Why India's Bail Scheme For Poor Prisoners Isn't Working

“We [family] make around Rs 10,000-15,000 a month,” said Shishir, who has studied till grade VII and has not found a fulltime job since his release from jail. “The others [other eight accused] went out almost a year ago, but I could not find someone to stand surety and pay Rs 25,000 for my bail.”

He was released on cash bail under the Union government’s 2023 scheme for support to poor prisoners. In her 2023 budget speech, Finance Minister Nirmala Sitharaman announced that the required financial support will be provided for poor prisoners who are unable to afford the bail amount. The scheme provides financial aid of up to Rs 40,000 for undertrials and Rs 25,000 for convicts, based on the approval of the empowered committee chaired by district collectors.

The scheme is intended to solve for a problem: In December 2023, based on data submitted by the High Courts, 24,879 accused who were granted bail by trial courts continue to be in custody due to their inability to furnish bail bonds, said an October 2024 report by the Centre for Research and Planning (CRP), the Supreme Court’s research wing. Allahabad (6,158), Madhya Pradesh (4,190), Bihar (3,345) and Bombay (1,661) high courts identified the most number of undertrials in prison due to their inability to furnish bail bonds, it said.

While the scheme is intended to support such prisoners, it has not been effectively utilised and implemented by states. An October 2024 home ministry advisory said that “many states/UTs are yet to take full advantage of this scheme despite MHA’s [Ministry of Home Affairs] repeated persuasion in this regard”. The report indicated the limited impact of the scheme.

This is further evident in the responses to right to information (RTI) requests filed by IndiaSpend in Delhi and eight states--Maharashtra, Uttar Pradesh, Odisha, Madhya Pradesh, Chhattisgarh, Kerala, Bihar, West Bengal. We sought information on formation of empowered committees at the district level and oversight committees at the state level, their meetings, and the number of prisoners identified for release under the scheme in 2023 and 2024. We filed the first request on August 16 and received the latest response on October 17. As of November 14, we received responses from six of these:

• Maharashtra was the only state which has released prisoners--10 undertrials and one convict--through the scheme.
• Odisha said that it had identified four convicts and three undertrials by July 31, but no one had been released as on the date of the response.
• Delhi identified 103 eligible convicts and undertrials; two jails said they did not release any prisoners, one said it did not maintain such data.
• Three Bihar jails shared data of custodial populations released through legal aid from District Legal Services Authority (DLSAs), but not through the scheme.
• Kerala did not share data on those released under the scheme, but said that the prisons department had opened a zero-balance subsidiary account for the scheme’s implementation
• West Bengal has not implemented the scheme as yet
• Uttar Pradesh, Madhya Pradesh and Chhattisgarh did not share data by the time of publishing.

Experts said that the scheme will not lead to the reduction of prison congestion or have a significant impact in tackling related issues. In any case, they said, the scheme may not be as impactful since existing court orders and processes are adequate for the purpose if implemented properly.

Odisha held 84 district level meetings, did not release prisoners

India is reported to have 573,220 prisoners, based on the latest government data released in 2023. The majority of them--more than three in four--are undertrials like Shishir, and 30% are educated below grade X.

Despite the Undertrial Review Committee’s (UTRC) campaign in July 2022 to identify undertrial prisoners eligible for release, and the country adding 11 prisons since 2021, India’s prisons were filled to 131% of capacity. One of the reasons to formalise the scheme was to reduce this congestion.

The scheme guidelines say that the National Crime Records Bureau (NCRB), as the central nodal agency (CNA), will provide funds. An ‘empowered committee’--to be constituted in each district with the District Collectors and Magistrates (DCs/DMs) as chair--will assess the financial support required in each case and will draw bail money from the CNA account.

Oversight committees at the state level including the Principal Secretary (Home/Jail), Secretary (Law), Secretary-State Legal Services Authority, Director General/Inspector General (Prisons) and Registrar General of the High Court, will decide in cases where the bail amount is more than Rs 25,000 or Rs 40,000, respectively, for convicts and undertrials

Those accused under the Prevention of Corruption Act, Prevention of Money Laundering Act, The Narcotic Drugs and Psychotropic Substances Act, Unlawful Activities Prevention Act or crimes against women are not eligible for the scheme.

The NCRB, which is the nodal agency, said that information was not available with the Central Public Information Officer.

Delhi had constituted 16 empowered committees, one for each of its jails, and an oversight committee, RTI replies show. As noted above, data were received from only three jails.

