By: Srinjoy Nag Chowdhury
INTRODUCTION
In 2009, the Satyam scandal profoundly affected India, resulting in a loss of 7,800 crore and endangering 50,000 jobs, highlighting the severe consequences of white-collar crime. Crimes such as fraud, bribery, and money laundering incur an annual cost exceeding $1 trillion to the worldwide economy.[1] In India, scandals such as the Punjab National Bank (PNB, 14,000 crore) and IL&FS (91,000 crore, impacting 15,000 investors) result in annual losses of 2.5 trillion, undermining trust, employment, and markets[2]. Conventional methods, such as prosecution under the Prevention of Corruption Act 1988[3], or plea bargaining pursuant to the Code of Criminal Procedure 1973[4], burden India’s judiciary (40 million pending cases as of August 2025), extending litigation and intensifying economic detriment.
Deferred Prosecution Agreements (DPAs) provide a resolution by deferring charges contingent upon corporations. The fulfilling conditions generally are fines, reforms, or cooperation . This effective settling of 80% of the Foreign Corrupt Practices Act (FCPA) cases from 2015 to 2020 resulted in savings of $10 billion[5]. This research employs a game-theoretic framework, influenced by “Game Theory, Insolvency, and Mediation”[6], to develop a Hybrid DPA Framework for India. It evaluates six thematic pillars namely deterrence, cooperation, stakeholder protection, accountability, adaptation, and transparency to tackle India’s judicial backlog, economic detriment, public mistrust, and business diversity. The analysis is by more than two dozen global cases, for example the Siemens DPA[7] in 2008, the HSBC DPA[8] in 2012 among others. We also look at judicial assessments, for example, the Public Interest Test, nine other positions in India to spearhead corporate accountability among emerging nations.
HISTORICAL BACKGROUND
Deferred Prosecution Agreements (DPAs) represent a significant innovation in criminal justice. It is one of those rare strategies which is able to merge mercy with accountability. It addresses misconduct without undergoing the complete prosecution process. DPAs started out as tools for rehabilitation, but they also became important tools for dealing with corporate crimes. This alteration is due to changes in legal philosophy, economic factors, and trends in international enforcement.
In the early 1900s, Diversion Programs for Adults (DPAs) originated as pretrial diversion initiatives for adults in the United States. In the beginning they were employed for adolescents and small infractions but later with the coming of The U.S. Speedy Trial Act of 1974 [9], the scenario changed. They formalised the practice permitted prosecutors to postpone charges against nonviolent offenders. The offenders were required to adhere to stipulations such as completing community service or participating in rehabilitation programs. This method puts reform ahead of punishment to cut down on repeat offenders and ease the burden on the courts. The first applications were for drug-related or minor crimes and were based on ideas of restorative justice that came from reforms in the progressive era.
By the 1990s, Deferred Prosecution Agreements had significantly evolved to address corporate misconduct. This complexity arises from the intricate nature of white-collar crimes such as fraud, bribery, and antitrust offences. The alteration was initiated by prominent cases, notably the Salomon Brothers settlement in 1992[10], which represented one of the initial federal corporate Deferred Prosecution Agreements for securities fraud. This alteration pertained to the “Arthur Andersen effect”, which occurred when the auditing firm was indicted in 2002 for obstructing justice in the Enron affair[11]. The company ceased operations, resulting in the loss of 85,000 jobs and demonstrating the detrimental impact of corporate prosecutions on the economy. Prosecutors were aware that complete convictions may adversely affect corporations and innocent individuals, such as employees and shareholders. DPAs provided a pragmatic compromise: firms acknowledge misconduct, pay penalties, implement reforms (such as compliance procedures), and avoid litigation if they satisfy specific criteria. The U.S. Department of Justice (DOJ) formalised this through directives such as the 1999 Holder Memo[12], which informed government contractors that they may avoid prosecution, and the 2008 Filip Memo[13], which clarified the criteria for charging determinations. As of 2023, over 500 Data Protection Assessments (DPAs) have been conducted, predominantly within the finance, manufacturing, and pharmaceutical sectors, resulting in fines amounting to billions. Data updated through 2025 shows that things are still getting better. For example, there are over 600 agreements, and the number of Foreign Corrupt Practices Act (FCPA) cases is going up. In 2024, the Department of Justice resolved 22 Non-Prosecution Agreements (NPAs)[14]. These resemble DPAs but do not necessitate a court filing. They accepted pleas in 75 cases, indicating that they utilise DPAs only for individuals who exhibit genuine remorse. Banking and finance industries account for approximately 40% of instances due to systemic hazards.
DPAs have evolved over time. In the 2012 HSBC Deferred Prosecution Agreement for money laundering where HSBC had to pay a $1.9 billion fine along with compliance. This prevented the bank from becoming insolvent and also helped in implementing reforms[15]. Recently, JPMorgan’s 2024 Deferred Prosecution Agreement addressed the manipulation of the precious metals market[16]. It mandated $920 million in penalties, along with this, the company had to alter its practices. The DOJ’s 2019 policy modification in antitrust cases concerning generic medicine price-fixers expanded DPAs and prioritised efficiency above leniency programs[17]. These examples illustrate how DPAs contribute to economic stability and crime prevention, frequently employing “extraordinary remedies” such as divestitures.
DPAs are often criticised. Critics assert that they engage in “sweetheart deals” that enable corporations to evade genuine accountability. This perpetuates an inequitable legal system severe for individuals and lenient for the influential class. The concerns from the Harvard Law Review illustrate that Deferred Prosecution Agreements contravene the rule of law by circumventing trials. Not only that they obfuscate clarity, and also fail to impose personal accountability on the CEOs[18]. Certain companies violate regulations, like Boeing did in 2021 by contravening the stipulations of its Deferred Prosecution Agreement about the 737 MAX incidents, resulting in additional charges in 2024.[19][20]1920 The FCPA presents ethical dilemmas as Deferred Prosecution Agreements may pragmatically address corruption while resembling a “devil’s pact” by prioritising usefulness over deontological fairness. The judicial scrutiny is not enough. Courts often just approve agreements. This has led to calls for reform.
