Wenefrida and Asep Suryahadi (Edited by
Kate Weatherley), 2008, The State of Local Governance and Public
Services in the Decentralized Indonesia in 2006: Findings from the Governance
and Decentralization Survey 2 (GDS2), Research Report, SMERU Research Institute.
Suhendra,
Maman, and Hidayat Amir, 2006, “Fiscal Decentralization in Indonesia: Current Status and Future
Challenges”, in Public Finance Journal,
Ministry of Finance, September.
Taslim,
Hermawi, 2007, Etnosentrisme dalam
Politik Lokal (Ethnocentrism in Local Politic), available at http://www.reformasihukum.org/konten.php?nama=Pemilu&op=detail_politik_pemilu&id=502
USAID
Democratic Reform Support Program, 2009, Desentralisasi
2009, Membedah Reformasi Desentralisasi di Indonesia (Decentralization 2009, Dissecting Decentralization
Reform in Indonesia), Stock
Taking Study, Bappenas.
World Bank Institute, The Worldwide Governance Indicators (WGI) Project. Available online
at http://info.worldbank.org/governance/wgi/index.asp
Yudhoyono,
Susilo Bambang, 2008, Keterangan Pemerintah
tentang Kebijakan Pembangunan
Daerah di Depan Sidang
Paripurna Dewan Perwakilan Daerah Republik Indonesia (Presidential Speech
on Regional Development Policy in front of Plenary Session, Regional Council
Representatives, Republic of Indonesia), Jakarta, August, 22.
Zulkieflimansyah,
2009, Dari Politik Dinasti, Nepotisme
Kekuasaan ke Budaya Partisan (From Politics of Dynasty and Cronyism of
Power to Participant Culture), in Banten
Raya Post, 10
Januari 2009.
[1] For deeper understanding on the relationship
between decentralization and good governance, see Johnson, Ronald W. and Henry P. Minis, Jr., no
year, Toward Democratic Decentralization:
Approaches to Promoting Good Governance, Research Triangle
Institute. According to Johnson and Minis Jr., decentralization or devolution
offers the most potential for obtaining governance benefits such as
accountability, problem solving, and citizen participation.
[2] In a more
contemporary debate, however, Oyugi (2000) concludes that whether
decentralization contribute to good governance or not will depend on the unique
circumstances prevailing in the individual country implementing
decentralization program. There is no direct relationship between
decentralization and the existence of democracy (including good governance);
one may exist without the other.
[3] World Bank
Institute, The Worldwide Governance
Indicators (WGI) Project. Available online at http://info.worldbank.org/governance/wgi/index.asp.
[4] Government Regulation No. 6/2008 concerning
Guidance on the Evaluation of Local Government Performance. The result of 2011
evaluation is done based on 2009 data and reports. The best three provinces are
North Sulawesi, South Sulawesi, and Central Java. The best ten of district
governments are: Jombang and Bojonegoro (East Java province), Sragen (Central
Java), Pacitan (East Java), Boalemo (Gorontalo), Enrekang (South Sulawesi),
Buleleng (Bali), Luwu Utara (South Sulawesi), Karanganyar (Central Java), and
Kulon Progo (Yogyakarta). The best ten of city governments are: Surakarta and
Semarang (Central Java), Banjar (West Java), Yogyakarta, Cimahi (West Java),
Sawahlunto (West Sumatera), Probolinggo and Mojokerto (East Java), Sukabumi and
Bogor (West Java). Province and city/district government achieving low
performance within 3 years consecutively, will have special examination from Dewan Pertimbangan Otonomi Daerah (Council
of Regional Autonomy).
[5] DID program was started in fiscal year 2010,
when 9 provinces and 45 district/city governments were awarded with incentive
funds. Criteria on financial management performance include improvement of
financial report evidenced by Unqualified Opinion from the Audit Board (Badan Pemeriksa Keuangan), and growth of
local revenue (Pendapatan Asli Daerah)
above the national average. Criteria on education development comprise gross
enrollment rate, and ability to reduce gap between regional and national IPM.
Criteria on welfare and economic development embrace economic growth which is
higher than average of economic growth at national level, ability to reduce
poverty rate, ability to reduce unemployment rate higher than national average,
and fiscal capacity. The objective of DID program are: 1) to encourage local
government in managing its finances resources better, and 2) to motivate the region in setting annual budget on time. The total budget allocated for regional incentive funds
reach IDR 1.3 trillion per year.
[6] Uganda also shared experience in terms of unprecedented creation of new districts in the beginning of
decentralization wave. The number of districts in Uganda has burgeoned
at an alarming rate from 33 districts in 1986 to 80 by 2007 which is more than
double in terms of growth, with 24 new districts created in the last three
years alone (Larok, 2008).
[7] In the beginning of Independence, there were 6
provinces and 99 district/city government. Thus, the increase amounted to 550
percent for province and 502 percent for district/city governments.
