06th April, 2026 Report of Study on Micro, Small, and Medium Enterprises in the Insolvency and Bankruptcy Code, 2016 (1.68 MB)

MSMEs IN THE IBC March 2026 Rohit Prasad Professor, Economics & Public Policy

S. Veena Iyer Associate Professor, Accounting & Finance MANAGEMENT DEVELOPMENT INSTITUTE GURGAON GURGAON Management Development Institute

2

Contents ACKNOWLEDGEMENTS ................................................................................................................. 5 EXECUTIVE SUMMARY .................................................................................................................. 6 CHAPTER 1: INTRODUCTION ........................................................................................................ 8 1. SCOPE OF STUDY AND OBJECTIVES ............................................................................................. 9 2. DATA AND METHODOLOGY ...................................................................................................... 10 2.1. Data for the pre-admission phase analysis ....................................................................... 11 2.2. Data for the post-admission phase analysis ...................................................................... 14 3. FINDINGS OF THE STUDY ........................................................................................................... 14 4. LIMITATIONS AND ORGANIZATION OF THE STUDY .................................................................... 15 CHAPTER 2: THE IBC PROCESS FOR RECOVERY AND RESOLUTION ........................... 17 1. INTRODUCTION .......................................................................................................................... 17 2. RECOVERY AND REDRESSAL MECHANISMS FOR MSMES ......................................................... 17 2.1. The TReDS (Trade Receivables Discounting System) ...................................................... 17 2.2. MSME Samadhaan portal and MSME Facilitation Council ............................................ 18 3. THE IBC ROUTE ......................................................................................................................... 19 3.1. Filing of Petitions.............................................................................................................. 19 3.2. Withdrawals, With and Without Settlement ...................................................................... 20 3.3. Other disposals pre-admission: Rejection and Dismissal of Petitions, or Reserved Orders

20 3.4. Withdrawal after Admission into CIRP............................................................................. 21 3.5. Mainstream CIRP Process resulting in Resolution or Liquidation .................................. 22 4. HOW THE PROCESS HAS FARED ................................................................................................. 22 4.1. Efficiency and Recovery Rates .......................................................................................... 25 5. IBC FOR MSMES ...................................................................................................................... 26 5.1. MSMEs as Creditors ......................................................................................................... 26 5.2. MSME as Corporate Debtors ........................................................................................... 31 6. CONCLUSION ............................................................................................................................. 33 CHAPTER 3: SETTLEMENT IN THE IBC ................................................................................... 36 1. INTRODUCTION .......................................................................................................................... 36 2. THE OPTIONS FOR SETTLEMENT ............................................................................................... 36 2.1. Section 12A ....................................................................................................................... 36 2.2. Regulation 30A, within Section 12A.................................................................................. 37 3. WHY CDS WANT TO SETTLE .................................................................................................... 38 4. A GAME THEORETIC VIEW ....................................................................................................... 39 5. THE IMPERATIVE OF FREQUENT SETTLEMENT FOR THE RATIONAL MSME OC ...................... 40 5.1. Allowing for Risk-Aversion ............................................................................................... 43 5.2. The Role of Regulation 30A .............................................................................................. 44 6. WHEN WOULD THE IBC BE TOO SETTLEMENT-FRIENDLY? ...................................................... 46 7. THE NATURE OF ENTITIES THAT GO IN FOR A SETTLEMENT ...................................................... 47 8. SETTLEMENT IN BANKRUPTCY: VALID OBJECTIVE OR SUBVERSION? ..................................... 56 8.1. International Perspective .................................................................................................. 56

3

CONCLUSION ............................................................................................................................. 57 CHAPTER 4: THE QUASI-ABSOLUTE PRIORITY RULE (QUASI-APR) WITH MSME ORIENTATION .................................................................................................................................. 60 1. QUASI-APR RULE WITH MSME ORIENTATION: HOW IT WORKS ............................................ 60 1.1. Rationale 1: Efficient Resolution ...................................................................................... 61 1.2. Rationale 2: Robust Auction Design ................................................................................. 62 1.3. Rationale 3: Avoidance of Voting Anomalies ................................................................... 64 1.4. Rationale 4: Reduction of Delays ..................................................................................... 65 2. REDESIGN OF IBC AUCTION ..................................................................................................... 65 3. COMPARING QUASI-APR WITH ACTUAL IBC PAYOUTS .......................................................... 66 4. CONCLUSION ............................................................................................................................. 66 5. APPENDIX 1: CONSEQUENCES OF THE CURRENT AUCTION FORMAT- A THEORETICAL STUDY

67 5.1. Theoretical Model ............................................................................................................. 67 5.2. Optimal Bidding Strategy .................................................................................................. 68 5.3. Impact of number of bidders and OC claim share on the allocation ratio ....................... 68 5.4. Impact of the number of bidders on payouts ..................................................................... 69 5.5. Impact of OC claim share on payouts ............................................................................... 70 6. APPENDIX 2: CONSEQUENCES OF THE CURRENT AUCTION FORMAT- AN EMPIRICAL VALIDATION ...................................................................................................................................... 70 6.1. Data, sources and constructs ............................................................................................ 70 6.2. Empirical tests and results ................................................................................................ 71 7. APPENDIX 3: IS THE QUASI-APR ALREADY IN USE- AN EMPIRICAL TEST............................... 74 CHAPTER 5: PRE-PACK INSOLVENCY AND RESOLUTION PACKAGE (PPIRP) ............ 76 1. INTRODUCTION .......................................................................................................................... 76 2. ANALYSIS OF PPIRP CASES ...................................................................................................... 77 3. LOW UTILIZATION OF PPIRP ..................................................................................................... 78 3.1. Do alternative avenues of government credit support and guarantees make the PPIRP a less preferred option for MSME CDs? ......................................................................................... 80 4. CONCLUSION ............................................................................................................................. 81 CHAPTER 6: POLICY RECOMMENDATIONS .......................................................................... 83 1. STRENGTHENING EVIDENCE ON PRE-ADMISSION OUTCOMES ................................................. 83 2. MANDATING CLAIM RECORDS IN NESL ................................................................................... 83 3. FACILITATING PETITIONS FOR MSME OCS .............................................................................. 84 4. REFORMING THE RESOLUTION AUCTION THROUGH QUASI-APR WITH AN MSME ORIENTATION .................................................................................................................................... 84 5. SUGGESTIONS FOR MORE EFFECTIVE PPIRP ............................................................................. 85 BIBLIOGRAPHY ............................................................................................................................... 87 GLOSSARY......................................................................................................................................... 89

4 List of Figures Figure 2.4.1: Initiation and exit of IBC cases filed with the NCLT ...................................................... 23 Figure 2.4.2: OC Initiated Cases: Pre- and post-admission outcomes .................................................. 24 Figure 2.4.3: FC Initiated Cases: Pre- and post-admission outcomes ................................................... 25 Figure 3.2.1: Outcomes of cases petitioned under the IBC ................................................................... 38 Figure 4.5.1: Impact of number of bidders on the allocation ratio........................................................ 69 Figure 4.5.2: Impact of OC claim share on the allocation ratio ............................................................ 69 List of Tables Table 1.2.1. MSME definitions as amended over the years .................................................................. 11 Table 1.2.2. Constructs and variables used for analysis ....................................................................... 12 Table 1.2.3. Data sourced and used for the study ................................................................................. 13 Table 2.5.1.1. Distribution of OCs in our sample ................................................................................. 28 Table 2.5.1.2. Solvency matrix for OC-applicants ............................................................................... 29 Table 2.5.1.3. DSO-DPO matrices for OC-applicants .......................................................................... 29 Table 2.5.1.4. DSO-DPO matrices for MSME OC-applicants with 45-day cut-off ............................. 30 Table 2.5.2.1(a). Solvency matrices of CDs petitioned by the FCs versus the OCs ............................. 31 Table 2.5.2.1(b). ANOVA for solvency rating across OC-initiated and FC-initiated CDs separated by MSME status ......................................................................................................................................... 32 Table 3.5.1: The Business Case for Filing if Admission into CIRP is Inevitable ................................. 41 Table 3.5.2. Pairwise Pre-Admission Settlement Parameters Needed to Break Even (Probability vs. Recovery), by Risk Preference and Reg. 30A ...................................................................................... 45 Table 3.7.1a. Solvency matrices for OC-applicants – pre-admission settled vs. others ....................... 48 Table 3.7.1b. ANOVA across OC-applicants settling under Reg.30A and other OC-applicants ......... 48 Table 3.7.2. DSO-DPO matrices for OC-applicants settled under Reg.30A versus the rest ................ 49 Table 3.7.3a. Solvency matrix for CDs that settled versus those that went through the full CIRP in OC versus FC-initiated cases ....................................................................................................................... 50 Table 3.7.3b. ANOVA for financial health of CDs that settled versus those that went through the full CIRP ...................................................................................................................................................... 51 Table 3.7.4a. Financial health of CDs, settled u/s 12A and Regulation 30A ........................................ 51 Table 3.7.4b. ANOVA of financial health between Reg.30A and Section 12A settled CDs ............... 52 Table 3.7.4c: DSO-DPO analysis comparing large CDs that settled versus those that went into full CIRP .............................................................................................................................................................. 53 Table 3.7.4d. DSO-DPO analysis comparing MSME CDs that settled versus those that went into full CIRP ...................................................................................................................................................... 54 Table 3.7.5. Pairwise correlation between Solvency and Liquidity scores across sub-sets of CDs ...... 55 Table 4.6.1. Distribution of Cases by OC% share (Amounts in Rs Billion) ......................................... 71 Table 4.6.2. Paired t-test comparison for OCRRActual and OCRRProp .................................................... 72 Table 4.6.3. Regression of Log (OC Recovery Rate/Overall Recovery Rate) on Log (OC claim share) .............................................................................................................................................................. 74 Table 4.7.1. Paired t-test comparison for (OCRRActual – OCRRAprop) ................................................... 74

5 ACKNOWLEDGEMENTS We acknowledge the constant support provided by the Insolvency & Bankruptcy Board of India in conducting this study. We thank Shri Ravi Mital, Chairperson, IBBI for entrusting us with this mandate and providing support. We are grateful to Mr. Jayanti Prasad, Whole time member, IBBI; Mr. Bhushan Kumar Sinha, Whole time member, IBBI; Mr. Jitesh John, Executive Director, IBBI; Mr. Ravinder Maini, Executive Director, IBBI and Mr. Shiv Anant Shanker, Chief General Manager, IBBI for their guidance and insights. We thank Ms. Namisha Singh, Assistant General Manager and Mr. Aniket Sharma, Manager for their patience in answering our questions and making data available to us. We thank our experts across organizations for their valuable insights, and participants of the Focus Group Discussion for their inputs. We thank Ms. Geetanjali Vijayran and Ms. Lipika Dasgupta for their excellent research assistance. Any errors or omissions are our own.

6 Executive Summary This report studies how the Insolvency and Bankruptcy Code (IBC), 2016 has played out for the Micro, Small and Medium Enterprises (MSMEs) since its inception. MSMEs are particularly vulnerable to liquidity stress, which can quickly cascade to insolvency. This report examines MSMEs in their capacity as Operational Creditors (OCs) as well as Corporate Debtors (CDs) in cases filed under the IBC. We have supplemented the post-admission case-level data provided by the Insolvency and Bankruptcy Board of India (IBBI) with financial data sourced from the Prowessdx and ProwessIQ databases provided by the Centre of Monitoring Indian Economy (CMIE), and qualitative information from expert interviews and focus groups of relevant stakeholders. We assess and determine firm characteristics using scores for solvency and liquidity using frequency matrices, correlations, and significance tests using ANOVA and paired t-tests. We also use theoretical and MS-Excel modelling to understand possible behaviour, which are tested using regression analysis on empirical data. We find that the pre-admission stage of the IBC has significance for both creditors – more so for the OCs – and debtors, particularly MSMEs. Petitions filed at this stage signal creditors’ intention to recover overdue payments, triggering a response from the counterparty (CD), whose response in turn signals not only their intent but also their ability to settle. The dynamics at this stage, most importantly, signal the OCs’ and CDs’ financial health to each other and the larger ecosystem they operate in. Our analyses of post-admission data provided by the IBBI confirm that firms that settle (pre- or post-admission) are financially healthier than those that enter the full CIRP; those that settle earlier are healthier. This applies to both OCs and CDs. With close to 80% of the petitions getting disposed of pre-admission, these dynamics are important to understand and take cognizance of for the IBC and the MSME ecosystem to be more meaningful and effective for MSMEs.
In this spirit, we recommend more detailed data capture at the pre-admission stage to better understand the behaviour of key players; while most creditors may be looking to settle and recover their dues, we aver that the IBC-as-a-recovery mechanism is not in conflict with the IBC-as-a-resolution mechanism. Indeed, the objective of the IBC is not resolution for the sake of resolution. Instead, it is the optimisation of firm value either through resolution, liquidation, or settlement. We discover a difference in CDs’ behaviour towards OC dues versus FC dues. This, coupled with FCs’ priorities and regulations that govern them, results in FCs petitioning CDs much later than an average OC applicant. We see a significant difference in financial health between CDs petitioned by FCs versus those petitioned by the OCs, the latter having better solvency and liquidity scores, which we believe, represents the value erosion caused by the delay in petitioning by the FCs. This result also

7 holds for MSME CDs. Strengthening the settlement mechanism will help conserve the asset value of these CDs and significantly improve payoffs for all parties involved. Speeding up the process can prevent MSME OCs and CDs, who are typically stressed for liquidity (we find that only a minority receive and/ or pay their dues within 45 days on average), from slipping into insolvency. An effective way could be to mandate that undisputed OC invoices above a threshold, say Rs. 1 crore, be uploaded to the NeSL, which serves as the basis for all IBC petitions and admissions. It will enable quicker pre-admission settlements as well as quicker admissions, thus shortening the process time. To make the IBC more affordable, cost-effective, and a relevant option for MSME OCs, we also recommend a process for aggregating claims and partial waiver of post-admission, IBC process- related expenses.
We analyse the IBC auction process and distribution of the proceeds among the OCs and FCs. A twin-objective function confronted by the bidders – maximizing the asset value and hence, overall realization, coupled with minimizing FC haircut – leads to sub-optimal plans, which opens up litigation, resulting in delays. We move to a single objective function in which the bidder (resolution applicant) with the highest overall offer for the combined claims of the FCs and the OCs wins the auction. A quasi-APR formula (Iyer & Prasad, 2025), with an MSME orientation is used to distribute proceeds among the creditor groups. This approach maximizes the asset value and reduces the need for voting on the amounts involved, and can speed up the process significantly. Our analysis of the Pre-packaged Insolvency Resolution Process (PPIRP) reveals an inherent lack of goal congruence and misalignment of incentives among the key players, i.e. the MSME CD, OCs and the FCs that comprise the CoC. The current framework is heavily tilted towards the CD and the OCs (for valid reasons), while giving the CoC the power to decide on both admission and resolution, creating a potential conflict of interest and potentially leading to actions that increase the probability of worse outcomes for the MSME CDs. This makes the PPIRP unfavourable to both FCs and the MSME CDs. We recommend reconsidering the clause requiring non-impairment of OCs, and suggest a principles-based design that results in payouts for and some sacrifice by both the FCs and the OCs, while still appreciating the latter’s need to remain strong for their own and the MSME CD’s sake. We also recommend speeding up the PPIRP by separating the main resolution process from the investigation of avoidance transactions. We suggest steps to increase awareness and trust in the PPIRP among MSME CDs through effective communication. While we do not make any recommendations outside the IBBI’s domain as part of this study, for MSMEs to effectively leverage the IBC, we recommend taking an ecosystem view, which will mean stepping outside the confines of IBBI’s purview to harmonize frameworks, platforms, rules and laws across the RBI, NCGTC, direct and indirect taxes, and the Ministry of MSMEs.

8 Chapter 1: Introduction Micro, Small, and Medium Enterprises (MSMEs) are central to India's economy, contributing ~30% to GDP, 49% to exports, and employing over 110 million people. Their widespread presence supports regional development, income distribution, and inclusive growth. MSMEs are affected by financial distress both as debtors and creditors. The Insolvency and Bankruptcy Code (IBC), 2016 has reshaped India’s insolvency regime - originally focused on large firms, but increasingly accessed by MSMEs. As corporate debtors (CDs), MSMEs can avail of schemes under RBI’s support framework to avoid going into bankruptcy via the Pre-Packaged Insolvency Resolution Process (PPIRP) or the Corporate Insolvency Resolution Process (CIRP) under the Insolvency and Bankruptcy Code (IBC) to revive by declaring themselves bankrupt and coming to a resolution with their creditors on the latter’s dues. As operational creditors (OCs), they can avail the Trade Receivables Discounting System (TReDS) platform for liquidity, and also file cases under Section 9 of the IBC - often using it as a trigger to recover dues before admission. They also have access to institutional redressal mechanisms such as the MSME Samaadhaan, and Facilitation Councils under the MSMED Act, 2006. The extent to which MSMEs, particularly as OCs, have used the IBC route indicates that these mechanisms have not fully addressed their problems with delayed payments and defaults.
MSMEs are not recognized as a distinct class under the IBC. When they act as creditors, they are generally subsumed within the broad category of OCs–typically as suppliers of raw materials, components, logistics, IT services, maintenance, or contract labour to larger enterprises. MSMEs, as OCs, encounter all the systemic disadvantages of the class of OCs, often in more acute form, thus, suffering a double marginalization.
• Information asymmetry: MSME OCs rarely receive early warning of debtor distress and often become aware of insolvency proceedings only after key steps are underway. • Limited legal capacity: Micro and small enterprises (MSEs) rarely have access to in-house legal expertise or professional advisers, leaving them ill-equipped to navigate complex resolution procedures or file claims effectively. • Discretionary recognition: Claims are frequently denied or reduced due to poor documentation, procedural lapses, or disputed deliveries—challenges that MSMEs lack the resources to contest. • Collective action problem: Fragmentation prevents MSMEs from organizing collectively, unlike financial creditors (FCs) who negotiate through syndicates or trustee banks. • Weak enforcement: Outside the IBC framework, MSMEs already struggle with delayed payments and ineffective legal remedies. Resorting to the IBC often compounds the risk: when their claims are excluded in resolution, they lose both recovery prospects and valuable time.

9 While their claims may appear modest in absolute terms, recovering these dues is often critical to the MSMEs’ survival. Unlike diversified or well-capitalized creditors, even a single unpaid invoice from a large buyer can trigger severe cash flow disruptions. Thus, MSMEs face a paradox: their economic exposure to debtors is disproportionately high, yet their legal and institutional leverage under the IBC is extremely limited. Case law over the years since the IBC’s inception has recognized the differentiated but skewed status of OCs vis-à-vis FCs, and has tried to reduce it while ensuring the priority of the FCs, as underlining the spirit of their respective contracts with the CDs. The Supreme Court has reiterated the commercial wisdom of the Committee of Creditors (CoC) comprising entirely of FCs and that the economic functions performed by OCs and FCs differ in ways that directly affect their respective rights and recoveries in insolvency, through landmark cases (Essar Steel India Ltd. v. Satish Kumar Gupta; K. Sashidhar v. Indian Overseas Bank, both in 2019). However, the case laws have also recognised the need for stronger OC protection, which has been instrumental for amendments such as Section 30(2)(b) and the pre-pack framework for MSME CDs in the IBC. The motivations of OCs, particularly MSMEs, to use the IBC route despite the cost, complexity, and time involved, as well as the poor payout percentages, are an interesting and important area for study. Understanding why and how OCs and CDs (and FCs too) use or do not use the IBC’s institutional mechanism, and their outcomes, will inform policy recommendations to further strengthen this much- needed set of frameworks to revive and rehabilitate businesses in the interests of sustaining a thriving entrepreneurial climate.

1. Scope of study and Objectives The objectives of this study include investigating the following questions. MSMEs as Operational Creditors: Investigate MSMEs' usage of recovery tools such as TReDS, the IBC-as-threat (Section 9), and the mainstream IBC route • What are their incentives to use the different mechanisms: IBC-as-a-threat, TReDS, mainstream IBC in isolation or in specific combinations with each other? • How are MSMEs leveraging Section 9 of IBC (operational creditor route)-as-a-threat mechanism to pressure large debtors into settlement? What percentage of such filings are withdrawn before admission? Are other OCs using this route as well? • How do MSMEs fare in liquidation waterfall and resolution plans when compared to large OCs and secured financial creditors?

10 MSMEs as Corporate Debtors: A. To examine the characteristics, motivations and behaviour of CDs through the following questions. • How do MSME CDs compare with large CDs at the time of admission into IBC? • How do the behaviour and characteristics of CDs petitioned by FCs versus OCs differ? How about MSME CDs? B. MSMEs’ awareness, attitude and behaviour towards the PPIRP: • What is the scale and effectiveness of the PPIRP for MSME corporate debtors? • What barriers (costs, stigma, complexity) prevent greater participation in the PPIRP? Besides, the study will aim to provide an assessment and recommendations as described below. • To benchmark India’s approach to MSME creditor rights against global best practices. • To develop policy and institutional recommendations aimed at assessing the role of the PPIRP under the IBC. • To recommend policy and institutional reforms to enhance efficiency and equity. 2. Data and Methodology This study adopts a mixed-methods approach, combining stakeholder interviews, mathematical modelling, financial simulations, statistical analysis of case-level data, and a review of global best practices to derive its policy conclusions. The analysis is organised into two broad phases of the insolvency process: the pre-admission stage, before a case is admitted into the Corporate Insolvency Resolution Process (CIRP), and the post-admission stage, after admission into the CIRP. Within these phases, the study examines the use and outcomes of the IBC for the FCs and the OCs, and within them, the MSMEs versus large firms. It also examines how the IBC has played out for CDs petitioned by the FCs and the OCs, and in the process, the characteristics that each of these stakeholders might display.
As part of our qualitative data collection, we interviewed representatives from relevant stakeholder groups – six resolution professionals (RP), an insolvency advocate, an MSME CD undergoing CIRP, two regulatory body CXOs (one each from TReDS and the NTGTC), two FC representatives, and a Focus Group Discussion (FGD) with a group of senior executives of a public sector bank. These inputs informed our theory-building for modelling and hypothesizing. They also helped us better understand and connect our results.
For our quantitative analysis, we have utilized data made available by the IBBI through their website, as well as supplied to us specifically for this study.