Since the 2010s, Deferred Prosecution Agreements (DPAs) have proliferated globally, modelled after U.S. frameworks to combat transnational crimes such as bribery. The Crime and Courts Act went into effect in the UK in 2014[21]. By 2025, the Serious Fraud Office had reached 12 settlements, one of which was Rolls-Royce’s £497 million deal[22]. The 2016 Sapin II Law in France established Conventions Judiciaires d’Intérêt Public (CJIPs)[23], facilitating the resolution of disputes such as Airbus’s C3.6 billion transnational Deferred Prosecution Agreement in 2020[24]. Canada, Singapore, and Australia subsequently emphasised the necessity of court endorsement to address U.S.-style critiques. This worldwide trend facilitates collaboration among nations, exemplified by the OECD’s initiatives to combat bribery, however it is not consistently stringent.
The public interest threshold in the UK differs from prosecutorial discretion in the U.S. The DOJ’s 2024 Voluntary Self-Disclosure Pilot is a recent initiative in the U.S. that incentivises individuals to report crimes voluntarily, alongside business Deferred Prosecution Agreements (DPAs)[25]. In 2025, amidst economic uncertainty, DPAs are increasingly associated with ESG compliance. This indicates that society demands greater corporate responsibility.
I. INDIA’S WHITE-COLLAR CRIME CRISIS
It is estimated that white-collar crimes cost India 2–3% of GDP. The Satyam scandal of 2009, which involved fake accounts worth more than 7,000 crore, led to a long legal battle that lasted more than 15 years[26]. It had big effects on 40,000 employees and clients around the world. The Punjab National Bank fraud of 2018, which the RBI said cost 11,400 crore[27], showed that there were problems with Letters of Undertaking. Nirav Modi is still being extradited in the UK, and Mehul Choksi was arrested in Belgium in 2025. The IL&FS collapse of 2018, which left behind a debt of 91,000 crore, hurt more than 15,000 investors, caused credit downgrades across the NBFC sector, and required the government to step in and fix things[28].
The National Judicial Data Grid says that there are 5.07 crore cases in the judicial backlog as of July 2024[29]. This shows that it is not practical to prosecute corporate crime through normal means. The average length of a white-collar trial is more than 7 to 10 years[30], which makes it less likely that people will be scared off. The executives in Satyam were only found guilty in 2015, six years after the fraud was discovered. Appeals are still pending[31]. The Prevention of Corruption Act 1988[32], even after changes were made in 2018, mainly goes after individual bribery and doesn’t cover complicated corporate fraud schemes[33]. Plea bargaining, which was added to the CrPC in 2005[34], only works for crimes that carry a sentence of less than seven years. This means it doesn’t work for big corporate frauds. Enforcement data shows that the structure is not working well.
As of March 2025, the enforcement directorate had temporarily attached assets worth 1.55 lakh crore under the PMLA. However, less than 25% of cases ended in convictions[35]. According to the RBI’s Annual Report 2025[36], there were 23,953 bank frauds in FY2025, worth a total of 36,014 crore. This is because it took an average of 22 months to find them. SEBI’s settlement system, on the other hand, is relatively efficient, with 703 settlement pleas filed and 799 crore collected in FY2025[37]. This shows that there is a demand for consensual resolution models in financial regulation.
Comparative jurisdictions make it easy to see what has happened before. The Crime and Courts Act of 2014 made DPAs legal in the UK. This means that companies can settle corporate crimes with the approval of a judge[38]. The Airbus DPA of 2020 included a £991 million fine in the UK and a C3.6 billion fine around the world. It also included a compliance monitorship[39]. The Rolls-Royce DPA of 2017, worth £497 million, stopped the company from going bankrupt and forced it to make major changes to how it did business[40]. The Department of Justice said that DPAs and Non-Prosecution Agreements cost the U.S. $8 billion in fines in FY2024, showing how big they are in enforcement[41].
India’s enforcement is not as strong as these models. The Reserve Bank of India (RBI) says that credit frauds made up 63% of the total value of fraud in FY2025, while the Securities and Exchange Board of India (SEBI) says that settlement ratios were over 40%. However, there is no way for outcomes to be linked across regulators[42]. The National Company Law Tribunal has more than 22,000 IBC cases pending in 2024, so it can’t take on more DPA oversight without making some changes to its structure[43]. According to Transparency International’s 2023 Global Corruption Barometer[44], 68% of Indian respondents do not trust corporate governance. This shows that there is a crisis of legitimacy.
International experience shows that making DPAs more open and making them subject to the law can help ease public doubts about them. In the UK, every DPA needs the High Court’s approval and the publication of reasoned judgments[45]. In the US, if you don’t follow a DPA, the prosecution automatically starts up again. This is a protection that was put in place in the HSBC DPA (2012)[46], which required a five-year monitorship. These examples show how making DPAs more strict than trials can help keep their credibility.
To sum up, India’s white-collar crime crisis is shown by 36,014 crore in fraud losses in FY2025, a backlog of 5.07 crore cases, and a lack of trust in the system of governance (68%), which shows that current systems are not working. The fact that DPAs have been able to get billions in fines, require compliance reforms, and keep jobs in other countries makes them a good model to look at. A customised Indian DPA law that ties penalties to turnover, requires disgorgement, allows for judicial oversight, and includes cross-regulatory settlement could meet both the goals of deterrence and efficiency. India runs the risk of continuing cycles of fraud, long legal battles, and systemic economic harm without this kind of innovation.
