## Wednesday, May 18, 2016

### Understanding judicial delays in India: Evidence from Debt Recovery Tribunals

by Prasanth Regy, Shubho Roy and Renuka Sane

The Insolvency and Bankruptcy Code passed by the Lok Sabha last week envisages Debt Recovery Tribunals (DRTs) as the adjudicating authority for individuals and partnership firms. When originally set up, DRTs were expected to resolve cases within a limit of 180 days. But experience tells us that judicial delay is as much of a problem in the DRTs as with other courts.

There is a great uproar about judicial delays in India today. We are now a country where it is widely felt that we know how to run elections but we don't know how to run courts. A commonly touted solution is to hire more judges. However, if the institutional arrangements in which judges operate is faulty, adding more judges is not likely to help.

In fact, little is known about why delays occur. The present approach of collecting information about judicial delays concentrates on a few macro-level statistics of pendency and disposal rates. There is very limited information or research that uses micro data about the time taken for judicial proceedings, and the reasons for the delay. In this article, we bring a novel research strategy to bear on understanding the issue of judicial delays. While this strategy is general, we apply it to Debt Recovery Tribunals (DRTs), and present our findings.

### 1  Debt recovery tribunals

Debt Recovery Tribunals (DRTs) are statutory bodies established under the Recovery of Debts Due to Banks and Financial Institutions Act, 1993. They were created because the existing mechanisms of debt recovery were ineffective. DRTs were expected to enable the "expeditious adjudication and recovery of debts" within one hundred and eighty days of filing the case. Subsequently, DRTs have also been tasked with enforcing some provisions of the Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Act (SARFAESI), and of the Insolvency and Bankruptcy Code, 2015.

DRTs have not been successful in realising the objective of efficient disposal of debt-related disputes. We knew, at the time of the creation of DRTs, that the previous debt-resolution mechanisms were slow and ineffective. The DRTs were meant to be an antidote to this problem. But now, more than 20 years after DRTs were set up, we know (Committee on Financial Reforms, 2009) that the DRTs have not been any faster than other courts in resolving issues.

### 2  Questions

If DRTs suffer from long delays, they will not be able to fulfil the new Insolvency and Bankruptcy Code-related responsibilities placed on them. So how can DRTs work better and faster? An answer to this question depends upon the answers to many other questions: Who causes delays at DRTs? What is the immediate cause of these delays, and what is their root cause? What is the role of the tribunal in enabling delays? And importantly, what are the incentives of the different stakeholders that lead to delays? Answering these questions is key to designing and implementing reforms to make our courts work better.

### 3  A dataset about the working of debt recovery tribunals

We hand-constructed micro data for a sample of cases disposed by the Debt Recovery Tribunal (DRT) in Delhi. We obtained complete information for 22 cases disposed between February and April 2014. The earliest of these cases had been instituted in 2006, and the last in 2013. We captured information about who filed the case, against whom, and the key issue at stake. At the end of each hearing, an order is passed. We analysed each hearing and each order. This yielded observations of 474 orders over these 22 cases, an average of 21.5 orders per case. We captured data about the date of the order, the content of the order and the date of the next order. If there was an adjournment, we noted who asked for it and why.

The data files have been released here.

### 4  How large are the delays

When a hearing resulted in an avoidable and unpenalised adjournment, we classified it as a failure. Most often, these adjournments were requested by one of the parties, but sometimes they are caused by the tribunal itself. On average, each failure adds about a month of delay to the case.

The cases we examined went on for about 2.7 years on average. This conceals a lot of variance — the duration varied from as few as 5 months to as many as seven and a half years. See Figure 1 below:

 Figure 1: Histogram of the case duration

How much of this was necessary? In other words, if the system had functioned well, how much delay could have been avoided? It turns out that of these 474 hearings, 279 hearings (about 60%) were failures. These failures induced half the delay.

Can we envision a world where trial failures are eliminated? At first blush, it seems that we could reduce the duration of the average case by roughly half if we are able to eliminate trial failures. The full gain is, however, larger than this. If there were fewer trial failures, cases would finish sooner, freeing up slots on the judicial dockets. This would let other cases have more frequent hearings. So the delay would drop by more than half if these failures were avoided.

### 5  Who causes delays: borrower or lender?

At DRTs, there are broadly two kinds of cases: those filed by the lender, and those filed by the borrower. The cases filed by borrowers typically ask the tribunal to stop action taken by the lenders under SARFAESI. In these cases, one would expect that the borrowers would have an incentive to delay: they already have use of the money, and they would like to repay as late as possible. This is indeed what's seen in the data. For the cases filed by borrowers, adjournments due to the borrower account for 46% of the total time lost, while about 21% is lost due to the lender, and about 17% due to the tribunal (see Figure 2).

 Figure 2: Who delayed?