11 2.1. Data for the pre-admission phase analysis For the pre-admission phase, we conducted statistical analysis to address three key questions. First, what types of OCs initiate insolvency petitions, and whether MSME OCs differ systematically from non-MSME OCs in this regard? Second, what are the characteristics of CDs against whom FCs file cases, as compared to those against whom OCs—particularly MSMEs—initiate proceedings, and whether MSME status matters for these patterns? Third, what are the characteristics of OCs and CDs that reach settlements prior to admission, and how these differ from cases that proceed further in the process? We supplemented the IBBI data provided to us with financial information on the CDs and OCs for 10 years – from 31 March 2015 to 31 March 2025 – queried from the Prowessdx database using the Corporate Identification Number (CIN). This part of our study required us to segregate the CDs and OCs into MSMEs and the rest. We estimated the MSME status of the CDs and OCs as of the date of admission using annual Sales and Net Plant & Equipment (NPE) data. This identification was based on the MSME definitions prevailing on the date of admission/ commencement of the process (see Table 1.2.1 for the classification). The financial data was patchy in places, especially for the CDs. Our propositions required us to estimate the health of the OCs and CDs on the date/ time of admission.
Table 1.2.1. MSME definitions as amended over the years Effective date Micro Small Medium 1 April 2025:
(https://www.dcmsme.gov.in/Notific ation-S.O-no-1364-E-dated- 21.03.2025-Revised-Definition.pdf) Investment: Not more than Rs. 2.5 crore and Annual Turnover not more than Rs. 10 crore Investment: Not more than Rs. 25 crore and annual turnover not more than Rs. 100 crore Investment: Not more than Rs. 125 crore and Annual Turnover not more than Rs. 500 crore 1 July 2020: Introduced turnover limit in addition to investment. Removed separation of manufacturing and service firms. (https://dcmsme.gov.in/State- Gov/Notification_%20S%20O%202 119.pdf) Investment: Not more than Rs. 1 crore and Annual Turnover less than Rs. 5cr. Investment: Not more than Rs. 10 crore and Annual Turnover less than Rs. 50cr. Investment: Not more than Rs. 50 crore and Annual Turnover less than Rs. 250cr.

12 Until 30 June 2020: Investment in Plant, machinery & equipment: (https://dcmsme.gov.in/publications/ circulars/GazNot/Recommendation_ of_Advisory_Committee.pdf) For mfg. firms: Not more than Rs. 25 lakhs For service firms: Not more than Rs. 10 lakhs For mfg. firms: Rs. 25 lakhs to Rs. 5 crores;
For service firms: Rs. 10 lakhs to Rs. 2 crores For mfg. firms: Rs. 5 crores to Rs. 10 crores;
For service firms: Rs. 2 cr. to Rs. 5 cr. We constructed Solvency scores and Liquidity scores to measure financial health. Customer payment delays impact liquidity, especially for MSMEs. This affects solvency by causing losses of business and relationships with vendors. Consequently, it can also cascade through the supply chain. Net worth was used to determine the Solvency score, and Days Sales Outstanding (DSO) and Days Payables Outstanding (DPO) were used to assess liquidity. Table 1.2.2 details the constructs used to determine size and financial health, and the categorizations created.
We estimated the size and health of firms based not only on the data as of the year of admission but also on the data leading up to that date. We have been conservative in estimating the health of firms in such situations, for example, preferring to label a firm as “Solvent” instead of “Healthy", if a continuous streak of improving net worth was not visible.
Table 1.2.2. Constructs and variables used for analysis Construct Variable used Definition and classification codes Type of firm by size MSME vs. large firm Applied the limits as defined in Table 1.2.1 as of the date of admission (or date of commencement of CIRP for these set of cases) Financial health Solvency of the firm determined by its net worth Classified into four levels: 1= Healthy, for those firms with consistent increasing net worth; 2= Solvent, for cases with steady but not growing net worth; 3 = Reducing net worth, nearing insolvency; and, 4= Insolvent, with negative net worth Delay/ default in customer receivables
Days’ sales outstanding (DSO) 1 = Low debtor or creditor days (<100 days); 2 = Medium duration (upto 360 days); 3 = High duration, beyond 360 days. An additional classification of MSMEs was performed using the 45-day statutory cut-off to define the Low category (see Section 5 of Ch.2 and Section 7 of Ch.3 for details). Delay/ default in payment to OCs
Days’ purchases outstanding (DSO)

13 Type of business Manufacturing vs. service firm Classification based on NIC description. All wholesale, retail and trading establishments are classified as services, as are transport, hotel and airline businesses. We have used samples of CDs and corporate OCs (i.e., excluding sole proprietorships and partnerships) to examine patterns and behaviour. The data covers CDs petitioned by OCs and FCs and settled at various stages of the bankruptcy process, if not, concluding with resolution or liquidation. The CINs of all these CDs in the IBBI database were fed into the Prowessdx database for retrieving financial data. The samples comprise the CDs for which financial data, as described above, was accessible.
Since the CIN for OCs was not readily available, we manually extracted the CIN for 326 OC applicants, with 928 cases between them, ongoing or concluded at various stages, for whom financial data were available for only 116 OCs1. Table 1.2.3 details the data on OCs and CDs available and used in our study. The OC numbers in column 5 across categories are not additive; some OCs had filed multiple cases and appear in more than one case status.
Table 1.2.3. Data sourced and used for the study Case outcomes OC-initiated cases in the universe (IBBI list) FC-initiated cases in the universe (IBBI list) Observations with data available for CD Observations with data available for the OC applicant Cases settled u/s 12A 803 347 115 OC- initiated + 50 FC-initiated 106 OCs (151 cases) Cases settled under “Appeal-Review-Settle” mode (ARS) 863 NA 186 (all OC- initiated) 25 OCs (34 cases) Cases under CIRP (resolved/ ongoing) 1061 (383 resolved) Approx. 550 resolved 69 OC- initiated + 147 FC- initiated 21 OCs (25 cases) Cases closed through liquidation 1172 1224 117 OC- initiated + 147 FC- initiated # 25 OCs (34 cases) #Due to the very large number of cases having undergone liquidation, we have considered only the OC and FC-initiated cases with OC claims exceeding Rs. 1 crore. 1 The Prowessdx database provided sufficient financials for only 83 OCs, for the remaining we had to query the ProwessIQ and Tracxn databases. We have used the 83 sub-sample for all the tests, except where the results might suffer from a sampling bias. More details in Chapter 2, Section 2.5

14 We have tested our logic using frequencies of the sample elements, followed by statistical significance tests using ANOVA, correlations, paired t-tests and regression models. 2.2. Data for the post-admission phase analysis For the post-admission phase, the focus shifted to developing a theoretical understanding of how the design of the IBC auction mechanism, combined with the CoC’s voting rules, influences resolution outcomes. We constructed a mathematical model of bidder behaviour and tested its predictions against observed data. In parallel, we benchmarked actual outcomes against a normative distribution framework, specifically the quasi-APR rule, to assess the fairness of payouts across creditor classes. Finally, we conducted simulation exercises to examine whether an OC would have an economic incentive to invoke the IBC in the absence of any realistic prospect of pre-admission settlement, drawing on observed outcomes from both the pre-admission and post-admission stages. Data published by the IBBI on CIRP resolutions (as of 30 September 2024) were used to study the treatment of OCs in this process. Further details on the methodology used are provided in Chapter 4. 3. Findings of the Study This study highlights several features of the insolvency landscape that are crucial for understanding how MSME OCs and CDs fare under the IBC.
First, a substantial portion of economically meaningful outcomes occur before cases are admitted to the CIRP. Settlements, withdrawals, and dismissals in the pre-admission phase shape the experiences of most MSME creditors, making it essential to analyse this stage as part of the insolvency ecosystem rather than treating it as peripheral. Second, our analysis shows that the CIRP is neither accessible nor effective for many MSME OCs. Micro and small enterprises (MSE)—often financially fragile and lacking the bandwidth for prolonged legal processes—rarely experience favourable outcomes through the IBC. Since these entities form the overwhelming majority of the MSME sector, this is an area of significant policy concern. Even for stronger MSMEs that avail the IBC, it is not the CIRP itself that makes the process attractive. Instead, the prospect of pre-admission settlement at relatively high recovery rates is what creates meaningful economic value. Without this settlement possibility, the CIRP alone would be unattractive even for healthy, well-capitalised MSME OCs. Third, our examination of the IBC auction framework reveals a structural defect arising from dual and conflicting objectives faced by the resolution applicants. These are, a) preserving the value of the going concern by optimizing OC allocations, and b) maximising their chances of winning the auction by allocating as much as possible to FCs, who alone vote on resolution plans. This tension distorts

15 incentives and produces outcomes in which OCs, especially those with medium or large claim shares, face decreasing ratios of share in payout to their claim share. The empirical pattern—declining allocation ratios as OC claim share rises—is aligned with the predictions of our auction model. Based on this, we recommend transitioning from a dual-objective system to a single-objective auction design, and suggest possible pathways towards this. Fourth, the IBC draws no formal distinction between MSME OCs and other OCs. In practice, this can disadvantage MSME OCs, whose smaller claim sizes, weaker bargaining power, and greater dependence on timely payments place them in more vulnerable positions. To mitigate this, we propose changes to the process and rules. Finally, the PPIRP for MSME CDs offers an additional channel that may reduce costs and expedite resolution. However, its effectiveness depends on FCs' willingness to engage constructively and MSMEs’ ability to negotiate viable base resolution plans. This is thwarted by the misalignment of incentives arising from the current design. We make recommendations that can achieve better goal congruence thus, aligning incentives, and also to increase awareness of and trust in the PPIRP among MSME CDs.. Overall, the findings underscore the need for a more MSME-sensitive insolvency architecture—one that addresses pre-admission dynamics, mitigates auction distortions, and explicitly protects smaller creditors whose health is vital for sustaining India’s supply-chain-intensive economy. 4. Limitations and organization of the study The primary limitations of this study have been the absence and non-accessibility of systematic data, particularly on the pre-admission stage of the IBC process. Because virtually no structured information is available on settlements, withdrawals, or dismissals before admission, we were compelled to use Regulation 30A cases as a proxy for understanding pre-admission behaviour, and further make assumptions within the cases settled u/s 12A to proxy for Regulation 30A settlements. Access to comprehensive pre-admission data would greatly enhance the robustness and precision of our findings. This study’s focus is MSMEs in bankruptcy cases, and most MSMEs, particularly MSEs are privately held and small firms. While Prowessdx provides data for a good number of unlisted firms, this starts to dwindle as we move towards MSMEs, and within them, the smaller firms. This tends to introduce an inadvertent large-firm sampling bias in our tests. We have tried to mitigate this bias by either consciously gathering data for the remaining firms (typically smaller) from alternate databases. Further, since CIN for OCs were not available from the IBBI database, these had to be manually collected. Given the lack of data for private and small firms as described above, it was necessary to pull

16 out CINs at least four times the final sample size. Using the names to gather these posed its own problems due to inconsistent spellings and data entry errors.
This study is organized into six chapters. Following this Introduction, we have Chapter 2 on the frameworks available to MSMEs and the processes therein. Chapter 3 discusses settlements in the IBC and whether IBC is used by OCs as a threat mechanism for early settlements. Chapter 4 models and tests the IBC as an auction mechanism and proposes alternative treatments. Chapter 5 discusses the PPIRP available to MSME CDs, and Chapter 6 lays out our policy recommendations.

17 Chapter 2: The IBC process for recovery and resolution

1. Introduction The MSMED Act, 2006 provides alternative routes for MSMEs to recover payments from their vendors. Filing a petition under the IBC, 2016 is an optional, last resort. They are expected to exhaust these alternative modes before using the IBC route. Even if the invoice generated by an MSME is not recognized or acknowledged by the buyer, it does not nullify the MSME’s right to claim payment under the MSMED Act, 2006. If the goods or services are delivered and no written objection is raised by the buyer within 15 days, the delivery is deemed accepted. This means the buyer is still legally obligated to make payment within the prescribed time frame—45 days if there is a written agreement, 15 days otherwise.
   This chapter examines the first line of remedies available to MSMEs to recover dues from their customers, and why they might knock on IBC’s door. Based on the IBC process and possible outcomes for the MSMEs, we use empirical tests on IBC case data to discern trends and patterns of similarities and differences across MSME and non-MSME OCs and CDs, and between CDs in FC-initiated versus OC-initiated cases. These enable us to understand the incentives and behaviour of the parties involved, which inform policy recommendations.
2. Recovery and redressal mechanisms for MSMEs These include recovery mechanisms that improve process and accessibility to customer dues for MSMEs (e.g. TReDS), and redressal forums to put up grievances related to delayed payments (e.g. MSME Samadhaan portal). Besides, the government and regulators, such as the RBI, introduce regulations and nudges to encourage MSME customers to pay on time. For instance, dues outstanding to MSMEs for more than 45 days must be disclosed separately by listed companies in their financial statements. The RBI directs buyers to pay interest, compounded monthly, on dues beyond 45 days to MSMEs at a rate three times the RBI rate. The Finance Act, 2023, has disallowed payments to micro and small enterprises that are delayed beyond the due date (45 days or an earlier date, if agreed, or 15 days, if no agreement exists) for computing income for income tax purposes.
   2.1. The TReDS (Trade Receivables Discounting System) The TReDS was launched by the Reserve Bank of India (RBI) in 2016 to address delayed payments to MSMEs, through factoring their receivables. Delayed payments to MSMEs were estimated at around Rs 10.7 lakh crore, with micro and small enterprises bearing 80% of this burden. The total amount financed through TReDS grew from approximately Rs 950 crore in FY2018 to over Rs 2,33,000 crore in FY2025 (Jain et al., 2025). Across five platforms, close to 70,000 MSMEs and 4,000 buyers were

18 registered at the end of November 2025 (RBI Statistics). Jain et al. (2025) find that participation in TReDS reduces the receivables cycle of MSME suppliers by an average of 23 percentage points; second-level effects include an increase in sales, salary expenses and fixed asset acquisition, highlighting cash flow easing. Challenges around adoption and effectiveness range from a lack of awareness and digital capabilities to the process itself, such as the need for the buyer to approve the MSME seller and their invoices on the portal, which creates a dependency that could victimize the MSME seller.
Further, banks and other financiers that the buyer has pre-defined credit limits with bid for the accepted invoice (factoring unit), and the winning bidder discounts the MSME seller’s invoice. Over 80% of transactions on TReDS constitute reverse factoring – cases where the buyer may not have predefined limits with any lenders and benefits from the TReDS platform by effectively stretching their cash flow runway at rates lower than those for direct working capital financing. So long as this induces buyers to pay their sellers on time, reverse factoring in itself may be a win-win in terms of effect, not so much in intent. Another major limitation voiced by the CEO of one of the leading TReDS platforms is that, since financiers bid for invoices based on buyer creditworthiness, lower-tier MSME sellers may not benefit. Considering that Micro enterprises comprise the majority of MSMEs and of the delayed receivables, trade credit insurance or guarantees to financiers (Jain et al. 2025), and ‘with recourse’ factoring may extend the TReDS coverage to the bottom rungs. These limitations, particularly the MSME seller’s dependency on the buyer to get validated on the platform, is possibly on of the reasons the TReDS has mostly benefited the larger MSMEs; the Small, followed by the Medium enterprises, comprise a majority of the invoices transacted. Most of these payments were perhaps likely to occur anyway; the inducement of lower cost of capital may have marginally increased the incidence of payments to MSMEs. Lower-rung MSMEs still find it difficult to avail of this benefit2.
2.2. MSME Samadhaan portal and MSME Facilitation Council The institutional baseline policy set up for redressal of non-payments for MSMEs largely comprises the MSME Samadhaan portal, an online platform launched by the Ministry of MSMEs in 2017. Matters registered on this portal with supporting documents and evidence are referred to the MSME Facilitation Council of the respective state for conciliation; if unresolved, proceed to arbitration, which must be concluded within 90 days. Since its inception until 15 December 2024, MSEs had filed over 2,16,000 2 https://www.m1xchange.com/rs-25-lakh-crore-msme-credit-gap-treds-has-sparked-hope-for-small-businesses-but-whats- holding-it-back/#:~:text=Over%20half%20a%20million%20invoices,according%20to%20the%20RBI%20data.

19 applications on this portal involving more than Rs.47,600 crore. Of these, close to 44,000 cases were disposed of, and more than 20,000 settled mutually; together, these involved more than Rs.15,000 crore outstanding (MSME Annual Report 2024-25)3. The remaining current stock in the pipeline is a live proxy for liquidity blocked with buyers. While the Samadhaan portal, along with the MSEFC, is a potentially cost-effective and quicker redressal mechanism, it has not sufficiently mitigated delays or recovered dues for MSMEs. Given that nearly 40 percent of the outstanding dues on the portal are from PSUs and government agencies, the skewed balance of power is a significant reason4. Other reasons include capacity constraints, court process for enforcement5, and fear of repercussions from large customers. 3. The IBC route Though not the most accessible or cost-effective, MSME OCs have resorted to the IBC to recover their dues from customers. While the objective of the IBC has been the resolution and revival of struggling firms, seen from a creditor’s lens – whether an FC or an OC – it is a recovery mechanism. It is therefore interesting and important to understand the process not only for its mechanics but also for the behaviours it induces, which in turn shape how it is used by the various stakeholders. 3.1. Filing of Petitions Creditors or debtors can approach the National Company Law Tribunal (NCLT) to file a petition; FCs filing under Section 7, OCs under Section 9, and CDs under Section 10, provided the minimum default amount is Rs. 1 crore, as per the current threshold notified by the government in March 2020. Since the Code came into force in 2016, the NCLT has received a rapidly growing number of petitions. By March 2025, more than 39,232 insolvency applications had been filed, covering defaults across sectors such as manufacturing, infrastructure, real estate, finance, and services (IBBI newsletter, 2025). 68% of these petitions came from OCs, including MSMEs, who often rely on the Code as a powerful tool to recover dues from larger corporates. An OC is required to issue a written demand to the CD under Section 8 of the IBC, specifying the amount due and the reason for the debt, for the CD to either pay the debt or dispute the claim with valid grounds within 10 days, pending which the OC can file a petition under Section 9. This provision ensures that the debtor has an opportunity to settle the claim before formal insolvency proceedings begin, promoting early resolution and reducing unnecessary litigation. By 31st March 2025, out of the 3 https://msme.gov.in/msme-annual-report-english-2024-25 4 https://swarajyamag.com/economy/beyond-gem-portal-why-msmes-are-still-waiting-for-their-money-from-government- buyers#:~:text=Since%20the%20launch%20of%20the,since%202017%20exceeds%202.5%20lakh. 5 https://smestreet.in/infocus/delayed-payments-to-msmes-the-hidden-liquidity-crisis-india-must-urgently-fix-10533283

20

total petitions filed by creditors, 30,745, or 79%, amounting to Rs. 14 lakh crore, were settled before admission into the IBC; 73% of the petitions settled before admission, amounting to Rs. 4.68 lakh crores of outstanding credit, were filed by OCs. 3.2. Withdrawals, With and Without Settlement A striking pattern since the Code’s introduction is that the majority of cases conclude even before admission. A large proportion of pre-admission withdrawals involve settlements between creditors and debtors. In these cases, the CD and the creditor-applicant resolve the dispute—often through repayment of the outstanding dues or a negotiated partial settlement. In such a situation, the NCLT dismisses the petition after the settlement is placed on record and acknowledged. However, once a case is admitted into the CIRP, it cannot simply be rejected or dismissed on account of settlement; instead, withdrawal of the petition must be sought under Section 12A of the IBC (see Chapter 3 on Settlements). This distinction underscores the procedural importance of timing: settlements made before or during filing prevent admission, while those made after admission require a formal withdrawal process. Withdrawal without settlement usually occurs for procedural or strategic reasons, such as incomplete or defective petitions, lack of supporting documentation, jurisdictional objections, or a shift in the petitioner’s commercial priorities. In some instances, OCs may also decide to pursue recovery through alternative legal forums. 3.3. Other disposals pre-admission: Rejection and Dismissal of Petitions, or Reserved Orders Cases filed under the IBC can be rejected by the NCLT at the admission stage for several reasons, primarily to ensure that only genuine and legally compliant petitions proceed to the CIRP. Common grounds for rejection include non-fulfilment of eligibility criteria, incomplete or defective petitions, disputed debts, failure to meet statutory default thresholds, multiple filings or abuse of process. In case a pre-existing settlement between parties is reached before filing, and the debtor can demonstrate that the claimed dues have already been paid or otherwise resolved, NCLT will typically reject the petition at the admission stage. Specifically, Jayadev et al. (2025) found that a significant proportion of applications were rejected due to disputes over debt or procedural non-compliance. These statistics highlight that the IBC framework, while enabling creditors to seek timely resolution, also filters out petitions that do not meet legal or procedural requirements, thereby preventing frivolous or invalid claims from proceeding in the insolvency process6. By law, this decision should be made within 14 days, but in practice, due to large case backlogs, it often takes much longer.

6 Petitions denied on procedural grounds are referred to as dismissed rather than rejected. Common grounds for dismissal include non-prosecution (e.g., applicant does not appear), non-compliance with procedural directions of NCLT, defective filing

21 An order is “reserved” when the bench has completed hearing arguments from all sides but does not issue a ruling immediately, instead taking time to deliberate and draft a reasoned judgment. In the context of insolvency cases, order reservation often occurs in complex matters involving multiple creditors, contested defaults, or voluminous documentation requiring close scrutiny. It may also apply when the tribunal is asked to evaluate the legitimacy of a settlement proposal or verify procedural compliance. The practice of reserving orders ensures that the NCLT delivers decisions that are well- reasoned, legally sound, and consistent with the principles of natural justice, thereby balancing the interests of creditors seeking repayment with the debtor’s right to fair adjudication. 3.4. Withdrawal after Admission into CIRP Once an insolvency petition is admitted, the CIRP formally begins. However, admission does not entirely close the door to withdrawal. Under Section 12A of the IBC and Regulation 30A of the IBBI rules, withdrawal remains possible even after the commencement of CIRP. If the CoC has not yet been constituted, the applicant may request withdrawal directly through the Interim Resolution Professional (IRP). After the CoC is formed, withdrawal can still take place, but only if it is approved by at least 90% of the CoC’s voting share. This framework reflects the Code’s design to encourage negotiated settlements and consensual exits, even after proceedings are underway.
There are some cases closed by way of (i) appeal to higher courts viz. NCLAT/High Court/Supreme Court, (ii) by virtue of getting reviewed by higher courts, or (iii) settlement allowed by higher courts. These cases, classified as “Appeal–Review–Settled” (ARS) are distinct from closures under Section 12A of the Code, where withdrawal takes place with the approval of 90% of the CoC. In ARS cases, the closure happens not through the CoC route under Section 12A, but rather through judicial intervention at appellate or higher forums. The ARS process can take place at any stage of the CIRP – pre-COC, post-COC, post-expression of interest, after receipt of resolution applications, etc.
In such a case, if the higher court rejects the appeal or finds the settlement defective, the CIRP does not automatically revive. Indeed, the IBC contains no mechanism for automatic restoration of a terminated insolvency process. Instead, the continuation or revival of CIRP depends entirely on the specific directions of the appellate forum. Only if the higher court expressly sets aside the NCLT’s closure order and remands the matter for further proceedings does the CIRP resume from the stage of admission. In the absence of such a direction, the higher court’s disposal order remains final, and the aggrieved creditor must initiate a fresh application under Section 7, 9, or 10 to trigger a new CIRP. Thus, not cured in time (formatting, documents missing), withdrawal by applicant before admission, parallel proceedings or jurisdictional issues, and so on.

