Set the ‘caged parrot’ free now
New Delhi, December 03, 2014
First Published: 23:00 IST(3/12/2014)
Last Updated: 13:31 IST(4/12/2014)
The new CBI director, Anil Kumar Sinha, takes over the reins at a particularly grim time for India’s premier investigation agency. The agency’s reputation has deteriorated owing to its handling of high-profile cases including the Aarushi Talwar case and the 2G, coal mining and Commonwealth Games scams.
Mr Sinha’s predecessor recently had the ignominy of being told by the Supreme Court to recuse himself from the 2G case. It is difficult to recall when the announcement of a CBI investigation last invoked expectations of rigorous inquiry and eventual justice.
An ambitious country like India cannot continue to have its leading investigation agency dogged by controversy and inefficiency. The Narendra Modi government must realise that having a hobbled CBI will in the end undermine its own reputation. The problem and the potential fixes are fairly clear. The CBI is undermined repeatedly by political interference despite strictures from the courts.
There is certainly scope for officers to demonstrate more integrity and resist interference, something that is woefully lacking, particularly when strong governments reign.
But there can be no substitute for structural solutions. The CBI lacks financial autonomy to insulate itself from political interference, fill vacancies and plot its own modernisation programme. The agency also has no control over the appointment of its senior officers.
The CBI leadership has repeatedly petitioned for autonomy to little avail. Under the UPA rule, the Centre refused to grant the ex-officio powers of a secretary to the CBI director on the grounds that it would set a precedent and “create heartburn” in other organisations. Respected jurists and policing experts have long argued that without a separate legal framework, the attempt to change the CBI from within will be fruitless. They reckon that the CBI ought to have statutory status like the Election Commission or the CAG and have its own charter that flows from a separate piece of legislation.
These are arguably tough questions for politicians wary of relinquishing authority and creating new power structures. The Modi government must, however, set in motion processes where such debates can take place. Creative handling at the Centre is imperative if the states are to follow suit and reflect constructively. A good start would be to hold Mr Sinha to high standards of delivery and accord him the degree of freedom his agency sorely needs.
Corruption: A Review
, Dundee Discussion Papers in Economics
Andrew Wedeman (2005), “Anticorruption Campaigns and the Intensification of Corruption in China”,Journal of Contemporary China, Volume 14, Number 42, February, 2005 , pp. 93-116(24)Gardiner J () “Defining Corruption” in A.J.Heidenheimer, Johnston.M., eds., Political Corruption: AHandbook (25-39)Jain, A. K. (2001), “Corruption: A Review”, Journal of Economic Surveys, 15: 71–121John R. Heilbrunn(2004), “Anti-Corruption Commissions: Panacea or Real Medicine to FightCorruption?”, World Bank , Stock No. 37234Kaushik Basu, Sudipto Bhattacharya, Ajit Mishra, (1992) “Notes on bribery and the control of corruption”, Journal of Public Economics, Volume 48, Issue 3, August 1992Mauro, Paolo. "Corruption and Growth" Quart.J. Econ., Aug. 1995,11093), pp. 681-712Pranab Bardhan,(1997), “Corruption and Development: A Review of Issues”, Journal of EconomicLiterature, Vol. 35, No. 3 (Sep., 1997), pp. 1320-1346Thomas Herzfelda, Christoph Weiss (2002), “Corruption and legal (in) effectiveness: an empiricalinvestigation”, European Journal of Political Economy, Vol. 19 (2003) 621–632Wade. R. "The Market for Public Office: Why the Indian State Is Not Better at Development," WorldDevel., Apr. 1985,13(4), pp. 467-97
question to see how the overall national perception is changing and how the perception in ruralarea is changing.
:The findings of this paper correspond well to the popular belief that very few corrupt officials arecaught in India. One important thing to curb corruption is to make it less appealing to thepotential corrupt officials. This can be done by increasing the fine or by increasing thepunishment rates. The fine need not be always in monetary terms. Higher fine in terms of socialstigma of corruption can also help in this matter. There is a wide difference in the performancestate anti-corruption bodies. The paper looks into parameters on which the performance of different bodies can be compared and specific improvements can be done in those areas. Role of courts is important for good performance of a state in anti-corruption efforts. The increase intrials completion rate in courts has been coupled with low conviction rates which should belooked into more detail. The widely used index for corruption measurement in a country isCorruption Perception Index by Transparency International. The index is based on perception.There is a need to check how this index correlates with ground reality.
small number, even though there is high conviction rate the anti-corruption effort in thestate remains poor.
