Good governance and corruption in the health sector: lessons from the Karnataka experience

Abstract

Strengthening good governance and preventing corruption in health care are universal challenges. The Karnataka Lokayukta (KLA), a public complaints agency in Karnataka state (India), was created in 1986 but played a prominent role controlling systemic corruption only after a change of leadership in 2001 with a new Lokayukta (ombudsman) and Vigilance Director for Health (VDH). This case study of the KLA (2001–06) analysed the:

• Scope and level of poor governance in the health sector;

• KLA objectives and its strategy;

• Factors which affected public health sector governance and the operation of the KLA.

We used a participatory and opportunistic evaluation design, examined documents about KLA activities, conducted three site visits, two key informant and 44 semi-structured interviews and used a force field model to analyse the governance findings.

The Lokayukta and his VDH were both proactive and economically independent with an extended social network, technical expertise in both jurisdiction and health care, and were widely perceived to be acting for the common good. They mobilized media and the public about governance issues which were affected by factors at the individual, organizational and societal levels. Their investigations revealed systemic corruption within the public health sector at all levels as well as in public/private collaborations and the political and justice systems. However, wider contextual issues limited their effectiveness in intervening. The departure of the Lokayukta, upon completing his term, was due to a lack of continued political support for controlling corruption.

Governance in the health sector is affected by positive and negative forces. A key positive factor was the combined social, cultural and symbolic capital of the two leaders which empowered them to challenge corrupt behaviour and promote good governance. Although change was possible, it was precarious and requires continuous political support to be sustained.

Governance, corruption, leadership, capital, health care sector, India

KEY MESSAGES

• Interventions against widespread corruption in the health sector should consider a multisystem approach which includes the political and justice system, the media and governance education of citizens.

• An effective anti-corruption agency requires a committed and powerful leadership, adequate resources, robust capability to investigate senior government officials and deal with internal governance issues, and the authority to propose institutional reforms.

• Governance in the health sector is the result of positive and negative forces at the individual, organizational and societal level.

• A shift towards good governance requires the interaction of leaders, followers and system changes.

Introduction

Globally, approximately US$3 trillion is spent annually on health care, with average losses from corruption of up to 10% (Transparency International 2006) and more in some countries (Lindelow et al. 2006). Corruption sets inappropriate incentives resulting in poor usage of resources, loss of trust in public services and health worker demotivation. It affects both the poor through informal payments and wealthier citizens through unnecessary treatment and investigations. Practices such as the use of substandard medicines and the sale of HIV-positive blood endanger society at large.

Promoting good governance and controlling corruption are global challenges. The health care sector is prone to poor governance and corruption due to uncertainty about future illness, asymmetry of information between different actors and the complexity of the system due to the large number of actors and interactions (Savedoff and Hussmann 2006). These factors increase the risk of human error, misjudgement, mismanagement including poor oversight, and corruption; together these constitute poor governance. Whilst the precise links between poor management and corruption are ill understood (Lewis 2006), mismanagement can facilitate corruption, and corruption can hide behind mismanagement. Misjudgement due to future uncertainty has to be differentiated from acts of corruption and mismanagement.

Research into health sector corruption has focused on measurement of specific practices and analysis of vulnerability in specific areas (Vian 2008). There is less coverage of the challenge of generalized poor governance and systemic corruption due to a dysfunctional legal framework and where the principal (government institution) does not act in the public interest.

Governance and corruption have been identified as key issues in India (Wade 1982; Sangita 1995; Anon 2003; Sanjay 2003; Sudarshan 2005; Transparency International India 2007; Solberg 2008). The former Prime Minister of India, Rajiv Gandhi, stated that only 15% of government money targeted at the poor reaches them (Sangita 1995).

The Karnataka Task Force on Health and Family Welfare (TFHFW) report described systemic corruption as the prime challenge for the State health system, widening health inequalities and distorting policy implementation (TFHFW 2001). Table 1 presents selected Karnataka State indicators. Since 1993 the Karnataka Panchayat Raj Act has decentralized political and administrative authority at three levels and provides for citizens’ participation in the social sector. There are large health inequalities and out-of-pocket expenditure for health services is a main cause of rural indebtedness. India and Karnataka State have a history of governance initiatives (Sangita 1995; Johnston and Kpundeh 2004), with selected ones listed in Table 2. The first Administrative Reform Commission (ARC) proposed ombudsman-like Lok Pal (central) and Lokayukta (state) institutions as public complaints agencies in 1966 (Sangita 1995; Venkatachala 2004). A bill for the central Lok Pal was introduced eight times between 1968 and 2001 but not enacted by the central Parliament (GoK 2006a). Seventeen out of 28 Indian states have established a Lokayukta (Abdul Kalam 2004) though most are not widely known. As yet there is no uniform authority for the different state Lokayuktas. The Karnataka Lokayukta (KLA) was established in 1986 to be headed by a retired Supreme Court Judge or High Court Chief Justice who is not a legislator and has no other office. The KLA has the authority to investigate complaints from citizens about public maladministration and initiate prosecution on criminal offences (Puliani 2005). Some public servants such as judges are excluded from investigation and the authority is confined to the exercise of administrative functions. Six months after the creation of the KLA, the original ‘suo moto’ authority of the KLA to investigate suspected offences of senior government officials without a written statement of a citizen under oath (affidavit) was removed (Table 2).