22 restoration is discretionary and judicially determined—not automatic—underscoring the procedural finality that attaches to a closure once ordered by the NCLT. 3.5. Mainstream CIRP Process resulting in Resolution or Liquidation For cases that continue beyond admission, CIRP follows a structured process. The NCLT appoints an IRP, and imposes a moratorium on all debt recovery actions, and calls for creditors to submit claims. The IRP verifies these claims and forms the CoC, which then decides whether to confirm the IRP as the Resolution Professional (RP) or appoint a replacement. An Expression of Interest is issued. Resolution applicants—including potential investors, buyers, or in some circumstances even existing promoters— submit plans for restructuring the distressed company. For approval, these plans must secure at least 66% of the CoC’s voting share.
Liquidation of the CD occurs when resolution efforts fail or when the CoC decides that the concern is not fit for resolution. Normally, liquidation involves either the piecemeal sale of the assets or a slump sale. Introduced in 2018, “liquidation as a going-concern” was a unique type of liquidation under the IBC where instead of unwinding and selling individual assets, the entire business could be sold in one go to a buyer. A misnomer, this was not the regular going-concern sale for continuity, which happens under resolution. The buyers usually sold all or some of the assets, retaining the shell or the name to run an entirely different business. It also encountered implementation difficulties: delays, litigation over liability transfer, complications regarding employee rights, regulatory approvals (licenses, permits), and uncertainty about the viability of the “going concern” business. Since the IBC abolished this provision in October 2025, the liquidation regulations have only included sales of assets through standalone assets, slump sales, or asset parcels.
By March 2025, 2,758 companies had been ordered into liquidation, a figure that outpaces resolutions by nearly 2.3 times (IBBI, 2025). This imbalance underscores the persistent difficulty in reviving distressed businesses, particularly smaller enterprises that often lack viable buyers or restructuring options. According to the IBBI reports, across all closed CIRPs until 31 March 2025, around 57% of cases ended in resolution, settlement, or other forms of closure, while 43% went into liquidation. Liquidation involves selling the company’s assets and distributing proceeds according to the IBC’s priority waterfall, where insolvency costs and secured creditors take precedence. 4. How the process has fared Figure 2.4.1 shows the outcomes of insolvency petitions filed under the IBC. It reveals that most insolvency petitions filed under the IBC - particularly those initiated by OCs - are resolved even before they enter a formal CIRP. Out of a total of 39,053 cases filed, only 8,308 (approximately 21%) were admitted into CIRP, while the vast majority were disposed of at the pre-admission stage.

23 Figure 2.4.1: Initiation and exit of IBC cases filed with the NCLT Source: Created by authors from NCLT and IBBI data For OC-initiated matters specifically, the pattern is even more striking. Of an estimated 26,376 petitions filed by OCs, only 3899 (about 15%) proceeded to admission, whereas approximately 22,477 cases (or 85%) were resolved prior to admission through rejection, dismissal, or withdrawal - either with or without settlement.
Of all admitted cases at the NCLT, 1,194 have culminated in successful resolutions. These cases represent the ideal scenario envisaged by the Code—where the business is preserved, creditors receive payments according to the plan, and the CD continues as a going concern under new management. A substantially larger number - 2,758 cases have resulted in liquidation. The predominance of liquidation indicates the structural challenges faced by distressed firms, the limited time available for revival, and the frequent deterioration of asset value during CIRP. Another 1,154 cases were closed through Section 12A withdrawals, where the original applicant withdrew the insolvency petition. These withdrawals typically reflect settlements negotiated after admission, often motivated by the debtor’s strong desire to avoid loss of control of the enterprise. A further 1,276 cases fall under the category “Appeal-Review-Settled” (ARS), reflecting matters where CIRP was halted due to intervention by higher judicial authorities (such as NCLAT or the Supreme Court), or where the parties reached an out-of-court settlement, or where the case was withdrawn as infructuous for reasons other than Section 12A. Finally, 1,926 cases remain ongoing, indicating that they are still under CIRP and have not yet reached a definitive outcome.

24 Within OC-initiated admitted cases, as represented in Figure 2.4.2, outcomes show substantial attrition: 863 were concluded through Appeal-Review-Settlement, 803 withdrawals occurred under Section 12A/ Regulation 30A mechanisms (with 245 occurring before constitution of the CoC via Regulation 30A and 558 after CoC formation through Section 12A), and 1,172 ended in liquidation.
Figure 2.4.2: OC Initiated Cases: Pre- and post-admission outcomes Source: Created by authors from NCLT and IBBI data More than 6% of FC-initiated cases at the NCLT find resolution, as against 1.5% OC-initiated cases, due to double the proportion of FC-initiated cases entering CIRP (33% vs. 15% for OC-initiated ones), and finding resolution (almost 19% compared to 10% for the OC-initiated ones) (Figure 2.4.2 and Figure 2.4.3). One plausible reason for this pattern could be that a significantly higher proportion of OC-initiated cases get settled under Section 12A or through appeal (ARS) – almost 43% of admitted cases, compared with 19% of FC-initiated admitted cases. Interestingly, the proportion of OC-initiated admitted cases ending up with liquidation (30%) is marginally lower than the corresponding figure for FC-initiated cases (33%).

25 Figure 2.4.3: FC Initiated Cases: Pre- and post-admission outcomes Source: Created by authors from NCLT and IBBI data Two strong inferences emerge from these distributions. 1. Particularly for OCs, the NCLT functions less as a venue for actual insolvency resolution and more as a credible enforcement and negotiation platform, where the threat of admission into CIRP catalyzes settlement between CDs and OCs. 2. A larger proportion of settlements among OC-initiated cases could be because the CDs in these cases are healthier than in FC-initiated ones, and prefer to settle with their OCs than go into full CIRP. 3. OC-initiated cases also saw a lower proportion of resolutions, while the proportion of liquidations is the same as for the FC-initiated cases. If our above argument holds, it implies that the remaining pool of CDs comprises the weaker ones, where liquidation is more probable than resolution. Alternatively, these cases are of relatively less interest to the CoC, leading to the ‘quicker-and-easier’ solution of liquidation rather than resolution. We examine these inferences in the following chapters. 4.1. Efficiency and Recovery Rates
Analyzed in detail in Chapter 4, we find the recovery outcomes under IBC to be mixed. In 2024, the average recovery for creditors through approved resolution plans was around 27% of admitted claims, reflecting the challenges in realizing full value from distressed assets (CRISIL Ratings, Aug 2024). However, there has been a modest improvement in efficiency, with the resolution-to-liquidation ratio rising from 0.46 in FY2022 to 0.61 in FY2024 (IBBI, 2024). Some high-profile cases, such as Essar Steel and Bhushan Steel, have demonstrated that large, viable companies can generate significant recoveries for creditors when successfully resolved. At the same time, smaller cases, especially those involving MSMEs, frequently yield minimal returns. As of 31 March, 2025, the average recovery for

26 creditors through approved resolution plans improved to 32.8% (IBBI newsletter, January-March 2025). 4.2. Delays and Backlogs Despite its success in strengthening creditor rights and encouraging early settlements, the IBC process has struggled with timelines. The law stipulates that CIRP should conclude within 180 days, extendable up to 330 days, including litigation delays. Yet by 2025, the average duration of CIRP cases had stretched to about 713 days, nearly double the prescribed limit. This delay is largely attributable to the overwhelming case burden on the NCLT. As of March 2025, there were close to 15,000 cases pending across 15 NCLT benches, with approximately 7,000 of these related to IBC alone (NCLT Case Status Report, March 2025). The mounting backlog has triggered debates about increasing NCLT capacity, appointing more members, and reforming procedures to ensure that the Code’s promise of a time-bound resolution is upheld; these numbers are significantly lower than a year ago, perhaps indicating action taken on this front. 5. IBC for MSMEs The COVID-19 pandemic brought out the vulnerability of MSMEs in more ways than one. It also highlighted the inaccessibility and lack of usefulness of the mainstream CIRP for them due to cost, time and complexity issues. The result was an attempt to design a simpler, less formal, and time-bound framework that catered to the MSME CDs and ‘favoured’ OC interests, in ways the mainstream CIRP could not. This led to the promulgation of the Pre-packaged Insolvency Resolution Process (PPIRP) in April 2021. MSME OCs however, continue to file IBC cases with the NCLT (they cannot directly use the PPIRP), and MSME CDs continue to be petitioned under the mainstream CIRP. The PPIRP process, track record and challenges are detailed in Chapter 5. In this section, we specifically examine MSMEs in the mainstream CIRP. 5.1. MSMEs as Creditors In 2020, the IBC was amended, raising the minimum claim amount a creditor must file to Rs. 1 crore. For the outstanding dues from a single corporate debtor to exceed Rs. 1 crore, an OC’s scale of business cannot be too small. Secondly, the OC applicant has to incur upfront costs of approximately Rs. 5 lakhs for initial IRP expenses, in addition to the nominal NCLT application fee. As an upfront outlay, an OC will need to be of a certain minimum size and liquidity to afford this cost and risk a potential loss of such amount. We examine whether the CIRP, in its present form, is therefore more relevant to relatively large (non-MSMEs, for the purpose of this study) and financially healthy OCs as applicants. We classified our sample of 83 OCs with more readily accessible financial data (titled Initial Set in Table 2.5.1.1) into MSME and larger firms as described in Table 1.2.1 in the previous chapter. We

27 found that only 24 were MSMEs and 59 were non-MSME firms. Further digging into the 24 MSME OCs revealed that the majority – 15 firms – were medium-sized, and not SMEs. While this indicates that, typically, an OC-applicant will be a non-MSME, this result could suffer from a possible sampling bias, given the low proportion of our OC CINs that yielded financial data from the Prowessdx database; out of 326 CINs, we got sufficient data for only 83 firms, which is slightly more than a quarter.
The Prowessdx database obtains data from MCA filings, where firms provide complete financial statements, along with all Schedules and Notes to Accounts. Many small, privately held firms, particularly MSMEs, file only the AOC-4 Form with the basic financial statements. These are not captured in the Prowessdx database in the interest of standardization. However, these are captured in the Prowess-IQ database. Clearly, our first cut of 83 firms would have missed some smaller firms, introducing a large-firm bias in our sample. We then checked for the remaining CINs in the Prowess- IQ database. Only 58 firms were found with some financial data. Only eleven of these 58 had all the necessary data for our analysis. Another 22 had data required for solvency and liquidity analysis, but not enough to categorize the firms as MSME or otherwise – information on PPE could not be separated. At this stage, we improvised to categorize these 22 firms as scientifically and objectively as possible. We will call this set of 33 OCs the ‘impaired sample’ – those for which data was difficult to access. These are described in two parts in Table 2.5.1.1 – the Second set comprising the 11 firms with data available in ProwessIQ, and the Third set for which we had to improvise, as described below. Their conscious inclusion helps us overcome any large-firm sampling bias present in the original sample of 83 OCs. To estimate the Net PPE for the Third set, we used the ratio of Sales to Net PPE as in the MSMED Act, 2006, to categorize MSMEs. This ratio was 5x in the definitions effective 1st July 2020, and 4x in the revision effective from 1st April 2025. Prior to 1st July 2020, MSME categorization was based only on Net PPE thresholds, separate for Manufacturing and Service industries, as explained in Table 1.2.1. Using a Sales to PPE ratio of 5x, we imputed the Net PPE for all 22 firms for which Net Sales were available. Using these imputed figures to categorize them into MSMEs and large firms, we found all 22 firms fell in the MSME category7.
7 To ensure the robustness of this approach, we computed the median Sales to Net PPE ratio for the initial set of 83 firms. This came to 17.74 times. Our ratio of 5x therefore, has an upward bias, if any, for the Net PPE of these firms thus, confirming their MSME status.

28 Table 2.5.1.1. Distribution of OCs in our sample Date of admission Until June 2020 Since July 2020 Total OCs Until June 2020 Since July 2020 Non- MSME MSME Micro/ Small Non- MSME MSME Micro/ Small Initial set 83 21 62 16 5 0 43 19 9 Second set 11 7 4 2 5 4 1 3 2 Third set 22 5 17 0 5 2 0 17 16 Total 116 33 83 18 15 6 44 39 27 As suspected, our initial set of firms showed a large-firm bias as their data was more readily accessible, though the 326 OC-applicant CINs searched for were randomly selected. The Second and Third sets comprised a larger proportion of MSMEs; the data availability and accessibility reduced as the firms became smaller. The Third set comprised entirely of MSMEs, though this result may be partially influenced by the assumptions we have used to impute Net PPE for most of the firms in this set. The interesting result, though, was that the overall sample was similarly split between MSMEs and non- MSMEs, both for the period before July 2020, and after. The proportion of Micro/ Small firms was marginally higher at 33% (24 of 83 firms) than in the earlier period, at 19% (6 of 33 firms). While the introduction of the Rs.1 crore floor may have precluded many Micro and Small firms from petitioning under the IBC, the redefinition of MSMEs at the same time brought many more firms into this ambit. This cohort has been further expanded since the update in MSME limits in 1st April 2025. Though the majority of our sample (53%) are non-MSMEs, the margin above 50% is not statistically significant. Moreover, 33% out of the 47% MSMEs in the post-July 2020 sub-sample were MSEs, unlike the pre-2020 cohort, which predominantly comprised medium enterprises (28% out of 47%). This result is at least partially attributed to the increase in thresholds for MSMEs, which would have pushed many erstwhile Medium enterprises into the Small enterprises category. Many firms that were earlier categorized as non-MSMEs and had claims of at least Rs. 1 crore, now got counted among MSMEs. Since the 1st April 2025 redefinitions, we can expect to find some of the borderline Micro enterprises accessing the IBC for their outstanding claims. We examined the financial health of the OC-applicants at the time of admission using solvency and liquidity scores. As described in Table 1.2.2 in Section 1.2, we have coded each OC into one of four solvency categories based on its net worth on and prior to the date of admission. The results in Table 2.5.1.2 for our initial set of 83 firms show that 87% of the OC-applicants were healthy or at least solvent as on the date of admission. However, to ensure robustness, we use our ‘impaired sample’ of 33 OCs, which are predominantly MSMEs, to ensure our results do not suffer from sampling bias.

29 Table 2.5.1.2. Solvency matrix for OC-applicants Solvency Panel A: Initial set of OCs Panel B: Second and third sets (impaired sample) MSME non- MSME Total MSME non- MSME Total Healthy 7 35 42 15 2 17 Solvent 12 18 30 12 1 13 Distressed/ Reducing NW 1 3 4 1 0 1 Insolvent 4 3 7 2 0 2 24 59 83 30 3 33 In Panel A, though the majority of both MSME and non-MSME OC-applicants were at least solvent, a majority of MSME OC-applicants were solvent, not necessarily healthy, as opposed to the larger OC- applicants where the Healthy firms were almost twice as many as the Solvent ones. This might indicate the relative vulnerability of MSME creditors when their dues are not paid on time. Our impaired sample in Panel B also has majority of the firms Healthy or Solvent. Taking the entire sample of 116 OCs, a simple one-way ANOVA shows that the relatively better solvency of non-MSME OC-applicants is statistically significant at 95% confidence level.
We further examine MSME OC-applicants’ vulnerability by analyzing their liquidity health using DSO- DPO combinations, as described in Table 1.2.2. Rising receivables can disguise a strained business model that can ultimately risk insolvency (The Economist, 2025). Table 2.5.1.3 describes the DSO- DPO matrices for OCs, separately for MSMEs and non-MSMEs. Table 2.5.1.3. DSO-DPO matrices for OC-applicants Panel A: MSME OC-applicants Panel B: Non-MSME OC-applicants Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 5 2 5 12 1 1 1 3 Medium 1 6 7 14 1 16 9 26 Low 0 10 14 24 0 4 28 32 Total 6 18 26 50 2 21 38 61 The matrices in Table 2.5.1.3 include all the OC applicants in our study, including the second and third sets. Among both MSME and non-MSME OC applicants, the Low-Low combination had the highest share, indicating receivables and payables outstanding for less than 100 days. More than 60% of the non-MSME OC applicants had Low DSO, while over half had Low DPO. These proportions fell to 52% and 44% respectively, for the MSMEs, indicating their relative vulnerability. While the proportion of non-MSME OC applicants in the High DSO and DPO categories was in single digits, among

30 MSMEs, 12% had an average DSO of more than a year, and 28% took more than a year on average to settle vendor dues, indicating a cascading effect. 14% of the MSME OC applicants mapped to the Low DSO-High DPO combination, implying that though they received dues within 100 days on average, they took over a year to settle their vendor dues. These proportions were significantly lower for the large firms. Whether this is indicative of causality due to liquidity problems, or wilful delays, requires further investigation. The MSMEs in the Second and Third sets (impaired sample, comprising smaller firms) surprisingly showed better liquidity health as measured by DPO. 40% of the MSMEs in the initial set of OC- applicants took more than a year on average to pay their vendor dues, which came down to 13% in the impaired sample. Could it be that the smaller MSMEs lacked the bargaining power against their vendors that the larger ones had? This requires further analysis with a larger sample. To understand this better, we reworked the DSO-DPO matrix for the MSME sub-sample by recalibrating the Low, Medium and High categories. The MSMED Act, 2006 requires MSME receivables to be settled within 45 days. DSO or DPO upto 45 days were categorized as Low, while those between 45 and 100 days were Medium, and the rest, Large (Table 2.5.1.4). Essentially, the previously ‘Low’ categorized MSMEs were now bifurcated at the 45-day cut-off. All the Medium and High categorized MSMEs move into the High category in Table 2.5.1.4 (observe all 24 MSMEs (6 High and 18 medium) are now in the High DSO, and similarly for DPO, and the 26 MSMEs categorized as Low DSO now are split between Medium (18) and Low (8)). Table 2.5.1.4. DSO-DPO matrices for MSME OC-applicants with 45-day cut-off Debtor days (DSO) High (>100 days) Medium (45-100 days) Low (<45 days) Total Creditor days (DPO) High 14 9 3 26 Medium 7 2 1 10 Low 3 7 4 14 Total 24 18 8 50 We find that only 16% of the sub-sample falls in the Low-DSO group, and 28% in the Low-DPO group; only 8% of the MSME OC-applicants have Low DSO and DPO – they receive their dues and pay their vendors within 45 days on average. This illustrates the precarious liquidity position of MSMEs and perhaps, also their lack of negotiating power with both their customers and vendors. This aligns with our results for solvency too, with non-MSMEs showing a statistically significant healthier position compared to MSMEs. It’s a matter of time before MSMEs’ liquidity constraints transform into distressed solvency.

31 5.2. MSME as Corporate Debtors The behaviour of CDs towards financial borrowing is likely to differ from their behaviour towards operational credit. The repercussions of defaulting on financial debt are more severe – hikes in interest rates, reduced credit ratings, stricter covenants, and similar penalties, as banks and FCs are required to report defaults to the Credit Information Companies (CIC). They could also initiate processes such as SARFAESI to recover assets. We believe that a CD would avoid delaying or defaulting on a bank interest payment or repayment, unless it is in a very poor financial condition. This is observed particularly among MSMEs, for example, the cases of Enn Tee International Ltd.8 and Shree Rajasthan Syntex Ltd.9 that invoked the PPIRP under the IBC. They had continued to pay dues to the bank for 3- 5 years despite losses, until factors such as volatile prices or Covid-19 hit their operations to the point of no return. On the other hand, it is relatively less costly to default on an OC. Therefore, it would default on OC dues even when it is in a relatively healthy state. By that logic, a CD against whom an OC has filed an insolvency petition is likely to be financially stronger than one against whom an FC has filed an insolvency petition. Similar analyses as conducted for OC-applicants were done for the CDs, too. For CDs, we were able to obtain richer data – larger samples across stages of the CIRP. The solvency matrices for CDs in Table 2.5.2.1 show a clear, significant difference in financial health between those petitioned by the FCs and those petitioned by the OCs. Table 2.5.2.1(a). Solvency matrices of CDs petitioned by the FCs versus the OCs Solvency Panel A: All CDs in OC-initiated cases Panel B: All CDs in FC-initiated cases MSME non-MSME Total MSME non-MSME Total Healthy 42 46 88 2 5 7 Solvent 87 58 145 45 27 72 Distressed 28 25 53 21 17 38 Insolvent 113 88 201 88 139 227 270 217 487 156 188 344 In both sets, more than 50% of the CDs are distressed or insolvent, with a higher proportion among the FC-initiated CDs (77%) than among the OC-initiated CDs (52%). The distribution in Table 2.5.2.1(a) is similar across MSMEs and non-MSMEs in Panel A (OC-initiated cases), while in Panel B (FC- initiated cases), we notice the proportion of solvent plus healthy firms is more among the MSMEs (30%) than the non-MSMEs (17%); most of these are just solvent, with firms in the Healthy category a 8 Source: https://ibbi.gov.in//uploads/order/45fabe0bfa7f088f5a279603b3f57ca7.pdf 9 Source: https://ibbi.gov.in/en/claims/order-process/L24302RJ1979PLC001948

32 miniscule percentage. Table 2.5.2.1(b) shows that this difference is statistically significant at the 1% significance level across the MSME and non-MSME sub-samples. Table 2.5.2.1(b). ANOVA for solvency rating across OC-initiated and FC-initiated CDs separated by MSME status Panel A: non-MSME CDs Panel B: MSME CDs Groups Count Average Variance Groups Count Average Variance OC-initiated 217 2.71 1.45 OC-initiated 270 2.79 1.32 FC-initiated 188 3.54 0.70 FC-initiated 156 3.25 0.85 F-statistic 62.872 F-statistic 18.606 p-value 0.000*** P-value 0.000*** Another reason FC-initiated CDs are financially weaker than OC-initiated ones on the date of admission is the time that elapses before the former are petitioned before the NCLT. More than a year passes before these debts become NPA on the banks’ books and are then declared as ‘Loss assets’. Thereafter, banks try other recovery mechanisms before resorting to the IBC10. The CD’s financial situation deteriorates further during this period. While MSMEs are typically more vulnerable and financially weaker than their larger counterparts, we examined whether the delay by the FCs led to greater or lesser loss of value for firms than did their stature as MSMEs or non-MSMEs. We find that an MSME CD against whom an OC has filed an insolvency petition is likely to be financially stronger than a non-MSME CD against whom an FC has filed an insolvency petition. The average solvency score is 2.79 for an OC-initiated MSME against a worse 3.54 for an FC-initiated non- MSME firm (Table 2.5.2.1b), where a higher score indicates worse financial health. This difference is statistically significant at a 1% significance level.
One would then expect the financial health of CDs with FC-initiated cases to be the worst of the lot. Surprisingly, we find that among the FC-initiated CDs, the MSMEs’ average solvency score is better and statistically significant at 1% significance level, than non-MSMEs.
A DSO-DPO analysis (Table 2.5.2.2) shows that a) OC-initiated CDs have better liquidity than FC- initiated CDs, and b) non-MSME CDs have better liquidity than MSME CDs. The former have a higher frequency in Low DSO and Low DPO, and a significantly lower proportion of firms in the High-High cells.
10 While not all FCs are banks, they constitute the majority of FC claims. This may differ for small firms, where NBFCs or other non-financial organizations could constitute FCs.

