Capital Rules Made Banks Safer and Balance-Sheet Capacity More Expensive. Finance Moved Accordingly.

A bank that has discovered it has been using rather too little of its own money invites a perfectly sensible response: make it use more.

The response has a certain rectitude.

It is also the sort of rectitude that changes the price of the drinks, the membership list, and the crowd on the pavement outside.

Prudential regulation did precisely that.

It required banks to hold more credible loss-absorbing funding, and more usable liquidity, against the risks they had chosen to carry.

This made the club considerably harder to kill.

It also made its balance sheet less casual, which is what safety looks like once Finance has had a week to prepare the deck.

The second effect arrived with the reliability of a waiter appearing the moment the credit committee begins to discuss strategic flexibility.

A loan, security, derivative, commitment or liquidity facility that consumes more capital or liquidity becomes more expensive to carry inside a bank. So it is repriced, reduced, sold, securitised, or housed somewhere subject to a different regulatory costume.

The economic risk has not necessarily changed its manners. It has merely been introduced to a different door.

None of which is an indictment of the dress code. A room in which members must bring some money of their own is generally preferable to one in which they borrow the furniture from depositors and call it innovation.

It is an account of what the dress code did, and of why the pavement became so unexpectedly busy.

First, the Club Became Harder to Kill

Capital is not a vault of idle cash.

It never was.

It is funding that absorbs loss before fixed claims payable at par on demand.

Common Equity Tier 1 goes first.

Additional Tier 1 and Tier 2 qualify only where they carry specified loss-absorbency features.

Which is why the Basel minima are expressed against risk-weighted assets rather than against a picturesque pile of notes in a branch basement.

The headline requirement is 4.5% CET1 of risk-weighted assets, before buffers and local add-ons. That does not mean each loan sits beside its own small reserve of cash in a hard hat.

Risk-weighted assets, eligible regulatory capital and accounting equity are three different measures doing three different jobs.

Above the minimum sits the capital conservation buffer, also in CET1. Fall into it and dividends, buybacks, discretionary bonus payments and certain capital-instrument payments are restricted.

The buffer is there to be used in stress. It is not there to be used without consequence, a distinction capable of making a board meeting markedly less festive.

The leverage ratio supplies the backstop, at a Basel minimum of 3% of Tier 1 capital against an exposure measure that takes in on-balance-sheet, derivative, securities-financing and off-balance-sheet exposures.

Risk weights are useful because a portfolio is not a single risk. The leverage ratio is useful because a flattering risk weight is not a hall pass for unlimited balance-sheet enlargement.

One is risk-sensitive.

The other is a chaperone.

Liquidity received its own inspection.

The Liquidity Coverage Ratio requires enough high-quality liquid assets to meet modelled net cash outflows through 30 calendar days of significant stress, under a scenario that assumes retail deposit run-off, partial loss of unsecured wholesale funding and a multi-notch downgrade.

The Net Stable Funding Ratio asks a longer structural question: whether available stable funding matches what the balance sheet actually requires.

Neither ratio confers immortality on your balance sheet.

A bank can be solvent and still fail a morning’s payment test. It can be perfectly liquid for the day and still lack the capital to absorb the losses.

Nor is every asset with a price an asset available in stress. A government bond that has been pledged supplies rather less usable liquidity than its market value suggests.

Legal encumbrance is an inconvenient quality in collateral, much like an umbrella that belongs to somebody else.

Those requirements made the regulated bank less fragile.

They also turned capital, stable funding, liquid assets, collateral mobility and reporting capacity into scarcer inputs.

The price of carrying risk had changed before any supervisor issued a fine. Which is how prudential regulation generally gets its work done, while everyone is watching enforcement.

The Six Ways a Balance Sheet Answers Back

A bank meeting a more expensive requirement has six responses: raise equity, retain earnings, reprice assets, reduce exposures, securitise or sell the risk, and migrate the activity to a differently regulated vehicle.

Six.

