State & Pace: Criminal Justice in India

Turns out that not just crash courses happen in four weeks, but policies too are made in India in four weeks. Some of them with policy goal as big as fast tracking of justice system in the country.  The Supreme Court’s or Chief Justice of India’s words wouldn’t have probably made a visible impact in a pre-NDA government regime. But these are major affairs now, for these hold the same promise of “good governance” and “efficiency” that BJP government has been promising the country.

High speed winds blow through the justice system this year. Besides the debate on the need and constitution of Judicial Appointment Commission, two other events last week are worth noting. First, Supreme Court’s instruction to the Center to formulate a policy within four weeks to speed up trial in criminal cases, adding that it is not a good sign of democracy and good governance. Second, the Chief Justice of India’s Independence Day address and his comment –

A curious and tragic paradox is that our prisons house more undertrial prisoners than convicts. In almost all central prisons, more than 50 per cent are undertrial prisoners; in district prisons more than 72 per cent are undertrial prisoners. The process itself has become a punishment. As head of judiciary, I cannot feel more pain than that.

– R.M. Lodha,  Chief Justice of India, Independence Day Address

In earlier posts – Indian prisons at a glance & prison reforms in India: a quick history, the issues affecting the system were mentioned. With the recent speech of Chief Justice and the Supreme Court instruction to the center we see current state of these issues and their gravity. The Chief Justice further added –

I was shocked to learn that two lakh criminal trials are pending all over the country which are more than five years old and 40,000 trials are pending which are over 10 years old. In the Supreme Court by the end of the year, over 65,000 cases are pending. These are peanuts compared to what we are seeing all over the country.

In an earlier project, we looked at the numbers for prisoners and crime rate from the NCRB database for the period 2001 – 2010. This post is to plug in some graphs which support the comments by the Chief Justice and also give a sense of the magnitude of the problem.  The other intent of the project was to understand and articulate the kind of criminal justice system that India exhibits, with a theoretical reference to Packer’s models of criminal justice. . Herbert Packer proposes that a criminal process is a function of a value system. This value system characterizes the kind of criminal process that a society has – it can be to control crime (repression of criminal conduct) or it could be oriented towards following a due process which is geared towards fact finding.[1] These two models – crime control and due process, are examined in case of India using empirical evidence. Each of these models has a set of characteristics. I examine if the prison and crime statistics support the characteristics of either of the model.


Graph 1: Prison capacity and prison strength in India during 2001-2010
Graph 1: Prison capacity and prison strength in India during 2001-2010
Graph 2: This graph shows how prison occupancy has changed over the decade (2001-2010) vs. the percentage change in prisoner strength in the same period.

When seen together graph 1 & 2 show a broader picture of how prisoner strength and occupancy has moved over a ten year period. This appears to be a shorter time frame to analyze. However, this is the best dataset available from a government agency (NCRB) for those interested in prisons and criminal justice to look at. It is interesting to note that from 2005 prison strength has shown negative change, which appears to be in line with Chief Justice’s observation on very low conviction rate. The dipping blue trend line in graph 2 could either be that more prisoners were released during the period 2005-2010 versus those who were sent to jail. Or also in a derived way, that conviction rate has shown a drop in the same period. More datapoints would be needed to say the latter.

Graph 3 below shows the percentage of convicted prisoners and undertrials as a part of the total prison population. The second quote above gets an empirical ground with this one.


Graph 3: Percentage of undertrials and convicted prisoners in India during the period 2001-2010.

Next, we move to crime data and look at the cases listed for investigation vs. cases investigated. NCRB records this data for all the states. The graphs here are national figures. Graph 4 shows the cases listed for investigation vs cases investigated. This is a rough measure of police capacity and efficiency. The cases taken up for investigation from the listed is perhaps the only impressive figure in the analysis. It reaches upto 80%.

Graph 4: Cases listed for investigation vs. cases investigated. All India figures from 2001-2010

Then we move on to see that with such a high percentage of undertrials in the prisons across the country, how many of them have access to legal aid. This in our understanding appeared to be one of the major reasons for so many people languishing in prisons without a conviction. Also, this points to the idea of an “accessible justice system” and is a broader measure of that access. 2010 is the year when over 20% of undertrials get access to legal aid. It doubled from 10% to over 20% within a decade. This also builds a ground for more lawyering for the poor and the vulnerable in the country.

Graph 5: Undertrial prisoners with access to legal aid between the period 2001-2010.

In countries with an even higher prisoner population and high rates of incarceration there is a very serious debate going on about the cost of keeping a person in prison. This has also been effecting the sentencing given by judges, where they have often been found to be carefully thinking about the cost of keeping a person in prison if sentenced for imprisonment vs. giving him a sentence which gives a non-prison punishment. We were curious to see how much does it cost in India to keep a person in prison. Graph 5 shows the cost per prisoner per year incurred by the government of India. Approximately Rs. 19,500 was spent per year on each prisoner. Looking at the poverty line consideration of Rs 32/day (which was criticized widely to be too low an estimate; makes Rs 11,680 per year) a prisoner does way too well. And if we consider a dollar a day income i.e USD 1/day (makes Rs 21,900 per year) then too the prisoners aren’t doing too bad.

Graph 5: Cost incurred per prisoner per year between 2001-2010

Let us now look at the conviction rate for IPC crimes for the same period. Graph 6 shows the same. Since the number of undertrials are high and conviction rate very low at about 36% in 2010 it appears that the to be characteristic of a due process model. Due process model relies heavily upon investigations and prosecution. Empirical evidence on investigations indicate that as soon as cases are registered they are followed up. There is little lag in the cases registered and cases investigated in a year. It is difficult to enforce either of the model for India though the system shows a lot of characteristics similar to the due process model.

Graph 6: Conviction rate for IPC crimes between year 2001-2010.


Finally, where does this analysis leave us? If prison system is to be reformed, then it should be examined from the view of procedural and rule based bottlenecks in the system The analysis here demonstrates that overcrowding of prisons is made into a key problem and point of action when reforms are attempted. Whereas, in fact, data suggests that prison occupancy is only 123% which does not appear to be a serious problem of space. If the movement of prisoners into and out of the jails is smoothened out then we can hope to reduce the burden on the prison system. Also, the reason for computing the expenditure on a prisoner in a year is to offer an estimate of what it costs versus other options of sentencing a convict. With this figure, sentencing must be seen in relation with the other likely alternatives that are cheaper and also make effective correction strategy.


[1]Two Models of the Criminal Process, 113 U. Penn. L. Rev. 1 (1964). Herbert L. Packer



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.