Odisha constituted empowered committees in all 30 districts and held 84 meetings. Its oversight committee had met once. The state has also identified one undertrial prisoner whose bail amount is more than Rs 40,000, but the case was rejected by the oversight committee. Two other convicts were identified whose bail amount was more than Rs 25,000, and their cases were under consideration as on July 31.

Kerala had constituted both district- and state-level committees but did not share information on meetings. In October, it was reported that the Kerala High Court had asked the state government and Director General for Prisons to furnish information on the current status of the zero balance subsidiary account opened under the scheme including the amount presently available in the account. The RTI said that a zero balance account had been opened and NCRB had been requested to map it based on scheme guidelines.

Maharashtra did not share any information on meetings held, but the state had demanded and utilised Rs 2.09 lakh in 2024-25. It released 11 prisoners. West Bengal had not implemented it as on September 26, 2024.

In Bihar, only Bennipetti jail in Madhubani district said that it had constituted committees in the state, although it did not specify if it was at the state or district level.

The RTI data shared by Betia, Muzaffarpur and Samastipur jails were related to prisoners released through legal aid support, and not specifically identified for the scheme. The jail did not share any other data.

“The SOP and guidelines for the support for poor prisoners have only been shared recently with Bihar jails,” said Praveen Kumar, founder and director of the Patna-based Law Foundation. The guidelines and SOP were signed on October 14 by the additional secretary and director (administration).

IndiaSpend has asked the MHA for comments on the scheme and its implementation. We will update this story when we receive a response.

Scheme being housed in collectors’ office makes it a non-starter

Vijay Raghavan, professor, Centre for Criminology and Justice at the Tata Institute of Social Sciences (TISS), Mumbai and project director of TISS’s Prayas, which works on social work intervention in criminal justice, was part of the committee constituted by the home ministry to finalise the scheme.

Section 436 of the erstwhile Criminal Procedure Code (now replaced by Section 478/479 of the Bharatiya Nagarik Suraksha Sanhita) said that if person is in jail unable to give bail even after a week of his arrest has passed, that person should be considered indigent and be released on personal bond, said Raghavan. “If this were implemented sincerely, the cash bail scheme may not be required. The discretion is with the presiding magistrate or judge.”

In January 2023, in the Sonadhar case, the Supreme Court order noted that, based on data furnished by the amicus curiae, there were 5,000 undertrials in India who were in jail despite being granted bail, of which less than half were provided legal assistance and less than one third had been released.

The Supreme Court in its order also issued a seven-point guideline for releasing prisoners. Points six and seven recommended that bail conditions may be modified or relaxed if bail bonds are not furnished within one month from the date of grant of bail, and the local courts may not impose local surety because it leads to delay in release of prisoners on bail.

Bail conditions for an accused can include the option for the accused and their sureties to enter into a bond of a certain amount so fixed by the judge, which may be forfeited upon violation of the conditions, said the CRP report.

The issue of local surety is a significant problem, particularly for prisoners who are not from the local area or jurisdiction. IndiaSpend had reported how poverty and lack of local community ties affect migrants under trial.

In the 1978 case of Moti Ram v. State of Madhya Pradesh, the Supreme Court had asked magistrates to consider fair and affordable bail conditions, and release on personal bonds.

There is a tendency for courts to “insist on monetary security with surety where only release on bail is provided for under the Code, without using the option of releasing an indigent accused on personal bond”, IndiaSpend had reported in March 2023.

When the scheme is housed in the DC’s office, “it becomes a non-starter”, said Raghavan. “It is low on the priority of the DC's agenda. The DLSA should lead this process. But the Union government was concerned about releasing the funds from CNA which requires DC's involvement.”

“The scheme does not complement the legal provisions or the bail jurisprudence which calls for liberal bail policies,” said Madhurima Dhanuka, a criminal justice reform expert. “... the prudent step is to ensure that such circumstances where financial support [is needed] for securing bail do not arise in the first place.”

The UTRC, which was established based on a 2015 Supreme Court order, has district level committees headed by the District & Sessions Judge, with the District Magistrate, Superintendent of Police and Secretary, DLSA as members. The committee reviews cases of prisoners and ensures they are not subjected to prolonged detention. It is similar to the role of committees in the 2023 financial support for poor prisoners.

Social workers needed for due diligence

TISS’s Prayas has been working with custodial populations in different prisons in Maharashtra for years, and has social work and legal fellows who since February of this year have been following up on cases of prisoners who can benefit from the scheme.