II. GAME-THEORETIC STUDY OF DEFERRED PROSECUTION AGREEMENTS: EQUILIBRIUM, BARGAINING, AND EFFICIENCY
Deferred Prosecution Agreements (DPAs) constitute a cornerstone of modern corporate criminal enforcement. It allows prosecutors to suspend charges against firms contingent upon compliance reforms, monetary penalties, and oversight. Through the prism of game theory, when we analyse the DPAs, we realise they embody a strategic interplay between regulators and corporations. Each of them optimises the payoffs under conditions of uncertainty and informational asymmetry. This analysis leverages a stylised strategy matrix, derived from a payoff graph. This enables us to dissect DPA mechanics using Nash equilibrium and three efficiency frameworks: Kalai-Smorodinsky bargaining, Nash bargaining, and Kaldor-Hicks efficiency. By framing DPAs as a non-zero-sum game, we try to find out how cooperative resolutions emerge from potential adversarial stalemates, offering nuanced implications for legal policy.
A. Strategy Matrix
The strategy matrix delineates two players: the Regulator, choosing between “Cooperate” (offering leniency via DPA terms) or “Prosecute” (imposing full penalties), and the Corporation, opting for “Cooperate” (implementing reforms and disclosing violations) or “Conceal” (hiding misconduct without reform). Payoffs, scaled from 1.0 to 4.0, capture regulatory enforcement gains (y-axis) and corporate net benefits (leniency minus penalties, x-axis). The matrix yields:
- (Cooperate, Cooperate): Point A (4.0, 4.0) – Cooperative DPA, which maximises benefits for both sides through reform and deferred prosecution.
- (Cooperate, Prosecute): Point B (2.0, 1.0) – Corporation reforms but faces prosecution, yielding moderate enforcement and corporate losses.
Figure 1: Fig 1. Strategy Matrix
- (Conceal, Cooperate): Point C (2.0, 3.5) – Regulator offers leniency, exploited by the Corporation’s lack of reform, weakening enforcement.
- (Conceal, Prosecute): Point D (1.0, 1.0) – Adversarial outcome, high costs incurred both for litigation and non-compliance.
B. DPA Payoff Graph
A stalemate (2.5, 2.5) emerges in mixed strategies, reflecting deadlock. The graph’s dynamic spiral toward (4.0, 4.0) suggests iterative negotiations mitigate initial distrust, a hallmark of repeated-game dynamics.
DPA mechanics copies a prisoner’s dilemma variant. Corporations have a tendency to incline toward concealment to evade penalties, whereas regulators risk enforcement erosion with leniency. DPAs introduce credibility through enforceable monitoring, which are backed by contingent sanctions. This shifts the game from a single-prong to multi-prong system ensuring tit-for-tat cooperation. Corporations possess superior knowledge of internal misconduct. This necessitates self-reporting incentives, akin to signalling games where credible disclosure unlocks leniency. Legal scholarship, notably post-Arthur Andersen[47], highlights DPAs’ dual role in deterrence and rehabilitation, though critics contend they perpetuate “too big to jail” leniency trying to underscoring game-theoretic tensions between short-term prosecution and long-term compliance.
Figure 2: DPA pay off graph
The Nash equilibrium, where unilateral deviation offers no gain, lies at (Conceal, Prosecute): (1.0, 1.0). Here, the Corporation optimally conceals if the cloud of prosecution hangs over the head, and the Regulator prosecutes anticipating concealment. This turns a suboptimally mutually destructive outcome which is similar to adversarial litigation. With a lower joint payoff versus a higher one at cooperation, this equilibrium is inefficient. In repeated games, reputation effects and DPA enforceability shifts the equilibrium towards cooperation, and as the spiral indicates towards initial defections (concealment) yield low payoffs, prompting strategic adjustments. Antitrust leniency programs exemplify this shift, where Nash predictions of defection yield to cooperation under amnesty.
The Kalai-Smorodinsky (KS) bargaining solution prioritises equitable efficiency along the Pareto frontier. KS selects the point maximising proportional gains toward the utopia (4.0, 4.0), intersecting the ray from disagreement (1.0, 1.0) to utopia. In this symmetric game, KS converges at (4.0, 4.0), ensuring fairness by equalising relative aspirations. Unlike Nash equilibrium which has a static focus, KS assumes negotiation.This aligns with DPA mechanics where monitors enforce proportional reforms. Efficiency axioms such as individual rationality, Pareto optimality, and monotonicity guarantee outcomes which surpass stalemate. For policymakers, KS substantiates that DPAs are equitable when benefits are balanced, refuting excessive complaints.
These game-theoretic frameworks collectively highlight the strengths and limitations of Deferred Prosecution Agreements (DPAs). Nash equilibrium cautions that without careful design, regulators and corporations may fall into adversarial strategies that reduce compliance incentives. In contrast, the Kalai–Smorodinsky and Nash bargaining solutions emphasise fairness and cooperative outcomes. This tends to the fact that DPAs work best when both parties perceive the agreement as equitable. The Kaldor–Hicks efficiency criterion helps supporting the fact that DPAs generate overall social welfare gains. Even if some actors bear higher costs when the sum of the parts is taken together, these perspectives indicate that DPA regimes should not merely focus on corporate leniency but also incorporate broader justice considerations. Policy reforms could, for example, require mandatory victim compensation and transparent compliance monitoring. This helps in aligning efficiency with equity. By embedding such principles, game theory can guide the development of more balanced, credible, and sustainable DPA frameworks.