Chart 2: Performance of Bihar(Source: Various reports of National Crime Records Bureau).
Another interesting case is that of Karnataka. As shown in the figure, the cases registeredin state of Karnataka shows a decline from year 2000 to year 2004. But the graph of trialscompleted shows increasing pattern in the same period. While from the year 2004 casesregistered shows upward trend trials completed show a downward trend. This might bebecause of low number of cases coming to courts gives more hearings per year to casesso that results in most trials are completed.
Chart 3: Performance of Karnataka
(Source: Various reports of National Crime Records Bureau)
In case of West Bengal, in last ten years there is almost no activity from state anti-corruption Bureau. The average number of cases registered is 0.9. It seems the anti-corruption Bureau is not active in the state. The one party dominance and difference in
Percentage of cases charge sheeted to cases registered –
The anti-corruption bureau investigates a case after it’s registered. At the end of investigation it is decided where to file a charge sheet or not. This filtering of cases between two stages can be due to false accusation which are discarded or inefficiencies in investigation of ACB. This ratio varies drastically across states. From 26% in Tamilnadu to 92% in Gujarat. This can be because of difference in criterions for charge sheeting in various states. This is a ratio of two parameters from the same state. As it is ratio it is independent of size of state. So it can be directly compared with other states without the need for normalization. There is a quantitative problem while calculating this ratio. The two variables in the ratio do not strictly match in their time periods. As it is possible tha tsome cases registered this year would be charge sheeted next year and vice a versa. It is assumed that when we take data for many years this effect due to non-perfect mapping is evened out.
Ratio of trials completed to cases sent for trial –
This parameter gives how fast trials are completed in courts. As the famous saying goes‘Justice delayed is Justice denied’. A case which is dragged in the court can be bad in two ways, an innocent has to suffer for a long time as well as person who has actually committed crime gets lot of time before his judgment. Many a times delay in court procedures is cited as major road block for anti-corruption efforts..There is a huge difference between the values of this parameter across states. While many states have shown increasing trend in this ration a lot remains to be done in states like Bihar. Again the variables in ratio are not strictly mapped to each other. But the explanation given inthe above parameter applies here as well.
Conviction rate is the ratio of number of cases convicted to trials completed. It is not clear whether high conviction rates can always considered be to good or bad. Japan,famously, has got very high conviction rates in tune of 99.97% . But this is not considered to be very good phenomenon because innocent people might be getting punished too often. But similarly very low conviction rates mean that guilty are acquitted.So it might be considered that extreme low or high rates of conviction is bad but in middle range higher rate of conviction is better. Indian states show wide variety in this parameter as well. There are states with very low conviction rates, but the causes of this low rate cannot be said to be only because of courts. How strongly case is presented by the anti-corruption bureau and the public prosecutor is also important. As none of the states in India shows very high conviction rates it can be assumed that a high conviction rate is a desirable characteristic.
2. Comparing performance of different states:
The other source of data for corruption is National Crime Records Bureau. Every year it publishes ‘Crime in India’
report which has summary of various criminal activities. It provides details of cases registered under prevention of corruption act 1988 from various states. The data is available from 1989 to 2010 the reports provide information of numerous parameters. There are various stages from case registration to conviction. Data provides details of the seintermediate stages.
This is shown graphically in the figure.After the case is registered an inquiry is carried out by Anti-corruption Bureau (or respective anti-corruption authority). After the investigation, it is decided whether there should be charge sheet drawn or not. After case is charege sheeted it is sent to court for trial. At the end of trial the person is convicted or acquitted. The important parameters which help us see the effectiveness of the any corruption efforts are
Number of cases registered in a year-
This forms an important parameter to judge how well the state Anti-Corruption Bureau unit is performing as well as the awareness and activeness of the people of the state. The cases registered by the anti-corruption bureau are on the complaints of citizens. So apart from functioning of anti-corruption bureau how vigilant citizens are also plays a role. But these should not be considered as mutually independent factors because if anti-corruption body is effective people feel comfortable to lodge the complaint. So amongst the twoanti-corruption bureau plays important role. The size of states in India varies widely. So itis important to compensate for variation of differences in states. The normalizing factorthat can be used is the state expenditure as it represents the economic activity of the stategovernment.
all corrupt central government employees only 0.96 % of them are punished according to 2010data.