33 Table 2.5.2.2. DSO-DPO matrices for CDs in OC- and FC-initiated cases, separated by MSME status Panel A: Large CDs Panel B: MSME CDs Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 31 22 17 70 69 29 12 110 Medium 12 44 27 83 16 43 19 78 Low 4 13 40 57 3 14 40 57 47 79 84 210 88 86 71 245 Panel C: Large CDs Panel D: MSME CDs Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 45 9 16 70 58 19 6 83 Medium 20 28 14 62 14 10 11 35 Low 3 16 19 38 6 3 11 20 68 53 49 170 78 32 28 138 The results did not change significantly when the thresholds were changed to 45 days to demarcate the Low and Medium categories (results not shown here). Though the proportion of Low DSO was similar for both OC-initiated and FC-initiated CDs (15% and 14%), the proportions of High DSO and DPO were significantly lower for the OC-initiated ones. These results indicate that OC-initiated CDs are likely to be in a financially stronger position than the FC-initiated ones. It may be worth investigating whether credit guarantees offered to FCs for MSME loans play a role here. On the one hand, the existence of such guarantees eases the flow of credit to MSMEs, thereby preventing them from falling into financial distress. On the other hand, such guarantees can create very real moral hazard incentives for both the lender and the borrower. To pre-empt moral hazard behaviour by FCs, the National Credit Guarantee Trustee Company Ltd. (NCGTC) rules require lenders to exercise due diligence both at the time of lending and recovery, before and after claiming the bad loans from NCGTC’s relevant programme. Typically, lenders will exhaust all other recovery options – debt restructuring, SARFAESI – before resorting to IBC, both to save time and maximize recovery and to comply with NCTGC requirements. The comparatively healthier status of MSME CDs perhaps, also points to the effectiveness of the credit guarantee schemes as well as the guardrails placed to check moral hazard behaviour. 6. Conclusion Overall, the NCLT process under IBC has functioned less as a straightforward pathway to insolvency resolution and more as a funnel. Tens of thousands of petitions are filed, yet most do not cross the

34 admission stage, with a majority being withdrawn or settled beforehand. A smaller share proceeds into CIRP, and an even smaller fraction results in successful resolution plans, with a significant number of cases ending in liquidation. Settlements, both pre- and post-admission, play a critical role in enforcing credit and payment discipline. The current lack of structured data and analysis of the pre-admission dynamics understates the effectiveness of IBC as a recovery and deterrence mechanism.
Recommendation 1: We recommend that the IBBI collect, classify, standardize, and make available, data and information on pre-admission settlements, recoveries and disposals beyond the currently available IAAA format. This can include the nature of the petitioner (FC, OC), MSME status of the CD and the OC-applicant, settlement terms where available, and the financial condition of the CD. Such data would allow policymakers to better understand why weaker firms eventually enter CIRP while stronger firms tend to settle earlier, and would enable a more accurate assessment of the IBC’s overall impact. Delays in admission, primarily due to disputed claims and invoices, erode value for all the parties. The NeSL platform is a robust authentication mechanism that presently captures FC claims; while FCs are mandated to upload their loans on the NeSL platform, OCs have the option to do so. As of December 2025, 3,192 entities had submitted details of financial debts to corporate debtors amounting to nearly Rs.334 lakh crore. Though the number of entities submitting operational debts was similar (3,186), this amounted to only a little over Rs.83,000 crore. Notably, operational debts by FCs are included in the latter; the point being that plausibly, only a small fraction of OC debts gets recorded in the NeSL database. While over 41,000 unique debtors defaulted on nearly Rs.20 lakh crore of financial debt, the corresponding figures reported for operational debt were 2,560 and close to Rs.16,000 crore, respectively. We have reason to believe that reporting to NeSL is not a priority for OCs, especially when defaults are not threatening their solvency. Recommendation 2: We recommend that OCs be mandated to record all accepted invoices of at least Rs. 1 crore (the threshold can be reduced in due course to include more dues) with the NeSL in a structured, searchable format. Systematic recording of acknowledged claims would reduce informational frictions, speed up dispute resolution, and lower transaction costs at the pre-admission stage, particularly for MSMEs. MSMEs, both OCs and CDs, are more vulnerable to liquidity stress, and speeding up settlement through process improvements can prevent them slipping into insolvency.

35 Nonetheless, the IBC has altered India’s credit culture fundamentally, giving creditors—especially small OCs—the power to compel repayment, encouraging debtors to settle dues quickly, and establishing a structured process for either the revival or closure of distressed companies. Differential attitudes of firms towards their dues owed to OCs versus the FCs appear to be a significant factor inducing OCs to use the IBC to compel repayment. A positive fallout of this is healthier OCs, which perhaps may not have been the case but for this mechanism playing out. The statistics illustrate both the deterrent effect of IBC in preventing defaults from escalating and the ongoing challenges of ensuring efficient, timely, and value-maximizing outcomes.

36 Chapter 3: Settlement in the IBC

1. Introduction The Insolvency and Bankruptcy Code, 2016 (IBC) was introduced to create a time-bound, creditor- driven insolvency resolution framework in India. Its central feature, the CIRP, shifts the control of the debtor company from its promoters to an Independent Resolution Professional (IRP) and seeks resolution plans from creditors and potential investors. This was a radical departure from earlier debtor- in-possession regimes and aimed to redress India’s historically weak credit recovery environment. However, alongside resolution, the Code has also been settlement-friendly. A significant share of insolvency filings never progress to the mainstream CIRP because they are either withdrawn or settled at early stages. The Code and its regulations contain detailed provisions governing the withdrawal of insolvency applications, primarily through settlements pre-admission and under Section 12A of the IBC and Regulation 30A of the CIRP Regulations, 2016, post-admission. This essay explores the framework, practice, and implications of these provisions. It starts by enumerating the different ways in which settlement is possible, highlights the desirability of a settlement window for CDs and OCs, particularly MSMEs, explores whether the IBC is too settlement-friendly, and presents empirical analysis on the nature of entities that go in for a settlement. The chapter ends by discussing the role of settlements in the context of international bankruptcy laws.
2. The Options for Settlement Settlement is possible both before and after admission into the CIRP. Before admission, settlement can occur through negotiation between the petitioner and the CD, with the court's approval. After admission, cases can be withdrawn under Section 12A – either before the formation of the CoC, using Regulation 30A, or after the formation of the CoC, using the core formulation of Section 12A. There is also an “Appeals-Review-Settlement” process (hereafter, ARS) which allows a case to be appealed to a higher court – either the NCLAT or the High Court – after admission to the CIRP. It is useful to understand the nature of Section 12A and the Regulation 30A window. 2.1. Section 12A Section 12A of the IBC and the associated Regulation 30A create two different tracks for withdrawal after admission

Before the constitution of the CoC, the OC can withdraw under Regulation 30A. • The applicant may apply for withdrawal through the IRP. • No CoC approval is required, since no CoC exists.

37 • The NCLT may approve the withdrawal directly. 2. After the constitution of the CoC, • The applicant must route the withdrawal application through the Resolution Professional (RP). • The CoC must approve the withdrawal by a 90% voting share. • The RP then files the withdrawal with the NCLT. Further, if the withdrawal is sought after the issuance of an invitation for resolution plans (Expression of Interest), the applicant must provide reasons justifying such late-stage withdrawal. Thus, section 12A, read along with Regulation 30A, creates a graded framework of withdrawal: more flexibility at early stages, greater creditor control at later stages.
2.2. Regulation 30A, within Section 12A Regulation 30A cases withdrawn before the constitution of the CoC are few in number and probably represent errors in judgment of the CD, who thinks the NCLT would dismiss the OC petition rather than accept it. In the words of an RP, “If Regulation 30A is removed, there would be an increase in pre- admission settlements, since both operational creditors and corporate debtors will prefer to settle earlier”. Our study shows that cases settled after the formation of the CoC involve CDs with worse financial health at admission than those that settled earlier. Cases that get admitted mostly involve CDs in genuine distress, which may not have the financial resources to settle. They, hence, do not have much to lose by entering the insolvency process; rather, it may provide them a chance at survival. Therefore, their value after admission is more or less the same or even greater than their value with a pre-admission settlement. Hence, the CD is willing to get admitted. Similarly, we also find that an OC-initiated case that is admitted often involves an OC who is relatively weak compared to those whose cases are settled pre-admission (refer to Section 7 of this chapter for details). Cases in the CIRP can also be withdrawn at any stage through a successful appeal to the NCLAT or the High Court. The appeal could result in a dismissal of the case or a withdrawal with or without settlement. The bench of the higher court can also pass an order asking for the case to be re-admitted to the CIRP. Thus, cases petitioned at the NCLT can be studied across a continuum as shown in Figure 3.2.1.

38 Figure 3.2.1: Outcomes of cases petitioned under the IBC 3. Why CDs Want to Settle Under normal circumstances, bargaining between the OC and a CD is skewed in the latter's favour. The OC often has a lower tolerance for delayed payments, and the debtor can usually find another vendor in case the relationship breaks down. However, after the enactment of the IBC 2016, this dynamic has changed.
The CD effectively faces three possible outcomes as an insolvency petition proceeds before the tribunal: settlement, dismissal, or admission into the CIRP. As hearings progress, the CD begins to discern the likely direction of the bench—whether the case is headed toward dismissal or toward admission. If the bench appears inclined toward admission, the CD’s choice narrows to a stark trade-off between settlement and admission. Settlement is financially costly but allows the CD to retain control of the company. Admission, by contrast, is catastrophic: once a case is admitted, control of the company shifts to the IRP, and the promoters lose management authority. The company’s value also tends to decline sharply due to negative market perceptions and disrupted stakeholder confidence. Under Section 29A, in addition to other grounds for disqualification, the promoter cannot be a resolution applicant if a year passes after admission, which is usually the case. For this reason, in cases that seem headed toward CIRP, CDs are generally eager to settle before admission to CIRP, even at some cost. The OC, however, confronts a very different calculus. For them, the same three outcomes—settlement, dismissal, and admission—carry distinct implications. Admission typically yields a median recovery of barely 5 percent (Iyer & Prasad, 2024). Settlement, by contrast, is highly attractive, as it offers immediate payment. Dismissal is disappointing but not catastrophic: it likely means the creditor will At NCLT petition stage • Mutual, bilateral negotiations and settlement between OC and CD, approved by NCLT U/s 12A - Reg. 30A • Before C0C formation • Mutual settlement, under IRP supervision • NCLT approval U/s 12A - after CoC formation • RP oversees settlement process • 90% CoC approval required • NCLT gives final approval CIRP process commencement • Result: Resolution or Liquidation Appeal to NCLAT or High Court possible at any stage – Can result in dismissal or settlement Pre-admission Post-admission to IBC

39 not recover from this particular debtor, yet the loss is manageable since the defaulting firm is usually only one among several accounts. Hence, even a full default, while financially straining, may rarely pose an existential threat. This is particularly true for medium-sized suppliers who maintain diversified customer portfolios.
Given these considerations, the negotiation in a case seemingly headed for admission becomes a bargaining game between a CD with everything at stake and a creditor negotiating from a position of limited exposure and diminished expectations. In this situation, a CD should always settle unless they miscalculated the bench's intent by mistakenly assuming the case will be dismissed.
4. A Game Theoretic View A simple exercise using the game-theoretic concept known as the Nash Bargaining Solution (NBS) can illustrate why settlement is far preferable to admission for a healthy CD and the amount that the CD would be willing to pay.
The NBS provides a systematic way of determining how two rational parties might divide the gains from cooperation when each has the option to walk away. It is based on the idea that the eventual agreement will maximize the product of each party’s incremental benefit over what they would receive if negotiations broke down—their so-called disagreement point. In the context of insolvency, this framework can be used to model the bargaining interaction between a CD seeking to avoid admission into the CIRP and an OC insisting on payment. By assigning payoffs to each side under settlement and under admission, the NBS helps reveal just how much the CD might rationally be willing to concede to avert the catastrophic consequences of admission. The following is a simple bargaining problem between a CD and an OC. If they reach an agreement, the OC receives a payment of Rs.X, and the CD pays the same amount. If no agreement is reached and the case is admitted into the CIRP, both parties incur losses. The OC bears only a modest loss— primarily legal and procedural costs—say around Rs.10 lakh. The CD, however, faces a vastly greater loss: admission into CIRP implies loss of control, reputational damage, and a sharp decline in enterprise value. Given that the OC’s claim is greater than Rs. 1 crore, the CD would have revenues significantly in excess of Rs. 1 crore. Therefore, we can safely assume that the loss of value would be of the order of Rs.10 crore. Under the NBS, the cooperative outcome is found by splitting the surplus that can be saved through agreement. The agreement itself generates no surplus, since the creditor’s gain is equal to the debtor’s loss. Hence, in this case, the total bargaining surplus is the difference between the OC’s and CD’s losses from non-settlement, Rs. 10 lakhs and Rs. 10 crore, respectively. The extra loss avoided by the debtor (Rs.9.9 crores) is divided equally under the NBS. Accordingly, the CD would be willing to pay Rs.4.45

40 crore to the OC in settlement. This amount often exceeds the original claim itself, showing that a rational, solvent CD would prefer to settle rather than risk admission into CIRP. The analysis above demonstrates that frequent pre-admission settlements are an inevitable consequence of the IBC 2016. Of the more than 30,000 cases withdrawn before admission until 31 March 2025, over 22,000 were OC-initiated (NCLT official website). However, there are also a significant number of OC- initiated cases that got admitted into the CIRP – out of around 8,300 cases admitted to IBC, close to 3,900 were OC-initiated, of which around 1,660 were either withdrawn under Section 12A or closed through the ARS route (IBBI Quarterly Newsletter, Jan-March 2025).
5. The Imperative of Frequent Settlement for the Rational MSME OC Let us also understand that if the possibility of frequent settlements with reasonably high recovery rates did not exist, a rational MSME OC would rarely have reason to use the CIRP. To do so, we analyse the incentives for an OC with the threshold claim of Rs. 1 crore to file a petition under the IBC.
It is fair to assume that the decision of an OC (and, indeed, of an FC as well) to initiate insolvency proceedings under the IBC is fundamentally an economic one. Filing a case before the NCLT involves high legal and procedural costs. While these are meant to be repaid to the petitioner, it is fair to assume that an MSME-OC petitioner, particularly, does not factor in the possibility of repayment in the distant future when making plans. The data presented in Table 3.5.1 illustrates the expected payoffs for an OC filing a petition against a CD under two distinct scenarios, both involving the minimum claim threshold of Rs.1 crore – one with Regulation 30A in force and the other, without. In each of these scenarios, we find two cases that achieve similar returns – one with a high settlement probability and one with a high recovery rate.
Consider an OC with an outstanding claim of Rs.1 crore against a CD. To use the IBC route, the OC must incur substantial costs: around Rs.50,000 to file the notice, roughly Rs.5 lakh in lawyer’s fees over ten hearings, and about Rs.10 lakh as its share of IRP-related costs if the case is admitted into CIRP. Together, this amounts to an expected outlay of about Rs.15.5 lakh11. For an OC whose cost of capital is around 15 percent, the question is whether the expected return from going to the NCLT is high enough to cover both the direct legal costs and the opportunity cost of funds tied up in the process. Our calculations show that this is true only if there is a meaningful probability of settlement before admission and if the settlement comes with a reasonably high recovery rate. We assume that the probability of admission into CIRP is the same as the current frequency of admission, i.e. 0.15, and the probability of pre-admission disposal (rejection, dismissal or withdrawal with or without settlement) is 11 Figures based on interviews.

41 0.85 (Figure 2.4.2). Within the admission branch, we use realistic IBBI-based outcome probabilities: roughly 30 percent of admitted cases end in liquidation (with zero recovery for OCs), 10 percent in resolution (with a median repayment of around 5 percent), 7 percent in withdrawals before CoC via Regulation 30A (with 86% percent repayment), 14 percent in withdrawals under Section 12A after CoC formation (with 4% percent repayment12), and about 22 percent in ARS routes, where OCs typically recover little or nothing13. Discounting a two-year resolution recovery at 15 percent and using these probabilities, the expected return from the admission branch alone is strongly negative: in the base case with Regulation 30A in place, the OC expects to recover only about Rs.6.6 lakh in present value terms against costs of Rs.15.5 lakh, implying an expected loss of roughly Rs.8.9 lakh if we imagine that every filed case got admitted. The picture changes when we introduce the pre-admission settlement option, which is integral to how the IBC is used in practice. As mentioned in Section 3 of Chapter 2, there are four possibilities before admission: dismissal, rejection, withdrawal without settlement, and withdrawal with settlement. Unlike in the post-admission scenario, we do not have data on recovery outcomes in the pre-admission process (apart from knowing that the OC receives nothing in the first three cases). Hence, we seek combinations of parameter values that allow the OC to earn a return on their IBC expenses that matches their cost of capital.
Table 3.5.1: The Business Case for Filing if Admission into CIRP is Inevitable Column 1 Column 2 Column 3 Column 4 With Regulation 30a Without Regulation 30a Assumptions about OC claim and costs OC claim amount (Rs.) 1,00,00,000 1,00,00,000 1,00,00,000 1,00,00,000 Cost of filing notice
50,000 50,000 50,000 50,000 Lawyer fees per hearing 50,000 50,000 50,000 50,000 Number of hearings 10 10 10 10 total lawyer fees
5,00,000 5,00,000 5,00,000 5,00,000 IRP fees 10,00,000 10,00,000 10,00,000 10,00,000 Total cost (Rs.) 15,50,000 15,50,000 15,50,000 15,50,000 Assumptions about IBC outcomes A. Pre-admission outcomes Prob (pre-admission disposal) 0.85 0.85 0.86 0.87 Prob (pre-admission disposal via settlement) 0.20 0.30 0.20 0.24 Average recovery in pre-admission settlement 26% 20% 31% 26% B. Post-admission outcomes Prob (CIRP admission) 0.15 0.15 0.14 0.13 Prob (Liquidation) 0.30 0.30 0.30 0.30 12 For Reg.30A and Section 12A withdrawals, we calculate the average repayment rate by dividing the total settlements by total claims from the IBBI data.
13 In the ARS route there are additional legal expenses and additional return. We assume they cancel out.

42 Prob (Resolution) 0.10 0.10 0.10 0.10 Prob (Settlement u/s 12A using Reg.30A) 0.07 0.07 0.07 0.07 Prob (Settlement u/s 12A after CoC) 0.14 0.14 0.14 0.14 Prob (ARS settlement) 0.22 0.22 0.22 0.22 Realization in Liquidation 0.00 0.00 0.00 0.00 Median realization in Resolution 0.05 0.05 0.05 0.05 No. of years to resolution 2 2 2 2 Average discount rate applicable to OC 0.15 0.15 0.15 0.15 Median realization in Reg.30A settlement 0.86 0.86 0.00 0.00 Median realization in Section 12A settlement 0.04 0.04 0.04 0.04 Median realization in ARS settlement 0.00 0.00 0.00 0.00 Computing expected outcomes R1. From admission to CIRP Resolution proceeds (PV) 3,78,072 3,78,072 3,78,072 3,78,072 Proceeds from Reg.30A settlement 86,00,000 86,00,000 0 0 Proceeds from Section 12A settlement 4,00,000 4,00,000 4,00,000 4,00,000 Proceeds from ARS settlement (PV) 0 0 0 0 Expected realization from CIRP admission 6,55,627 6,55,627 95,767 95,767 Net realization from CIRP -8,94,373 -8,94,373 -14,54,233 -14,54,233 R2. From pre-admission disposal Net expected recovery from pre-admission disposal channel 20,50,000 14,50,000 25,50,000 20,50,000 Net expected recovery from petitioning under IBC 2,14,344 2,35,594 2,34,546 2,36,838 Return on costs incurred 13.83% 15.20% 15.13% 15.28% One such combination is the following: suppose that 20 percent of cases disposed pre-admission result in settlement14, and that in such settlements the OC recovers about 26 percent of its claim, i.e. around Rs.26 lakh. After netting out indicative costs, this yields a net recovery from pre-admission settlement of about Rs.20.5 lakh. When we weight this by the probability of pre-admission disposal (0.85) and combine it with the negative expected value from the admission branch (weighted by its 0.15 probability), the overall net recovery comes to about Rs.2.14 lakh, which corresponds to an internal rate of return of roughly 14 percent—a little short of the OC’s cost of capital. In other words, with a 20 percent chance of pre-admission settlement and a repayment rate of around 26 percent, a risk-neutral 14 Note that we are computing and reporting required probabilities of settlement within the cases disposed pre-admission. The required probability of settlement for the process as a whole would be the probability of disposal pre-admission multiplied by the required probability of settlement within the cases disposed pre-admission. For example, if the required probability within the cases disposed pre-admission is 20%, then the probability for the process as a whole is 0.85 (the probability of pre- admission disposal) multiplied by 0.2, i.e. 17%.

