When China Evergrande Group collapsed, the immediate questions were about the company. How had one of China’s largest property developers accumulated such an extraordinary debt burden? How had its reported financial position diverged so far from economic reality? And how had PwC continued to issue unmodified audit opinions while the underlying problems developed?
The regulatory investigations that followed produced unusually detailed answers. Chinese authorities concluded that Evergrande’s principal mainland subsidiary overstated revenue by RMB213.989 billion in 2019 and RMB350.157 billion in 2020. Hong Kong regulators later concluded that Evergrande’s reported profits of RMB33.5 billion and RMB31.4 billion in those years should instead have been losses of approximately RMB7.1 billion and RMB19.9 billion. PwC Hong Kong subsequently agreed to set aside HK$1 billion for eligible minority shareholders under an agreement with the Securities and Futures Commission, without admitting liability. Separately, the Accounting and Financial Reporting Council found serious auditor misconduct across audit evidence, professional scepticism, independence, going concern, review and quality management. (China Securities Regulatory Commission – Administrative Penalty Decision on PwC Zhong Tian; Hong Kong Securities and Futures Commission – PwC Evergrande Shareholder Compensation Agreement; Accounting and Financial Reporting Council – Evergrande Audit Sanctions)
That would already make Evergrande one of the most significant Big Four audit failures of recent years. But the case is becoming structurally more important. Evergrande’s liquidators are not only pursuing the PwC firms that conducted the audits. They are also pursuing PricewaterhouseCoopers International Limited, the entity that coordinates the global PwC network. In August 2026, the Hong Kong Court of First Instance refused to remove PwC International from the proceedings. It did not find PwC International negligent or establish that it owed Evergrande a duty of care. It decided that the allegations concerning its role in governing, monitoring and reviewing member firms were sufficiently arguable to require discovery. The case therefore raises a broader question: how long can a professional services network centralise control, capability and quality management while keeping liability and capital predominantly local? (Financial Times – Liquidators Can Pursue PwC Globally Over Evergrande Collapse; China Evergrande Group v PricewaterhouseCoopers – Hong Kong Court Judgment; DAC Beachcroft – Hong Kong Court Allows Evergrande Claim Against PwC International to Proceed)
The Audit Failure Was Already Institutional
PwC’s relationship with Evergrande stretched back to the company’s Hong Kong listing in 2009. By 2019 and 2020, this was not a new engagement where the auditor was still learning how the business operated. Yet regulators later found that contradictory evidence had already entered the audit process. In 2019, PwC’s working papers recorded 65 physical observations of property projects. Regulators found discrepancies in 57 of those records. In 2020, PwC initially selected 50 projects for observation, after which Evergrande negotiated the replacement of 37. Regulators subsequently found false revenue recognition in 32 of those replacement projects, representing RMB66.76 billion of false revenue. The economic reality was therefore not entirely hidden from the audit process. Important warning signs had already entered the system. (China Securities Regulatory Commission – Administrative Penalty Decision on PwC Zhong Tian)
The problems extended considerably beyond revenue recognition. China’s Ministry of Finance found serious deficiencies involving audit evidence, restricted cash, liabilities, consolidation and independence. Its penalty decisions describe PwC personnel assuming management responsibilities in connection with the preparation of consolidated financial statements and making unsupported consolidation adjustments that inflated Evergrande’s reported profits. Hong Kong’s Accounting and Financial Reporting Council reached similarly broad conclusions. The significance is not that several isolated procedures failed simultaneously. Several parts of the institutional system designed to prevent an incorrect audit conclusion failed together. (Ministry of Finance of the People’s Republic of China – Administrative Penalty Decision on PwC; Accounting and Financial Reporting Council – Evergrande Audit Sanctions)
The quality findings make that particularly important for the wider PwC network. Hengda represented more than 80 percent of Evergrande’s revenue, profit, assets and net assets during the relevant period, while PwC Hong Kong relied extensively on work performed by PwC Zhong Tian’s Guangzhou office. Following the regulatory findings, six partners and five staff were removed, the China Territory Senior Partner stepped down, PwC’s Global Risk & Regulatory Leader became interim Territory Senior Partner and network specialists supported a broader remediation programme. Once the failure became sufficiently serious, the response was therefore not purely local. PwC China and its Governance Board acted with support from the wider PwC network. (PwC – Network Statement Regarding the Evergrande Matter)
The Global Network Combines Shared Capability With Separate Liability