Table 1

Selected demographic, administrative, health service and health indicators of Karnataka State (adapted from TFHFW 2001)

Indicator	Year	Value
Demographic
Total population	2001	53 million
Number of inhabited villages	2001	27 066
Literacy rate in % of population (% of women)	2001	67(57)
Administrative
Number of Zillas (provincial level) Panchayats	2001	27
Number of Taluka (district level) Panchayats	2001	175
Number of Grama (local level) Panchayats	2001	5692
Health service
Public urban outpatient health care as proportion of total treatments received in public and private facilities	2001	0.30
Public rural outpatient health care as proportion of total treatments received in public and private facilities	2001	0.35
Public urban inpatient health care as proportion of total treatments received in public and private facilities	2001	0.49
Public rural inpatient health care as proportion of total treatments received in public and private facilities	2001	0.58
Private expenditures as proportion of total health care spending	1995/96	0.58
Institutional deliveries as proportion of total deliveries	1998/99	0.51
Mothers who received antenatal care as proportion of total pregnancies	1998/99	0.86
Health
Life expectancy at birth, male/female	1996–2001	62/65 years
Infant mortality rate, total/urban/rural per 1000 live births	1999	58/24/69
Maternal mortality ratio per 100 000 live births	1998	195

Table 2

History of selected central and state governance initiatives

Year	Level	Topic
1947	Central	Prevention of Corruption Act
1965	State	Mysore State Vigilance Commission
1966	Central	First Administrative Reforms Commission (1966–70) recommends public complaints agency with appointment of Lok Pal for central level and Lokayukta for each state
1984	State	Karnataka Lokayukta Act, came into force in January 1986
1985	State	Karnataka Lokayukta Rules
1986	State	Amendment to Karnataka Lokayukta Act in September 1986 with removal of ‘suo moto’ authority of the Lokayukta for senior public servants
1986	Central	Consumer Protection Act
1988	Central	Prevention of Corruption Act (PCA), Repeal of 1947 Act
1990	State	Efficiency Audit and Vigilance Bureau
1993	State	Karnataka Panchayat Raj Act, provides for citizens’ participation in the social sector at Grama, Taluka, Zilla level
2000	State	Karnataka Transparency in Public Procurement Act
2006	Central	Right to Information Act

In 1999 a new Karnataka State government was elected with a Chief Minister committed to fight corruption and improve state governance (Johnston and Kpundeh 2004; GoK 2006b). In 2001 the Chief Minister nominated a retired judge with 15 years experience in the High Court and 3 years in the Supreme Court of India for a period of 5 years to lead the KLA, and propose and implement an anti-corruption strategy. At the time the KLA agency had been criticized by the Karnataka High Court and Karnataka Administrative Reform Commission for its failure to hold governments accountable, assure effective redressal of grievances and improve public administration governance. There was a focus on minor corruption, a large number of pending cases, a low conviction rate and a failure of Legislators and State Ministers to declare their assets and liabilities before the KLA (GoK 2006a; GoK 2006b). Within six months the new (fourth) Lokayukta had created the post of Vigilance Director for Health, Education and Family Welfare (VDH) and appointed the Chairman of the TFHFW.

Subsequently the KLA became widely known and gained a reputation for independence and a strong will to fight maladministration. This paper describes and analyses the KLA approach (2001–06) on the following:

• What was the scope and level of poor governance in the health sector?

• How did the KLA define its objectives and implement its strategy?

• What factors affected public health sector governance and the operation of the KLA?

The next section explains the theoretical concepts followed by a description of the methods used, the findings and a discussion and conclusions on the future of good governance in India.

Background

There are no universally agreed definitions of governance and corruption. We define good governance as the exercise of power through institutions to steer society for the public good. Good governance should lead to inclusive, responsive and fair processes and outcomes, and public trust in a social system. Our governance definition draws on published concepts (Rosenau 1995; UNDP 1997; Huther and Shah 1998; Buse and Walt 2000; Graham et al. 2003; UNDP 2004; World Bank 2007; Iqbal and Shah 2008). We want to draw attention to the following aspects. We have replaced authority with power and state with society to emphasize that good governance requires real power and clear direction through norms and the commitment of a critical mass of citizens. The state is one key institution of society to serve its citizens. Other institutions comprise the system of rules, norms, values and policies which enable society to exercise power and establish a strategic steering process of its social system, including sub-systems, to serve its stated purpose. The steering process requires rule and control mechanisms such as clear responsibility, full transparency and accountability. Improved health system governance influences all health system functions and may lead to better performance and outcome (Siddiqi et al. 2009). However, governance is an inter-sectoral goal where changes can spread between systems and result in synergistic or antagonistic governance dynamics (Rosenau 1995).