43 creditor just about breaks even. If a creditor is risk-averse, they would need an even higher rate of return.
This logic also shows how sensitive the decision is to the probability and quality of pre-admission settlement. If the probability of pre-admission settlement is higher, for example, 0.3, then the OC can tolerate a lower settlement repayment rate of 20% and still earn a 15 percent return. Conversely, if the probability of settlement is less than 0.2, then the required settlement repayment rate has to be higher than 26 percent to compensate for the losses expected in the admission branch.
Most importantly, if we set the probability of pre-admission settlement effectively to zero, the OC would be staring at a strongly negative expected return. A rational OC with a Rs.1 crore claim would simply not use the IBC route if there were no realistic chance of settling before admission. The break-even threshold for filing under these assumptions, if admission is inevitable and Regulation 30A is in force, is around Rs.2.25 crore. If Regulation 30A is withdrawn, the break-even threshold for filing, if admission is inevitable, rises to around Rs.12.5 crores, effectively excluding smaller MSME OC- applicants from petitioning under the IBC. 5.1. Allowing for Risk-Aversion The calculations presented above implicitly assume that OCs behave as risk-neutral agents—that is, they evaluate the IBC option purely on the basis of expected monetary returns. In reality, most OCs, especially MSMEs, are risk-averse and often loss-averse: they weigh losses more heavily than equivalent gains and are reluctant to expose themselves to outcomes with high downside risk. If we incorporate even modest risk aversion into the evaluation of the IBC option, the breakeven conditions for using the Code become substantially more stringent. Since the post-admission branch of the IBC tree carries very large downside risk—high legal costs combined with low expected recovery—the OC would require higher settlement probabilities and higher settlement repayment rates in the pre- admission stage to compensate for these risks. Likewise, loss aversion amplifies the psychological cost of the worst-case scenario (admission followed by liquidation or ARS), further increasing the required pre-admission payoff to make the IBC route attractive. In short, once we recognise that typical OCs are risk-averse and loss-averse, the threshold for the “IBC threat” to become economically attractive is significantly higher. To see this concretely, consider the earlier example of an OC with a Rs.1 crore claim and an expected cost of Rs.15.5 lakh. Under risk neutrality, a pre-admission settlement probability of 0.20 combined with a settlement repayment rate of 26% was just enough to yield a 15% ROI overall. Now introduce mild risk aversion, for example, with a concave utility function such that the utility of a rupee gained is only 0.8 of the utility of a rupee lost. Under this assumption, the disutility of the worst-case outcome (a

44 Rs.15.5 lakh loss) is about 20% stronger than the utility of an equivalent gain. When we adjust the expected value calculation to reflect this, the OC now requires either: • a settlement probability of close to 0.3 (instead of 0.20) at a 26% repayment rate, or • a settlement repayment rate of 35% (instead of 26%) at a 0.20 settlement probability, in order to break even on a risk-adjusted basis. With moderate loss aversion, often modelled as losses counting twice as much as gains, the breakeven conditions become even tighter. These numbers clarify the central point: risk-averse or loss-averse OCs require substantially better pre-admission settlement terms than risk-neutral OCs, and therefore, the economic viability of the IBC for OCs is even more dependent on strong, credible, and high-quality settlement opportunities before admission.
5.2. The Role of Regulation 30A Our second set of calculations (Columns 3 and 4 of Table 3.5.1) explores what happens if Regulation 30A is removed (note we assume OCs are risk-neutral). With Regulation 30A in place, withdrawals after admission but before the CoC formation are a “do-or-die” moment for the CD: the case has already been admitted, the promoters are about to lose control, and the shadow of the CoC looms large. Unsurprisingly, when cases are withdrawn under Regulation 30A, recoveries for OCs can be very high; the current average repayment is 86 percent of the claim (Rs.86 lakh) in that branch. This high-recovery tail event precisely softens the blow of the otherwise poor outcomes in resolution, liquidation, and ARS, and is part of what makes the IBC threat credible. If Regulation 30A is removed, however, that high-recovery branch disappears. In Column 3, we assume that all cases in the erstwhile Reg. 30A branch would now be settled pre-admission at the sort of recovery levels one sees in ordinary pre-admission settlements. When we recompute the expected value from the admission branch under this assumption, the OC’s expected loss from admitted cases becomes even larger. To preserve an overall expected return of around 15 percent, the OC now needs a higher repayment rate in pre-admission settlements: with a 20 percent settlement probability, the break-even settlement rate required jumps to around 31 percent instead of 26 percent (Column 3). With a 26 percent settlement rate, the required settlement probability jumps to around 24 percent instead of 20 percent (Column 4).
Thus, Regulation 30A, by creating a window of high repayment, reduces the necessary settlement probabilities and repayment rates in the pre-admission phase for the IBC to be viable for OCs. Removal of Regulation 30A would require the process to be tweaked to facilitate higher settlement probabilities and repayment rates in the pre-admission phase.

45 The following table shows the combinations of settlement probability and expected recovery that allow an OC to break-even on the procedural expenses incurred.
Table 3.5.2. Pairwise Pre-Admission Settlement Parameters Needed to Break Even (Probability vs. Recovery), by Risk Preference and Reg. 30A OC Risk preference Reg. 30A Settlement probability Settlement recovery rate Description Risk-neutral Yes 0.20 26% Lower settlement probability is workable only with a higher recovery rate (to still roughly meet ~15% return). Baseline breakeven case.
Risk-neutral Yes 0.30 20% Higher settlement probability lets the creditor accept a lower recovery rate and still hit ~15% return. Risk-neutral No 0.20 31% Without the high-recovery 30A Reg, the same 0.20 probability needs a higher recovery rate to break even (~15%).
Risk-neutral No 0.24 26% Without 30A, holding recovery at 26% requires a higher settlement probability than with 30A to reach ~15%. Risk-averse (mild) Yes ~0.30 26% With risk aversion, keeping recovery at 26% requires higher settlement probability than the risk-neutral case to break even (risk-adjusted). Risk-averse (mild) Yes 0.20 35% With risk aversion, keeping probability at 0.20 requires a higher recovery rate than the risk-neutral case to break even (risk-adjusted). The overarching conclusion from this section is twofold. • First, an OC with a Rs.1 crore claim would not rationally pursue the IBC route if admission into CIRP were the only outcome, because the expected value of post-admission recoveries is strongly negative given current recovery statistics. The only way the economics make sense is if there is a reasonably high probability of pre-admission settlement at a substantial repayment rate. • Second, Regulation 30A, by creating a high-pressure post-admission settlement window with very high OC recoveries, improves the expected payoff from the IBC. Removing 30A would

46 lower the expected value of the admission branch and force OCs to demand even higher pre- admission settlement rates or higher settlement probabilities to break even. Put differently, without a meaningful chance of settlement—either pre-admission or under 30A—a rational OC would stay away from the CIRP altogether. This analysis reinforces the critical role of pre-admission settlements for MSME OCs in the functioning of the IBC. If the possibility of frequent settlement were not available, the IBC would become unattractive for them. The question that automatically arises: is the IBC too settlement-friendly?
6. When would the IBC be too Settlement-Friendly? A natural concern in any insolvency framework is whether the system becomes too settlement-friendly, to the point where filing an IBC petition becomes an arbitrage opportunity for OCs—especially MSMEs. A simple way to formalize this is to compare an MSME OC-applicant’s expected net recovery from filing (i.e., expected recoveries minus legal and procedural costs) to the total economic exposure involved in the case, which consists of both the expenses incurred and the portion of the claim that remains unrecovered. If the ratio of these two—the effective return on investment—exceeds the MSME applicant’s cost of capital, typically around 15%, then the MSME would, in principle, be better off filing IBC cases as a routine “profit-generating activity”. That would be undesirable from a policy perspective, as it would distort incentives and encourage strategic litigation. However, the calculations presented earlier show that we are nowhere near such a situation. Even under realistic assumptions about pre-admission settlement probabilities and repayment rates, the ROI for an OC measured as an MSME’s expected net recovery from filing divided by the total expenses incurred (not taking the portion of the claim that remains unrecovered) barely reaches the 15% threshold, and usually falls below it once risk aversion is accounted for. The return falls further in the likely event of a withdrawal of Regulation 30A.
In other words, the IBC today does not offer MSMEs any systematic arbitrage; rather, it offers a narrow and fragile pathway to break even, contingent on a favourable pre-admission settlement. One could also ask whether the IBC offers a simpler recovery process than other channels available to MSMEs. Here, when evaluating the usefulness of different recovery channels available to MSMEs, it is essential to ask what recoveries would have occurred in the absence of that channel. This counterfactual helps us distinguish between mechanisms that create genuine additional recovery and those that merely facilitate transactions that would have happened anyway.
In the case of the IBC, the answer is relatively clear: the settlements and repayments triggered by the threat of admission into CIRP would, in all likelihood, not have materialized through any other mechanism available to MSME OC-applicants. The IBC creates a unique pressure point—loss of

47 control for the CD and a time-bound adjudicatory process—that produces recoveries that simply would not have arisen under the pre-IBC regime. By contrast, for other channels such as the TReDS platform, this counterfactual is much less straightforward. As mentioned in Chapter 2, close to 80% of the TReDS transaction value comes from reverse factoring, in which the corporate buyer borrows from a financial institution at a concessional interest rate to pay its MSME suppliers. One could plausibly argue that many of these payments would have been made even without the TReDS platform, except that the CD might have had to borrow at a slightly higher interest rate or rely on its existing credit lines. The marginally higher interest rate would have reduced repayment rates, but only marginally so. In other words, with respect to the bulk of transactions on its portals, TReDS may often be facilitating and discounting transactions that were already likely to occur, rather than generating entirely new recovery events for MSMEs. This contrast highlights why, for OCs—especially MSMEs—the IBC plays a uniquely potent role: it produces recoveries that are, to a considerable extent, additional to what the normal course of business would have yielded. To that extent, the IBC does represent a unique opportunity for MSMEs. However, its pre-eminence is as much a reflection of the fact that other channels do not work well as it is of the fact that the IBC is excessively favourable. Further, there is a very important caveat: the IBC settlement route only works for a small segment of MSME OCs, as the following section demonstrates.
7. The nature of entities that go in for a settlement It is clear by now that the intent of a petitioning OC would be to persuade the CD to settle early rather than let the case go into the full CIRP; in fact, they would prefer a pre-admission settlement. A financially sound OC can afford the expenses of filing a case under the IBC and might also have the negotiating and persuasion power to get the CD to settle.
Simultaneously, a CD also has an incentive to settle cases filed against it under Section 7 or 9 of the IBC, especially if it has much to lose. Going through a CIRP involves loss of control over the company and a significant loss of value, besides reputational damage. CDs that are financially strong would have a greater incentive to settle at the earliest. The CDs admitted are those unable to settle pre-admission, and those that go through the full CIRP are those unable to use Section 12A either, for settlement. It is unlikely that a financially healthy OC or CD would not use the settlement mechanisms available pre- or post-admission. We evaluated the financial health of samples of OCs and CDs that settled versus those that did not, as of the date of admission of the case, by/ against them.
Analysis of OC-applicants: We separated the OCs who settled under Regulation 30A from the rest. The latter includes those that settled under Section 12A after CoC formation, or proceeded to the full CIRP. We use settlement under Regulation 30A as a proxy for the pre-admission stage, and assume

48 that all cases settled within 30 days of admission used Regulation 30A since the RP has 30 days to constitute the CoC, after which any settlement requires the consent of 90% of the CoC.
We first examined their solvency, by classifying the sampled firms into the four solvency categories as described in Table 1.2.2. The solvency matrices for OC-applicants are shown in Table 3.7.1a, where Panels A and B describe the solvency status of OCs that settled pre-admission and the rest, respectively. Table 3.7.1a. Solvency matrices for OC-applicants – pre-admission settled vs. others Solvency Panel A: OCs that settled under Reg.30A Panel B: OCs that settled later or did not settle MSME non-MSME Total MSME non-MSME Total Healthy 1 6 7 6 29 35 Solvent 5 4 9 7 14 21 Distressed/ Reducing NW 0 0 0 1 3 4 Insolvent 2 0 2 2 3 5 8 10 18 16 49 65 We found that majority of the OC applicants - 89% of Panel A and 86% of Panel B OCs - were healthy or at least solvent. While these proportions are quite close, we found the differences between the MSME and non-MSME sub-samples more informative. Though small, our Reg.30A sub-sample had 100 percent of the non-MSME OC-applicants healthy or solvent, while the corresponding percentage in Panel B was 88% (43 of 49 OCs). The results in the MSME cohorts though, showed a relatively more even spread of OC-applicants across the four categories, though the healthy and solvent applicants made up the significant majority.
Table 3.7.1b. ANOVA across OC-applicants settling under Reg.30A and other OC-applicants Groups Count Average OCs that settled under Reg.30A 18 1.00 OCs that settled later or did not settle 65 1.68 F-statistic 10.020 P-value 0.002*** An ANOVA of the difference in solvency ratings across the two categories in Table 3.7.1b shows a statistically significant difference at 99% confidence levels, with the Reg.30A settlements significantly stronger. The average solvency rating of even the sample OC-applicants that settled under Section 12A (a majority of our Panel B cohort were Section 12A settlers rather than those who went into the full process), was in the Healthy-to-Solvent range, though on average, less healthy than the Panel A cohort.

49 A liquidity analysis of the two sets of OC-applicants aligns with the solvency trends, as summarized in Table 3.7.2. The maximum number of cases in both sub-samples had the Low DSO-Low DPO combination, suggesting that most of these OC-applicants did not experience liquidity stress at the time of case admission, nor did they delay payments to their creditors. The High DSO-High DPO combination was a very small proportion in both cohorts. We did not find any significant difference in liquidity between the Reg.30A settlers and the rest of the OC-applicants. More than 60% cases in both sub-samples held less than 100 days’ sales in receivables. Table 3.7.2. DSO-DPO matrices for OC-applicants settled under Reg.30A versus the rest Panel A: OCs that settled under Reg.30A Panel B: OCs that settled later or did not settle Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 1 0 2 3 3 2 3 8 Medium 1 5 1 7 1 13 9 23 Low 0 0 8 8 0 5 25 30 Total 2 5 11 18 4 20 37 61 As in the previous chapter, we conducted a sub-test for the MSMEs, taking the 45-day cut-off for outstanding receivables and payables (results not shown here). Our limited sample shows greater liquidity strain among MSMEs, particularly among the non-settlers in Panel B. While five of the eight MSME OC-applicants that settled under Reg.30A recovered their dues within 45 days, only two of the twelve non-settling MSMEs managed it. Our data also shows that MSMEs take longer to pay their creditors (higher DPO) than the average time taken to receive dues from their customers; while only 3 MSMEs fell in the DSO>100 days bracket in Panel A, six of them took more than 100 days to pay their creditors, the corresponding numbers being 6 and 7 out of 12 for the Panel B MSMEs. This could reflect liquidity strain specific to MSMEs, or simple wilful delay in payments. More data and deeper investigation are required to establish this reasoning and possible causality. Analysis of CDs: We compared CDs that settled at any stage (Reg.30A, Section 12A, or ARS) with those that went through the full CIRP, using the solvency and DSO-DPO matrices as earlier. We also tested whether there was any significant difference between CDs that settled under Reg.30A and those that settled post-admission under Section 12A or ARS. Table 3.7.3a maps the solvency matrices for CDs that settled versus those that did not, at the point of admission, separately for OC-initiated and FC-initiated cases. We found the CDs that settled to be financially stronger than the rest, the difference being statistically significant. This was seen among both OC-initiated and FC-initiated cases. The data also shows that, while the majority of CDs in the OC-initiated cases in our sample settled, the majority of CDs in the FC-initiated cases underwent the

50 full CIRP. Since an average CD in FC-initiated cases is financially weaker than those in OC-initiated ones (as seen in Section 5.2 of Chapter 2), settling is not an easy option for them.
Table 3.7.3a. Solvency matrix for CDs that settled versus those that went through the full CIRP in OC versus FC-initiated cases
Solvency Panel A: CDs in OC-initiated cases that settled Panel B: CDs in OC-initiated cases that underwent full process MSME non- MSME Total MSME non- MSME Total Healthy 41 44 85 1 2 3 Solvent 58 37 95 29 21 50 Distressed/ Reducing NW 14 14 28 14 11 25 Insolvent 53 40 93 60 48 108 166 135 301 104 82 186 Solvency Panel C: CDs in FC-initiated cases that settled Panel D: CDs in FC-initiated cases that underwent full process MSME non- MSME Total MSME non- MSME Total Healthy 1 3 4 1 2 3 Solvent 17 5 22 28 22 50 Distressed/ Reducing NW 3 3 6 18 14 32 Insolvent 13 5 18 75 134 209 34 16 50 122 172 294 60% of the CDs in OC-initiated cases (Panel A), both MSME and large, were at least solvent. On the other hand, among the CDs that underwent the full CIRP (Panel B), less than 30% were solvent, both among MSMEs and large CDs. More than 50% of the CDs in FC-initiated cases that settled (Panel C) were healthy or at least solvent. The proportion of solvent and healthy CDs dropped precipitously among those that went for the full CIRP (Panel D), particularly for the non-MSME subsample. Less than 20% of the CDs were solvent in Panel D as against close to 30% in Panel B.
The ANOVA for the difference in solvency scores for the four sets of CDs in Table 3.7.3a is presented in Table 3.7.3b, part (a). The differences are statistically significant in both OC- and FC-initiated cases. Among the MSME CDs within these samples, we found similar results – those that settled were stronger at admission. However, though statistically significant, the F-statistic of the difference between the settled and full-process MSME CDs in FC-initiated cases (Panel D of 3.7.3b (b)) was far lower, indicating a lesser degree of difference than between the other sets.

51 Table 3.7.3b. ANOVA for financial health of CDs that settled versus those that went through the full CIRP
a) All CDs in our sample Panel A: In OC-initiated cases Panel B: In FC-initiated cases Groups Count Average Variance Groups Count Average Variance Settled 301 2.43 1.43 Settled 50 2.76 1.08 Full process 186 3.28 0.84 Full process 294 3.52 0.65 F-statistic 69.002 F-statistic 34.562 p-value 0.000*** P-value 0.000*** b) MSME CDs Panel C: In OC-initiated cases Panel D: In FC-initiated cases Groups Count Average Variance Groups Count Average Variance Settled 166 2.48 1.39 Settled 34 2.82 1.00 Full process 104 3.28 0.82 Full process 122 3.37 0.75 F-statistic 35.149 F-statistic 9.873 p-value 0.000*** P-value 0.002*** We further tested whether CDs that settled pre-admission (using Regulation 30A cases – those settled within 30 days of admission as proxy) were stronger than those that settled post-admission. Table 3.7.4a shows that 64% of CDs in OC-initiated cases that settled under Regulation 30A were healthy or at least solvent. This result was primarily driven by the large firms, where 71% were healthy or solvent; among the MSMEs, 56% were healthy or solvent (Panel A). Among the FC-initiated cases, a larger proportion of CDs, at 71%, were solvent (Panel C). This proportion held for both the MSMEs and non-MSME firms. We suspect the high proportion of healthy, solvent CDs in Panel C is due to the dominance of manufacturing firms; service firms were negligible in this group. The small number of observations used for analysis in Panel C though, reduces the confidence of results derived from it.
Table 3.7.4a. Financial health of CDs, settled u/s 12A and Regulation 30A a) OC-initiated cases Solvency Panel A: CDs settled under Reg.30A Panel B: CDs settled u/s 12A MSME non-MSME Total MSME non-MSME Total Healthy 9 16 25 5 7 12 Solvent 9 9 18 9 0 9 Distressed/ Reducing NW 5 2 7 5 3 8 Insolvent 9 8 17 13 6 19 32 35 67 32 16 48 b) FC-initiated cases

52 Solvency Panel C: CDs settled under Reg.30A Panel D: CDs settled u/s 12A MSME non-MSME Total MSME non-MSME Total Healthy 1 3 4 0 0 0 Solvent 4 2 6 13 3 16 Distressed/ Reducing NW 0 2 2 3 1 4 Insolvent 2 0 2 11 5 16 7 7 14 27 9 36 Among the CDs that settled u/s 12A (beyond Reg.30A), we found that only 44% were healthy or solvent as of the date of admission, for both OC-initiated and FC-initiated cases. This pattern of decrease in healthy and solvent firms as settlement duration increases supports the logic that healthy CDs that have more to lose are likely to volunteer to settle with their creditors at the earliest to avoid monetary costs and reputation damage. Our ANOVA results bear this out, as seen in Table 3.7.4b. While the Reg.30A CDs are significantly healthier than those settled outside it, u/s 12A, this difference is more statistically pronounced among the FC-initiated cases. As observed previously among the settled versus full process CDs, we found that the majority of CDs in OC-initiated cases settled under Reg.30A, whereas only a minority in FC-initiated cases settled early, further confirming a marked difference in their financial situation at admission. Table 3.7.4b. ANOVA of financial health between Reg.30A and Section 12A settled CDs a) All CDs Panel A: In OC-initiated cases Panel B: In FC-initiated cases Groups Count Average Variance Groups Count Average Variance CD-12A 48 2.71 1.53 CD-12A 36 3.00 0.91 CD-Reg.30A 67 2.24 1.46 CD-Reg.30A 14 2.14 1.05 F-statistic 4.144 F-statistic 7.776 p-value 0.044** P-value 0.008*** b) MSME CDs Panel C: In OC-initiated cases Panel D: In FC-initiated cases Groups Count Average Variance Groups Count Average Variance CD-12A 32 2.81 1.32 CD-12A 27 2.93 0.92 CD-Reg.30A 32 2.44 1.42 CD-Reg.30A 7 2.43 1.29 F-statistic 1.646 F-statistic 1.394 p-value 0.204 P-value 0.246 The subset of MSME CDs under OC-initiated and FC-initiated cases also showed better solvency (lower average score) for the Reg.30A cohorts, though the difference in mean solvency scores from those that

53 settled u/s 12A was statistically insignificant. Many MSMEs are unaware of the gravity and time-bound nature of an IBC petition, and miss the window for pre-admission or Regulation 30A settlement. This may be one of the reasons for the insignificant difference between the characteristics of the pre- and post-admission settler cohorts among the MSME CDs. We examined the liquidity situation of these sets of CDs using the DSO-DPO matrices. Table 3.7.4(c) shows that majority of the large CDs that had settled were in the Low and Medium combinations, both among OC and FC-initiated cases (Panels A and C), compared to the ones that went into the full CIRP; the latter were primarily concentrated in the Medium and High combinations for both DSO and DPO, showing relatively better liquidity situation of the settlers. Table 3.7.4c: DSO-DPO analysis comparing large CDs that settled versus those that went into full CIRP
a) OC-initiated cases against large CDs Panel A: Settled Panel B: Full CIRP Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 16 8 12 36 15 14 5 34 Medium 4 31 19 54 8 13 8 29 Low 2 8 32 42 2 5 8 15 22 47 63 132 25 32 21 78 b) In FC-initiated cases against large CDs Panel C: Settled Panel D: Full CIRP Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 3 0 2 5 42 9 14 65 Medium 1 0 1 2 19 28 13 60 Low 0 1 7 8 3 15 12 30 4 1 10 15 64 52 39 155 In contrast, the data on MSME CDs did not show such a clear demarcation, as in Table 3.7.4(d). MSME CDs that settled on OC-initiated cases were evenly distributed across the three levels for both DSO and DPO. Here too, while most were present in the Low-Medium DSO, their DPO was Medium or High. This pattern was visible among the MSMEs that settled on FC-initiated cases, too. This shows that while they were receiving payments within (or payments were delayed up to) 360 days, they were taking more than (or had outstanding exceeding) 360 days to pay their OCs. Delays of 3-4 months can significantly disrupt working capital management for MSMEs, given their lower resilience, leading them to delay and default on their payments.