PwC is explicit about its legal architecture. The PwC network is not a global partnership, a single firm or a multinational corporation. Its member firms are separate legal entities. PwC International is an English private company limited by guarantee and does not practise accountancy or provide services to clients. Member firms cannot bind PwC International or one another, and each remains responsible for its own acts and omissions. There are strong reasons for this structure. Audit and other regulated professional activities remain subject to national licensing, ownership, independence and supervisory requirements. (PwC – How the Global Network Is Structured)
Legal independence, however, does not mean institutional isolation. PwC International facilitates coordination across the network. Its Network Leadership Team sets overall network strategy and the standards to which firms agree to adhere. Its Global Board governs PwC International and approves network standards. PwC’s governance reporting also describes network-level oversight of enterprise risks, quality and compliance. These mechanisms are part of what allows clients dealing with PwC in different countries to expect something more coherent than a collection of unrelated local accounting firms. (PwC – Governance, Quality and Risk 2025)
The quality architecture goes further. PwC firms assess compliance with network standards annually, while network-level bodies oversee risk, quality, monitoring and remediation. There is nothing inherently problematic about this. A global audit network needs strong quality management precisely because a serious failure inside one member firm can damage confidence in the wider brand. But the stronger these mechanisms become, the more important the distinction between coordination and control becomes. Legal separation can remain entirely real while operational independence gradually declines. (PwC – Governance, Quality and Risk 2025; PwC – Global Transparency Report 2024)
The Claim Targets What the Global Organisation Actually Does
That is the boundary Evergrande’s liquidators are now targeting. They allege that PwC International developed and monitored network standards, possessed powers connected with supervised remediation and leadership changes, participated in or caused reviews of the relevant audit functions and assumed responsibilities that went beyond general coordination. The liquidators seek approximately RMB57.9 billion from PwC Hong Kong and PwC Zhong Tian in relation to the FY2017 to FY2020 audits. They are separately pursuing PwC International for approximately RMB38.1 billion, around US$5.6 billion, in relation to claimed losses arising from FY2018 to FY2020. (Des Voeux Chambers – Evergrande’s RMB38 Billion Claim Against PwC International Survives Strike-Out; DAC Beachcroft – Hong Kong Court Allows Evergrande Claim Against PwC International to Proceed)
PwC International contests that interpretation. It was not a party to Evergrande’s audit engagement letters, was not qualified to perform audit work in Hong Kong and received no audit fee from Evergrande. The August decision did not reject those arguments on the merits. It only determined that the allegations against PwC International were sufficiently arguable to proceed into discovery. Evergrande has not established that PwC International owed a duty of care, breached such a duty or caused the losses being claimed. (China Evergrande Group v PricewaterhouseCoopers – Hong Kong Court Judgment; DAC Beachcroft – Hong Kong Court Allows Evergrande Claim Against PwC International to Proceed)
That distinction matters. The court has not created global Big Four liability. What it has done is allow the inquiry to move beyond the formal legal description of the network and into the operating reality underneath it. The relevant question is no longer only whether the member firms are separate legal entities. It is what PwC International actually did, what powers it possessed and whether those activities were sufficiently connected to the failures being alleged. That is a much more consequential test for a professional services network becoming increasingly integrated in practice.
The Boundary Has Been Tested Before, but the Institution Has Changed
Evergrande is not the first attempt to connect a local audit failure to a global accounting organisation. After Parmalat collapsed, plaintiffs pursued international entities including Deloitte Touche Tohmatsu and Grant Thornton International. Courts allowed some theories to proceed where the allegations involved meaningful control over member-firm behaviour. Later cases such as Otto Candies went the other way, rejecting attempts to impose vicarious liability where the necessary agency or joint-venture relationship had not been established. PwC also faced global-network allegations after the Satyam scandal, which ended in settlement without a final determination of the broader liability question. The precedents therefore cut in both directions. Formal separateness does not always end the inquiry, but common branding and shared standards are not enough by themselves. (In re Parmalat Securities Litigation – Global Network Liability; Otto Candies v KPMG – Delaware Court of Chancery Decision; Satyam Computer Services Securities Litigation – PwC Settlement)
What makes Evergrande more interesting is how much the operating infrastructure underneath the large networks has changed since those earlier cases. Cybersecurity has become a network-level concern. Data environments are more connected. Global delivery centres perform substantial amounts of professional work outside the member firm owning the client relationship. Audit technology is increasingly standardised. Artificial intelligence requires investment, data governance and common infrastructure that become progressively harder to reproduce independently across every national partnership. The legal architecture may still be organised around separate firms, but an increasing share of the operating architecture is not.
