# Legal entity rationalization: simplifying the corporate structure
When HSBC's Group CFO Georges Elhedery presented the bank's Q3 2024 results, buried in the slide deck was a number that didn't make headlines but should have: the bank had eliminated over 1,000 legal entities across its restructuring program, contributing meaningfully to the $1.5 billion in cost savings targeted by 2026. Each dormant subsidiary that disappeared wasn't just a line on an org chart, it was $75,000 to $200,000 in annual carrying costs, a quarterly Pillar Two filing, a local audit, a board meeting that had to be documented, and a director who needed D&O insurance.
The dirty secret of large multinationals is this: most CFOs don't know how many legal entities they actually own. When Deloitte surveyed 200 multinational CFOs in 2023, 42% admitted they couldn't produce an accurate, current count of their group's legal entities within 48 hours. GE famously discovered it had over 1,000 "zombie" entities during its breakup, subsidiaries acquired decades earlier through deals nobody could remember, sitting on the books, generating compliance costs and zero economic value.
This lesson is about one of the highest-ROIROIReturn on Investment: the ratio of net profit to the cost of an investment. A 300% ROI means each dollar invested returns $3.View full definition → projects in corporate finance, and one of the most avoided.
Every legal entity in your group structure carries a recurring annual cost that most finance teams systematically underestimate. The true loaded cost of maintaining a single legal entity in a developed market typically runs $50,000 to $150,000 per year, and in regulated industries or complex jurisdictions, it can exceed $500,000.
Let's decompose that number, because it's where most rationalization business cases get built or destroyed:
Now multiply by 500, 1,000, or 3,700.
The RBS legal entity rationalization program, executed between 2014 and 2019 under CFO Ewen Stevenson and his successor Katie Murray, remains the most studied example in European banking. RBS started with approximately 2,500 legal entities following its disastrous acquisition spree. The program reduced this to under 900 by 2019.
The reported direct savings exceeded £200 million per year, but the more interesting number is what RBS didn't report publicly: senior executives later acknowledged that the program freed up enormous amounts of senior finance, tax, and legal capacity to focus on strategic priorities. The Chief of Staff to the CFO at the time described it as "getting 30% of our brain back."
If the ROIROIReturn on Investment: the ratio of net profit to the cost of an investment. A 300% ROI means each dollar invested returns $3.View full definition → is so obvious, why doesn't every CFO do this? Three reasons:
1. No P&L owner. Legal entity costs are scattered across tax, legal, finance, treasury, and local country teams. No single executive sees the total bill.
2. Fear of stranded liabilities. Every entity might contain a hidden tax exposure, an unresolved litigation, an environmental obligation, or a pension promise. Dissolving the wrong entity can crystallize a $50M liability that was effectively dormant.
3. The work is unglamorous. Rationalizing 500 entities is 500 separate legal projects, often in 30+ jurisdictions. No CFO gets promoted for it; many get blamed if something goes wrong.
The European bank referenced in our hook, widely understood in the industry to be ING Group following its 2015 program, used a four-stage methodology that has since become the de facto standard. Pfizer applied a similar framework after its Wyeth acquisition (reducing 400+ entities), as did Diageo following its Mey Içki and United Spirits deals.
You cannot rationalize what you cannot count. The first deliverable is a single source of truth: a master entity register containing, for every legal entity:
This list takes longer than CFOs expect, typically 2-3 months even for well-organized groups. At Siemens, when then-CFO Ralf Thomas initiated entity rationalization in 2018, the census alone identified 87 entities that no internal system had recorded as still being active.
Every entity gets sorted into one of four buckets:
| Bucket | Criteria | Action |
|--------|----------|--------|
| Keep | Active operations, regulatory license, material tax attribute | Retain, optimize |
| Merge | Same jurisdiction, same business, no regulatory barrier | Statutory merger |
| Dissolve | Dormant, no liabilities, no attributes | Liquidate |
| Sell/Distribute | Non-core, marketable | Divest |
The "Keep" bucket should be challenged aggressively. A useful test: "If we were building this group from scratch today, would we create this entity?" If the answer is no, it goes into Merge or Dissolve regardless of how long it's been around.