54 Table 3.7.4d. DSO-DPO analysis comparing MSME CDs that settled versus those that went into full CIRP
a) OC-initiated cases against MSME CDs Panel A: Settled Panel B: Full CIRP Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 22 23 10 55 47 6 2 55 Medium 7 22 15 44 9 21 4 34 Low 3 11 33 47 0 3 7 10 32 56 58 146 56 30 13 99 b) In FC-initiated cases against MSME CDs Panel C: Settled Panel D: Full CIRP Debtor days (DSO) Debtor days (DSO) High Medium Low Total High Medium Low Total Creditor days (DPO) High 6 6 3 15 52 13 3 68 Medium 3 1 4 8 11 9 7 27 Low 2 0 4 6 4 3 7 14 11 7 11 29 67 25 17 109 Upon resetting the ‘Low’ DSO and DPO categories for MSMEs at the 45-day mark, we found that only 31 of the 146 in Panel A and 6 of the 99 in Panel B received their dues within 45 days, on average. The trend for DPO was also similar – 21 of the 146 in Panel A and only 5 of the 99 in Panel B were in the Low-DPO category. The split was similar for the FC-initiated cases in Panels C and D. The average DPO score was higher than the average DSO score for all four panels in Table 3.7.4(d) (higher scores indicate longer delays and liquidity stress). Paired t-tests (not shown here) revealed this difference to be statistically significant at the 5% level for the OC-initiated CDs, but insignificant for the FC-initiated ones. This is an interesting result that provides greater insights into MSME behaviour and distress timelines.
MSME CDs, in their initial stages of stress, delay OC dues while honouring FC dues to manage their liquidity. At this stage, DPOs will increase and diverge from DSOs. If the OCs petition the CD at this stage, the case could either lead to settlement (Panel A) or CIRP proceedings (Panel B). In either case, the average DPO will be higher than the DSO. Assuming the OCs do not petition the CD, and the liquidity stress continues to build for the CD, the DSO is likely to stretch, and the CD is likely to not only delay and default on OC dues but also on FC dues. Therefore, when the FC initiates a case against the CD at the NCLT, the DSO and DPO are both high, and not significantly different. By the time the FC initiates a case, the CD is in a worse liquidity and solvency situation, and is less likely or able to settle, as seen in Table 2.5.2.2 in the previous chapter. Our data shows that OCs more often

55 petition MSMEs than FCs do. The reasons for this could range from the need for OCs to recover their dues, while FCs enjoy credit guarantees, and would not be interested in pursuing small quantum recovery through the IBC route. Finally, as expected, the averages were lower (better liquidity scores) for the MSMEs that settled than those that went into full CIRP.
For MSMEs, particularly, liquidity stress is more severe than for larger firms that have multiple avenues to manage liquidity. We believe liquidity stress in MSMEs can quickly cascade into solvency stress. To test this, we examined the association between liquidity and solvency by computing pairwise correlation coefficients among the Solvency, DSO, and DPO scores. The results are shown in Table 3.7.5. We found that the correlation between solvency and liquidity was stronger in large firms than among MSMEs, especially the Solvency-DPO score pairs. The strong DSO-DPO correlations evidence that delays in receipts can cascade into delays in payments, leading to defaults. Consistent bad debts over time can erode equity and lead to insolvency, reflected in the positive, statistically significant Solvency-DPO correlations. The correlations clearly illustrate the severity of liquidity issues for MSMEs, even when their balance sheet may be solvent.
Table 3.7.5. Pairwise correlation between Solvency and Liquidity scores across sub-sets of CDs No. of pairs Solvency-DSO Solvency-DPO DSO-DPO I. Large firms (with 100-day cut-off for Low DSO and DPO) OC-initiated, settled 128 0.277*** 0.327*** 0.415*** OC-initiated, continued 94 0.217** 0.278*** 0.574*** FC-initiated, settled 15 0.267 0.517** 0.567** FC-initiated, continued 150 0.109 0.182** 0.294*** II. MSMEs (with 45-day cut-off for Low DSO and DPO) OC-initiated, settled 141 0.169** 0.313*** 0.432*** OC-initiated, continued 89 (0.071) 0.123 0.520*** FC-initiated, settled 29 (0.249) (0.143) 0.387** FC-initiated, continued 97 0.036 0.019 0.364*** We re-computed the correlations for MSMEs using the 100-day cut-offs, as for the large firms, to examine whether the results differed. We found that the correlation coefficients increased across almost all combinations, with a marginally higher proportion of cases showing statistical significance. Greater alignment between solvency and liquidity at a 100-day cut-off means MSMEs unable to collect their dues within 100 days or thereabouts are on the brink of insolvency. The regulation that requires customers to pay MSME vendors within 45 days, therefore, has merit. The overall results helped us infer some interesting patterns among MSME CDs: a) A majority of the MSME CDs have DSO and DPO higher than 45 days, indicating liquidity issues at the time of admission. The proportion is higher among the non-settlers.

56 b) MSMEs take longer to pay their vendors than their customers take, i.e. DPO on average is higher than DSO, the difference being more significant in the initial period of liquidity stress. c) OCs are likely to initiate a case against CDs in the initial stages of liquidity stress – the CDs are more solvent and stronger here, and are more likely to settle. By the time FCs, on average, initiate cases, CDs’ liquidity stress is more severe (High DSO-High DPO-Low solvency or insolvent), with the case more likely to undergo the full CIRP. 8. Settlement in Bankruptcy: Valid Objective or Subversion? Globally, insolvency law has multiple objectives: • Maximizing value of the debtor’s estate. • Balancing interests of different stakeholders like promoters, FCs and OCs • Providing collective resolution to avoid the race to the courthouse. • Promoting rescue where possible, or orderly liquidation where not. Settlement fits within these objectives when it maximizes value and reduces costs. If the debtor and creditor can resolve their dispute quickly, without running through an expensive CIRP, the system as a whole benefits. But settlement becomes problematic when it undermines collective action — e.g., when a dominant creditor and debtor collude to withdraw proceedings to the detriment of smaller creditors. We turn to analysing settlements in jurisdictions other than India for international perspectives on settlement within a bankruptcy resolution framework. 8.1. International Perspective Different jurisdictions treat settlement differently: • United States (Chapter 11): Settlements are very common. Creditors may use the threat of Chapter 11 filing to induce restructuring negotiations. Courts encourage pre-packaged plans or “pre-packs” precisely to promote negotiated outcomes. Settlement is seen as consistent with bankruptcy’s goals, provided it respects collective rights once the case is filed. Once admitted, the case is in collective mode; dismissal/withdrawal requires creditor involvement (via objections or votes) and court approval. • United Kingdom (Administration & Schemes of Arrangement): Creditors often use statutory demand or winding-up threats as leverage to obtain settlement. The system explicitly tolerates this, as long as the filing is in good faith. English courts have occasionally struck down petitions deemed to be “abuse of process” where the creditor’s intent was purely coercive and not genuinely insolvency-related. Once admitted, the case is in collective mode;

57 dismissal/withdrawal requires creditor involvement (via objections or votes) and court approval. • Singapore: The framework (which draws from both US and UK traditions) also recognizes the role of pre-packs and schemes. Settlement is encouraged, but safeguards exist to protect minority creditors. Once the court admits a case, it is collective, and settlements cannot bypass creditors. • Civil Law Jurisdictions (e.g., France, Germany): Preventive restructuring frameworks allow amicable settlements supervised by courts or mediators, often confidentially, to avoid stigma. Here settlement is explicitly seen as a policy goal — rescuing viable firms outside of formal bankruptcy. Once proceedings are admitted/commenced, they cannot be withdrawn unilaterally. Settlements require court oversight and/or creditor consent. Internationally, therefore, the consensus is: • Settlement is a valid and even desirable objective of insolvency proceedings. • But it must be regulated to prevent opportunism and prejudice to minority creditors. • Early-stage settlements (before collective creditor rights are engaged) are generally accepted. • Once the collective process starts (post the CoC formation in India, post-commencement in the US/UK), collective creditor rights kick in and settlements must respect majority creditor consent. 9. Conclusion IBC jurisprudence has consistently drawn a sharp conceptual distinction between debt recovery through pre-admission settlement and insolvency resolution, emphasizing that the Code is not a substitute for ordinary recovery mechanisms. However, one should keep in mind that the emphasis on resolution reflects an underlying focus on maximising the value of viable assets. The core objective of the IBC is to optimise firm value through resolution, liquidation, or settlement. With this perspective, the sharp lines between settlement and resolution blur.
Empirical evidence from the functioning of the Code presented earlier in this chapter shows that firms that settle their dues at the threshold stage are typically those with stronger balance sheets, viable business models, and ongoing operations. For such firms, entry into the CIRP is not a neutral, procedural step but a value-degrading event. The initiation of insolvency proceedings triggers loss of control, reputational damage, disruption of commercial relationships, and heightened uncertainty, all of which can rapidly erode enterprise value. If these firms fail to settle thus, entering CIRP, they face a high probability of either liquidation or resolution at a significantly lower level of viability.

58 In this sense, settlement operates as a sorting mechanism, enabling fundamentally viable firms to remain outside formal insolvency while reserving the CIRP for cases where collective resolution or liquidation is genuinely warranted. Preventing avoidable entry into CIRP through timely settlement therefore preserves value, sustains productive capacity, and aligns with the Code’s objective of maximising the value of assets, even if this occurs outside the formal resolution process.
One could ask an even more fundamental question. Why should healthy firms be held accountable to OCs in the first place? If OCs were not allowed to file petitions under Section 9, these healthy firms could continue to operate without fear, and the question of settlements that would allow them to avoid CIRP would not arise. Is the aim to transfer resources from healthy CDs to struggling OCs?
Our empirical results indicate that pre-admission settlements under the IBC predominantly involve financially strong OCs. In distributive terms, such settlements are therefore largely welfare-neutral: they do not involve a transfer from a weaker to a stronger party (which would reduce welfare), nor do they alleviate liquidity constraints of a vulnerable creditor (which would increase welfare). The welfare significance of pre-admission settlement instead lies elsewhere. Its primary effect is to impose payment discipline on otherwise viable CDs, thereby correcting incipient payment indiscipline and reducing the likelihood of future insolvency. By inducing timely performance of operational obligations, settlement functions as a preventive mechanism. This implies that the normative objective of the pre-admission settlement framework in OC-initiated cases is not recovery but the enforcement of payment discipline that preserves firm viability and avoids inefficient entry into the CIRP. The design of pre-admission settlement rules should therefore be evaluated by how effectively they promote timely compliance and deter strategic non-payment. This perspective yields an important implication for policy design. Recommendation 1: Non-payment of small dues to a large number of OCs that cumulatively amount to a significant sum is as strong a signal of payment indiscipline as non-payment of large dues to a single OC. To capture this signal, the insolvency framework should permit aggregation of multiple small, admitted operational claims into a single petition to cross the Rs. 1 crore threshold. Allowing such aggregation aligns the pre-admission process with one of its core objectives, namely, enforcing payment discipline and preventing future insolvency. In practice, small operational claims are held not only by MSMEs but also by large firms. However, we do not recommend restricting aggregation to MSMEs, as regulations should not incentivise firms to remain artificially small. Aggregation should be permitted irrespective of creditor size, subject to the

59 safeguard that no individual claim exceeds the prescribed individual threshold. One could place a restriction that one aggregated claim cannot include more than 10 OCs. Thus, each OC will, on average, have a claim exceeding Rs. 10 lakhs, the earlier threshold for OC filings.
The analysis of the economic rationale for OC filings in Section 5 of this chapter indicated that getting an adequate return on the costs of filing a petition when the underlying claim is small is challenging. This leads to the following policy recommendation:
Recommendation 2: The economic calculation for a rational OC shows that their cost of filing a claim triples on account of the money they have to pay post-admission to the IRP. If this cost is waived, the IBC framework will become more meaningful for MSME OC applicants. Hence, we recommend that for MSME OC-applicants, all IBC process-related expenses post-admission, especially payment to the IRP, be initially borne by the IBBI, to be later recovered from the CIRP proceeds. In case of an aggregate claim, all such expenses up to the proportion of the value of claims held by MSME OC- applicants should be borne by the IBBI.
Finally, it is noteworthy that in settlements, OCs are being paid what is contractually owed, or, often, a fraction thereof. Therefore, this is not rent extraction; it is merely the performance of a private contract.

60 Chapter 4: The Quasi-Absolute Priority Rule (Quasi-APR) with MSME Orientation This chapter proposes a redesign of the IBC auction to convert it from a multiple-objective auction to a single-objective auction. The redesign will also make voting within the CoC unnecessary in most cases. By definition, an auction awards the highest bidder. We suggest an auction design that accepts the resolution plan of the highest bidder, subject to eligibility and compliance requirements. The quasi-APR (Iyer & Prasad, 2025) thereafter determines the split of proceeds among the financial and operational creditors, thereby eliminating the need for voting on these decisions. The CoC’s commercial wisdom will come into play in negotiating the resolution design – instruments used, payout duration and form, terms and conditions, among others. We believe this new format can result in higher recovery rates, reduced time to resolution, and fewer litigations.
Although this chapter analyses the resolution auction from the perspective of operational creditors as a whole, it squarely falls within the scope of an MSME-focused study, since MSMEs inherit—and often experience more acutely - the structural challenges faced by operational creditors. Hence, by focusing on the interests of OCs as a whole, as well as ensuring equity between large OCs and MSME OCs, we squarely address MSME interests.
The specific rule proposed is the quasi-APR with MSME orientation.

1. Quasi-APR Rule with MSME Orientation: How it Works Under the absolute quasi-APR rule, FCs are given priority only up to the liquidation value, and any remaining value is distributed proportionately across the residual claims of all creditors. The following example illustrates how it works.
   Consider a corporate insolvency with the following characteristics: • Winning Bid: Rs.100 • Total admitted claims: Rs.250 • FCs (FCs): Rs.180 • OCs (OCs): Rs.70 • Large OCs: Rs.40 • MSME OCs: Rs.30 • Liquidation value of the firm: Rs.70

61 Step 1: Allocate liquidation value • FCs receive Rs.70 (equal to the liquidation value) • Remaining distributable value: 100 - 70 = 30 Step 2: Compute residual claims • FC residual claim = (180 - 70 = 110) • OC claim = 70 • Total residual claims = 180 Step 3: Distribute remaining Rs.30 proportionately • FC share = 110 180 (30) = Rs.18.33 • OC share = 70 180 (30) = Rs.11.67 Final payouts: • FCs: (70 + 18.33) = Rs.88.33 • OCs: Rs.11.67 The quasi-APR with MSME orientation gives large OCs and MSMEs a recovery rate proportional to their claims. • Large OCs receive 40 70 (11.67) = Rs.6.67 • MSME OCs receive 30 70 (11.67) = Rs.5.00 1.1. Rationale 1: Efficient Resolution To understand the rationale for quasi-APR, it is important to distinguish between liquidation and resolution. In liquidation, assets are sold piecemeal, and creditors are paid according to a strict statutory waterfall. In resolution, by contrast, the firm continues as a going concern, often generating value that exceeds what would have been realised in liquidation. This additional value - sometimes referred to as the “resolution premium”- exists only because the firm is preserved rather than dismantled. The quasi-APR offers a pragmatic and economically grounded approach to distributing value in bankruptcy resolution. Its central premise is simple: FCs should retain full priority up to the liquidation value of the firm, reflecting their security interest, but any value created beyond liquidation should be shared proportionately among creditors. This approach preserves contractual priority while recognising the economic realities of reorganisation.

62 A useful way to understand claims to this additional value is through the concept of a call option. A call option gives its holder the choice to benefit from upside gains arising beyond a certain threshold. In bankruptcy resolution, once the liquidation value has been secured, any further upside created by reorganisation resembles an option payoff. While FCs hold secured claims up to the liquidation value, OCs, such as suppliers, service providers, and employees, also have an implicit stake in the firm’s continuation. Their ongoing cooperation, firm-specific knowledge, and relationship capital are often essential for generating the surplus value associated with reorganisation (Berger & Udell, 2002). In this sense, OCs hold an implicit call option on the firm’s upside beyond liquidation. Strict application of the Absolute Priority Rule (APR) under which FC claims are settled first in resolution effectively assigns much of this upside to FCs, even though OCs play a material role in creating it. The quasi-APR rule addresses this imbalance by drawing a clear line at the liquidation value. FCs are fully protected up to that point, but once that threshold is crossed, the residual value is distributed proportionately across creditors. This preserves the FCs' security interests while acknowledging that the surplus created by reorganisation is a collective product. Importantly, quasi- APR does not mandate equal treatment between financial and OCs; it merely ensures that OCs are not excluded from the upside created by resolution.
While the quasi-APR rule addresses the claims of OCs as a class, it does not, by itself, resolve distributional concerns within that class. OCs are a heterogeneous group, comprising both large firms with diversified exposures and MSMEs, whose financial resilience is far more limited. MSMEs are structurally more vulnerable to under-allocation: they are less diversified, more liquidity-constrained, and more dependent on uninterrupted cash flows. As a result, inadequate allocations are likely to impose disproportionate costs on MSMEs.
Giving OCs as a class a proportional payout from the residual proceeds does not ensure MSMEs receive a proportional payout. Hence, we introduce the additional stipulation that MSME OCs receive the same recovery rate as large OCs. This approach preserves the neutrality of the quasi-APR with respect to the resolution premium across creditor classes while preventing discriminatory outcomes within the OC category, ensuring that MSMEs are not systematically left behind despite their greater vulnerability to payment delays and losses. 1.2. Rationale 2: Robust Auction Design There is a fundamental game-theoretic rationale for recommending the quasi-APR rule, rooted in the well-established field of auction design. Over the past several decades, insights from auction theory have been used to redesign complex real-world auctions, most notably spectrum auctions by aligning

63 bidder incentives with socially desirable outcomes. These redesigns have led to substantial improvements in both allocative efficiency and revenue realization, not by changing participants’ objectives, but by changing the rules of the game. The application of game-theoretic modelling to the bankruptcy auction also points to the desirability of the quasi-APR rule.
The current IBC resolution process embodies an unusual auction structure. On the one hand, the CoC, composed entirely of FCs, acts as the auctioneer and determines the winning resolution plan through a voting process. On the other hand, the bidders are not just financial claimants but prospective owners of the firm, who must operate the business as a going concern after the auction concludes. This dual role creates an inherent tension: bidders must satisfy FCs to win the auction, but they must also ensure sufficient cooperation from OCs to preserve the firm’s value post-resolution. This tension gives rise to a counterintuitive and economically troubling outcome. In standard bargaining settings, an increase in a party’s claim share or economic importance would be expected to increase its bargaining power and, consequently, its share of the surplus. However, in the IBC auction framework, our theoretical model predicts precisely the opposite for OCs. As the claim share of OCs increases, bidders face stronger incentives to systematically under-allocate value to them. This occurs because higher OC claims increase the cost of maximising the value of the going concern from the bidder’s perspective: allocating proportionately to OCs diverts resources away from FCs, whose approval is decisive for winning the auction. As a result, bidders rationally respond by giving OCs a share of payout that is less than their claim share. Further, this under-allocation increases as the OC’s claim share increases (see Appendix 1 for details on the model). Under-allocating to the OCs reduces the overall recovery rate for all parties combined, since the firm value suffers when OCs’ participation in operations is disrupted.
Importantly, this theoretical critique is not merely abstract. The empirical analysis strongly supports the model’s predictions. We observe that, in the data, the under-allocation increases as the OC claim share increases, precisely as the model suggests. This alignment between theory and evidence reinforces the conclusion that the observed under-allocation is not accidental or case-specific, but rather a systematic consequence of the auction’s institutional design (see Appendix 2 for empirical results). This outcome is perverse. OCs are less well-treated precisely when their economic exposure is larger and when, in principle, their cooperation should matter more for sustaining the firm as a going concern. The auction design thus generates an incentive structure in which greater OC importance is associated with weaker outcomes, a result that runs counter to both intuitive bargaining logic and efficient firm continuation.

64 1.3. Rationale 3: Avoidance of Voting Anomalies One of the most distinctive features of the IBC’s bankruptcy auction is that the “winning” resolution plan is not chosen by selecting the highest economic bid, but rather by a voting process among members of the CoC. A plan is approved only if it receives at least 66% of the voting share by value. While this system is grounded in sound economic reasoning that the FCs are best placed to evaluate viability, it gives rise to an unusual form of collective decision-making, where a creditor’s voting power is not always aligned with the size of its claim. In some configurations, small creditors can wield disproportionate influence, enabling them to block or dictate outcomes, even when the majority of the value lies with large secured creditors. In others, large creditors can exercise disproportionate power. To illustrate this formally, we rely on a classic tool from cooperative game theory, the Shapley-Shubik Power Index (SSPI) (Shapley & Shubik 1954)15.
The SSPI is a method for measuring the actual decision-making power each member has in a voting system. It does not simply ask, “How many votes does a member hold?” but rather, “In how many possible sequences of voting does this member play the pivotal role – the vote that turns a losing coalition into a winning one?” This index shows that voting power can differ dramatically from economic weight. We demonstrate this using a simple example.
Suppose there are three voters, with weights: • A: 40% • B: 40% • C: 20% Threshold = 50%. It might seem like A and B are the most powerful, and they are, but the SSPI reveals that all three have equal weight. No individual voter can win on their own, even with 40% of the vote share. Any two voters combined can win: A and B, A and C, B and C. And, of course, all three voters combined end up winning. Thus, all three voters are symmetrical in terms of power structure.
However, if the threshold becomes 66%, then C has no power at all. No matter where C appears in the voting order, A and B must jointly support the proposal for it to pass.
15 Shapley, L. S. and M. Shubik, "A method for evaluating the distribution of power in a committee system," American Political Science Review 48 (1954), 787-92.

65 The key insight is that power does not scale linearly with claim size. To illustrate the anomalies in the IBC voting process, consider a stylized CoC composed of: • Three large secured creditors holding 20% each (total = 60%) • Eight smaller unsecured creditors holding 5% each (total = 40%) • Threshold for approval = 66%. Even if all three large secured creditors (60%) unanimously support a resolution plan, the motion does not pass. They fall short of the 66% threshold. This alone creates a structural vulnerability: the largest and typically most sophisticated creditors cannot carry a proposal on their own. They must rely on at least two of the eight smaller creditors. This opens the door to hold-up behaviour, where small creditors, despite having little economic exposure, can effectively block or force modifications of resolution plans. There are also other configurations of voting power in which large creditors have a disproportionate share of power. By eliminating the necessity of voting, the quasi-APR rule avoids the anomalies of the voting mechanism. 1.4. Rationale 4: Reduction of Delays Although Section 12 of the IBC mandates the completion of CIRP within 180 days, extendable by a further 90 days, i.e. a total ceiling of 270 days, the actual average duration is 713 days, with almost all proceedings breaching the statutory timeline. This defeats the very purpose of time-bound resolution and often results in value erosion. By moving to a single-objective format and doing away with the need for voting, the quasi-APR promises to significantly reduce delays.
2. Redesign of IBC Auction The root of the anomalies of the current system including under-allocation to OCs, and delays, lies in the multi-dimensional nature of the bidding problem. Bidders must simultaneously choose (i) how much total value to offer and (ii) how to divide that value between FCs and OCs. Because only FCs vote, the allocation decision becomes a strategic instrument rather than a reflection of underlying economic contributions. This embeds distributive conflict directly into the auction, distorting price discovery. The consequence is systematically biased outcomes against OCs and a reduction in the value of the resolved firm. Moving to a quasi-APR rule collapses this dual objective into a single objective by fixing the rule according to which any given bid is allocated across creditor classes. Bidders are required to specify

66 only a single scalar bid—either the total amount they are willing to commit or, alternatively, the amount they are willing to allocate to financial creditors. Once the bid is submitted, the quasi-APR rule mechanically determines the distribution across creditor classes. As a result, ranking bidders by total bid, by payout to financial creditors, or by payout to operational creditors yields the same ordering; all three metrics are monotonic transformations of one another. The auctioneer therefore, needs to evaluate bids along only one dimension. By taking the OC allocation “out of the bidding space,” quasi-APR restores a more standard auction structure in which bidders focus on valuation and feasibility rather than distributive gaming. 3. Comparing Quasi-APR with Actual IBC Payouts When actual payouts under the IBC are benchmarked against the quasi-APR rule, we find that, in the aggregate, IBC outcomes conform to quasi-APR distributions. In this sense, the proposed rule is not a departure from prevailing practice, but rather a formalisation of the allocation pattern that emerges on average in the existing framework. At the same time, there are important variations around this aggregate result. Specifically, when the OC claim share is very low, OCs tend to receive more than the amount implied by quasi-APR. At intermediate and high claim shares, their recoveries fall below the quasi-APR benchmark. These patterns highlight both the approximate alignment of current outcomes with quasi-APR on average and the scope for reducing variability and anomalies through a clearer rule- based framework (see Appendix 3 for details of the empirical tests comparing the quasi-APR with current allocations). 4. Conclusion The quasi-APR should be understood not as a redistributive intervention, but as a mechanism-design correction. It preserves FC priority, eliminates a perverse incentive that penalises OCs as their economic exposure grows, and maximises recovery rates. The quasi-APR with MSME-orientation protects against inequity within the OC classes without challenging the priority of FCs over the liquidation value.
At the same time, quasi-APR is best understood as a floor rather than a rigid rule. In some cases, especially where OCs are small, resolution applicants may voluntarily offer more than the quasi-APR minimum to ensure the continuity of operations. A strict, mechanical application of quasi-APR in all cases could inadvertently reduce recoveries for some OCs who currently receive higher payments. For this reason, quasi-APR is most effective when used as a benchmark or minimum entitlement rather than an inflexible distribution mandate.