PwC itself provides a useful example. PricewaterhouseCoopers IT Services exists to provide shared technology services to PwC firms, including hosting, cloud infrastructure, information security and application support. It is a separate legal entity owned and sponsored by several PwC firms on behalf of the network. This does not turn PwC into a multinational corporation, but it shows how strategically important capabilities can sit outside the member firm contracting with the client. The historical bargain therefore becomes harder to maintain cleanly: professional ownership and liability remain local while parts of the system on which local delivery depends become progressively more shared. (PwC IT Services – Shared Technology Structure)
The Missing Global Balance Sheet
The financial architecture makes the tension visible. PwC’s member firms reported aggregate gross revenues of US$56.9 billion for the year ended 30 June 2025. Those are not the revenues of PwC International. They are the combined revenues of separate member firms across the network. PwC International, meanwhile, is registered at Companies House as a private company limited by guarantee without share capital and classified as a non-trading company. Its latest accounts were filed as micro-company accounts. (PwC – Financial Performance 2025; Companies House – PricewaterhouseCoopers International Limited; Companies House – PwC International Filing History)
That is not evidence of something improper. It is how the architecture works. Revenues, assets, profits and partners remain primarily inside the member firms, while PwC International coordinates the network without consolidating those economics onto its own balance sheet. Under normal conditions, the structure is highly efficient. Global coordination can expand without requiring a single global partnership or a single global capital pool. The tension only becomes visible when the entity coordinating the network itself faces a claim measured in billions.
A global balance sheet is also not the only way to absorb global risk. The Financial Times reports that PwC International’s 2025 financial statements state that insurance limits exceed its potential loss exposures and that former PwC executives expect a successful claim to engage L&F Indemnity, an in-house insurer funded by member firms. The details of those arrangements are not public, but they suggest that the network may already possess something economically analogous to a global risk-bearing layer. This connects directly to Case Study 35: EY, Wirecard and the Real Economics of Public-Interest Audit. Wirecard demonstrated the asymmetry between the annual economics of a public-interest audit and the liability environment that can emerge when the audit fails. Evergrande moves that question one institutional level higher: what happens when responsibility itself begins moving beyond the local audit firm? (Financial Times – What Does the Evergrande Case Mean for the Big Four?)
What If Liability Actually Moves Up?
The stress test is unusually concrete. PwC Hong Kong and PwC Zhong Tian face claims of approximately RMB57.9 billion, roughly US$8.5 billion, while PwC International is separately being pursued for approximately RMB38.1 billion, around US$5.6 billion. These are claims, not awards. But assume that the liquidators eventually succeeded on a material part of their case. A judgment against PwC International would still not automatically become a judgment against PwC US, PwC UK or every other member firm. The legal separation would remain relevant.
The economic consequences would then depend on the risk architecture behind PwC International. If insurance fully responds, the structure may prove considerably more resilient than the size of the claim initially suggests. PwC would effectively have mutualised catastrophic risk without making every member firm directly liable for every other firm’s professional work. Liability, capital and legal ownership do not necessarily have to sit inside one company if contractual arrangements connect them effectively.
But that outcome would also reveal that the economics of a major failure are not entirely local. If a global risk pool is ultimately funded by member firms, risk is already being shared economically even where legal liability remains separated. If available cover proved insufficient or disputed, the question would become sharper. Would member firms contribute additional capital, support a settlement or restructure the global entity? Or would they preserve the legal separation and leave the consequences with PwC International? The first route would demonstrate economic mutualisation. The second would test how replaceable the institution coordinating the brand, standards and network governance actually is.