We define corruption as the deliberate betrayal of public trust and the undermining of the public good for private gain which destroys universal ethical norms in society and leads to negative effects on social cohesion and steering of social systems. Public trust as a fundamental and volatile resource for the sustenance of a social system diminishes because of corruption. Our concept includes the responsibility of public office holders (World Bank 2007), private providers of public services (Transparency International 2006) and the actions of citizens weakening public functions and trust, e.g. through tax evasion or fraudulent electoral practices (ICAC NSW 2010). Corrupt practices are always intentional acts in contrast to misjudgement and mismanagement. However, systemic corruption is often beyond individual control so that the term abuse implying full personal choice is misleading (NORAD 2009).

The facilitating context for good governance comprises institutions, interests, policies (Huther and Shah 1998) and the balance of power. A theoretical understanding of power and corruption is necessary to analyse good governance measures. The model of principal–agent (P–A) with appropriate incentives is only useful when the government as principal is honest and the legal framework effective (Galtung 1998; Andvig et al. 2001; Leruth and Paul 2007). Galtung (1998) proposes a modification of the model (Figure 1) with P1 as the state and its collectivity of citizens as the sovereign. Government politicians are the principals (P2) who delegate tasks to public servants as agents (A2), but they can also become agents (A1). In this case Bourdieu’s theory of capital can help to explain the exercise of symbolic power by the collectivity of citizens as sovereign. Bourdieu lists four types of capital which enable the owners to exercise and compete for power and influence in society: economic (command over economic assets), social (network of relationships), cultural (knowledge, skills and education) and symbolic which originates from any of the other forms of capital and includes the law, moral values and ethical principles in society (Bourdieu 1997; Bourdieu 1998). His symbolic capital theory suggests the potential to transform the human desire for power into a positive force for good governance.

Figure 1

Two-level principal-agent-client model (modified from Galtung 1998)

Awareness and recognition by citizens is essential for the development of symbolic capital. Symbolic capital exerts ‘social gravity’ which holds society together. The State is the ideal place for the accumulation of symbolic capital (Bourdieu 1998). Individuals who promote the public good can embody symbolic capital. Bourdieu argues that morality, especially in politics, only has a chance if society creates the institutional means which promote behaviour based on moral values (Bourdieu 1998) and therefore institutionalize good governance behaviour.

Scandinavian societies are often mentioned as examples of good governance and a low level of corruption. Rothstein and Uslaner (2005) argue that public trust is rooted in universal and redistributive social policies which promote social solidarity and a perception of a shared fate among citizens. Based on the Scandinavian experience they see an initial equality in society and honesty in government as the starting point to develop universalistic social policies. We suggest that the widely recognized policies and institutions be treated as the symbolic capital of egalitarian Scandinavian society which empowers citizens, generates public trust and promotes good governance. Poor and inegalitarian societies may find it difficult to implement systemic changes and overcome the combined challenges of inequality, general mistrust and dysfunctional government institutions (Rothstein and Uslaner 2005).

Methods

The study was initiated by the former Lokayukta and VDH who were keen that the experience of the KLA during the period 2001–06 was documented and, as far as possible, evaluated. The Indian civil society organization Karuna Trust and the autonomous academic Indian Institute of Management, and two European institutions, of whom one provided the principal investigator, collaborated.

A study design for this evaluation was developed (Yin 1998) with the public complaints agency KLA as the unit of analysis. A participatory workshop in November 2006 in Bangalore brought together health professionals, academics and activists to identify factors impacting on corruption (De Koning and Martin 1996). The former Lokayukta (Ombudsman) and his VDH (the main KLA officers in the period studied) participated as the two key informants. The principal investigator and three local research assistants including the VDH collected the data and the latter reviewed the study report. The information gained from these key informants was triangulated and complemented through interviews, archival records and reports about KLA operations and site visits.

Given the sensitive nature of the evaluation, involvement of the former VDH was crucial to facilitate data collection. Interviewees and site visits were proposed by the VDH and selected because of their willingness to participate. Citizens were contacted in the vicinity of public health facilities to discuss their governance experience. All interviewees were asked not to reveal the names of persons involved in illegal activities and confidentiality was assured.

Forty-four semi-structured interviews were conducted with representatives of civil society organizations (CSO), public servants, health professionals and citizens. Two urban General Hospitals, the Health Service Administration of one Zilla (Province) including one rural Health Centre and one District (Taluka) Hospital were visited. Interview questions were based on 10 health sector functions, starting with an open question and prompting interviewees with examples from key informants of selected types of poor governance practice and their actors. Annual and inspection reports of the KLA, government documents, and newspaper articles were analysed. The principal investigator recorded events and ideas in a research diary (Hughes 1996). We used the force field concept to summarize the identified positive and negative governance forces at the individual, organizational and societal level.

No formal ethical clearance was sought, in part as the two Indian partners had no active ethical review process. However, both institutes endorsed the evaluation approach which is also supported by the Indian ‘Right to Information Act’ of 2006. The methodology complied with the ethical standards of the University of Leeds.

The design has obvious limitations given the difficulty of studies in this field including the identification of, and access to, respondents and differentiation between corruption and misjudgement. We did not attempt to measure specific corrupt practices. There is a risk of bias from various sources. Most obviously systemic governance issues extend beyond public health care into the political, legal and private sector; however, we had no access to leading politicians, senior government administrators or representatives of the private-for-profit sector. The dual role of the VDH as a researcher and key actor also presented potential bias, but we attempted to deal with this through triangulation. Despite these potential biases we consider that the results provide valuable insights into a rarely documented but key challenge to the health sector, and a specific institutional response.