67

Note that when bidders offer more than the quasi-APR to the OCs, the auction becomes a multi- objective auction. Two bids may no longer be unambiguously ranked: one bidder may offer a higher total bid but a lower allocation to financial creditors as compared to another. It is only in this situation that voting or subjective evaluation becomes necessary. However, the data shows this is likely to happen only in a small proportion of cases when the OC claim share is very small.
In sum, the quasi-APR rule with MSME orientation offers a balanced approach to insolvency resolution. It respects the contractual and statutory priority of FCs, recognises the economic contribution of OCs to going-concern value, and reduces distortions arising from anomalous auction design and the use of the voting mechanism. By simplifying the process, it also can be expected to significantly reduce delays. While not without limitations, it provides a coherent framework for sharing the gains from reorganisation in a way that supports both value maximisation and systemic stability. 5. Appendix 1: Consequences of the Current Auction Format- A Theoretical Study 5.1. Theoretical Model In the stylized version of this model, bidders (Resolution Applicants) estimate a certain value for the distressed company. This estimated value is unique to each bidder and is likely based on their core competence and business strategy. The value represents the bidder’s expectation of the going- concern value under the assumption that the firm’s operational network continues to function smoothly. We assume that the realized value of the going concern may be equal to, or lower than, the estimated value. Any contraction in value is assumed to arise from reduced cooperation by OCs, which in turn depends on the extent to which their realized payouts deviate from a claim-proportional reference point. We therefore define an allocation ratio as the ratio of the OCs’ share in total repayments to their share in total admitted claims. For example, if OCs account for 10% of total claims but receive only 7% of the total payout, the allocation ratio is 70%. The corresponding allocation shortfall is defined as one minus this ratio; in this example, the shortfall would be 30%. Firm value erosion is greater when a given allocation shortfall is exercised on a larger claim share. If there is a 30% shortfall on a claim share of 20%, i.e. OCs are only given 14% of the total payout, then the erosion in value is higher than when there is a 30% shortfall on a claim share of 10% with OCs being given 7% of the total payout. This is because a higher share of operational debt is often associated with greater reliance on OCs, thereby leading to a greater contraction of value for a given allocation shortfall. An alternative explanation for our assumption is that a larger claim share represents a higher level of initial trust, leading to greater value erosion in the event of a shortfall.

68 Finally, large firms with higher estimated value experience greater absolute value erosion reflecting scale effects. Thus, firm value contraction is proportional to the allocation shortfall, the OCs’ claim share, and the estimated value. If there is a 30% shortfall on a claim share of 20%, and the estimated value is Rs. 100 crore, then the contraction in value is 30% of 20% of Rs. 100 crore = Rs. 6 crore. Note that if there is complete fairness, there would be no value contraction. Our specification should not be read as a normative statement about how value ought to be distributed between creditor classes. Rather, it is a behavioural assumption about how bidders think OCs respond to different allocation outcomes. In Appendix 2, we test the predictions of our model against the observed data – lower allocation ratios for OCs and lower recovery rates with higher OC claim shares

• and find that it is a plausible way of thinking about bidders’ incentives.
  5.2. Optimal Bidding Strategy The IBC was designed to align with commercial considerations, with bid amounts and allocations to OCs determined as rational business decisions. What is the optimal bidding strategy in case of two- dimensional bids that comprise the bid amount as well as an allocation ratio (or, equivalently, the share allotted to the FC)? The bidder’s problem is modelled as a game with asymmetric information using principles of game theory. Next, we find the bidder’s optimal strategy and use it to examine the impact of parameters of the problem like 𝑛𝑛, the number of bidders, and 𝑘𝑘, the claim share of the OC, on the allocation ratio and the payouts. Recall, there are two considerations in the mind of a bidder – increasing the realized value by increasing the allocation ratio, and increasing the chance of winning by allocating as much as possible to the FC, i.e. by increasing the shortfall with respect to the OCs. We first look at the impact of number of bidders and OC claim share on the allocation ratio. 5.3. Impact of number of bidders and OC claim share on the allocation ratio Competition in the auction increases as the number of bidders goes up. This leads us to surmise that the allocation ratio would fall as the number of bidders increases because bidders would become keener to cater to the needs to the decision maker, i.e. the FC. We have also argued above how bidders face stronger incentives to systematically under-allocate value to OCs when they have an increasing claim share.
  The theoretical results are as follows: The optimal allocation ratio chosen by the bidders is always strictly in between zero and one, i.e., bidders are never completely fair, nor do they every completely ignore the OCs. Further, the allocation ratio declines continuously as the number of bidders goes up. It also declines continuously as the OC claim share increases.
  The following figures shows how these effects operate.

69 Figure 4.5.1: Impact of number of bidders on the allocation ratio Figure 4.5.2: Impact of OC claim share on the allocation ratio Next, we explore the impact of the number of bidders and the OC claim share on the overall payout as well as the payouts to the FC and the OC.
5.4. Impact of the number of bidders on payouts As the number of bidders increases, the total payout and aggregate recovery rate improve, and financial creditors also experience higher recoveries. In contrast, the recovery of operational creditors follows a 0 Decreasing Allocation Ratio Allocation Ratio 1 n 0 Decreasing Allocation Ratio Allocation Ratio 1 k = 1 k

70

non-monotonic pattern: it rises with the number of bidders up to a point, but beyond a threshold begins to decline. As explained above, this counterintuitive outcome is not driven by fundamentals, but by the design of the auction itself. Because bidders must simultaneously compete for approval by financial creditors while internalising the post-resolution costs of dealing with operational creditors, increased competition intensifies incentives to prioritise allocations to financial creditors. As a result, greater bidder competition, while beneficial in aggregate, can reduce recoveries for operational creditors. 5.5. Impact of OC claim share on payouts As the operational creditor claim share increases, the total payout and aggregate recovery rate decline, and the payouts to financial creditors also fall. In contrast, payouts to operational creditors display a non-monotonic pattern: it rises up to a threshold and then declines. The analysis of the current IBC auction format shows that as the OC claim share or competition increases, the consideration of winning the auction constrains bidders into reducing the allocation ratio. In the process, they also reduce the realised value of the company and the overall proceeds on offer for the resolution of the company.
This result is counterintuitive. If one assumes that a higher OC claim share corresponds to a higher level of bargaining power, the allocation ratio should increase as the OC claim share increases. The reason for our result is the unusual format of the auction in which bidders have to balance dual objectives.
6. Appendix 2: Consequences of the Current Auction Format- An Empirical Validation In this section, we conduct an empirical estimation to test if the central conclusions of the theoretical model hold. This helps us validate the model's usefulness as a source of policy recommendations. 6.1. Data, sources and constructs We use data on CIRP resolutions published by the IBBI on successful resolutions up to 30th September 2024. The dataset of 1,068 resolved cases as of 30th September 2024 was systematically cleaned to ensure accuracy and consistency before analysis. We cleaned the dataset for missing values, zero admitted claims of either OCs or FCs, and outliers where the recovery rate for OCs exceeded 100%. Since the quasi-APR uses the liquidation value (LV) as an anchor and applies the division rule to the ‘resolution premium’, cases where the liquidation value was unavailable or the premium was negative, were excluded. Also excluded were cases in which FC claims were lower than the LV, since this left them with no residual claims and therefore no role in post-liquidation negotiations or distributions16.

16 This adjustment is required only to test for the Quasi-APR. Including these cases does not change our results.

71

After applying these filters, 638 cases remained. For each case, we used the combined realizations of FCs and OCs as a proxy for the total proceeds available for distribution. These realizable amounts are reported net of all priority payments, including insolvency resolution process costs and liquidation expenses. 6.2. Empirical tests and results We begin by comparing the actual recovery rates of OCs (OCRRActual) against OC recovery rates in proportional pay-outs (OCRRProp) across all resolved cases, and then refine the analysis by partitioning cases in the manner described above.
Table 4.6.1 presents the descriptive statistics of the dataset. A key observation is that the average OCRRActual, defined as recoveries to OCs as a percentage of their admitted claims, is substantially lower than the overall recovery rate (RR%), measured as total proceeds relative to total admitted claims. The median figures further highlight this disparity: while the median overall recovery rate stands at 21.75%, the median OC recovery rate is only 4.4% (not shown in the table). This indicates that OC recoveries are heavily skewed toward the lower end, with the average of 15.67% driven upward by a small number of unusually large recoveries. Table 4.6.1. Distribution of Cases by OC% share (Amounts in Rs Billion) OC% Class
No. of Cases Average total Claims Average OC Claims Total OC Claims Average OC share of claims (OC%) Average Recovery Rate (RR%) Average OC recovery rate OCRRActual Low (<4.903%) 212 24.71 0.41 86.92 1.73% 28.86% 24.65% Medium (4.903%-21%) 213 12.74 1.33 283.29 11.72% 31.26% 11.53% High (>21%) 213 6.42 2.08 443.04 43.79% 28.11% 10.86% Overall 638 14.61 1.27 811.83 19.11% 29.03% 15.67%

Table 4.6.1 further exposes the imbalance in recovery outcomes for OCs, this time using the percentage share of OC claims as the lens for analysis. In the Low-OC% share class (below 4.903%), OCs secure an average recovery rate (OCRRActual) of nearly 25%—the highest across all categories, close to the overall recovery rate. However, this relative advantage is short-lived: in the Medium class (between 4.903% and 21%) and High class (above 21%), recoveries drop steeply to just 10–12%, even though the overall recovery rate (RR%) in these larger cases remains much higher, at 28–32%. The OCs’ share of total claims remains capped at 12% in the Low and Medium groups and reaches 43% only in the High class, reinforcing their position as structural minorities within the insolvency framework.

72

The OC recovery rate can be lower than the RR%, if and only if it is lower than the FC recovery rate (FCRR). Thus, the overall pattern is straightforward: OCs have lower recovery rates than FCs. However, this difference is not uniform. When OCs hold only a small share of the claims, their recovery rates are broadly comparable to those of FCs. It is only in cases where OCs hold medium- or large- sized claim shares that their recoveries fall significantly below those of the FCs.
Yet another perspective is to compare the OCRRActual with what they would get if the resolution proceeds were divided between OCs and FCs according to their share of the claim, the ‘proportional division rule’. The OCRRActual is lower than the recovery rate under proportional division if and only if the OCRRActual is lower than the FCRR, or if it is lower than the overall RR%. This is the lens we use to examine whether the allocation ratio of OCs decreases with an increasing claim share.
To test the statistical significance of our descriptive statistics, we turn to the two-tailed paired t-test, a tool designed for exactly the kind of comparison we seek. Each bankruptcy case gives us two perspectives: the actual recovery rates of OCs received, versus the recovery rate if they were paid in proportion to their share of the total claim. Our question is simple yet fundamental: are the two pay- outs essentially the same, or do they diverge in a meaningful way? The paired t-test allows us to ask whether the average difference between actual and normative pay-outs (proportional pay-out, in this case) is statistically different from zero. We compute the mean difference of the variable (OCRRActual minus OCRRProp) for three different size classes of claim share percentage- low, medium, and high, and use paired t-tests to test for the significance of the differences, in Table 4.6.2.
Table 4.6.2. Paired t-test comparison for OCRRActual and OCRRProp

Aggregate data Low OC% class Med. OC% class High OC% class Mean Difference -0.134 -0.031 -0.197 -0.173 t-statistic -12.401 -1.221 -13.757 -14.469 p-value 0.000*** 0.223 0.000*** 0.000*** df 637 211 212 212

This discrete partitioning of cases reveals a structure that mirrors the qualitative predictions of the auction model. The space of OC claim shares naturally divides into two distinct regions. The t-tests show that in the region of low OC claim shares, the actual OC recovery rate is not significantly different from the recovery rate under the proportional rule. However, in the regions of medium and high OC claim shares, the actual OC recovery rate is significantly lower than the normative.

73

While this comparison helps identify broad regions of relative advantage or disadvantage for OCs, it does not tell us how the allocation ratio evolves as the OC claim share increases. The scatter plot of the allocation ratio versus the OC claim share is as follows:

To investigate this more precisely, we conduct a regression analysis relating the allocation ratio to the continuous variation in the OC%.
We use the specification in Equation 4.1.
𝑙𝑙𝑙𝑙𝑙𝑙( 𝑂𝑂𝑂𝑂 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂𝑂 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅) = 𝛼𝛼+ 𝛽𝛽 (𝑂𝑂𝑂𝑂 𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐𝑐 𝑠𝑠ℎ𝑎𝑎𝑎𝑎𝑎𝑎 %) + 𝜖𝜖
(4.1), where 𝛼𝛼 is the intercept term, 𝛽𝛽 is the coefficient on the OC claim share %, the explanatory variable, and 𝜖𝜖 is the error term.
We specify the empirical relationship between the OC allocation ratio and the OC claim share in a log- log form, rather than in levels, for both conceptual and empirical reasons. Conceptually, the theoretical model predicts a non-linear, hyperbolic relationship between claim share and allocation: changes in claim share have a large effect on allocation outcomes at low levels, but the marginal effect diminishes as claim share increases. A log-log specification captures this curvature naturally, allowing proportional changes in claim share to be associated with proportional changes in allocation outcomes. Finally, the log-log specification improves statistical performance by reducing heteroskedasticity and limiting the influence of extreme observations, particularly when claim shares are very small or very large. For these reasons, the log–log formulation provides a parsimonious and theoretically consistent way to test the model’s predictions.

74

The results of the regression in Table 4.6.3 indicate a significant negative relationship between the allocation ratio and the OC claim share, as predicted by the theoretical model. Empirical consistency with the model’s predicted sign, despite the absence of the number of bidders, the other theoretically relevant independent variable, provides encouraging support for our conceptualisation of the underlying mechanism. Table 4.6.3. Regression of Log (OC Recovery Rate/Overall Recovery Rate) on Log (OC claim share)

Estimate Std. Error t-statistic p-value Intercept -0.195 0.041 -4.717 0.000 *** Log (OC Claim%) -0.46 0.035 -13.209 0.000 *** Observations 578 Adjusted R² 0.23 F-statistic
174.5***

7. Appendix 3: Is the Quasi-APR Already in Use- An Empirical Test The Quasi-APR blends elements of the Absolute Priority Rule (APR) and proportional distribution. Under this approach, the APR operates until the liquidation value (LV) of assets is fully allocated. FCs (assumed secured in our study) receive first claim on an amount equal to the LV, credited against their dues. Beyond this stage, the distribution shifts to a proportional framework to address distribution of the residual proceeds, described in Appendix 2.
   This analysis uses a new variable, defined as the difference between the actual and normative OC recovery rates (OCRRActual – OCRRAprop), to capture the extent of deviation from the Quasi-APR benchmark. The mean difference of this variable is then tested for statistical significance at the aggregate level and across subsamples classified by the percentage share of OC claims (OC%) divided into low, medium, and high classes using the t-test.
   Table 4.7.1 shows that at the overall level, OCs get pay-outs in line with the normative pay-out. However, when we partition cases into low, medium, and high claim shares, and test, the results change.
   Table 4.7.1. Paired t-test comparison for (OCRRActual – OCRRAprop)

Aggregate data Low OC% class Medium OC% class High OC% class Mean Difference 0.0003 0.0998 -0.0456 -0.053 t-value 0.0275 4.086 -3.455 -5.256 p-value 0.9781 0.000*** 0.000*** 0.000*** df 637 211 212 212 Significance levels: ‘’: 1% level; ‘’: 5% level, ‘’: 10% level

75

Our tests reveal that the OCRRActual exceeds the normative payout when the OC claim share is low, but falls below the normative payout when the OC claim share is medium or high. Thus, the IBC auction is more than fair to OCs when OC claim share is low, but unfair otherwise. These details are masked by the analysis at the overall level which indicates alignment of the actual and the normative recovery rate.

76

Chapter 5: Pre-pack Insolvency and Resolution Package (PPIRP)

1. Introduction The PPIRP was introduced under the IBC via the IBC (Amendment) Ordinance, 2021 on 4th April 2021, in the wake of the Covid-19 pandemic that affected MSMEs disproportionately. The CIRP is designed for larger firms, and the Rs.1 crore threshold brought in 2020, effectively excluded the smaller MSMEs. Ghio (2025), in their study of U.K. Insolvency laws, highlighted that when large-firm frameworks are applied to small firms, they mostly result in liquidation rather than rescue, driven by delays that tighten cash cycles, further pushing them into distress.
   The primary reprieve the PPIRP provides MSME promoters as compared to the CIRP is its debtor-in- possession (DIP) feature blended with a Creditor-in-Control (CiC) model, which is aimed at minimizing disruption to the firm’s operations and conserving firm value (Dhage, 2024). Available to CDs who have defaulted at least Rs.10 lakhs to apply for initiation of the PPIRP, it allows the promoters of the defaulting CD, if eligible under Section 29A of the IBC, to participate in resolution by presenting a Base Resolution Plan (BRP), which is further put up for challenge in the interest of value maximization. Thus, it opens the door for promoters to take possession of their company in consultation and agreement with the creditors, offering the firm a fresh start.
   In the interest of quick resolution within 120 days, the PPIRP attempts to offer a relatively informal set- up for the promoters of CDs to negotiate resolutions with the FCs. This process is undertaken under the supervision of an RP, with the resolution requiring the consent of at least 66 percent of the CoC. An important design guardrail of the PPIRP for OCs is the requirement that the BRP should not impair them, and their interests are preserved. Alternative resolution plans in the form of a Swiss challenge are called for if such impairment is found. Dhage (2024) concludes that the PPIRP’s compressed timeline, pre-negotiated plans, and minimal disruption to operations can shorten receivable lock-ups and preserve cash flows not only for the applicant-CD but also for the OCs in the supply chain, thereby mitigating liquidity strain for small suppliers of a distressed buyer. The framework has the potential to improve predictability and lower frictions for these OCs. The PPIRP can preserve firm value better than the CIRP because of its DIP and quick resolution features (Thakkar & Agarwal, 2022), though initiation of a Swiss Challenge to determine the best resolution plan can delay the process, further tightening credit for the MSME CD and its OCs (Kavitha, 2022).
   Despite these intentions, the PPIRP has been sparsely used by the CDs; so far, only 14 cases have been filed under it, with eight finding resolution. Possible reasons could range from the onerous documentation and process-intensive design, including CoC-formation and court involvement, even in the Pre-Initiation phase, that is supposed to enable the parties to quickly negotiate and settle in a

77

relatively informal set-up (Tripathi, 2024), to the availability of multiple government schemes and guarantees to support MSMEs that either pre-empts a bankruptcy situation or dissuades them from using it. We examine these possibilities more closely in this chapter.
2. Analysis of PPIRP cases As of 31 March 2025, of the 14 cases filed under the PPIRP, eight concluded in resolution, one was withdrawn, and five were ongoing. Three were disposed of within 200 days, three others in less than 400 days, while three cases, including the one withdrawn, took more than 400 days to be disposed of. The PPIRP of GCCL Infrastructure and Projects Limited took 721 days. This clearly shows that while the intent of the PPIRP is noble, implementation challenges abound. Based on data provided by the IBBI, OCs recovered 100% in all cases, while FCs recovered between 5% and 90%; their median recovery was 25.56%. This shows that preventing OC impairment as part of the resolution is working. However, a closer look reveals that this may not be so in all cases. Anand and Sharma (2025) use the Amrit India Ltd. PPIRP to demonstrate that despite the regulation designed to protect OC claims, unsecured creditors suffered 91.42% impairment, while secured creditors realized 39.37%. They document that the secured creditor received Rs.5 lakhs against a claim of Rs.12.70 lakhs, as documented with IBBI. However, the contingent creditors received Rs.2.20 lakhs against their claim of Rs.25.63 lakhs; it appears that only Rs.2.20 lakhs of their claims was admitted, which translates into 100% realization. They attribute this disparity to CoC-primacy (also Kavitha, 2022), which retains creditor-class asymmetries (as in the CIRP), shaping outcomes.
Our examination of the PPIRP orders of another CD, Sudal Industries Ltd. showed that while the FCs took haircuts in order of their seniority, with the unsecured FCs getting barely anything, the court ordered the OCs, including the employees and workmen dues to be paid as a going concern, with no sacrifice or concession sought from them. Whether the OC claims were fully admitted is unclear, with the figure in the IBBI database at approximately 50% of the claims in the judgement1718.
However, in general, an examination of the eight resolved PPIRP cases reveals a genuine intent to award relatively favourable outcomes to the OCs. In contrast, FCs made a full recovery in only two cases.

17https://efiling.nclt.gov.in/ordersview.drt?path=VJOAJEt%2B4FEFG5aaDWryLnb5hX4tkiRcaO9oQTZ7zg0Jzg4x9bADEl vCUAKZjiB2l0ay1zj8K50soItvo3IO862lBA8KfwwKI6lyB%2B8eXLcONjZ%2BuFTTGotiiDE5wWJ9tzgckBHFqdahJJq9f 9%2FHjyY5V8OU8fy9HlCd5woWWPcocDs5lEB9bsCZSu1S2RZA 18 We plan to examine the remaining six resolved cases to get more insight into the outcomes.