According to Raghavan, given the resource and time constraints faced by jail and legal authorities, it is important that the scheme provides for the presence of social workers to identify those prisoners who really need the support, based on their family financial and social circumstances.

“I had suggested during [MHA] Committee meetings that social workers should be attached to the DLSA or the prison on a pilot basis, to carry out due diligence of cases before they are considered for release,” Raghavan said. “Usually the prisoner's family is poor, or the family does not know the process, or they live in a far off location. My suggestion to appoint social workers was appreciated by the committee members, but not taken on board fully.”

The final guideline did not make it mandatory to include a social worker or conduct a pilot. It said that the DLSA “may take” the assistance of civil society representatives, social workers/ NGOs, district probation officers or revenue officers, and that reviews should be completed within a period of 10 days.

Paralegal volunteers in jail, usually prisoners, identify and share information about the case status of a prisoner with jail authorities, who share a list with the DLSA. Based on those who are eligible under the guidelines, a due diligence is done by social work fellows of Prayas who visit homes and families of the prisoners. Presently the work is mostly limited to Thane, Taloja and Alibaug jails. Shishir was identified in February 2024.

While more eligible prisoners had been identified initially, some of them managed to pay bail and furnish surety. Finally, due diligence was done for those prisoners who were finalised by the DLSA before it was sent to the DC, said Saugata Hazra, Prayas’s legal fellow placed at Thane’s DLSA. “There were two meetings (April and August),” he said. “Most of the prisoners were in their 20s and the families were poor or did not want to take up the responsibility.”

Pravin Patil, Prayas’s social work fellow placed in Thane jail, who did the due diligence in Shishir’s case has been part of the team for 10 months. During due diligence, the social workers access copies of documents like Aadhaar or rent which are available with the court. The final list is approved and sent to DC, where it is signed off after a review meeting.

“Often, if the person is a habitual offender, the family does not want to take the responsibility. We inform them [and the prisoner] that after release they cannot abscond or miss court hearings,” said Patil.

Because of the requirement that the DC clear the final list, there are delays in releasing the prisoners whose due diligence has been approved by the DLSA. In Shishir’s case, he was identified in February but released only in mid-October.

While Prayas’s existing work and familiarity with the legal system in Maharashtra fit into the scheme’s requirement for due diligence, other states may not necessarily be able to replicate the process.

Praveen Kumar of Law Foundation in Patna said that the process is bureaucratic because it makes the DCs chair and they have the final say, and not the DLSAs. “It will require involvement of civil society organisations to invest time and resources.”

The organisation helps release prisoners through personal recognizance bond or PR bonds, where the prisoner themselves undertake that they would appear for court hearing. Praveen Kumar said that in Bihar, judges ask for surety which means someone has to stand as a guarantor. This means that there is additional effort needed to place a modification application in court to convert surety to cash bail, if the prisoner is to benefit from the scheme.

“I think the impact of the support for poor prisoners scheme is going to be limited because in Bihar cash bail is not prominent unlike in some of the metropolitan areas,” Kumar said.

Scheme won’t reduce prison congestion

One of the claims of the scheme is that it will reduce overcrowding of prisons, but experts do not believe this will happen. The scheme must be revised to align with the ground realities of India’s criminal justice system and with the UTRC that already exists, they said.

Valay Singh, lead, India Justice Report, feels that the scheme is “extremely imperfect” for ensuring decongestion of prisons or reducing the number of undertrials languishing in prisons despite bail. The focus must shift to unnecessary arrests and detention, and to the rejection of bail despite multiple Supreme Court orders.

“...there is already a mechanism of the UTRC whose mandate is to identify prisoners who can be released,” said Singh. “It is not clear why it has been left out of this scheme.”

Of the 73,635 undertrial prisoners recommended by UTRC for release, 48% were released in 2023, as were 352 convicts of the 2,188 recommended.

Dhanuka also does not agree that the scheme has any role to play in reducing prison overcrowding. “Lack of construction of prisons in accordance with increase in prison population, and the increasing duration of trials are in my view the major cause of overcrowding, which is not talked about enough,” she said.

Although courts seem concerned about accused absconding during court hearings, she said that courts must rely on PR bonds given improved technology and possibility of tracking released people, ensuring privacy-related legal requirements are met.

Amritha Moorthy, a research intern at IndiaSpend, contributed to the story.

We welcome feedback. Please write to respond@indiaspend.org. We reserve the right to edit responses for language and grammar.