Results

Perceived scope and level of poor governance and corruption

Table 3 presents examples of poor governance practices reported to the KLA after 2001. These types affect 10 health sector functions and cover a wide range of responsibility from politicians to users.

Table 3

Reported poor governance practices and responsibility

Health sector function	Poor governance practice (selected examples)	Responsibility
Medical education and training	Examiners at public and private colleges charge students to pass or ‘not to fail’ examinations regardless of student’s real academic achievement	Directors and teachers in training institutions, users
Financing	Unnecessary operations done for poor people [Below Poverty Line (BPL)] in private sector hospitals with reimbursement by government False BPL certification of persons who are above poverty line	Health facility directors, health professionals, politicians, users
Infrastructure development and maintenance	Corrupt tendering processes	Politicians, administrators, health professionals, companies
Material resource management (selection, procurement, distribution)	High ranking bureaucrats and politicians organize price fixing with suppliers of dialysis machines and medicines Free government medicines supplied to private pharmacies and sold to patients without prescription Companies set up in third party names but owned by officials and politicians who decide on procurement	Politicians, bureaucrats, directors of health facilities, health professionals, pharmaceutical companies, wholesalers
Oversight	Director/Administrative Medical Officer of hospitals does not monitor the budget (not trained in hospital management) Users misinformed and not made aware of their rights	Director/Administrative Medical Officers
Policy and strategic planning	Political influence of Ministers to start training colleges without required conditions such as infrastructure, teaching beds and teachers	Politicians, professional councils, universities
Regulation and inspection	Substandard medicines released into/not removed from market Blood supplied without HIV testing and no action taken Regulatory bodies of medical colleges bribed to approve institution without required norms	Drug control officer, pharmaceutical companies, professional councils, health professionals
Service delivery	Health facility staff demanded a payment of 150 Rupees (girl) or 200 Rupees (boy) to ‘show the newborn’ to the parents or siblings Unnecessary operations such as Caesarean sections performed for payment ECG machines made non-functioning at government hospitals so that physician could charge patients for ECG tests in private clinic Unjustified referral of patients to colluding pharmacists, blood banks or diagnostic centres	Health professionals, private for-profit service providers
Service use	Illegal sex determination is requested, with sex selective termination of pregnancy Orthopaedic surgeons provide false results of disability assessments and get a percentage of benefit claims	Health professionals, users
Workforce management	Absenteeism from public sector Doctors not respecting requirement to reside in location of health facility Transfer, promotion or further education is linked to payment of bribes to senior staff which have to be shared with administrators and politicians	Politicians, senior bureaucrats, directors, health professionals

Though poor governance was widespread, there were several instances where it was impossible to differentiate between mismanagement, misjudgement and corruption. Corrupt practices were reported at all levels. Indeed, interviewees stated that the KLA itself was corrupt before the changes in 2001. Complaints from the public were either not investigated or false complaints were framed to intimidate professionals. Government employees did not trust anybody to deal with their complaints about poor human resource management in the Department of Health (DoH). The systemic challenge was summarized by a government doctor:

“The system doesn’t recognize honesty, so the doctors think let me at least make money…. Corruption has become mainstream.”

One health centre medical officer spoke positively about the leadership in his province expecting different behaviour:

“The Zilla Health Officer never collects any money from the health facilities.”

KLA objectives and strategy

After the new Lokayukta’s inauguration, the Chief Minister announced the plan to amend the Lokayukta Act based on proposals from the Lokayukta to strengthen the KLA organization and authority and turn the KLA into a functioning institution (GoK 2006b). For the first time a Lokayukta took the oath in Kannada, the state language, and spoke to journalists about how media and citizens could help him to discharge his duties.

The new Lokayukta attributed the previous failure of the KLA to the following (GoK 2006b):

• Authority of the KLA undermined by Legislators and Ministers not declaring their assets and liabilities;

• Political corruption ignored as the root of administrative corruption;

• Absence of reporting between investigating KLA and prosecuting police;

• Sanctions for prosecution even in clear cases poorly drafted with no conviction.

The Lokayukta summarized the KLA objectives to (GoK 2006b):

• Make the KLA transparent and accessible for the public and remove corrupt officers;

• Educate the public about good governance and determinants of corruption;

• Target senior public servants’ governance behaviour;

• Propose an amendment of the KLA Act to the Government to restore ‘suo moto’ authority against senior public servants, remove gaps and inconsistencies hindering effective trials, and allow for effective action against Legislators and Ministers not declaring assets and liabilities;

• Advocate for a central Lok Pal and uniform Lokayukta Act, confer constitutional status on the institution, make recommendations of the institution binding and open to challenge in the central Supreme and national High court.