Not a selection of fashionable talking points, and not a menu from which you may abstain. A seventh response, in which the arithmetic is treated as optional, has historically produced a great deal of documentation and no durable business model.

Raise equity means asking owners for more loss-absorbing funding. Retain earnings means keeping profits that might otherwise have left as dividends or buybacks, which the conservation buffer makes very concrete indeed once CET1 slips into its range.

Both make the bank sturdier.

Both alter returns to shareholders, who tend to notice when the institution’s old leverage is quietly converted into governance.

Repricing is the quieter response.

A bank does not need a speech about social priorities in order to charge more for an exposure that consumes more capital, funding, liquidity or operational capacity.

It need only adjust the spread, the fee, the maturity, the collateral terms, the commitment size, or its willingness to quote at all.

A mortgage approval is constrained by expected default losses, regulatory capital, liquidity needs, funding cost, risk appetite and expected profitability. It is not mechanically funded by whichever depositor happened to walk into the branch that morning.

Reducing exposures is plainer still.

Decline the application.

Cut the line.

Limit the concentration.

Shorten the commitment.

Run the asset down.

The effect is not that credit has developed a moral objection to your business. The bank has decided that the return after capital, liquidity, funding and expected loss no longer justifies the balance-sheet use.

This is the moment at which ample market demand is introduced to approved risk appetite.

Securitising or selling changes the direct holder of an exposure. Migrating the activity changes the legal perimeter carrying it.

Neither is improper in itself.

Regulatory arbitrage is something more specific: shifting an activity or a legal structure to obtain weaker treatment without materially changing the underlying risk or service. A change of label deserves rather less admiration than a change of economics.

The functional perimeter is what decides which is which. Taking repayable funds, arranging investments, advising, dealing, operating a payment service, issuing insurance, managing client assets or running trading infrastructure can each trigger authorisation.

A group affiliate, software vendor or distributor may sit outside one entity’s prudential boundary while still shaping the control, the service or the risk transfer inside it.

A logo is a poor risk map.

It has never been much good at contracts either.

Liquidity Made the Pavement Busier Too

Capital pricing is only half the story.

A bank can originate a loan and create the matching deposit at drawdown, expanding both sides of its balance sheet at once.

The moment the customer pays a seller banking elsewhere, however, the originating bank has to settle through reserves or other settlement funds.

Treasury then acquires an entirely practical interest in deposits, wholesale borrowing, interbank funding, repo, liquid securities, correspondent arrangements and eligible central-bank facilities.

The loan on your books is still the asset. The cost of replacing the funding and the settlement liquidity decides whether it is still an attractive one.

So the liquidity ratios do more than improve a line in a prudential return. Treasury runs daily and intraday forecasts of inflows, payments due, collateral calls, reserve needs, committed but undrawn facilities, and assets that can genuinely be sold or pledged.

That last phrase carries more weight than it looks. A pledge converts a handsome market value into a rather less useful payment plan.

Liquidity rules therefore reprice commitments and off-balance-sheet capacity as well as drawn loans. A legally binding undrawn facility is not yet a loan asset, but it is a plausible future payment, a funding demand and a capital use.

A pipeline of approved but undrawn mortgages becomes a treasury event the moment borrowers draw quickly into more expensive funding.

Your mortgage department calls this pipeline.

Treasury calls it Tuesday.

The response can be entirely sensible and still be distributional. More liquid assets reduce the prospect of a forced sale at a punitive discount. They also earn less than an illiquid loan or a private asset.

The forgone return reduces the firm’s margin first, then reappears in borrower pricing, investor returns, product design, or a decision not to offer the service at all.

The benefit is visible when the stress arrives.

The bill is visible every quarter.

Which is why these debates so often resemble an argument between a fire-insurance policy and a kitchen renovation.

The Cost of Looking Respectable

The balance-sheet price is not the only price.