This is where rationalization either creates or destroys value. Before eliminating any entity, you must answer:
Pillar Two has actually *increased* the value of rationalization for many groups. Maintaining sub-scale entities in low-tax jurisdictions now creates top-up tax exposure with limited offsetting benefit. Several Fortune 500 tax directors have privately confirmed that Pillar Two accelerated their rationalization timelines by 18-24 months.
Sequencing matters enormously. The general rule: clean before you cut. Settle intercompany balances, transfer or terminate contracts, novate guarantees, and distribute or liquidate assets *before* filing for dissolution. A premature dissolution filing can crystallize liabilities you intended to extinguish.
A typical large program processes 15-30 entities per quarter once the engine is running. The full program for a 1,000-entity reduction typically takes 36-48 months.
Knowledge check
1. According to HSBC's Q3 2024 disclosure cited in the lesson, approximately how many legal entities did the bank eliminate as part of its restructuring program targeting $1.5 billion in cost savings by 2026?
2. In the 2023 Deloitte survey of 200 multinational CFOs, what percentage admitted they could not produce an accurate, current count of their group's legal entities within 48 hours?
3. GE's breakup revealed it owned over 1,000 'zombie' entities. In corporate finance terminology, a 'zombie entity' most precisely refers to:
4. Select ALL correct answers regarding the recurring annual carrying costs of maintaining a single legal entity, as discussed in the lesson.
Sélectionnez toutes les réponses correctes.
5. Select ALL correct answers about why legal entity rationalization is described as 'one of the highest-ROI projects in corporate finance, and one of the most avoided.'
Sélectionnez toutes les réponses correctes.
Most failed rationalization programs fail in tax, not in legal execution. Three specific traps deserve attention.
In 2019, a US technology company (publicly disclosed only as a $40B+ market cap firm in its 10-KKThe average number of new users each existing user generates through referrals. Above 1.0, growth compounds on itself and becomes exponential.View full definition → footnote) dissolved a Luxembourg holding company carrying approximately $280 million in tax-loss carryforwards. Under the planning memo, those losses were considered "stranded" because there was no foreseeable income against which to offset them.
Three years later, the group restructured its European IP holdings, and would have generated exactly the kind of income those losses could have absorbed. The opportunity cost, calculated at Luxembourg's 24.94% combined rate, exceeded $70 million.
The lesson: NOLs are options, and options have value even when out-of-the-money. Before extinguishing tax attributes, run sensitivity analysis on plausible future restructuring scenarios.
Holding company structures often exist because of specific treaty benefits, a Dutch BV that reduces withholding tax on dividends from Brazil, a Singapore entity that provides treaty access to Indian royalties. Eliminating these entities without restructuring the underlying flows can permanently increase the group's withholding tax cost by 5-15% on affected cash flows.
Under the EU's Corporate Sustainability Reporting Directive (now in force for large undertakings reporting on FY2024 onward), every in-scope legal entity may have its own sustainability reporting obligations. Rationalization that consolidates entities into a single reporting entity can actually *reduce* CSRD burden. But mid-program, you can also accidentally pull entities into scope by merging a sub-threshold entity into a larger one. Model the reporting perimeter before, during, and after.
The CFO needs a credible business case to authorize spending $15-40 million on a multi-year program. Here is the structure that wins board approval:
Annual run-rate savings calculation:
For a group eliminating 500 entities at $80,000 average loaded cost: $40M gross savings, ~$32M net run-rate.
One-time costs:
Payback: typically 2-3 years on direct costs; ROIROIReturn on Investment: the ratio of net profit to the cost of an investment. A 300% ROI means each dollar invested returns $3.View full definition → over a 7-year horizon often exceeds 300%.
Beyond the dollar savings, sophisticated CFOs sell three additional benefits to their boards:
1. Faster M&A integration capacity, a simpler structure absorbs acquisitions more efficiently
2. Reduced Pillar Two compliance burden, fewer jurisdictional calculations
3. Improved enterprise risk profile, fewer dormant entities means fewer hidden liabilities
For the CFO ready to act on Monday morning, here is the specific sequence:
1. Commission the entity census this quarter. Demand a single-page report listing every legal entity, its purpose, and its annual cost. If your tax, legal, and finance teams cannot produce this in 60 days, you have already identified your first problem. The total annual cost will shock you, and it is the number you