The Lokayukta believed he could not function effectively without technical expertise. He created three new posts of Vigilance Director of the Police, of the Bangalore Municipal Corporation, and of Health and Education (VDH) based on section 15(3) of the KLA Act. Against strong political resistance, he appointed the (former) Task Force chairperson of the TFHFW into the VDH position. The VDH was a medical doctor with extensive health experience who had been managing the Karuna Trust, a respected CSO, for many years (Karuna Trust 2006). He had lived on donations without an official salary for the past 20 years to demonstrate his service to the public good. The Lokayukta agreed to pay him a nominal amount of 1 Rupee per month, as a salary was required for an official employee of the KLA. This facilitated the employment, since no budgetary objections could be raised by the Government. The collaboration of the Lokayukta and the VDH focused on health rather than education, because corruption appeared to be more frequent and serious in the former and particularly affected the poor (Sekhar and Shah 2006). The Lokayukta stayed in office until 2006 but was not reappointed, reportedly due to a lack of political support for his activities. This led to the resignation of the VDH.

The strategy of the Lokayukta considered related systems and organizations and the collectivity of citizens as the sovereign of the Government. Figure 2 maps the important KLA interventions and their expected impact on governance control. The central section presents the institutions of KLA, Police, Prosecution and Judiciary whose functions are essential to deal with serious maladministration cases. The term maladministration rather than malgovernance is used, because the KLA authority [in contrast to ICAC Hong Kong (ICAC 2007)] is limited to the public sector (excluding the judiciary), publicly owned companies and registered CSOs. The KLA can investigate (square dot arrow) complaints of citizens (long dash dot arrow) and cases referred by the government in organizations (left and right section of map) which either control and provide health care services or provide resources for health care. After investigation the KLA can opt to: Options 2, 3 and 4 can be combined in different ways. The process depends on the permission and follow-up of the competent authority and integrity of the judiciary.

1. Close the case, or,

2. Recommend redressal of injustice for the complainant,

3. Propose vacation of office for the accused public servant,

4. Initiate prosecution.

Figure 2

Map of KLA strategy and interventions to improve health sector governance in Karnataka State

The Lokayukta approached organizations to support his efforts (green arrows). The KLA established a beneficial collaboration with the private media informing citizens about corruption scandals and conflicts with the Government such as over the reintroduction of the ‘suo moto’ authority to facilitate the investigation of senior government officials. This authority was important, because low-ranking public servants and citizens were afraid to provide an official statement (affidavit) about corrupt behaviour against a powerful public servant. The autonomous Karnataka Drug Logistics Society was created with the assistance of the European Union and was a major improvement, making procurement of medicines more transparent and accountable to the public. This was not initiated by but was welcomed by the KLA.

During their term, the Lokayukta and VDH attempted to make themselves accessible to citizens, KLA and health service employees. Interviewees described them as independent, proactive and approachable. Their approach was perceived as simple, fair and transparent, and many newspaper articles commented on their commitment to the common good. The Lokayukta and VDH visited all 27 Zilla (provincial) and 175 Taluka (district) Panchayats (administration). They investigated between 100 and 200 complaints on each visit and mobilized citizens and the private media to report these activities. The Lokayukta described the role of the media in the control of corrupt behaviour as:

“It is the truth that they [corrupt officials] cannot deny.”

The Lokayukta and VDH used their position to mobilize citizens through CSOs to control local public health facilities through governing boards and the Government through the elected legislators. The approach was successful in some local facilities but failed at the state level due to lack of political accountability, as will be discussed later.

Table 4 summarizes how complaints under the KLA Act, with and without affidavit, increased sharply after their first year in office. Serious complaints, as described in Table 3, were investigated and typically were found not to be due to misjudgement. The number of convictions increased only during the fourth year, because the initial KLA aim was public education, rather than legal action. However, prosecution was commenced against some (mainly junior) corrupt officials based on KLA findings. Barriers within the political and judicial system explain the low conviction figures and were beyond the Lokayukta’s authority. The sharp increase in pending cases (with affidavit) can be linked to inadequate numbers of KLA staff and failure to implement KLA reforms. The inconsistencies of some figures are due to the weak monitoring system within the KLA.

Table 4

Reported complaints, activities and results of the KLA between 2001 and 2005

	Complaints under KLA Act with affidavit (written declaration made upon oath)					Cases referred by government					Complaints received without affidavit			Cases investigated by police wing of KLA				Manner of disposal of police wing cases			Court convictions
Period	A	B	C	D	E	A	B	C	D	E	B	C	E	A	B	C	E	F	G	H	I	J
2001–2002	907	1958	880	124	1985	25	24	22	7	27	3085	788	2986	426	241	231	436	10	125	96	18	1
2002–2003	1854	7256	4928	115	4182	32	32	24	13	27	8185	7233	3938	441	173	245	369	2	168	75	10	0
2003–2004	4271	7732	4461	92	7450	30	38	15	6	47	7913	7842	1931	369	109	195	283	4	145	46	19	0
2004–2005	7450	7096	4712	56	9778	47	27	32	2	40	6988	7965	954	283	105	158	230	12	123	23	41	1

Key:

A: Cases pending at start of period; B: Cases received during period; C: Cases disposed during period on enquiry/investigation; D: Action on cases recommended to competent authority; E: Cases pending at end of period; F: Departmental enquiry recommended but no prosecution; G: Prosecution launched; H: Cases closed; I: All court convictions; J: Convictions linked to health sector.

