Good governance and anti-corruption strategies go together and are consequently the key for a well-rounded development agenda. Voluminous literature on corruption shows that poor governance, manifested by different forms of corruption, is a major deterrent to investment and growth. Public awareness of the detrimental impact and severity of the problem has increased markedly as more and more policy makers, businesses, civil society organisations and media have begun to confront the issue openly. The unprecedented public response to public agitations is indicative of the public wrath over corruption prevalent in both the higher and lower echelons of the government.
India’s performance – as is well known and also reflected in the Global Integrity Index 2007 – indicates a huge gap between anti-corruption policies and practices. The legal and institutional framework to curb corruption is well-developed and the country receives high scores in terms of anti-corruption law and institutions. Though a well-developed legal and institutional framework exists – with institutions, including the Central Bureau of Investigation (CBI), the Comptroller and Auditor General (CAG), and the Central Vigilance Commission (CVC) – evidently its effectiveness is quite poor. Despite some successes ranging from a tough stance on practitioners of corrupt practices by some elements of the judiciary, and the institution of the Right to Information (RTI) Act, the control on corruption seems to have withered away. Being an illegal activity, data on corruption is difficult to generate. But from recent events and evidence, it would be safe to say corruption is increasing and India is far from controlling, what appears to be, its exponential growth.
The police force is a key element in ensuring criminal justice. Investigations, under norms stipulated by the Code of Criminal Procedure, lead to a final report that can either lead to a no-offence situation or a chargesheet1. The National Crime Records Bureau annually publishes data related to various economic offences across the country. A closer look at this data reveals that the proportion of investigated cases has increased significantly from 66 per cent in 1990 to 91 per cent in 2012. However, chargesheeted cases, as a proportion of total investigated cases, have grown by a meagre 8 percentage points over the past 22 years. Similarly, the conviction rate (convictions as a proportion of completed trials) increased from 32 per cent in 1990 to 37 per cent in 2012 – remaining largely stagnant2. Despite the exponential growth in corruption, a modest increase in chargesheeted cases, or the ones that saw conviction, clearly reflects the government’s failure in keeping a check on corruption.
Meanwhile, according to the CBI’s 2012 Annual Report, bribery related cases against public servants accounted for around 20 per cent of the total cases registered with it. About 6 per cent of the total cases were registered for possession of assets disproportionate to known sources of income. Further, the CBI’s Crime Report for 2011 shows that 31 per cent of the total cases registered against the public servants were against Group A officers. However, Group A officers who underwent trial, accounted for a mere 12 per cent of the total trials undergone by public servants. On the other hand, 22 per cent of the total cases registered with the CBI are registered against Group C and D officers. Additionally, Group C/D officers, who underwent trial, accounted for 40 per cent of the total trials undergone by public servants3 – indicating that the more influential an officer, the greater the delays in trials being conducted.
The levels of corruption vary significantly from State to State, and so do the efforts of the government to tackle corruption. In view of this,we attempt to rank some of India’s States on the basis of the strides made by the State governments in combating corruption. The data available from administrative sources are quantitative and naturally lack some of the qualitative characteristics, such as, how well investigations were conducted and how well cases were administered. Be that as it may, analysis of this data can still provide invaluable insights into how the Indian State treats corruption.
Measuring anti-corruption efforts
One of the most important constraints in developing effective anti-corruption strategies is the intrinsic difficulty of measuring corruption. Corruption indicators can be of two broad categories: One, perception-based indicators, which are usually surveys of experts, investors and households; and two, fact-based indicators, which rely on such data as the existence of anti-corruption laws, legal prosecutions of corrupt acts and the existence of anti-corruption agencies4. Fact-based indicators have an advantage as they are less subjective and less sensitive to respondent and interviewer biases; however, these have an inherent limitation of data availability and accuracy as administrative data sometimes does not measure adequately the underlying activity. For instance, though the extent of corruption itself is impossible to infer from secondary data, the data on action taken against those accused is far more fertile in terms of insights it can generate.
It can also be argued that a good index on corruption, or anti-corruption, should contain actionable indicators that look not just at the outcomes but also at processes, and are able to separate the sometimes nuanced distinction between the change in law and change in practice. Indicators should also be able to capture corruption across various verticals, and across all geographies.