The Institutional Choices
One possible response is to move back toward a truer federation. The global organisation could focus more narrowly on brand, methodology and coordination while reducing intervention powers and operational dependency. That could create cleaner liability boundaries, but it would move against many of the forces reshaping professional services. Regulators want stronger quality management. Cybersecurity rewards common infrastructure. AI rewards scale. Multinational clients expect consistency. A weaker centre might therefore create a cleaner legal structure while producing a less capable institution. A second route is to preserve legal independence while making economic risk mutualisation more explicit through shared insurance pools, captive structures, catastrophe reserves or regional risk arrangements. That would accept that legal liability can remain local while extreme economic risk is shared more broadly.
A third route is a form of Alternative Practice Structure at global scale. In a conventional APS, the regulated attest practice remains in a separately owned professional entity while tax, advisory and other non-attest activities sit inside another organisation capable of accepting external capital. Personnel, technology, premises and back-office services can then be supplied to the attest entity through contractual arrangements. I explored the underlying economics in The Regulated Trust Layer: How Private Equity Is Separating Audit From the Economic Platform Around It. Applied globally, the same logic could allow locally owned audit entities to sit around a separately capitalised platform containing technology, data, AI, global delivery and other shared capabilities. (AICPA & CIMA – Alternative Practice Structures)
That architecture has obvious attractions, but it does not make the liability question disappear. It changes where the boundary sits. I have not identified a public precedent in which a modern private-equity-backed APS has yet faced an Evergrande- or Wirecard-scale audit failure that conclusively tested how catastrophic liability travels across the attest firm, the non-attest platform, shared personnel, technology infrastructure and contractual service arrangements. If substantial parts of the system producing an audit sit outside the regulated entity, plaintiffs and regulators may eventually ask whether some responsibility follows those capabilities as well. An APS may create a cleaner ownership architecture. Whether it creates a cleaner liability architecture remains unproven. Full global integration would align ownership, control, capital and liability more clearly, but professional ownership requirements and national audit regulation make that the least immediately practical option. There is no structurally clean answer.
What Happens Next?
The immediate litigation will turn on facts that are not yet public. Discovery may show how PwC International monitored PwC Hong Kong and PwC Zhong Tian, which issues were escalated, what reviews were performed, how remediation mechanisms operated and what intervention rights existed in practice. It may also provide greater visibility into insurance and the contractual arrangements connecting PwC International to member firms. The difference between powers theoretically available and powers actually exercised may become particularly important. (Des Voeux Chambers – Evergrande’s RMB38 Billion Claim Against PwC International Survives Strike-Out; DAC Beachcroft – Hong Kong Court Allows Evergrande Claim Against PwC International to Proceed)
Several outcomes remain plausible. PwC International could establish that it owed Evergrande no duty of care. The case could settle without producing a decisive ruling on global network responsibility. Or the litigation could establish that some conduct at international level created responsibility independent of the member firms signing the audit opinions. None of these outcomes would automatically make every Big Four network globally liable for every local failure. The case remains highly fact-specific.
Discovery may nevertheless matter even if PwC ultimately wins. It could expose in unusual detail how one of the world’s largest professional services networks governs quality, allocates responsibilities, finances shared infrastructure and insures network risks. Future plaintiffs and regulators will study that evidence. Leaders of increasingly complex Alternative Practice Structures should as well. The legal architecture is different, but the institutional question is similar: once regulated work depends on capabilities sitting elsewhere, how confidently can responsibility still be contained inside the entity formally signing the opinion?
Closing Thoughts
Evergrande began as a financial reporting scandal. It became an audit failure. The regulatory investigations showed that the problems extended beyond individual audit procedures into independence, review, monitoring and quality management. The litigation against PwC International now moves the question another institutional layer higher. It is no longer only asking whether the audit failed. It is asking which parts of the wider institution had enough responsibility for the systems surrounding that audit to become legally relevant when the failure occurred.
For decades, the Big Four have combined separate national firms with common brands, methodologies, professional standards and increasingly shared infrastructure. That structure has been remarkably successful. Evergrande does not prove that it is broken, nor that global liability is inevitable. But it exposes a tension that becomes harder to avoid as the networks integrate operationally. Control can move without ownership. Liability can move without revenue. Risk can be mutualised without creating a global corporation. Alternative Practice Structures can separate regulated ownership from the economic platform surrounding it.