The Lokayukta was generally perceived as having controlled corrupt practices within the KLA itself—an important achievement. The interventions of the KLA were seen as having promoted good governance in general (e.g. increased citizens’ complaints) and assisted local good governance initiatives (e.g. such as setting performance targets) effectively within the health sector. However, the Lokayukta was unable to introduce the proposed institutional changes because of political resistance and his short period in office. A deputy Indian prime minister has described the limitations of the political system; Legislators started their political career with a lie under oath stating that they had not spent more money on their election than stipulated by the Election Commission (GoK 2006a). Interviewees confirmed poor governance practices among elected Legislators. The Lokayukta proposed legislative changes to the executive Government with no success. A KLA public servant reported the statement of a politician relieved about the departure of the Lokayukta and VDH:

“Who will investigate me – nobody can investigate me now!”

Factors affecting governance and operation of the KLA

We now examine factors affecting governance and the KLA operation at two levels: the health facility and wider system factors.

Governance at the public health provider level

Our observations, interviews and document analysis identified a number of factors seemingly affecting governance and the effectiveness of KLA interventions. The factors act as forces promoting good or poor governance at the level of the individual, organization and society (Table 5). For example, at the individual level the perception that rules will be enforced seemed to be an important influence on good governance, while at the society level weak representation of citizens was a factor which many interviewees associated with poor governance decisions within specific health facilities.

Table 5

Positive and negative factors affecting governance at the public health provider level

	Governance factors
Level	Positive	Negative
Individual government agent	Management training Passion and commitment to management Same moral standards for all citizens reporting governance problems Perception of enforcement of procedures	Financial problems Perception of weak sanctions Low salary
Organization	Honest leadership committed to the common good Organizational transparency Accountability Adherence to procedures Complexity reduction Effective board of citizens	Leaders without adequate training in management or corruption prevention Inadequate financing Lack of autonomy Vague procedures Large space for discretion Poor education of citizens
Society	Effective and well-resourced anti-corruption agency Functional government institutions such as fair process for election and changes in power Government transparency Independent media	Weak representation of citizens Government policies favouring particular groups such as private sector Inadequate vetting of political candidates Inadequate public financing of election process

At the wider level our research identified a number of systemic themes.

Political system

The political system including senior administration was described by most interviewees as key to governance:

“Top level corruption should stop, only then corruption will stop.” (Medical superintendent)

The Chief Minister, who had nominated the Lokayukta, expressed publicly his ‘helplessness’ to introduce the proposed institutional KLA reform against political and administrative opposition (GoK 2006b), highlighting the difference between formal authority and real power. Altogether three different Chief Ministers between 2001 and 2006 promised and failed to implement the proposed changes of the KLA (GoK 2006b). Interviewees reported that many politicians saw an election campaign as a financial investment which had to be recuperated once elected. Health workforce management was targeted for this purpose so that employment, transfer, promotion and development decisions were linked to informal payments. These costs were passed on from service providers to service users:

“To enter Government services, jobs are purchased at different rates. Ministers directly make money from this. It is a market economy, so people then try to make more than what they spent to get the job. This is a start for corruption.” (CSO officer)

Justice system

The Lokayukta was aware of the paramount importance of the central and state justice system. He tried to address several factors which he considered undermined the effective use of the justice system to investigate and penalize corrupt practices:

• The absence of a central Lok Pal institution to investigate at the central level and of a uniform central Lokayukta Act weakened the state institution.

• Gaps and inconsistencies in the KLA Act such as the absence of ‘suo moto’ authority to investigate senior politicians and administrators without a written and sworn statement and the inability to oblige Government to follow up KLA recommendations undermined its authority.

• Lack of clarity existed between the functions of the central Prevention of Corruption and the state KLA Act.

• Corrupt practices within the KLA, the Police and the Judiciary undermined the authority and legitimacy of these institutions.

These factors were serious handicaps for the effectiveness and credibility of the KLA. Government sanctions for prosecution were forthcoming with junior public servants, but were withheld in high level corruption cases such as a University Vice-Chancellor. Government failed to act on KLA recommendations for institutional reforms and to remove the suspension of judicial proceedings against public servants obtained in the High Court. Three successive governments used legal inconsistencies to weaken the authority of the KLA.

The Lokayukta tried to mobilize senior politicians and civil society at the national and state levels for a more effective KLA and legislation. This created considerable awareness among citizens and health professionals, e.g. the complicated issue of ‘suo moto’ authority which was widely discussed in newspapers and mentioned by most interviewees. However, it was not translated into legislation due to a lack of political will and power.

The Lokayukta targeted corrupt practices within the KLA and the Police which assisted the KLA with investigations (Figure 2). He decentralized KLA functions to the provincial level to improve access for citizens. He appointed a special Vigilance Director for the Police. The functioning of the KLA improved as demonstrated in Table 4. The efforts against corrupt practices in the police were described as less successful than those in the health sector. An interviewee explained that the appointed Vigilance Director (Police) was ‘playing safe’ and was less co-operative than the VDH.

Role of private sector

Poor governance practices in the public sector were influenced by the private sector. Interviewees stated that national and international private companies bribed senior public servants. The private and political need for money made public functionaries vulnerable to corrupt offers. Some professionals moved government equipment and supplies to their own private business and referred public sector patients to their private practice. Furthermore, some public facilities were transformed into private enterprises. Public health professionals justified corrupt practices on the basis of their low income in comparison with the private sector.