Prudential and conduct rules require legal interpretation, licensing, compliance officers, risk specialists, financial-crime staff, operations, internal audit, training, reporting, incident response, and systems for identity checks, transaction monitoring, communications capture, suitability fields, reconciliation, complaints, surveillance and retention.

Risk, compliance and internal audit are three separate control functions, each needing enough independence to challenge the rest of the firm without being invited to approve its own homework.

All of it is fixed cost, incurred before a single compliant transaction completes.

One additional customer is cheap once the machinery exists. A new product, market, algorithm or distribution channel is not: it wants a new policy, a data field, a validation, a control test and a report.

That does not make the work useless.

It means the work has scale economics.

A large incumbent spreads it across a very large balance sheet. A smaller entrant meets the same cathedral of evidence with rather fewer parishioners.

The result is a slightly awkward bargain.

Compliance makes a large bank less nimble and less leveraged while making its moat deeper.

The institution dislikes the expense.

It is markedly less troubled by the discovery that a would-be competitor must also hire the lawyers, build the systems, establish the governance, file the returns, retain the records and submit to examination before it may compete at all.

An entry barrier is never introduced to you as an entry barrier. It arrives in a sensible font, under the heading Control Framework.

Who Got the Better Table

Depositors and senior creditors did, from a bank carrying more loss-absorbing resources ahead of their claims and a larger stock of liquid assets against modelled stress.

The benefit is not mystical.

It is a more credible promise to pay, in a business built entirely on promises to pay.

Taxpayers did, in the narrower but consequential sense that capital and resolution planning are designed to put losses on shareholders and creditors before public solvency support.

The Financial Stability Board’s Key Attributes aim at orderly resolution that protects critical functions while allocating loss privately.

That promises no world without public consequences.

It does make the private loss-bearing order less accidental, which is a considerable improvement on discovering it over a weekend.

Large incumbents did too, in a more complicated fashion. The ratios, the calculations, the reporting systems, the independent control functions and the supervisory relationship all cost real money.

They also make it very hard to enter a regulated activity without substantial capital, systems, governance and evidence.

You pay the dues and acquire a fence.

Both things are true before lunch.

Creditors benefit from information as much as from capital. The Basel Framework gathers common standards for capital, leverage, liquidity, large exposures and disclosure. Comparable reporting costs the firm money and occasionally invites an unflattering interpretation.

It also means a creditor need not rely on rumour and private access when deciding whether to lend. For a business whose raw material is confidence, that is not a trivial service.

Who Paid the Dues

Shareholders paid first, through lower leverage, more retained earnings, and the higher return demanded on equity and long-term loss-absorbing instruments than on insured deposits and short-term borrowing.

Replacing cheap short-term funding with equity raises the annual financing charge before anyone counts the lower probability of failure or the tighter funding spreads that follow. That is arithmetic, not an invoice issued by the regulator.

Marginal borrowers paid wherever a loan, facility or relationship was repriced, reduced or declined because it consumed scarcer capital, liquidity or operational capacity.

The burden is not evenly spread.

A risk weight, a funding need, a collateral term and an undrawn commitment can change the relative price of exposures that look broadly identical from the borrower’s side of the desk.

The bank calls this allocation.

The borrower calls it an increased margin.

Smaller firms paid through the fixed burden of compliance, and through the pleasure of competing against institutions already equipped with governance, data, reporting, audit and funding infrastructure. Some of that activity then discovers a legal structure where the relative economics are friendlier.

Which is not proof that regulation failed.

It is proof that the price changed, and that finance, a creature of considerable adaptability, noticed.

So the verdict is neither triumphal nor cynical.

Capital and liquidity rules make regulated banks harder to kill. They do not make loans free, liquidity costless, risk models infallible, or equivalent economic exposures incapable of moving house.

They turn risk-bearing capacity into an explicit and expensive product, then oblige everybody else to decide whether to pay for it, reshape it, or take it somewhere the dress code is not enforced.

When regulation changes a balance-sheet price, look for the same economic risk wearing a different legal costume.