None of these mechanisms necessarily fails. But none makes the boundary problem disappear. As capability becomes more integrated across legal entities, the institution eventually has to decide where responsibility sits and what capital or risk-bearing mechanism stands behind it. Evergrande may therefore matter far beyond PwC. The existing Big Four network model is undergoing one of its most serious liability tests. At the same time, private-equity-backed firms are constructing a different answer through Alternative Practice Structures. Those structures have not yet faced a comparable catastrophic test. The profession may ultimately discover that neither federation nor APS eliminates the boundary problem. They simply draw the boundary in different places.
What This Means for Boards
Boards and global leadership teams should not interpret Evergrande only as an audit-quality problem or litigation risk. The broader governance question is whether operational control, professional accountability and risk-bearing capacity remain aligned as the institution becomes more integrated. The legal organisation chart is only one part of that assessment. Leadership needs to understand who can require remediation, control shared platforms, set mandatory standards, replace leadership and direct responses when something goes seriously wrong.
They should also understand the extreme downside architecture. Which insurance responds to a multibillion-dollar loss? Who ultimately funds it? What happens if limits are exhausted? Are member firms required to provide additional support? As Case Study 35: EY, Wirecard and the Real Economics of Public-Interest Audit demonstrated, the economics of public-interest audit cannot be understood purely through the annual engagement margin. The downside sits at institutional scale.
Boards operating through Alternative Practice Structures face a related question. Who employs the people, owns the technology and data, controls budgets and provides the infrastructure required by the attest firm? What happens when professional obligations conflict with the economic interests of the surrounding platform? Until one of these structures faces a major professional failure, nobody should assume that every liability boundary will work exactly as intended. Evergrande therefore leaves the profession with a larger question than who ultimately pays one claim:
When the institution increasingly operates across legal boundaries, where does responsibility ultimately stop?
Continue Exploring
The Bigger Picture
The Professional Services Transformation Theory
Why changing economics, ownership structures, operating models and governance are forcing professional services firms to rethink how the institution itself is designed.
Understand the Liability Economics
Case Study 35: EY, Wirecard and the Real Economics of Public-Interest Audit
Why the economics of public-interest audit can look attractive in normal years while creating institutional-scale downside when something goes badly wrong.
Explore the Alternative Structure
The Regulated Trust Layer: How Private Equity Is Separating Audit From the Economic Platform Around It
Why regulated audit entities are increasingly being separated from the capital, technology, people and infrastructure surrounding them, and what that may mean for future professional services models.
Follow Where Control Moves
The Platform Gravity Problem: Why Control Over Shared Systems Increasingly Shapes Power Inside Professional Services Networks
Why control over shared technology, data and infrastructure can redistribute institutional power even when formal legal ownership does not change.
Apply This to Your Own Institution
Future of Professional Services Board Review
A private outside-in case study of your firm, region or network, using the same research approach behind my public case studies to identify what stands out, which comparisons matter and the questions I would ask if I were sitting on your Board.
Sources
Company Publications
- China Evergrande Group – Change of Auditor, January 2023
- PwC – Network Statement Regarding the Evergrande Matter
- PwC – How the Global Network Is Structured
- PwC – Governance, Quality and Risk 2025
- PwC – Global Transparency Report 2024
- PwC – Financial Performance 2025
- PwC IT Services – Shared Technology Structure
Independent Reporting
- Financial Times – Liquidators Can Pursue PwC Globally Over Evergrande Collapse
- Financial Times – What Does the Evergrande Case Mean for the Big Four?
Industry, Regulation & Legal Analysis
- China Securities Regulatory Commission – Administrative Penalty Decision on PwC Zhong Tian
- Ministry of Finance of the People’s Republic of China – Administrative Penalty Decision on PwC
- Accounting and Financial Reporting Council – Evergrande Audit Sanctions
- Hong Kong Securities and Futures Commission – PwC Evergrande Shareholder Compensation Agreement
- China Evergrande Group v PricewaterhouseCoopers – Hong Kong Court Judgment
- Des Voeux Chambers – Evergrande’s RMB38 Billion Claim Against PwC International Survives Strike-Out
- DAC Beachcroft – Hong Kong Court Allows Evergrande Claim Against PwC International to Proceed
- Companies House – PricewaterhouseCoopers International Limited
- Companies House – PwC International Filing History
- In re Parmalat Securities Litigation – Global Network Liability
- Otto Candies v KPMG – Delaware Court of Chancery Decision
- Satyam Computer Services Securities Litigation – PwC Settlement
- AICPA & CIMA – Alternative Practice Structures