[8] The detail of Peraturan Daerah (Perda) being canceled is as follows: 2002 (19
Perda), 2003 (105), 2004 (236), 2005 (126), 2006 (114), 2007 (173), 2008 (229),
2009 (876), 2010 (407), and 2011 March (114). Perdas canceled by central government were mostly about retribution
or levies.
[9] Public Sector Integrity Survey is annual
survey program by KPK aimed to identify significance and score of public
service providers’ integrity by using academic assessment to composing Integrity Index (Indeks
Integritas). Assessment of survey was conducted by combining two elements. Firstly, the experience of integrity (weight
0.667), reflects the experience of respondents to the level of corruption they
experienced. This element consists of two indicators i.e. practices of
corruption and perception to corruption. Secondly, the potential integrity
(weight 0.333), reflects the factors that could potentially lead to the
occurrence of corruption perceived by respondents. This element consists of
four indicators, i.e. work environment, administrative system (internal
business process), individual attitude, and prevention of corruption.
[10] Survey was held
merely over three service activities, i.e. building permit (Ijin Mendirikan Bangunan), trading
license (Ijin Usaha Perdagangan), and
issuance of residence identification card (Kartu
Tanda Penduduk).
[11] See KPK’s
official website at http://www.kpk.go.id/modules/news/article.php?storyid=1645. Moreover,
evaluation on the performance of local government, more specifically on the
local economic governance, has been conducted by Monitoring Committee on the
Implementation of Regional Autonomy (Komite
Pemantau Pelaksanaan Otonomi Daerah - KPPOD). In 2010, KPPOD has launched a
“Doing Business in Indonesia”, covering 14 city government, i.e. Balikpapan (East Kalimantan Province), Banda Aceh (Aceh), Bandung (West
Java), Denpasar (Bali), Jakarta, Makassar (South Sulawesi), Manado (North
Sulawesi), Palangkaraya (Central Kalimantan), Palembang (South Sumatra),
Pekanbaru (Riau), Semarang (Central Java), Surabaya (East Java), Surakarta
(Central Java), and Yogyakarta. Meanwhile, in 2011 KPPOD releases a Ranking of
Investment Climate and Local Economic Governance.
[12]
Simatupang’s
study (2009) serves a balanced evaluation showing that education outcomes improved with
decentralization, and that local governments are responding to local needs for
education services. Decentralization also brings improvement to health
services, as mortality rates and life expectancy are significantly improved
with decentralization. However, results indicate that decentralization does not
improve availability of health services, as only small percentage of municipalities
in Indonesia have access to health facilities. Likewise, proportional
evaluation is available from study conducted by USAID Democratic Reform Support Program (2009) and
Widyanti and Suryahadi (2008).
[13] On the
rationale of decentralization, read: Rondinelli and Cheema (1983), Smith
(1985), Litvack, Ahmad, and Bird (1998), Ford (1999), Hutchcroft (2001), Bardhan and Mookherjee (2006), Cheema and
Rondinelli (2007), etc.
[14] The
relationship between decentralization and nation-state building can also be
scrutinized in other parts of the world. In Congo, for
instance, democracy promotion (through decentralization) and state-building are
supposed as complementary goals (Barrios and Ahamed, no year). Similarly, the
case of Colombia reveals that the reform of the 1991 Constitution (i.e. granting decentralization)
installed local development as one of the primary strategies to recuperate the
nation-building project (Eslava, 2009).
[15] It is
interesting to pay attention on Eisenstadt and Roniger’s superior study (1984). They identify four main reasons
why kinship-based politics is much more
preferable in many countries. Firstly, the relatives are
more trusted and considered
not possible of doing betrayal. Secondly, the relatives are considered to have a higher loyalty,
especially in terms of maintaining dignity and
honor of the relatives.
Thirdly, the relatives have a strong level of solidarity, especially in helping a large
family clan from
the bankruptcy of power. Finally, the kinship system
is associated with the model of maintaining the prestige and honor
of the family.
[16] In the
multi-party system, the occurrence of dynasty or neo-patrimonial reflects the
failure of political parties to perform the function of political recruitment.
Also, it indicates the low competency of political parties in terms of
financial sources generation, cadre development, and organizational management.
It is worsened by immaturity of political culture from the majority of people.
[17] The phenomenon
of elite captures over bureaucracy and financial resources is delightfully
enlightened by Agus Dwiyanto in his newest book, Bringing Public Trust Back Trough Bureaucratic Reform, Jakarta:
Gramedia, 2011.
[18] Prasojo (2008)
mentions that costly election in local level encompasses registration and
re-registration cost due to invalid demographic data, procurement of goods and
services to support recurrent voting in every election, cost of political
campaign, etc. There is no efficiency paradigm at all in local election.