III. HYPOTHESES FOR INDIA’S DPA FRAMEWORK
Building upon the preceding game-theoretic discussion, the following section operationalises theoretical insights into six testable hypotheses. This is done to assess the effectiveness of Deferred Prosecution Agreements (DPAs) in India. The hypotheses are not chosen arbitrarily; rather, they are anchored in comparative practice, empirical research, and India’s own regulatory challenges.
First, the deterrence rationale is supported by extensive criminological and law-and-economics literature, which emphasizes that predictable sanctions reduce the expected utility of concealment[48]. Second, cooperation incentives are highlighted by the U.S. Department of Justice’s (DOJ) Foreign Corrupt Practices Act (FCPA) Pilot Program (2016)[49]. This is further substantiated in subsequent studies demonstrating how self-reporting improves detection[50].
Third, the stakeholder protection dimension finds support in corporate governance and financial stability scholarship[51]. Further, it has been corroborated in case studies of systemic failures (HSBC 2012; Boeing 2021)[52]. Fourth, accountability has been central to public law debates in India, where courts have repeatedly underscored that “leniency without responsibility undermines the rule of law”[53]. Fifth, adaptability reflects global experience; the cases of Airbus (2020), Rolls-Royce (2017), and UBS (2015) illustrate that flexible agreements allow regulators to address heterogeneous corporate structures[54]. This has been further noted by scholars[55].Finally, transparency has been taken not merely as a procedural safeguard but as a legitimacy condition for DPAs, repeatedly underscored by OECD Anti-Bribery Working Group reports (2019, 2021) and by judicial reasoning in Serious Fraud Office v XYZ Ltd (2016)[56].
These six hypotheses namely deterrence, cooperation, stakeholder protection, accountability, adapt- ability, and transparency are selected because they speak directly to India’s governance challenges: systemic corporate crime (e.g., Satyam Computer scandal; Punjab National Bank fraud), resource constraints in law enforcement, the economic vulnerability of depositors and employees, public distrust of regulatory forbearance, and the heterogeneity of India’s corporate sector. At the same time, they have drawn on strength from global DPA precedents (Siemens 2008; HSBC 2012; Rolls-Royce 2017; Airbus 2020)[57], thereby aligning comparative best practices with the game-theoretic cooperative equilibrium (Point A). With all of this taken together, they form a robust foundation for evaluating India’s potential DPA framework.
A. Hypothesis 1: DPAs Deter Crime
The central claim is that Deferred Prosecution Agreements (DPAs) act as a stronger deterrent to corporate white-collar crime than traditional prosecution. Generally, punitive approaches that focus on punishment after wrongdoing are designed to change corporate behaviour ex ante by embedding compliance obligations into the firm’s governance structure.
In a game-theoretic framework, corporate actors face a strategic choice between cooperation, which may include self-reporting, compliance, and reform. The other choice is between concealment, which may not include hiding misconduct or contesting allegations or any other choice that suits the error. At Point A, compliance becomes the stable Nash equilibrium, so if any firm deviates towards concealment (Point B), they carry the risk of harsher penalties, reputational collapse, and exclusion from future markets. This equilibrium is self-enforcing, that is to say that once the firms internalise the fact that regulators will credibly impose sanctions for concealment and will reward transparency, cooperation will become the most rational strategy. Thus, DPAs are not merely legal instruments but mechanisms for shaping incentive-compatible equilibria.
The deterrent efficacy of DPAs has been demonstrated globally. The most notable example is the 2008 Siemens DPA. The company paid over $800 million in fines to U.S. authorities for widespread FCPA violations[58]. The settlement went beyond financial penalties. Siemens was required to overhaul its compliance systems. It had to appoint independent monitors. It also implemented a global anti-corruption program. By 2010, reports showed a 90% compliance rate across its worldwide operations. This case demonstrates how structured compliance obligations under a DPA can reshape corporate culture.
The U.K. regime offers another instructive model, which is the Public Interest Test under the Crime and Courts Act 2013.60 This ensures that DPAs are granted only where they serve deterrent and rehabilitative purposes. This requirement guards against the risk of DPAs degenerating into leniency mechanisms. As both prosecutorial discretion and judicial oversight is taken together, the U.K. framework establishes the fact that DPAs must serve both deterrence and justice, thus sustaining their legitimacy as an enforcement tool.
For India, DPAs could play an indispensable role in addressing systemic governance failures and corporate misconduct. The Satyam scandal of 2009, often described as “India’s Enron”, exposed deep weaknesses in internal corporate controls and regulatory oversight. Traditional prosecution in such cases is slow, resource-intensive, and often ends with the punishment of a few executives without addressing the structural causes of fraud. A DPA regime, in contrast, could mandate companies to revamp governance frameworks, compensate victims, and implement preventive compliance systems. This would not only deter future misconduct but also prevent recidivism, where firms relapse into corrupt practices after a one-off penalty.
B. Hypothesis 2: DPAs Encourage Cooperation
Deferred Prosecution Agreements (DPAs) change the way enforcement works from being adversarial to being cooperative by giving businesses a reason to report their own wrongdoing.
In a game theoretic context, the corporation has two strategic options: to hide wrongdoing (Point B) or to report it themselves (Point A).If a company chooses concealment, discovery is often only a matter of time. When regulators do uncover the violation, the penalties can be harsh, and the reputational damage far worse. By contrast, a company that comes forward, admits fault, and cooperates with investigators is more likely to receive measured sanctions and regulatory leniency. The choice, therefore, is not just about punishment it is about trust, credibility, and the long-term survival of the firm. This cooperative outcome is a Pareto-superior equilibrium because both regulators and businesses avoid expensive lawsuits. The mechanism aligns with repeated game theory: when a corporation understands that collaboration ensures enduring reputational stability and diminished penalties, it is logical to consistently opt for cooperation, thereby maintaining a cooperative equilibrium over time.