“Government doctors come to the hospital at around 11am and collect cases for their private practice… The doctors in private sector are a bad influence on Government doctors. They mint money. This creates frustration among the Government doctors who think that if patients are willing to pay Rs. 50 to the private doctor, they can easily pay Rs. 20 [bribe] to the Government doctor.” (CSO officer)

Public health professionals questioned why private health care was praised by senior government officials during interviews and not controlled for corrupt practices when, for example, illegal antenatal sex determination and sex selected termination of pregnancies were widely practised in the private sector.

Role of public voice

A critical and committed citizenship seems to have played an important role in the move towards better governance, as shown by the large and increasing number of complaints with and without affidavit made by the public (Table 4). The creation of the KLA and the appointment of the Lokayukta in 2001 resulted from public concern about poor governance and election promises. The failure to reappoint the Lokayukta in 2006 also caused public demonstrations demanding his reappointment. The Karnataka State Human Rights Council (KSHRC) organized a mass opinion poll campaign on the internet. However, the public voice was insufficiently strong to influence political decision makers and a representative of civil society stated:

“How can you succeed, if there is no political will?”

Good basic education of a critical mass of citizens was seen by many interviewees as an important factor in detection and control of governance problems. One hospital visited in the research was perceived as having better governance and an active governing board. The citizens in the area were described by several interviewees as better educated with more economic resources.

KLA resources

The Lokayukta and VDH perceived that the KLA was poorly equipped. Clearly an assessment of appropriate resources requirement is difficult. The KLA budget comprised about 0.03% of State expenditure. Staffing (469 employees and 212 vacancies in 2006) was also perceived as inadequate, given a role of oversight of 616 365 employees including 6000 doctors and more than 32 government departments. The KLA has no official responsibility for the private sector. In comparison, the Hong Kong anti-corruption agency received 0.3% of the budget of the government of Hong Kong Special Administrative Region (HKSAR). With around 1200 staff members, it has jurisdiction over corruption-related matters in both the public and private sectors in HKSAR. Currently, there are some 160 000 civil servants and around 60 government departments in HKSAR (ICAC communication, 1 August 2008).

Discussion

Following the change of leadership in 2001, the KLA appeared to improve both its own function and to some degree the governance of health care. The collaboration of the Lokayukta and VDH was based on their common understanding of governance. Both saw governance as the responsibility of all citizens. They expected that good governance of the health system should lead to inclusive, responsive and fair outcomes, as suggested by Iqbal and Shah (2008), and regain trust of the public in the system. The KLA operated in a context of widespread public sector problems as described elsewhere (Sangita 1995; Anon 2003; Sanjay 2003; Rao 2005; Sudarshan 2005; Cameron 2006; Transparency International India 2007). These problems had not changed since 1990 despite the existence of the KLA for 15 years (Sangita 1995) and a short-lived government-initiated good governance coalition (Johnston and Kpundeh 2004).

The appearance of a committed new KLA leadership in 2001, combined with a strong and persistent movement of citizens, appeared to have been critical for the success of the KLA. Many authors have stressed the importance of committed leadership (Hock 2000; Larmour and Grabosky 2001; White 2001; Vian 2008) and critical citizens for good organizational and public governance (Hock 2000; Johnston and Kpundeh 2004; Joshi and Moore 2004; Savedoff and Hussmann 2006; Peters and Muraleedharan 2008). Such leadership needs to start at the political level; electoral reform is needed (Wade 1982; Sangita 1995) so that political power is well exercised and promotes good governance.

A theoretical understanding of power and corruption is necessary to develop and support appropriate good governance measures. The model of principal–agent with appropriate incentives is only useful when the principal is honest and the legal framework effective (Andvig et al. 2001; Leruth and Paul 2007). In our case, Bourdieu’s theory of capital helps to explain the relative success of the KLA in the period 2001–06 compared with previous Lokayuktas (Bourdieu 1997; Bourdieu 1998).

In 2001 the two KLA leaders combined considerable social capital from their networks, cultural capital from their legal and health care expertise, and symbolic capital from their visible commitment to the common good. The nominal salary drawn by the VDH, and the media and civil society reports about the actions of the Lokayukta, demonstrate their embodied symbolic capital (Rajendra 2006).

The Lokayukta and VDH used their combined capital to exercise power and improve governance in the health sector. Our analysis identifies the following factors as critical to control health sector corruption:

• Political and justice system,

• Administrative authority of the KLA,

• Promoting transparency on, and public voice and participation in governance issues,

• Anti-corruption coalitions,

• Universal and redistributive social policies.

The Lokayukta used his capital and tried to address the political and justice system at central and state level in order to improve governance in health and other sectors. However, his influence at the central level was limited and the KLA responsibilities in the political and justice sector were constrained. The link between the judicial and political systems appears to be critical for the promotion and protection of good governance in all other social systems (Transparency International 2007). Good governance of the justice system depends on a web of interdependent institutions such as the judiciary, police, prosecution, lawyers and enforcement agencies (Buscaglia 2007). A weak justice system which is neither independent nor able to enforce the rule of law on the political system can lead to mutually reinforcing corrupt practices and may deter honest candidates from entering the political system (Transparency International 2007). In this case study the elected political representatives played no active governance role, which caused a serious disruption of the circuit of governance accountability between citizens, legislative and executive government (Figure 2).