78

3. Low utilization of PPIRP The biggest concern on the PPIRP front is its extremely low utilization. Fourteen cases over four years clearly show that MSME CDs are not applying under the framework as intended by the regulators. The foremost reason for this low utilization is misaligned incentives of the key parties involved, which make it unattractive to both FCs and the MSME CDs. This is despite the PPIRP having been designed specifically to safeguard the MSME CDs’ and their OCs’ (likely smaller MSMEs) interests. The parties will prefer more attractive, less risky alternatives, if available. A distressed MSME CD should prefer the PPIRP option to an informal, negotiated settlement with its creditors, as it provides a comprehensive legal framework to put its house back in order and start afresh. Out-of-court negotiations with one or a few creditors might attract objections and legal notices from those left out or dissatisfied with the proposal, thereby turning out costlier operationally and administratively, besides evoking surprise petitions later from dissenting creditors, employees or other parties. Once an MSME CD admits itself under the PPIRP, no FC or OC can file a petition under Sections 7 or 9 of the IBC, nor can an FC initiate recovery under other routes, such as SARFAESI. The resolution plan that is approved and passed under the PPIRP becomes legally binding on all parties, and all claims against the CD stand extinguished (e.g. fresh claims made after BRP approval were quashed by the NCLAT in the case of Shree Rajasthan Syntex Limited). Most importantly, the promoter and incumbent management retain control of the CD, thereby giving a second chance to a team that has operated the CD in good faith and wants to continue. By mandating that a resolution plan be accepted only if it does not impair OCs’ claims, the PPIRP framework aligns with CDs’ interest in maintaining their relationships with OCs for business continuity.
   The tilt towards the CD (DIP mode vs. CiC mode) and the OC (non-impairment of OC claims) results in a high probability of large haircuts or delayed recovery for the FCs. This ‘unfairness’ that FCs face ab initio was the most common reason we encountered for their reluctance to a PPIRP. Creditors are sceptical (Jacob, 2025) and will agree to the PPIRP for negotiation and settlement only when they see a chance of reasonable recovery and revival of business.
   Credit guarantees that make good these haircuts, whether partly or fully, do not compensate for the side effects of accepting the haircuts in the first place – scrutiny, both internally and by agencies such as the CVC, particularly for public sector banks (PSBs). Since haircuts resulting from a market mechanism become more defensible, inviting competing proposals to the BRP within the PPIRP, as well as the CIRP, becomes a preferred option for the FCs. Practically, though, the small size and specificity of the MSME assets may not attract many bids, thus forcing the CoC to accept a low-value BRP (e.g. Sudal Industries Limited). The DIP model, coupled with exemption from Section 29A of the IBC, makes the

79

FCs feel a lack of control and wary of possible perverse behaviour by the promoters, such as asset- stripping or indulging in ‘avoidance transactions’. In such an ‘adverse’ environment, giving the CoC primacy and power at crucial junctures within the PPIRP framework may create a conflict of interest for the FCs. The CoC is the gatekeeper for the MSME CD's admission to the PPIRP and the approver of the resolution plan. With the design and payoffs skewed against the FCs, expecting the CoC to be objective and neutral, while also likely providing fresh funding to keep the MSME operational, is unrealistic. FCs have a strong incentive to block the MSME CDs’ admission into PPIRP, particularly if they expect to recover less than through SARFAESI or liquidation of the CDs’ assets.
In the event the case is admitted to PPIRP, the FCs’ incentives prompt them to seek plans beyond the BRP, triggering a Swiss Challenge that risks the promoter's control over their business. This, in turn, reduces the MSME CD’s incentives to file under the PPIRP. Though the PPIRP would be preferable to a CIRP for the CD, the risk of losing control might lead them to settle privately with OCs or seek a One-Time-Settlement under the RBI framework with their FCs outside the IBC. To be effective, the trust deficit arising in the current PPIRP framework needs to be addressed by aligning the incentives of all parties and making it more meaningful for them. Other reasons for the low interest in PPIRP are:

1. MSMEs have been found to avoid initiating PPIRP for fear of reputation risk that can lose them customers and vendors, thus escalating distress. MSMEs often operate in clusters where interconnectedness is high in both business and personal relationships. Submitting to bankruptcy proceedings is still seen by some as a ‘stigma’, which they seek to avoid. Interestingly, Menezes, et al. (2020) recommended a pre-insolvency online dispute resolution framework that would overcome the complexity, cost and stigma associated with traditional insolvency for small firms; the PPIRP mitigates some of these issues, not all.
2. Capacity constraints–institutional capacity at NCLT, number of insolvency professionals– have been found to cause delays and slow relief (Thakkar & Agarwal, 2022; Kavitha, 2022).
3. While Anand & Sharma (2025) found the PPIRP successfully resolving MSME CD stress within short timelines at low costs (approximately Rs.5 lakhs), they recommend procedural simplification to make the PPIRP more user-friendly and effective. Even the pre-initiation process, supposedly simpler and “informal”, is procedure-intensive.
4. Frictions caused by the statute design: A requirement that potentially pre-empts MSME CDs from using the PPIRP is the requirement for the CDs to declare avoidance transactions u/s 54C(3)(c), which is further verified by the RP, who has to then provide an independent opinion to the NCLT whether the value ‘lost’ needs to be ‘clawed back’. This can be a serious deterrent

80

at multiple levels. First, if the CD has indeed indulged in such transactions, they have no incentive to declare them, and risk investigation and loss of credibility. Second, an investigation by the RP, followed by possible litigation, can easily stretch the timelines beyond the desired 90+30 days, not to mention the increased costs involved. Third, this design undermines the inherent trust required among the parties – the CD, the RP, and the creditors – to ensure cooperation toward a mutually beneficial solution, creating an adversarial relationship between the RP and the CD’s management and vitiating the atmosphere required for a consensual solutioning process. 5. Misuse of PPIRP: Agrawal et al. (2025) document cases of promoters re-entering “through the back door,” by misusing the carve-outs that tilt the process toward debtor protection. PPIRP is invoked tactically in such cases, and pro-MSME clauses are misused to specifically stall recoveries and sideline OCs, thus weakening creditor rights. Instances of such misuse were echoed by our interviewees as well. Not only FCs, but MSME vendor receivables also get trapped and impaired in the process, aggravating their working-capital stress rather than relieving it. These practices end up weakening OC rights and reducing the usefulness of IBC for MSME-OCs too.
3.1. Do alternative avenues of government credit support and guarantees make the PPIRP a less preferred option for MSME CDs? The RBI and the government have regularly brought in schemes and regulations to ease financial burden on MSMEs, both direct (easy lines of credit, collateral-free credit) and indirect (incentivizing lenders with guarantees, easy restructuring, and loan provisioning concessions). Since 2021, when the PPIRP was promulgated, relaxations were introduced to address the Covid-19-induced stress, and the CGTMSE (Credit Guarantee Fund Trust for MSE) corpus was enlarged, with two additional guarantee funds announced in the 2023-24 budget.
We believe that while these mechanisms may temporarily reduce (or permanently, for some) or delay the need for an insolvency restructuring, they are not a substitute for the PPIRP. They are simpler and quicker, less visible from a stigma perspective, and, most importantly, allow incumbent management to continue at the helm without fear of losing control of the firm. Guaranteed credit can provide MSMEs with much-needed ‘liquidity-runway’, but it also increases their financial leverage, thus increasing solvency risk. However, to the extent a liquidity-strapped MSME can productively use this concessional credit to revive its operations and generate returns exceeding the cost of credit, these facilities can help it avoid potential insolvency caused by extended liquidity stress. To the extent MSMEs are unable to appreciate this opportunity and use it as easy credit, such alternatives are only deferring an inevitable insolvency and collapse.

81

Therefore, the concessional credit lines available to MSMEs through budgetary and guarantee schemes should be seen as a first line of defence, which they should prudently use to avoid having to resort to the second line of defence, i.e., the PPIRP. The NCTGC framework can be leveraged to make the PPIRP more effective, as detailed in the next section. 4. Conclusion The PPIRP was the result of the realization that large-company frameworks like the CIRP do not serve MSMEs, which are more vulnerable to financial shocks and are “to poor to go broke” under these systems, as exposited by Ghio (2025). Despite design changes intended to simplify and expedite MSME-friendly resolutions from both the CD and OC perspectives, it has failed to catch on, primarily because of misaligned incentives. With only fourteen cases filed and eight resolved since inception, there is an urgent need to review the framework and make changes that make it more meaningful and attractive to all stakeholders.
We provide two sets of recommendations with regard to the PPIRP – the first set to better help align the incentives for key stakeholders, and the second, to speed up and reduce bottlenecks in the process, making the framework more meaningful for all involved. I. Recommendation to better align the incentives of the FCs, OCs and the CD We recommend reconsidering the clause requiring that OCs suffer “no impairment” of their claims. While well-intentioned, this provision creates misaligned incentives and may not adequately address FCs’ interests, who remain the key decision-makers in deciding whether to admit a CD into the PPIRP. Instead, the study proposes replacing this clause with a clear and transparent allocation rule that applies symmetrically. It ensures payoffs and requires sacrifices for both sets of creditors, based on principles, effectively reducing discretion and hence, bias. Specifically, OCs should receive the higher of: a) a share of resolution proceeds proportional to their claim-share, or b) the entire liquidation premium (defined as the amount over and above the liquidation value), up to their claim amount.
This formulation better aligns incentives between FCs and OCs, reduces litigation and uncertainty, and ensures that FCs are not systematically disadvantaged in the PPIRP, thereby bringing the scheme to a standstill. II. Recommendations to simplify, without compromising on rigour and safeguards against misuse

82

1. While avoidance transactions should be discouraged and penalized in the interests of the creditors, we recommend separating the primary pre-pack process towards finding a resolution plan from investigating the avoidance transactions. The former can then be concluded within the desired time frame. The RP’s role should be more of an administrator (as was intended) to collect the information, ‘accept the CD’s declaration’ regarding such transactions with due diligence at only a high level, and pass it on to the NCLT for further independent investigation. This is easier said than done. For instance, the South Korean bankruptcy system has seen phenomenal adoption by small firms. This is despite having a very similar framework to the PPIRP. However, simplification and focus on speed have led to limited court review resulting in overlooking pre-filing misconduct that is classified as avoidance transactions under the IBC. To counter this, the penalties for non-disclosure by the CD should be severe, going beyond clawbacks. The creditors too, participate in finding a solution, trusting the information provided, and provide the new management (or the existing promoter, if the BRP is accepted) a clean slate to start on, without loss of firm value. If the independent investigation reveals illegitimate carve-outs by the earlier management, these can be clawed back, and used towards creditor claims, prioritizing the FCs.
2. Reasons for the low use of PPIRP go beyond the statute and its clauses. For any adoption, awareness is the first step. MSMEs lack awareness19, resulting in lack of trust in the process and fear of losing control over the business. Awareness can be built through a) outreach through sector-specific curated sessions in MSME clusters, newsletters, videos and interviews of promoters showcasing success stories, b) trainings organized jointly by IBBI and the MSME Ministry, and incentivizing MSMEs with ‘credits’ (that can be redeemed in the form of lower fees or rate of interest on a loan) for completing the trainings, and c) dedicated page on PPIRP on the Udyam website with all information, training, incentives, and access to an advisory help-desk and support.
   Effective legal reform must factor in behavioural, informational and institutional elements in the regulatory design. The behaviour of the players in an IBC process spans activities regulated by different entities, and unless these are seen in their entirety, the regulation will not fully address these cross-cutting implications.

19 NITI Aayog report: https://www.niti.gov.in/sites/default/files/2025- 05/Enhancing_Competitiveness_of_MSMEs_in_India.pdf

83

Chapter 6: Policy recommendations The objective of the IBC is not resolution for the sake of resolution. Instead, it is the optimisation of firm value either through resolution, liquidation, or settlement. With that perspective, this study identifies several structural gaps in the current operation of the IBC, particularly as they affect OCs and, within that group, MSMEs. The following policy recommendations are organised thematically, corresponding to the different stages of the IBC press and the chapters of this report.

1. Strengthening Evidence on Pre-Admission Outcomes A central finding of this study is that a substantial share of the economic impact of the IBC occurs before admission into the CIRP. Settlements, withdrawals, and dismissals at the pre-admission stage play a critical role in creditor recoveries and debtor discipline. However, current public reporting fuses almost entirely on post-admission outcomes, thereby systematically understating the effectiveness of the IBC as a recovery and deterrence mechanism. It is therefore recommended that the IBBI collect, standardise, and publish comprehensive data on pre-admission recoveries and settlements, including the nature of the petitioner (FC, OC, MSME status), settlement terms where available, and the financial condition of the CD. Such data would allow policymakers to better understand why weaker firms eventually enter CIRP while stronger firms tend to settle earlier, and would enable a more accurate assessment of the IBC’s overall impact (Chapter 2).
2. Mandating claim records in NeSL The study recommends that the National e-Governance Services Ltd. (NeSL) be mandated to record all accepted OC invoices of at least Rs. 1 crore (the threshold can be reduced in due course to include more dues) in a structured, searchable format. Systematic recording of acknowledged claims would reduce informational frictions, speed up dispute resolution, and lower transaction costs at the pre- admission stage, particularly for MSMEs. It will also speed up admissions, where pre-admission settlements do not happen (Chapter 2).

84

3. Facilitating Petitions for MSME OCs The current threshold requirements under the IBC, combined with the fragmented nature of MSME claims, limit effective access to the insolvency framework for small OCs. Individually, many MSMEs do not hold claims large enough to justify the cost of initiating proceedings, even when non-payment is persistent and economically damaging. To address this, the study recommends explicitly permitting aggregation of OC claims to cross statutory thresholds, subject to safeguards against abuse, including limiting the number of OCs in an aggregated claim to 10. Aggregation would allow similarly placed OCs to collectively initiate proceedings, thereby improving access to the IBC without lowering thresholds across the board. This suggestion is a simpler alternative to introducing a separate pre-pack type mechanism for MSME OCs. Further, the study recommends a waiver or reduction of procedural and filing charges for MSME OCs post admission to the CIRP. Such waivers may be structured as full or proportional, depending on the share of total value represented by MSMEs within the claim (Chapter 3). This approach recognises the vulnerability of MSMEs to liquidity risk, which get exacerbated by their disproportionate exposure to delayed payments, while preserving fiscal discipline in the system.
4. Reforming the Resolution Auction Through Quasi-APR with an MSME Orientation A key contribution of this study is the identification of structural inefficiencies in the current resolution auction design, arising from its multi-objective nature. Resolution applicants are required to simultaneously compete for approval from FCs while ensuring the firm's post-resolution viability, which depends critically on OC cooperation. This creates perverse incentives, including systematic under-allocation to OCs as their claim share increases and distortions driven by voting thresholds within the CoC. To address these issues, the study recommends adopting a) A single objective auction model where the bidder offering the highest overall recovery is awarded, coupled with b) a quasi-Absolute Priority Rule (quasi-APR) with an MSME orientation for distribution of proceeds. Under this framework, FCs’ priority is fully recognised up to the firm’s liquidation value, reflecting their security interests. Any value realised beyond liquidation is then distributed proportionately across creditor classes. The quasi-APR with MSME orientation further provides large OCs and MSMEs with recovery rates proportional to their claims.

85

Importantly, quasi-APR is proposed not as a rigid redistribution mechanism but as a rule-based floor that simplifies bidder incentives and aligns the auction more closely with standard principles of auction design used successfully in other high-stakes settings. This approach eliminates the need for voting unless a) there is a tie, or b) the bidders allot more than the quasi-APR warrants to the OCs (an event which would occur in a small proportion of cases).
It can be expected to
• reduce strategic manipulation • improve allocative efficiency • mitigate anomalies arising from voting rules, and
• reduce procedural delays
while preserving the core priority structure of the IBC (Chapter 4).
5. Suggestions for more effective PPIRP For the PPIRP process, the study recommends interventions to better align the key players’ incentives, eliminate bottlenecks, and simplify the process (Chapter 5).
1. The first of these is reconsidering the clause requiring that OCs suffer “no impairment” of their claims. While well-intentioned, this provision creates misaligned incentives, particularly for the FCs, who remain the key decision-makers in admitting a CD into the PPIRP.
Instead, the study proposes replacing this clause with a clear and transparent, principle-based allocation rule that applies symmetrically. OCs may receive the higher of:
a) a share of resolution proceeds proportionate to their claim-share, and
b) the resolution premium (defined as the amount over and above the liquidation value), up to their claim amount.
This formulation can reduce litigation and uncertainty arising from FCs being systematically disadvantaged in the PPIRP. 2. We recommend considering separating the primary pre-pack process towards resolution from the investigation of avoidance transactions. In order to safeguard against misuse of

86

this separation, the resolution can be made ‘provisional’ until the RP’s clean chit on avoidance transactions is received, within a specified time period. Penalties for non- disclosure of such transactions by the CD should be severe and go beyond mere clawback. 3. Finally, we recommend steps to increase awareness of the PPIRP among MSMEs through outreach, leveraging the Udyam website, making available and incentivizing training, and providing support platforms (details in Chapter 5, Section 4).

87

Bibliography Agrawal, P., Kajaria, S., & Mukhopadhyay, D. (2025). MSMEs and IBC – gaming the system? Insolvency & Restructuring International, 19(2). Anand, M. K., & Sharma, R. (2025). Pre-packaged insolvency in India: Balancing efficiency and equity through the first successful case of Amrit India Limited. International Journal of Dynamic Educational Research Society, 6(2), 94–111. Apte, P., & Das, S. K. (n.d.). Treatment of MSME insolvency under IBC. In Insolvency and Bankruptcy Regime in India—A Narrative (pp. 279–294). Insolvency and Bankruptcy Board of India. Berger, A. N., & Udell, G. F. (2002). Small business credit availability and relationship lending: The importance of bank organisational structure. The economic journal, 112(477), F32-F53. Chauhan, N., & Abidin, G. (2025). Insolvency and competition law: Navigating anti-competitive risks in pre-packaged processes in India. International Journal of Environmental Sciences, 11(4), 1596–1602. Dhage, D. (2024). Optimising resolution: Analysing pre-packaged insolvency’s academic efficacy in creditor–debtor relations. RGNUL Student Research Review, 10(1), 180–191. Ghio, E. (2025). The rescue illusion: Why insolvency law fails MSME (University of Edinburgh School of Law Research Paper Series, Working Paper No. 2025/19). University of Edinburgh. Insolvency and Bankruptcy Board of India. (2025). IBBI Newsletter. https://ibbi.gov.in/en/publication ICRA Limited. (2025, June). Structured Finance. Retrieved from https://www.icra.in/Rating/DownloadResearchSpecialCommentReport?id=6372 Iyer, V., & Prasad, R. (2025). Exploring a quasi-absolute priority rule for pay-outs to operational creditors in bankruptcy resolutions: Evidence from India. Global Finance Journal, 64. https://doi.org/10.1016/j.gfj.2025.101074 Jacob, M. (2025, February 25). Catalyzing MSMEs through a facilitative distress resolution process. IBC Laws Blog. https://ibclaw.blog/catalyzing-msmes-through-a-facilitative-distress-resolution- process-by-mathew-jacob Jain, T., Singh, R., & Subramanian, C. (2025). Impact assessment of the Trade Receivables Discounting System (TReDS). Receivables Exchange of India Ltd. Jayadev, M., Anand, A., Krishna, A., & Mohanan, S. (2025). Behavioral impact of IBC. Centre for Capital Markets and Risk Management, IIM Bangalore, for IBBI.

88

Kavitha, D. (2022). A panacea for small firm insolvencies? A critical review of the pre-pack scheme launched in India. International Journal of Law and Management, 64(5), 389–402. https://doi.org/10.1108/IJLMA-08-2021-0192 Menezes, A., Mocheva, N., & Shankar, S. S. (2020). ‘Under Pressure’: Integrating online dispute resolution platforms into pre-insolvency processes and early warning tools to save distressed small businesses. Vikalpa: The Journal for Decision Makers, 45(2), 79–92. https://doi.org/10.1177/0256090920941813 Shapley, L. S., & Shubik, M. (1954). A method for evaluating the distribution of power in a committee system. American Political Science Review, 48(3), 787–792. https://doi.org/10.2307/1951053 Thakkar, D., & Agarwal, K. (2022). Pre-packaged - Integration of debtor centric model with creditor in control model: Indian insolvency regime. Bank Quest – Indian Institute of Banking and Finance, 93(1), 5–16. https://ssrn.com/abstract=4752965 Tripathi, U. (2024). A critical analysis of India’s pre-pack regime for MSMEs. International Insolvency Review, 33(1), 89–110. https://doi.org/10.1002/iir.1517

89

Glossary Abbreviation Expansion ANOVA Analysis of Variance APR Absolute Priority Rule ARS Appeals–Review–Settlement BRP Base Resolution Plan CD Corporate Debtor CGSE Credit Guarantee Scheme for Enterprises CGTMSE Credit Guarantee Fund Trust for Micro and Small Enterprises CiC Creditor-in-Control CIC Credit Information Company CIN Corporate Identification Number CIRP Corporate Insolvency Resolution Process CoC Committee of Creditors Covid-19 Coronavirus Disease 2019 COVID-19 Resolution Framework RBI Resolution Framework for COVID-19 Related Stress CRISIL Credit Rating Information Services of India Limited DIP Debtor-in-Possession DPO Days Payables Outstanding DSO Days Sales Outstanding EOI Expression of Interest FC Financial Creditor FCRR Financial Creditor Recovery Rate FGD Focus Group Discussion FY Financial Year IBC Insolvency and Bankruptcy Code, 2016

90

IBBI Insolvency and Bankruptcy Board of India IIM Indian Institute of Management IRP Interim Resolution Professional LP Liquidation Premium LV Liquidation Value MCA Ministry of Corporate Affairs MCGS-MSME Mutual Credit Guarantee Scheme for MSMEs MSE Micro and Small Enterprise MSEFC Micro and Small Enterprises Facilitation Council MSME Micro, Small and Medium Enterprise MSMED Act Micro, Small and Medium Enterprises Development Act, 2006 NBS Nash Bargaining Solution NCGTC National Credit Guarantee Trustee Company NCLAT National Company Law Appellate Tribunal NCLT National Company Law Tribunal NIC National Industrial Classification NPA Non-Performing Asset NPAs Non-Performing Assets NPE Net Plant and Equipment OBP Original Bankruptcy Problem OC Operational Creditor OC% Percentage Share of Operational Creditor Claims OCRRActual Actual Operational Creditor Recovery Rate OCRRProp Operational Creditor Recovery Rate under Proportional Rule OCRRAprop Operational Creditor Recovery Rate under Quasi-APR PPE Plant, Property and Equipment

91

PPIRP Pre-Packaged Insolvency Resolution Process RA Resolution Applicant RBI Reserve Bank of India RBP Residual Bankruptcy Problem Reg. 30A Regulation 30A of the CIRP Regulations, 2016 RP Resolution Professional RR% Overall Recovery Rate Rs Rupees RV Realised Value SARFAESI Securitisation and Reconstruction of Financial Assets and Enforcement of Security Interest Act, 2002 Section 7 Section 7 of the Insolvency and Bankruptcy Code, 2016 Section 8 Section 8 of the Insolvency and Bankruptcy Code, 2016 Section 9 Section 9 of the Insolvency and Bankruptcy Code, 2016 Section 10 Section 10 of the Insolvency and Bankruptcy Code, 2016 Section 11A Section 11A of the Insolvency and Bankruptcy Code, 2016 Section 12A Section 12A of the Insolvency and Bankruptcy Code, 2016 Section 29A Section 29A of the Insolvency and Bankruptcy Code, 2016 Section 30(2)(b) Section 30(2)(b) of the Insolvency and Bankruptcy Code, 2016 Section 53 Section 53 of the Insolvency and Bankruptcy Code, 2016 Section 54A Section 54A of the Insolvency and Bankruptcy Code, 2016 Section 54C(3)(d) Section 54C(3)(d) of the Insolvency and Bankruptcy Code, 2016 SSPI Shapley–Shubik Power Index TReDS Trade Receivables Discounting System U.K. United Kingdom U.S. United States