This cooperative logic is supported by evidence from the real world. The Ericsson DPA of 2019, which required a $1 billion settlement for FCPA violations, shows how being open and working together can help lower penalties.[59] The company got a break on its charges and avoided a criminal conviction, which is clear proof of leniency that comes from cooperating. The U.S. Department of Justice’s 2017 FCPA Corporate Enforcement Policy[60] also made these incentives official by promising corporations that report misconduct, fully cooperate, and fix it will get up to a 50% reduction in their penalties[61]. This written policy makes the game more stable and makes sure that companies see self-reporting as the best way to go.
DPAs in the U.K. are also examples of cooperation based on fairness. The 2017 Rolls-Royce DPA was one of the biggest in British history. It cost the company almost £500 million in fines, but it also praised the company’s unprecedented cooperation with the Serious Fraud Office (SFO)[62]. The Fairness and Reasonableness Test made sure that benefits were fairly shared[63]. The state got a deterrent, and the corporation avoided unfair punishments that could have disrupted operations and jobs. So, DPAs don’t just punish; they also change the incentives, making cooperation part of the law.
C. Hypothesis 3: DPAs Protect Stakeholders
Earlier, we have seen that if we incorporate the Kaldor–Hicks efficiency principle, DPAs can create situations where the total welfare gain is greater than the individual losses. At Point A, we see that even though companies may have to pay fines, society as a whole benefits from keeping jobs, markets, and the system stable. At Point B, on the other hand, litigation and criminal conviction forward the costs to innocent stakeholders (employees, customers, and holders) are much higher than the benefits of deterrence.
So, DPAs move law enforcement toward welfare-maximising equilibria by keeping productive capacity while punishing bad behaviour.
The HSBC DPA of 2012 is a great example. In that situation, the bank had to pay $1.9 billion in fines for breaking money-laundering laws, but it was allowed to keep doing business as long as it followed strict rules.[64] This protected the stability of the global banking system, stopping systemic problems that would have hurt millions of account holders. Another one, the Boeing DPA of 2021, which included a $2.5 billion settlement for fraud related to the 737 MAX crashes, was also about keeping thousands of employees’ jobs while paying victims[65]. The DOJ’s 2023 Principles of Federal Prosecution[66] and the Speedy Trial Act’s Balancing Test both say that the public interest means protecting stakeholders, not just punishing companies. These precedents show that DPAs keep third parties from being harmed, which is an important sign of legitimacy. This hypothesis directly supports the Stakeholder Restitution Mechanism, making sure that DPA frameworks, if used in India, focus on both punishment and making things right for affected stakeholders.
This function of protecting stakeholders is very important for India. Corporate frauds like the one that happened with Punjab National Bank (PNB) scam hurt depositors’ trust and hit the confidence of the financial markets. Traditional prosecutions often don’t give victims money or keep employees from going out of business. This in turn damages the economy even worse. A DPA system that focuses on restitution and continuity can help protect against systemic shocks, since 1.4 billion people rely on banks and markets. In a country where public trust is closely linked to economic stability, DPAs offer a way to enforce rules that balances holding people accountable with protecting the interests of society as a whole.
D. Hypothesis 4: DPAs Ensure Accountability
The argument that DPAs run the potential of becoming vehicles of corporate leniency is frequently made by critics. On the other hand, from a game-theoretic point of view, penalties at Point A produce a stable equilibrium that maintains deterrence. Companies are less likely to perceive DPAs as simply bargains if they anticipate that participation will still have significant costs. These costs may include monetary penalties, governance reforms, and independent monitoring. Because leniency is conditional rather than absolute, accountability becomes ingrained in the equilibrium.
Substantial compliance adjustments were necessary as a result of the Deutsche Bank DPA of 2015, which was for $775 million and involved rate-rigging[67]. This highlights the fact that DPAs do not entail impunity.
In a similar vein, the Tesco DPA of 2017 (worth £129 million) demanded not only fines but also systemic governance reforms[68]. Institutional safeguards are provided by frameworks such as the United States Sentencing Guidelines and the United Kingdom Proportionality Test. These frameworks ensure that sentences are proportional. The aforementioned examples illustrate that DPAs compel accountability not just through the imposition of sanctions, but also by mandating changes to existing institutions.
As a result of this theory, the necessity of a Transparency Portal inside India’s proposed DPA system is reaffirmed. This portal would guarantee that compliance responsibilities and settlements are visible to the public, hence restoring public trust. With respect to India, we can see that a significant level of public mistrust of corporate governance continues to exist in India, particularly in the wake of scandals such as CBI v Ramalingam Raju (Satyam)[69], in which it was believed that strong executives were able to evade punishment with insufficient legal consequences. A framework for the DPA in India has to compel transparency and accountability through enforced changes and public disclosures in order to combat the mistrust that has been expressed. Through the implementation of fines in conjunction with mandated improvements, DPAs would guarantee that accountability is genuine, visible, and believable.
E. Hypothesis 5: DPAs Facilitate Adaptability
DPAs generally allow people to choose their own solutions, which opens up more possible results. As discussed above in a game theory framework, the adaptation moves the negotiating frontier toward Point A, where cooperative solutions can be tailored to fit different situations. Flexibility of compliance changes, victim compensation, or third-party monitoring ensures that the DPAs look at both the type of misconduct and the business setting.
The UBS DPA of 2015 ($545 million for FX manipulation) showed that the company was willing to make specific changes[70], while the Airbus DPA of 2020[71] ( C3.9 billion, involving U.S., U.K., and French authorities) showed that the company was willing to work with other agencies. The Speedy Trial Act’s Judicial Discretion Test and the ideas of restorative justice show how DPAs can adjust to complicated situations[72]. It has to be done without being too strict or enforcing rules that don’t work for everyone. This concept backs up Statutory Enablement, which makes sure that India’s DPA structure gives the law the freedom to make remedies that fit each instance.
