What it means
The logic is one of focus and confidence. A bank weighed down by non-performing loans spends its senior management time on collections rather than growth, and investors discount the entire institution because they cannot tell how deep the problem goes.
Splitting the balance sheet answers that. The good bank is left with a clean loan book that can be valued and funded normally, while the bad bank holds a defined pool of assets with its own funding, its own management and a mandate to maximise recovery over several years rather than to grow.
The structure comes in several forms. It can be an internal ring-fenced unit, a legally separate subsidiary, a full spin-off sold to investors, or a state-sponsored asset management company that buys distressed loans from many banks at once.
The hardest question is always the transfer price. Too high and the bad bank is subsidised, which becomes a public cost if the state is involved; too low and the parent takes a loss it cannot absorb, which is why bad bank creation is so often paired with a capital raise.
The results are not always grim. Because assets are transferred at heavily discounted prices, a patient workout entity that collects more than it paid can end up profitable, and several national asset management companies have eventually returned money to their governments.
In practice
Real-world examples.
Example
A regional lender with heavy exposure to a collapsed property developer moves $1,200,000,000 of construction loans into a ring-fenced workout subsidiary. Freed of those files, its commercial team resumes lending to small businesses within two quarters.
Example
A government facing widespread bank failures creates a national asset management company that buys distressed real estate loans from six institutions at an average of 38 cents on the dollar. It is given a fifteen-year mandate to sell the underlying property in an orderly way.
Example
A large universal bank creates an internal non-core unit holding shipping loans and legacy derivatives. It reports the unit separately each quarter so analysts can see the run-off progress without it clouding the core numbers.
Formula
Calculation
Loss on transfer = net book value of the assets - transfer price
Capital ratio = common equity capital / risk-weighted assets
A commercial bank holds $800,000,000 of non-performing loans against which it has already set aside provisions of $200,000,000, so the net book value on its balance sheet is $800,000,000 - $200,000,000 = $600,000,000. It agrees to transfer the portfolio to a new workout entity at 45 cents on the gross dollar, which is $800,000,000 x 45% = $360,000,000.
The additional loss on transfer is $600,000,000 - $360,000,000 = $240,000,000. The bank's common equity capital falls from $500,000,000 to $500,000,000 - $240,000,000 = $260,000,000, so it raises $300,000,000 of new shares, bringing capital to $260,000,000 + $300,000,000 = $560,000,000.
Removing the portfolio also cuts risk-weighted assets from $5,000,000,000 to $4,200,000,000. The capital ratio therefore improves from $500,000,000 / $5,000,000,000 = 10.0% to $560,000,000 / $4,200,000,000 = 13.3%, which is the whole point of the exercise.Case study
Seen in the real world.
Kestrel Regional Bank is a fictional institution created to illustrate how a bad bank works. Years of aggressive lending to holiday-home developers left it with a portfolio of loans on which borrowers had stopped paying, and its share price had fallen to a third of book value because nobody trusted the stated provisions.
The board approved a split. A subsidiary called Kestrel Asset Recovery took an $800,000,000 gross portfolio at $360,000,000, crystallising a further $240,000,000 loss, and Kestrel raised $300,000,000 from existing shareholders in the same week so the capital ratio ended higher than it started at 13.3%.
Over the next four years the recovery unit collected $455,000,000 against the $360,000,000 it had paid, a surplus of $95,000,000 that flowed back to the group. Meanwhile the clean bank grew its small business book at double digits because it could finally fund itself at normal rates. The illustrative lesson is that the loss was always there; the split simply made it visible and finite.
Watch out
Common mistakes.
- Believing a bad bank makes losses disappear. The losses are recognised immediately on transfer, and the structure only changes who manages the assets and how clearly the damage is disclosed.
- Assuming the assets are worthless. They are impaired rather than valueless, and a patient workout entity often recovers more than the discounted price it paid.
- Setting up the split without raising capital at the same time. The loss on transfer consumes equity, so a bank that does not pair the two steps can leave itself thinly capitalised.
Questions
People also ask.
Is a bad bank always a separate legal entity?
No, it can be an internal non-core unit reported separately, though a separate entity gives cleaner funding and governance.
Who funds the bad bank?
Depending on the design it can be funded by the parent, by outside investors buying into a spin-off, or by government-guaranteed bonds in a state-sponsored scheme.
Does the taxpayer always lose money?
Not necessarily, because several state-sponsored asset management companies bought assets at deep discounts and eventually repaid their funding in full.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%