The National Crime Record Bureau (NCRB) annually publishes data on cases investigated and prosecuted by anti-corruption and vigilance departments of States and UTs under the Prevention of Corruption Act and related sections of IPC. While the official government sources are widely used for policy making, one would also be sceptical of actions derived from insights based only on such statistics; action against corruption can take many forms, not merely investigations – like, administrative tightening, persuasion, moral suasion, transfers, threats of a poor performance appraisal, etc. However, we argue that States that act against corruption do it in an all-rounded manner and are likely to use whatever tools they have.
Moreover, only a few of the indicators on corruption that are available in the public domain are actionable, for instance, the scale and scope of investigation and prosecution of corruption offences. However, many offences are not brought to light, and it is also true that in areas where the extent of corruption is high, the reporting of offences also tends to be low. Despite these limitations in developing an index to measure anti-corruption efforts, one should not undermine the importance of having such a measure-based administrative data on investigations. There are ongoing anti-corruption efforts by the State government that deserve some recognition.
Different states have been investigating and enforcing anti-corruption laws to varying degrees and the indicators selected for measuring such efforts of the government are such that they bring out the complete flow of ‘catch, pursue and justice’. Corruption-related crimes are handled through various means. Not all are cognizable; some are handled through departmental investigations . Take the national data, for instance. In 2012, 3,531 such cognizable crimes were registered. Accounting for a backlog of 6,814 cases from earlier years, there were thus 10,337 cases for investigation. Of these, chargesheets were filed in 2,961 cases. There is a issue of time-lag in the sense that chargesheets may be filed at some future date. Nevertheless, for illustrative purpose, the probability of a chargesheet being filed is 28.6 per cent. Looking at the State-wise variation, Gujarat has the highest probability of cases being chargesheeted while, apart from the smaller States and Union Territories (UTs), probability of cases being chargesheeted is low in the States of Delhi and Rajasthan.
In 2012, there were 747 convictions. The probability of conviction once a chargesheet has been filed is 25 per cent. Again, State-wise variations are stark with Punjab having 98 per cent of the chargesheeted cases being convicted, while in Bihar, Jharkhand and Chhattisgarh less than 10 per cent of the cases are convicted. Thus, while corruption is a low-risk and high-gain activity in India, it is not so in all States. Thus, an aggregated index value is computed, using all such indicators that reflect the series of action of the government in punishing the guilty for being corrupt.
State-level variations are brought out with the help of an index using data for this decade. The rating is done for 21 large States. UTs and the smaller states of North-East are not ranked due to low reporting of data there. The index is built using the following measures:
a. Cases chargesheeted as a share of total cases investigated.
b. Completion rate of trials.
c. Proportion of cases convicted to trials completed
d. Proportion of gazetted and non-gazetted officers in public sector undertakings (PSUs)booked under corrupt practices out of total persons arrested
e. Ratio of gazetted to non-gazetted officers in PSUs booked under corrupt practices.
f. Total pending trial cases with CBI for more than 10 years out of total pending cases.
The data on various measures for the period 2000-12 showed several fluctuations. A sudden rise or fall in any one measure can be the result of various factors and may not necessarily reflect a spurt of action taken by the government. Therefore, we have aggregated 5-6 year period by taking the arithmetic mean of the ratios. Two time points are considered, one, reflecting the early 2000s, i.e., 2000-05, and the other reflecting late 2000s, i.e., 2006-12. Each of the indicators are normalised using the standard max-min approach and then aggregated with equal weights (See Table 2).
Gujarat leads the nation in its anti-corruption efforts. It has more than 90 per cent of chargesheeted cases out of those investigated, which is far above than that of any other State. It has shown rapid improvement from the early 2000s in all measures. Completion of trials in the State increased by 7 percentage points, while conviction rate increased by almost 26 percentage points. However, the State still needs to look at the large number of pending cases with the CBI for more than 10 years.