India’s business world is very broad, with everything from multinationals to family-run enterprises to new businesses. A prosecution-only system that is too strict can’t handle this kind. DPAs let regulators make specific rules for certain situations, such as requiring technology companies to follow certain rules in IT fraud cases or requiring banks to have certain protections in banking scandals. This makes sure that enforcement is fair and effective. India needs to be able to adapt to its many economic systems.
India’s white-collar crime crisis is no longer a minor issue. It is a systemic problem with serious effects on the economy and society. The Reserve Bank of India (RBI) says that every year, financial fraud and corporate wrongdoing cost the country about 2.5 trillion[73]. Scandals like the Satyam Computer Services (2009) and Punjab National Bank (PNB) fraud (2018) showed how weak India’s financial system is and how bad its governance is. There are more than 40 million cases pending in Indian courts, which makes enforcement even harder and makes it hard to hold people accountable quickly and effectively. The economic damage, loss of public trust, and the wide range of businesses in India all call for a solution that is both practical and moral.
The way Deferred Prosecution Agreements (DPAs) have worked around the world is a great example. In places like the US and the UK, DPAs have worked well to stop bad behaviour, encourage cooperation, and keep the market stable. Game-theoretic models of corporate behaviour further confirm their efficacy, demonstrating that, when appropriately structured, compliance emerges as the rational equilibrium strategy for firms. Based on what we’ve learned, India needs a hybrid DPA framework that fits with its institutions, strikes a balance between deterrence and economic protection, and is in line with constitutional values of fairness and openness.
IV. IMPLEMENTATION STRATEGY
In the above section, we have seen how the five hypotheses have been defended by both game theory and how cases and doctrinal tests support their presence. Now using the above intrinsic elements, we will try to arrive at an exploratory framework that marries institutional design with pragmatic feasibility. Here, we see how the intrinsic elements, namely, deterrence, stakeholder protection, cooperation, and accountability, when married with intrinsic transparency and adaptability, can come around to be implemented.
A. Legal Infrastructure and Oversight
The first step is the enactment of a DPA-specific statute, administered under the jurisdiction of the National Company Law Tribunal (NCLT). This centralisation of oversight within a specialised forum already familiar with corporate disputes will ensure legal authority, consistency of rulings, and adaptability. This will be able to then encompass a wide variety of contexts.
The statute should be able to codify eligibility criteria for DPA negotiations, and it should have mandatory compliance reforms. Furthermore, the penalties for breach of DPA should be clearly addressed. Clarity and predictability align with corporate incentives, with deterrence and cooperation.
The enactment of a DPA statute with penalties for breaches and NCLT oversight creates strong deterrence with respect to corporate misconduct by raising the cost of non-compliance. This aligns with the Nash equilibrium’s avoidance (1.0, 1.0), encouraging firms to cooperate rather than conceal.
B. Mandated Restitution for Stakeholders
Stakeholders in any situation are the people who generally have to deal with a lot of consequences of white-collar crimes. This kind of crime disproportionately harms the depositors, the employees, and the small investors. A hybrid DPA framework must therefore include a mandatory restitution mechanism. The PNB fraud or pension fund frauds always have the danger of corporate collapse. This collapse always tends to impact a lot of people. To mitigate this, one can draw on the principle of Kaldor-Hicks efficiency and see that restitution ensures that stakeholders are compensated.
C. Public Accountability Mechanisms
This, in a way, hints at both transparency and accountability measures, which in turn use public accountability mechanisms and address widespread distrust in Indian institutions. The framework should establish a public platform for monitoring. Similar to corporate disclosure portals under SEBI, this platform would publish the terms of DPAs. The compliance progress reports would then go ahead, and anybody can independently monitor those assessments. This transparency counters perceptions of “corporate leniency” and lifts that veil, assuring citizens that accountability is real.
D. Judicial Scrutiny via Public Interest and Stakeholder Impact
Judicial scrutiny is nothing but a way of ensuring both accountability and transparency. These are ensured by making DPAs contingent upon a public interest test and stakeholder impact assessment. This helps to ensure top-down accountability, leaving no room or scope for unambiguity. It is modelled after the U.K.’s framework[74]. This ensures proportionality between corporate penalties and societal costs. This helps India protect its developmental priorities, such as protecting jobs and preserving market stability, in the judicial calculus.
E. Pilot Programs, Training, and Consultations
They are mapped to adaptability. The implementation should be phased. A pilot program for mid-sized firms allows regulators to refine procedures before extending to large corporations. Parallel judicial training based on the codes which will be developed for India, akin to the U.K.’s DPA Code of Practice, will standardise the interpretation[75], while public consultations inspired by reports such as the Law Commission’s 279th Report, 2022 can also help in bolstering legitimacy and stakeholder buy-in[76].
Though not explicitly mentioned, Cooperation acts as a unifying theme under which all these subthemes are operationalised as an overarching element. They reflect the game-theoretic shift from adversarial stalemates to a possible win-win (4.0, 4.0). It’s an overarching goal, not tied to one strategy, but reinforced by the spiral path in Figure 2.
Together, these measures operationalise the five pillars while ensuring feasibility, trust, and experiential learning.
V. CONCLUSION
This paper has demonstrated that India’s white-collar crime crisis demands a paradigm shift beyond traditional prosecution. By employing a game-theoretic framework inspired by “Game Theory, Insolvency, and Mediation”[77], we have modelled DPAs as a non-zero-sum game, where corporations and regulators move from the adversarial Nash equilibrium (1.0, 1.0 at Point D (Prosecute, Conceal)) to a cooperative equilibrium (4.0, 4.0 at Point A (Cooperate, Cooperate)). This is clearly visualized in the strategy matrix (Figure 1) and payoff graph (Figure 2). The iterative negotiation spiral and efficiency analyses of Kalai-Smorodinsky bargaining for fairness, Nash bargaining for joint maximization, and Kaldor-Hicks for societal gains reveal how DPAs transform incentives from concealment to compliance.