The main strategies of the KLA were raising public awareness, and controlling and penalising poor governance behaviour. The last focus required considerable resources, but had limited success in obtaining convictions due to the complex judicial system with its own integrity issues (Transparency International India 2007), inconsistencies and the lack of political will and power. The absence of the ‘suo-moto’ authority for senior government officials and an official KLA mandate to propose administrative reform further limited the effectiveness of the KLA. In analysing the success of the Hong Kong anti-corruption agency ICAC, Doig (1995) stresses the strategic priorities of public education and structural reform to make the administration more resilient against corruption. ICAC Hong Kong and ICAC New South Wales (Australia) have their own system to investigate internal governance issues (ICAC 2007; ICAC NSW 2010). Both ICACs have specific units for prevention and education including research, and ICAC Hong Kong includes the private sector.

The Lokayukta and his VDH strengthened their public position through a combined strategy of transparency of actions, dissemination of information and public accountability. The media assured a continuous informal dialogue with the public voice. Robertson (2007) describes publicity on governance as the best disinfectant against contagious corrupt practices. The right of access to information, which has been introduced in India in 2006, and a public interest defence for media appear to be essential preconditions for tackling corruption. A recent proposal to improve governance in India’s health services also suggests the need for a combined approach with better public participation, rather than an isolated administrative approach focusing on rules (Peters and Muraleedharan 2008). It proposes consumer-oriented approaches which require a functioning legal framework and institutionalized co-production of good governance, with power and resources jointly used by the State and citizens’ groups to improve governance. This co-production approach can be described as a combined top-down and bottom-up governance approach (Rosenau 1995) or ‘citizens being actively involved to improve the functions of State agencies’ (Ostrom 1996).

The strategy of anti-corruption coalitions (Johnston and Kpundeh 2004) can be linked to the proposed institutionalized co-production of regulation. The ideas of a strong and joint political will of leaders and followers, with a vigorous civil society supportive of its members (Johnston and Kpundeh 2004) and of strengthened public authority through institutionalized co-production (Joshi and Moore 2004; Peters and Muraleedharan 2008) fit well with the theory of symbolic capital. One example was the creation of the autonomous Karnataka Drug Logistics Society. The Lokayukta and his VDH were aware of the social challenges of inequality, general mistrust and dysfunctional government institutions (Rothstein and Uslaner 2005), particularly the financing of political elections, but they had neither the authority nor the time to deal with them effectively. They encouraged the active involvement of citizens in the KLA activities and proposed more governance control at the decentralized Grama, Taluka and Zilla level. A change of KLA leadership in 2006 has threatened this approach, because it was not institutionalized and therefore dependent on the new leader.

This case study presents valuable international lessons and reveals several factors which can force a system towards either good or poor governance. Vian (2008) developed a framework for corruption with a focus on the individual being influenced by three factors (rationalization, opportunity and pressure to abuse). While individual corruption is an important issue, our concern is a system shift from poor to good governance. We see the individual as part of an organization and part of a society (Table 5) and present this through a force field diagram of good and poor governance with three levels. Specific factors exercise their effects at one or more levels.

In our case study the leadership of the Lokayukta and the VDH transformed the KLA organization into a functional government institution for good governance. As a consequence, a general shift towards improved governance was reported. The same factors exercised their positive force on different organizations of the health care system. Organizations with good governance as ‘islands of hope’ could be found in a context of systemic corruption. Committed leadership with management training, adequate, transparent and respected procedures, and an educated and committed citizenship could produce such a shift towards good governance in some organizations. Our case study demonstrates the important interaction between leaders, citizens and system changes. However, a sustainable governance improvement would require a persistent and strong citizens’ movement to achieve the necessary institutional changes.

Further research is needed to:

• Assess the usefulness of the force field analysis of governance;

• Search for critical factors at different levels of a health system and in the wider context to strengthen good governance;

• Identify the determinants of an effective anti-corruption agency in different political, cultural and socio-economic contexts;

• Determine what is the critical mass of factors to achieve system changes; and

• Investigate whether symbolic capital combined with other forms is a useful and measurable concept to improve understanding about committed leaders, and resilient and vulnerable societies and organizations.

Conclusion

Governance in the health sector is the result of positive and negative forces at the individual, organizational and societal level. Control of widespread corruption in the health sector requires a multisystem perspective which considers the political and justice system, the media and the education of citizens. The effectiveness of an anti-corruption agency depends on a committed and powerful leadership, sufficient resources, and the ability to investigate suspected senior government officials, to deal with internal governance issues and to propose institutional reforms. Interaction between leaders, citizens and system changes is essential for governance improvement.

Funding

The University of Leeds financed the study.

Ethical approval

The evaluation complied with the ethical standards of the University of Leeds.

Acknowledgements

We would like to thank the citizens and health professionals of Karnataka State for their frank contributions, the University of Leeds for financing the study and Rob Summers for his comments on an earlier draft.