In the cases filed by the lender, one would expect that the lender-petitioner would want a quick disposal of the case. After all, it has lent the money, and would presumably like it back as soon as possible. Besides, these lenders are sophisticated institutions that maintain proper documentation and have expensive lawyers in their service.

Remarkably enough, in these cases, the petitioner is the major cause of delays. The lender-petitioner accounts for 37% of the delay, compared to 20% caused by the defendant and 26% by the tribunal. Many of these delays are because the lender asks for adjournments while it locates and files documents. This clearly violates our expectation that the lender would want his money back quickly. Given that they have good documentation and good quality legal counsel, it is not clear why they would take so long to perform tasks that lie well within their control.

### 6  Looking deeper into the delays

This raises several questions. Why do lenders file cases in courts, and then ask for so many adjournments? Is it possible that their objective in filing the case is not to obtain a judicial mandate in their favour, but to exert pressure on the borrower to come to a negotiated settlement? The large number of cases that seem to be settled outside the court-room (about half of the cases we studied) would seem to indicate so.

It is likely that the cost of prolonging a legal fight is lower for a financial institution than for most borrowers. Financial institutions have legal departments and lawyers on retainers, while for most small borrowers, the legal system is a source of anxiety and expense. Perhaps banks delay, in the expectation that the borrower would wilt under the pressure and would be willing to come to a negotiated agreement.

But if we look at Figure 2, about 70% of the delays are due to the lawyers. The large number of times the lawyers (on both sides) ask for more time to file documents, as well as the many instances where one or the other (often both) lawyers are absent, or both lawyers request adjournments, suggests another possibility: that the lawyers are in no hurry to finish the case. If the incentives of the lawyers are perverse — for instance, if they get paid not on the basis of prompt resolution of the case in favour of their client, but on the basis of the number of hearings — then it is reasonable to expect that they would prefer to have more hearings.

An important weak link is the behaviour of the court. Those asking for repeated adjournments are imposing a cost on the judiciary. Judicial time and capacity are scarce public resources, and repeated delays are a waste of these precious resources. In spite of this, we have seen very little evidence of the court imposing penalties on the parties for causing delays. There is emotion about hiring more judges, but not about cracking down on delaying tactics.

Another facet of the data is how often the tribunal itself causes delays. Figure 2 shows that more than a quarter of the time lost is attributable to the court. The reasons are many: the registrar might be on leave, the judge may be attending a conference, or the bar association might have requested a holiday. Most other institutions find ways to ensure that work is not held up due to such reasons. As an example, a railway station or a stock exchange or a bank branch works all the time.

### 7  Conclusions

To summarise, our study has shown that more than half the time of a case is lost in avoidable and unpenalised adjournments. The parties to the case, the lawyers, and the tribunals, all participate in this delay. The study indicates that lenders may use court delays as a strategy to pressure the borrower to come to a negotiated agreement. Lawyers may have perverse incentives to draw out cases, and tribunals often contribute to delay themselves due to administrative reasons.

These results suggest two clear directions for reform:

1. Laws must provide incentives for litigants, lawyers, and judges to reduce litigation time. For an example, see here on how smart drafting of rules can create incentives to quickly resolve disputes.
2. Poor administrative processes also contribute to delays. Judicial time gets wasted because the administrative functions of a case (like serving notices) have not been completed. The solution to this is better administration, through investment in the court infrastructure, as well as through the separation of administrative and judicial functions of the tribunal.

Such reform should be grounded in a sophisticated understanding of the value (to a protagonist) and the cost (to society) of judicial delays, and it should be guided by the principles of law, economics, and public administration. The work we have presented above is a step in this direction. Future work will aim to derive rigorous inferences about the causes of delay, the incentives of the various stakeholders, and how these incentives could be modified.

### References

Committee on Financial Sector Reforms. A Hundred Small Steps: Report of the Committee on Financial Sector Reforms. Planning Commission, Government of India, 2009.

Martin A. Levin. Delay in Five Criminal Courts. The Journal of Legal Studies, Vol. 4, No. 1, pp. 83-131, January 1975.

Richard A. Posner. An Economic Approach to Legal Procedure and Judicial Administration. Journal of Legal Studies, 399--458, 1973.

Shubho Roy. Reducing delays in litigation by reshaping the incentives of litigants. Ajay Shah's Blog, January 8, 2016.

Pratik Datta and Ajay Shah. How to make courts work? Ajay Shah's Blog, February 22, 2015.

Pratik Datta. Transforming the operational efficiency of tribunals and courts. Ajay Shah's Blog, February 8, 2016.

Prasanth Regy and Shubho Roy are researchers at the National Institute of Public Finance and Policy. Renuka Sane is an academic at the Indian Statistical Institute. We thank Anirudh Burman, Pratik Datta, Kushagra Priyadarshi and Sanhita Sapatnekar for their participation and contributions in the early stages of this research project.

LaTeX mathematics works. This means that if you want to say $10 you have to say \$10.