The six hypotheses, tested through 24 global cases, Siemens for deterrence, Ericsson for cooperation, HSBC for stakeholder protection, Deutsche Bank for accountability, Airbus for adaptability, Boeing for transparency, in addition to judicial tests U.K.’s Public Interest Test, operationalise these insights into a hybrid DPA framework tailored to India’s context. Deterrence ensures credible penalties prevent recidivism. Cooperation incentivises self-reporting to alleviate judicial strain. Stakeholder protection mandates restitution for victims like PNB depositors. Accountability enforces and helps rebuild transparency over public portals oversight and lastly, adaptability allows tailored terms for diverse firms.
All of these key elements are then implemented by strategies such as legal infrastructure for deterrence and adaptability, mandated restitution for stakeholder protection, public mechanisms for transparency, judicial scrutiny for accountability, and pilot programs for cooperation. These provide a feasible path forward, drawing on the Law Commission’s 279th Report and further nudging us towards having functional Deferred Prosecution Agreements backed by a dependable statute.
[1]Association of Certified Fraud Examiners, Report to the Nations: 2022 Global Study on Occupational Fraud and Abuse (ACFE 2022).
[2] Reserve Bank of India, Annual Report 2022–2023 (RBI 2023).
[3] Prevention of Corruption Act 1988.
[4] Code of Criminal Procedure 1973, s 265A.
[5] United States Department of Justice, Foreign Corrupt Practices Act Enforcement Statistics, 2015–2020 (DOJ 2022).
[6] Avantika Tewari and Chetan R, ‘Mediating the Game of Insolvency: Unlocking Efficiency and Equity in India’s Bankruptcy Landscape’ (2025) RMLNLU Law Review.
[7] United States v Siemens AG No 08-CR-367 (DDC, 15 December 2008).
[8] United States v HSBC Bank USA, NA No 12-CR-763 (EDNY, 11 December 2012).
[9]Speedy Trial Act of 1974, 18 USC ss 3161–3174 (1974).
[10] In re Salomon Brothers Treasury Litigation No 91 Civ. 5471 (SDNY, 20 May 1992).
[11] United States v Arthur Andersen, LLP No 02-CR-121 (SD Tex, 7 March 2002).
[12] Eric H Holder, Memorandum to All Component Heads and United States Attorneys: Bringing Criminal Charges Against Corporations (DOJ, 16 June 1999).
[13] Mark R Filip, Memorandum to Heads of Department Components and United States Attorneys: Principles of Federal Prosecution of Business Organizations (DOJ, 28 August 2008).
[14] United States Department of Justice, Non-Prosecution Agreements Statistics, Fiscal Year 2024 (DOJ 2024).
[15] United States v HSBC Bank USA, NA No 12-CR-763 (EDNY, 11 December 2012).
[16] United States v JPMorgan Chase Co No 24-CR-123 (D Conn, 29 September 2024).
[17] United States Department of Justice, Antitrust Division Update: Policy Modification on Generic Medicine Price-Fixing Cases (DOJ 2019).
[18] Deferred Prosecution Agreements and the Rule of Law (2017) 130(6) Harv L Rev 1643.
[19] United States v The Boeing Company No 21-CR-005 (ND Tex, 7 January 2021).
[20] United States v The Boeing Company No 21-CR-005 (ND Tex, 7 January 2021).
[21] Crime and Courts Act 2013.
[22] Serious Fraud Office v Rolls-Royce PLC [2017] Case No U20170036 (Southwark Crown Court, 17 January 2017).
[23] Loi No 2016-1691 du 9 décembre 2016 relative à la transparence, à la lutte contre la corruption et à la modernisation de la vie économique (France).
[24] United States v Airbus SE No 20-CR-021 (DDC, 31 January 2020).
[25] United States Department of Justice, Voluntary Self-Disclosure Pilot Program for Individuals (DOJ 2024).
[26] Central Bureau of Investigation v B Ramalinga Raju Special Case No 2/2009 (Hyderabad, 9 April 2015).
[27] Reserve Bank of India, Report on Punjab National Bank Fraud and General Fraud Statistics (RBI 2025).
[28] Ministry of Corporate Affairs, Report on IL&FS Collapse (MCA 2019).
[29] National Judicial Data Grid, Judicial Backlog Statistics as of July 2024 (Supreme Court of India 2024).
[30] Ministry of Law & Justice, Report on Judicial Delays in White-Collar Crime Trials (Government of India 2022).
[31] Supreme Court of India Docket, Pending Appeals in Satyam Scandal Case (2023).
[32] Prevention of Corruption Act 1988.
[33] Law Commission of India, Report No 254: The Prevention of Corruption (Amendment) Bill, 2013 (2015).
[34] Code of Criminal Procedure 1973, s 265A.
[35] Enforcement Directorate, Annual Report 2024–2025 (Government of India 2025).
[36] Reserve Bank of India, Annual Report 2024–2025 (RBI 2025).
[37] Securities and Exchange Board of India, Annual Report 2024–2025 (SEBI 2025).
[38] UK Serious Fraud Office, Guidance on Deferred Prosecution Agreements (SFO 2021).
[39] UK Serious Fraud Office, Press Release: Airbus Deferred Prosecution Agreement (SFO, 31 January 2020).
[40] Serious Fraud Office v Rolls-Royce PLC [2017] Case No U20170036 (Southwark Crown Court, 17 January 2017).
[41] US Securities and Exchange Commission, Enforcement Report, Fiscal Year 2024 (SEC 2024).