Table 5: Probability of corrupt central government employee punished
officials 10% 25% 50%
4.82 1.93 0.96
5.51 2.20 1.10
4.69 1.88 0.94
3.08 1.23 0.62
As it is seen from the table 5, the probability of punishment is very low. As we have no surve ypointing at actual rate of corruption, we attempt to calculate an estimate based on judgment of former CVC Mr. Pratyush Sinha According to him the 80% government employees are corrupt
. That makes according, to 2010 data, the percentage of corruption employees punished 0.60 % or 0.006 in probability terms.Let us try to estimate a simple model,For a government official to be willing to take the bribe the expected gain from the bribe shouldbe positive. The official retains the bribe with probability ‘1-p’ while he gets caught with probability ‘p’. If he gets caught, he pays fine proportional to his bribe amount. Let for bribe ‘B’the fine is ‘f’ * ‘B’.So the condition for the official to take bribe is: (1-p)B - p(fB) > 0
> f This is the condition for official to take bribe. Now if we put our estimate of p i.e. p=0.006into this equation, we have the condition as 165.66 > f. So to stop the official the amount of fine should be more than 165.66 times the bribe. So for Rs. 10,000 bribe the fine should be Rs.16,65,666.But it should be considered that this result is coming out of the definition of the functional for mof fine which we have considered to be directly proportional to the bribe. The probability of punishment is assumed to be a constant function for all the employees. The fine is calculated in purely monetary terms. In reality fine involve social costs like stigmas associated with being caught; cost of imprisonment with cannot be easily converted into monetary terms. The corruption considered here is only a concerning bribe and not including other forms of corruption. But can be stated that the fines or probability of getting punished or both are much lower than what they should be to deter corruption.
Corruption has social acceptance”, in Mint, 6 Sept, 2011 as viewed on 9 sept,2011
are given. Assuming that trials completion rate and conviction rate for anti-corruption wing to be same as overall rates for CBI, The cases convicted by CBI are calculated.
Table 2: Punishments by CBI
Year 2010 2009 2008 2007Estimated cases punished by courts from Anti-corruption wing of CBI382 348 333 348
(Source : Various CBI annual reports)
Adding the two numbers from CVC and CBI, the total number of central government employees punished is obtained.
Table 3: Total number of punishments in Central Government
Year 2010 2009 2008 2007Total Punishments 12542 14771 12871 8643The total number of central government employees is obtained from Economic Survey of India 2010-11. But the data is available only till 2008. 2009 and 2010 data is obtained by linear extrapolation.Table 4: Number of central government employees Year 2010 2009 2008 2007
Govt Employees (in lakhs) 26.17 26.78 27.39 28.00
(Source: Economic Survey of India 2010-11)
From table 3 we know the number of punishments in the central government and from table 4 the total number of employees in central government is known. The formula used to calculate the probability of a corrupt central government official punished is Probability corrupt central government official punished =
But we do not have an estimate of total number of corrupt officials. To tackle this problem the percentage of corrupt central government officials is assumed. Based on assumption various probabilities can be obtained.Total Number of Corrupt Officials = Assumed percentage of corrupt central government employees * Total number of central government employees.The probabilities can be obtained for 10%, 25%, 50%, 75% and 90% assumed percentage of corrupt central government employees. As the probability numbers are difficult to comprehend the following table shows the numbers in percentage form i.e. we multiply the numbers by 100.According to table if we assume that in central government 50% employees are corrupt then of
State level bodies-
At state level, various states have passed lokayukta act. The functioning of office of Lokayukta is similar to CVC in center. It is an advisory body. Even thought states have not passed the act at same time and some states have taken very long time but currently many states have institution of Lokayukta. Apart from Lokayukta, states have their own anti-corruption departments. The relation between anti-corruption bureau and office of lokayukta is similar to that of one between CBI and CVC in center.After understanding the anti-corruption structure in India, it is important to see their performance. The paper attempts analyze their performance. The paper is divided into three parts. Part one tries to find the probability of getting caught of central government employees.Part two tries to compare performance of different state anti-corruption and vigilance bodies.Part three tries to see the Corruption perception Index for India and trying to compare with available primary information.
1. Percentage of Corrupt Punished:
In the theoretical literature concerning corruption, an important parameter is the probability of acorrupt official getting punished. To calculate this probability the data can be used from annual CVC reports
and annual CBI reports
. From their annual reports, the number of punishments delivered to government officials can be determined. The following table shows the punishments in various departments as per CVC reports.