Madhya Pradesh takes the second slot. It leads in various measures – the completion rate of trial is 15 percentage points higher than that of the all-India average. Similarly, conviction rate is at least 14 percentage points higher than that of the all-India average. Also, the State has the least number of cases pending with the CBI for more than 10 years. The index value of the State has increased 1.2 times compared to that in the early 2000s. This improvement is a resultant of various measures. Interestingly, in the last five years, more gazetted officers are being booked compared to non-gazetted officers. Clearly, it’s a reflection of better governance as the influential people in the State have equal probability of being caught as the less influential.
Andhra Pradesh, though among the top three performers, slipped from the top slot it had in the early 2000s. Though there has been some improvement in measures like conviction rate, rate of cases being chargesheeted, etc., but the State has declined miserably in catching the high-post officials. In the early 2000s, the State had seen more gazetted officers booked while in the last five years it has been almost negligible. This poses threat of more institutionalised corruption in the state.
Of the 21 States covered, 11 States showed an improvement in their index values from the early 2000s. Of these states, Gujarat, Madhya Pradesh, Bihar and Tamil Nadu are the top four achievers in terms of improvement. The success story of Bihar has been consistently followed by media and researchers. Undoubtedly, the governance in the State had a major role in its overall development. In the early 2000s, less than 6 per cent of the investigated cases were chargesheeted, while in 2012 it has increased to more than 31 per cent. Also, conviction rate out of the total trails completed during the year increased significantly in the last five years. More and more gazetted as well as non-gazetted officers are being brought to book. However, it still has grey areas. For instance, the completion rate of trials in the State is still amongst the lowest in India. A large number of cases are pending with the CBI for more than 10 years. Tamil Nadu has also put in place a very active vigilance and anti-corruption department. This is clearly reflected in the rising rate of completion of trials, conviction rate, etc.
Looking at the other end of the table, weakest State is Jharkhand. It fails miserably in all the measures of anti-corruption efforts. Proportion of chargesheeted cases to total cases investigated in an aggregated period of 2006-12 is only 39 per cent, far below the all-India average of 42 per cent. Completion rate of trial is even less than 1 per cent. With only five trials being completed in the last six years, the conviction rate out of the trials completed in Jharkhand is no measure of prosecution or justice. The State also has a large number of pending cases with the CBI.
Similarly, abysmally low figures in Assam, West Bengal, Uttar Pradesh and Karnataka put these States at the bottom five slots. What is even more worrisome is that these States were weak performers in early 2000s, and yet have shown no improvement. Of these, Karnataka had a major setback, falling by three positions, from 14th rank in the early 2000s to 17th rank in the late 2000s.
Other States with declining index values are Uttarakhand, Chhattisgarh, Rajasthan, Odisha and Punjab. Uttarakhand in the early 2000s was at the 8th slot, but it is now at the 16th rank. In six years, since 2006, only 35 cases in the State were investigated, of which merely 15 trials were completed and only seven convicted. Clearly, it fails in catching the guilty, leaving aside prosecution and justice.
Another story of a major setback is that of Punjab. Punjab was rated amongst the top two achievers in the early 2000s. More than 80 per cent of the cases in 2000 to 2005 were chargesheeted out of those investigated. The completion rate of trial was about 22 per cent, higher than the all-India average by 9 percentage points, and the conviction rate was higher than the all-India average by 11 percentage points. However, in the late 2000s, there has been significant decline in all of these measures. From 2000-05 to 2006-12, proportion of chargesheeted cases to cases investigated, decreased by 6 percentage points. Meanwhile, the completion rate of trials decreased by 2 percentage points and the conviction rate decreased by 3 percentage points. As a result, Punjab slipped to the 8th rank in the late 2000s.
Overall, there has been a trivial improvement in the index value from early 2000s to late 2000s. The weak States continue to remain weak with the exception of Bihar. Some of the top achievers in the early 2000s, like Punjab and Uttarakhand, slipped from their positions. Kerala, Himachal Pradesh, Haryana, Delhi, Maharashtra and Jammu & Kashmir grew marginally. The leaders in this effort are Gujarat and Madhya Pradesh.
Laveesh Bhandari is Founder Director, Indicus Analytics. Minakshi Chakraborty works at Indicus Analytics.