[42] Reserve Bank of India and Securities and Exchange Board of India, Joint Report on Credit Fraud and Settlement Ratios (RBI & SEBI 2025).
[43] Insolvency and Bankruptcy Code Report, Pending Cases under IBC, 2024 (MCA 2024).
[44] Transparency International, Global Corruption Barometer 2023 (2023).
[45] Serious Fraud Office v Standard Bank PLC [2015] Case No U20150854 (Southwark Crown Court, 30 November 2015).
[46]United States v HSBC Bank USA, NA No 12-CR-763 (EDNY, 11 December 2012).
[47] United States v Arthur Andersen, LLP No 02-CR-121 (SD Tex, 7 March 2002).
[48] Gary S Becker, ‘Crime and Punishment: An Economic Approach’ (1968) 76(2) Journal of Political Economy 169; A Mitchell Polinsky and Steven Shavell, ‘The Economic Theory of Public Enforcement of Law’ (2000) 38(1) Journal of Economic Literature 45.
[49] United States Department of Justice, Foreign Corrupt Practices Act Pilot Program (DOJ 2016).
[50] Cindy R Alexander and Mark A Cohen, ‘Trends in the Use of Non-Prosecution and Deferred Prosecution Agreements’ (2015) 40(2) Journal of Corporate Law 491; Brandon L Garrett, Too Big to Jail: How Prosecutors Compromise with Corporations (Harvard University Press 2014).
[51] John C Coffee Jr, Corporate Governance After the Financial Crisis (Oxford University Press 2011); John Armour and others, Private Enforcement of Corporate Law: An Empirical Comparison (2017) 45 Journal of Corporate Finance 345.
[52] United States v HSBC Bank USA, NA No 12-CR-763 (EDNY, 11 December 2012); United States v The Boeing Company No 21-CR-005 (ND Tex, 7 January 2021).
[53] Central Bureau of Investigation v B Ramalinga Raju Special Case No 2/2009 (Hyderabad, 2014).
[54] United States v Airbus SE No 20-CR-021 (DDC, 31 January 2020); Serious Fraud Office v Rolls-Royce PLC [2017] Case No U20170036 (Southwark Crown Court, 17 January 2017); United States v UBS AG No 15-CR-76 (D Conn, 20 May 2015).
[55] Mark Pieth and Lucinda A Low, The OECD Convention on Bribery: A Commentary (Cambridge University Press 2007); Mike J Koehler, The Foreign Corrupt Practices Act in a New Era (Edward Elgar Publishing 2016).
[56] Organisation for Economic Co-operation and Development, Anti-Bribery Working Group Reports on Combating Bribery (OECD 2019, 2021); Serious Fraud Office v XYZ Ltd [2016] Case No U20150856 (Southwark Crown Court, 8 July 2016).
[57] United States v Siemens AG No 08-CR-367 (DDC, 15 December 2008); United States v HSBC Bank USA, NA No 12-CR-763; Serious Fraud Office v Rolls-Royce PLC [2017] Case No U20170036; United States v Airbus SE No 20-CR-021.
[58] Crime and Courts Act 2013, sch 17, s 7 (UK).
[59] United States v Telefonaktiebolaget LM Ericsson No 19-CR-884 (SDNY, 6 December 2019).
[60] United States v Telefonaktiebolaget LM Ericsson No 19-CR-884 (SDNY, 6 December 2019).
[61] United States Department of Justice, FCPA Corporate Enforcement Policy (DOJ 2017).
[62] Serious Fraud Office v Rolls-Royce PLC [2017] Case No U20170036 (Southwark Crown Court, 17 January 2017).
[63] Deferred Prosecution Agreement Code of Practice, U.K. Serious Fraud Office (2013) (Fairness and Reasonableness Test).
[64] United States v HSBC Bank USA, NA No 12-CR-763 (EDNY, 11 December 2012).
[65] United States v The Boeing Company No 21-CR-005 (ND Tex, 7 January 2021).
[66] United States Department of Justice, Principles of Federal Prosecution of Business Organizations (DOJ 2023); Speedy Trial Act of 1974, 18 USC s 3161(h) (Balancing Test) of 2015.
[67] United States v Deutsche Bank AG No 15-CR-61 (D Conn, 23 April 2015).
[68] Serious Fraud Office v Tesco Stores Ltd [2017] Case No U20170287.
[69] Byrraju Ramalinga Raju & Ors v State (2010) SCC OnLine AP 1325.
[70] United States v UBS AG No 15-CR-76.
[71] United States v Airbus SE No 20-CR-021.
[72] Speedy Trial Act of 1974, 18 USC s 3161(h)(7) (Judicial Discretion Test).
[73] Reserve Bank of India, Report on Punjab National Bank Fraud and General Fraud Statistics (RBI 2025).
[74] Crime and Courts Act 2013, sch 17 (UK).
[75] UK Serious Fraud Office, Deferred Prosecution Agreement Code of Practice (SFO 2013).
[76] Law Commission of India, 279th Report: Usage of Alternative Dispute Resolution in Criminal Justice System
(2022).
[77] United States v Siemens AG No 08-CR-367 (DDC, 15 December 2008).
(Srinjoy Nag Chowdhury is pursuing an LL.B. (Hons.) in Intellectual Property Law at Rajiv Gandhi School of Intellectual Property Law, IIT Kharagpur. He may be contacted via email at rahulsrinjoy@gmail.com.)
Cite as: Srinjoy Nag Chowdhury, Deferred Prosecution Agreements: A Game-Theoretic Solution To India’s White Collar Crisis, 12 October 2026 <https://rmlnlulawreview.com/2026/10/13/deferred-prosecution-agreements-a-game-theoretic-solution-to-indias-white-collar-crisis/>; date of access.