Table 1: Punishments in Central government departments S. No. Nature of Penalty
2010 2009 2008 2007
Major Penalty 3853 4562 3258 3526
1 Cut in pension 56 121 67 992 Dismissal/Removal/Compulsory retirement 947 1132 819 8273 Reduction to lower scale/rank 1803 2053 1664 18764 Other Major Penalty 1047 1256 978 724
Minor Penalty 8307 9862 9011 4770
5 Minor Penalty other than censure 4550 5510 5252 14086 Censure 3757 4352 3759 3362
Total 12160 14424 12539 8296
(Source: Various CVC annual reports)
Apart from these punishments, CBI takes up cases of corruption. Anti-corruption wing of CBI handles cases related to corruption. The CBI reports mention the number of cases registered with its Anti-Corruption Bureau. The total number of trials completed at courts and conviction rate
One important measure to curb corruption has been effective anti-corruption agencies. Although anti-corruption agencies by themselves are not sufficient to tackle corruption, they form animportant tool in the arsenal of anti-corruption efforts. For understanding the condition of corruption in India it is important to see the current anti-corruption structure in India.On the central government level Chef Vigilance Commission (CVC) and Central Bureau of Investigation (CBI) are two main bodies to fight corruption.
Central Vigilance Commission-
A committee under Chairmanship of Shri K.Shantharam, MP, was formed to review the existing instruments for checking corruption in central government services and to suggest practical steps to make anti-corruption efforts more effective. On recommendation of committee the Government of India, through a resolution, set up The Central Vigilance Commission in 1964. In September 1997, an independent review Committee constituted by Central Government recommended conferring statutory status on the Central Vigilance Commission which was passed, in 2003, as Central Vigilance Commission Act, 2003.As per the 2003 act Commission consists of one Central Vigilance Commissioner as chairman and two Vigilance Commissioners as Members. The Central vigilance commission is assisted bya Secretary and two additional secretaries and other staff. The Commission also uses the services of Directors/Deputy secretaries as Commissioner of Departmental Inquiries. Each department/organization covered under the normal advisory jurisdiction of the Commission has a vigilance unit headed by the Chief Vigilance Officer (CVO). The CVOs act as an extended arm of the Commission.CVC also has a technical wing that provides assistance in formulating the opinion of commission about technical matters
Central Bureau of Investigation-
CBI was initially formed in 1941, at the time of second world war, with mandate to investigate cases of bribery and corruption in transactions with which the War and Supply Department was concerned. Its mandate was extended at many times and Delhi Special Police Establishment Act,1946 DSPE Act) was brought into existence. Even though it was initially established as anti-corruption body it was entrusted with additional responsibilities with passage of time. Currently CBI has three investigation divisions. Anti-Corruption Division, Economic Crimes Division,Special Crimes Division. With enactment of CVC act, 2003, the superintendence of investigations of offences carried out by CBI, under the Prevention of Corruption Act, 1988,vests with the Central Vigilance Commission
Corruption: How many of them are being caught and how different states are performing.
When India is going on a mission against corruption there is a pertinent question which is being ignored,how many of the corrupt are punished? The paper finds this using numbers from Central Vigilance Commission, Central Bureau of Investigation and State anti-corruption bodies. The paper tries to calculate the probability of a corrupt government official being punished (subject to assumption of corruption in the country) which is a very important parameter to see how the country is doing in anti-corruption efforts. The paper also compares different states in India in their anti-corruption efforts using the data from ‘National Crime Records Bureau’ on officials prosecuted under ‘Prevention of corruption act, 1988’.
Corruption, Anti-corruption, Bribery
Corruption has become a buzzword in India. But if we try to see how Corruption is defined there is no unanimity about the definition of corruption in literature. But for practical purposes the accepted definition which is used is ‘Corruption is use of public office for private gain’. This definition limits the scope of corruption and does not consider many aspects of corruption like nepotism and patronage. There can be difference in laws of countries in defining corruption due to cultural and social factors. For example something which is considered as a gift in a country can be considered as bribe in some other country.Although there has been a lot of work done on corruption in 80’s the majority of work has been focused on theoretical front. As the Corruption by nature an illegal activity, hence kept secret,the data on corruption is hard to come by and thus limiting the empirical research. Researchers relay on perception data for lack actual data on corruption. From 90’s there is also large amount of empirical research as well, but the majority of empirical research has been based on cross country analysis. This line of research has provided important insights into the aggregated eterminants and effects of corruption. It has tackled some of the important questions like relation between corruption and growth (Mauro-2000), Inflation and political institutions, but this has some drawbacks. Relying on perception data makes it liable to perception biases. These cross country studies by definition are unable to say anything about within-country variations.One of the important results coming out of this research is the important of Institutions as determinant of corruption .
...and I am Sid Harth