High-cost election can also be calculated from budget allocation. According to
Yudhoyono (2008), since June 1, 2005 until August 2008, there had been 414
regions (provincial and district/city level) completed election. If the average
cost for a local election is IDR 50 billion, the total cost of election in
Indonesia would reach IDR 20.7 trillion. Those costs do not include the costs
to be borne by political parties and candidate of governor/bupati/mayor/member
of DPRD, from campaign cost to tactical cost to influence people’s voice.
[19] Decentralization in China is reflected with the creation of
special economic zones, open coastal cities and development zones. Four special
economic zones were created in 1978 (Shenzhen, Zhuhai, Shantou and Xiamen in
Hainan province). Provinces Guangdong and Fujian were given extensive
independence and autonomy to develop their own region economically (authority
to approve foreign investment projects up to $30 million). Furthermore, in 1984
fourteen coastal open cities were created and certain inland cities (those
along the Yangtze River and bordering with Russia) are delegated powers like
those of the special economic zones (Basuki 2006). The Government of Indonesia
(GOI) has set Master Plan of Acceleration and Expansion of Indonesian Economic
Development 2011-2025, just in 2011. In this document, six economic corridors
have been established, i.e. Sumatra Economic Corridor (EC), Java EC, Kalimantan
EC, Sulawesi EC, Bali – Nusa Tenggara EC, and Papua – Moluccas Island EC.
[20]
Decentralization is one of the principles of state administration in addition
to deconcentration and assistance tasks (from Dutch term, medebewind). Philosophically, decentralization is granted within
the framework of unitary state, meaning that decentralization constitutes a
national government’s strategy to strengthen – not to weaken – the unitary
states of the Republic of Indonesia.
[21] The law stipulates that the Indonesian Civil
Service (PNS) is to
serve the public as well as the state; the law also provides prohibition on
political party membership (article 3).
[22] Brinkerhoff (2008) argues that fragile state is the state captured by elites, preys upon its citizens and maintains power
through a combination of patronage and repression. Citizens distrust and fear
the state, and hold low expectations that government has the ability, or the
desire, to meet their needs. Fragile states face deficits in fulfilling all
three of the core functions, i.e. assuring security, achieving effectiveness, and generating legitimacy. Meanwhile, Leftwich (2000)
proposes that the soft state is characterized by “a general lack
of social discipline in underdeveloped countries, signified by deficiencies
in legislation and, in particular, in law observance and enforcement, lack
of obedience to rules and directives handed down to public officials on
various level, often collusion of these officials with powerful persons
or groups of persons whose conduct they should regulate, and, at bottom, a
general inclination of people in all strata to resist public controls and
their implementation. Within the concept of the soft state belongs also corruption.
As a result, the soft state is incapable of promoting urgently needed
development.
[23]
To prevent
the decentralization from corruption hostage, some strategies are becoming
imperative, such as equipping decentralization with supervisory instrument and
rights by the community, strengthening anti-corruption initiatives and
platforms in local level. Legal reform, boosting leadership quality, and
maintaining cultural reform would also be top priority to prevent from
corruptive behavior.
[24] In the province
and district/city levels, payment of remuneration or performance incentives is
being charged to local budget. Since the budget capacity is diverse across
region, the amount of remuneration is also varied. As a result, the jealousy
among employees across regions couldn’t be avoided. The same jealousy has
happened before at the central level over the remuneration granted to Ministry
of Finance since 2007, while many ministries and agencies do not receive the
same policy treatment. Regarding the source of remuneration fund, government
does not need to provide or allocate separate funds, which would burden state
budget. Preferably, it should be taken from saving obtained from efficiency due
to reform. Otherwise, government may get collapse or even bankrupt.
[25] Korean experience
provides a good insight on how modern bureaucracy has been built based on
traditional culture. Korean government is principally based on the Confucian ideology
during the Yi-dynasty (1392-1910). The nine-level of government structure is
also adopted from bureaucratic model implemented by the Chosun Dynasty. See:
Lee (2007).
[26] Korean
experience in bureaucratic reform realizes a
policy of streamlining government functions and reducing its size. The new
administration reduced the number of cabinet members from 21 to 17 (Kim, 2000).
In addition, to precede regulatory reform President Kim Dae Jung has ordered
removal of 50% of regulation unconditionally. The purpose of this deregulation
is to simplify licensing procedures and provide more legal certainty to the
community, including the business sector. Regulatory reform is also intended to
reduce quasi-compliance and potential corruption (Choi, 2010).
[27] Just a matter
of fact, the number of ministries in Japan is 12, China 24, and Korea 17.
[28] According to
Law No. 39/2008 (article 25), LPNK is government institution which is formed to
carry out certain duties of state administration. Head of LPNK is directly
responsible to the President and is coordinated by the relevant minister.
[29] Grindle (2007)
defines “good enough governance” as minimum conditions of governance necessary to allow
political and economic development to occur.
http://ir.nul.nagoya-u.ac.jp/jspui/bitstream/2237/15868/11/7_Tri%20Widodo%20Wahyu.pdf