Cost of Capital | Why Bank Equity Is Expensive but Necessary

Capital is not a pile of cash

This distinction is foundational.

A bank can have strong capital and still face a short-term liquidity problem. It can also hold large cash balances while being economically insolvent if its assets are worth far less than its liabilities.

Capital is a balance-sheet funding and loss-absorption layer. Cash and central-bank reserves are liquid assets.

Read Bank Equity and Retained Earnings and What a Bank Liquidity Buffer Is Actually For.

Equity is last in line because it owns the residual

Shareholders own what remains after the bank’s liabilities are satisfied.

If assets rise in value and the bank earns well, shareholders benefit. If assets lose value, shareholder equity absorbs the first loss before many creditors are affected.

assets − liabilities = equity.

That residual position makes equity both risky and useful.

Why shareholders demand a higher return

A depositor with an insured account expects contractual repayment and may have scheme protection within defined limits. A senior bondholder has a contractual claim. An equity investor has no fixed repayment date and no guaranteed dividend.

The shareholder therefore demands compensation for:

• first-loss exposure;
• uncertain dividends;
• market-price volatility;
• business-model risk;
• regulatory change;
• macroeconomic and credit cycles;
• possible dilution from future capital raising;
• the possibility of losing most or all invested capital.

Cost of equity is not printed on a bank statement

Unlike a deposit interest rate or bond coupon, the cost of equity is not directly observable.

It is the return investors require to hold the bank’s equity at its current perceived risk. Banks and analysts estimate it using market prices, risk models, peer comparisons and other methods.

The Bank of England’s 2025–2026 capital work explicitly notes that cost of equity cannot be observed directly and that estimates are uncertain.

The bank’s overall funding cost is a mix

A bank finances assets through several sources:

• customer deposits;
• wholesale debt;
• secured funding;
• subordinated instruments;
• common equity;
• retained earnings.

Each source has a different price, maturity, loss position and regulatory treatment.

Deposits are often cheaper because they provide money-like utility

Customers hold deposits not only for return but for payments, liquidity, convenience and safety.

That makes deposits a distinctive source of bank funding. They can be cheaper than market equity even though the bank must invest heavily in branches, apps, payment infrastructure, cyber defence and deposit servicing.

More equity can raise average funding cost and reduce bank risk

Replacing some cheap debt or deposits with equity can increase the direct funding cost because shareholders require a higher expected return.

At the same time, a better-capitalised bank is less risky to creditors and can face lower debt funding spreads.

This offset is one reason the economic cost of more capital is usually smaller than simply comparing a deposit rate with a required equity return.

Capital pricing is a trade-off, not a morality play

Too little capital can make banking fragile and increase the probability and cost of crises.

Too much required equity, if badly calibrated, can raise the price of bank credit or push some activity outside banks.

The Bank of England’s July 2026 review of the bank capital framework describes this exact policy balance: the macroeconomic cost of higher capital comes partly through borrowing costs, while the benefit is a lower likelihood and cost of future banking crises.

Current evidence says the cost is real but not one-for-one

Bank of England research published in May 2026 revisited more than thirty years of UK evidence. It found that higher capital ratios are associated with some increase in corporate lending spreads, while the effect on mortgage spreads was not statistically significant over the long run in that study.

The important mechanism is not the exact UK estimate. It is that capital has an economic cost, but the pass-through is partial and shaped by competition, funding structure and borrower type.

Regulatory capital and economic capital are related but different

Regulatory capital is the amount and quality of capital required under prudential rules.

Economic capital is an internal estimate of how much loss-bearing capacity the bank believes a business or risk requires under its own models and risk appetite.

A business line can meet the regulatory minimum and still be unattractive internally if the bank believes its true risk is higher.

The Basel minimum is a floor, not a target

The Basel Framework currently sets minimum risk-based ratios including 4.5 per cent Common Equity Tier 1, 6 per cent Tier 1 and 8 per cent total capital relative to risk-weighted assets, before buffers and other applicable requirements.

Banks commonly operate above minimums because supervisors, markets, rating agencies and their own risk appetite expect headroom.

Capital headroom is strategic flexibility

A bank with capital only just above minimum requirements has little room for loss or growth.

A bank with a prudent buffer can absorb an unexpected credit shock, continue supporting customers and avoid emergency capital raising under stress.

Retained earnings are the quietest way to build capital

Profits not paid out as dividends or buybacks remain inside the bank and increase common equity, subject to accounting and regulatory adjustments.

This is why recurring profitability matters to prudential resilience.

Article 81 and Article 82 show how both interest and non-interest revenue can contribute to retained earnings.

Issuing new equity has explicit and implicit costs

A bank can sell new shares to raise capital. Existing shareholders may be diluted. The market may interpret the issue as evidence the bank is weaker than previously believed. Transaction costs arise.

Raising capital in calm markets is therefore very different from raising capital during a crisis.

The cost of capital enters loan pricing

A loan consumes some combination of funding, liquidity and capital.

The bank therefore needs a lending rate that covers:

• funding cost;
• expected credit loss;
• operating cost;
• liquidity cost;
• capital cost;
• tax and other expenses;
• an acceptable residual return.

Read Why Loan Rates Are Usually Higher Than Deposit Rates.

A risky loan should consume more economic attention even when its face value is identical

Two S$1 million loans can require different risk-weighted assets, provisions and internal capital because their probability of default, collateral, maturity and borrower quality differ.

The nominal loan amount does not tell the full capital story.

Read Risk-Weighted Assets.

Return on equity can reward leverage

If a bank earns the same profit with less equity, reported return on equity rises.

That can create a dangerous incentive to minimise capital too aggressively.

A high ROE can therefore mean superior franchise economics—or simply more leverage.

A simple ROE example

Bank A earns S$100 million on S$1 billion of equity: 10 per cent ROE.

Bank B earns the same S$100 million on S$500 million of equity: 20 per cent ROE.

Bank B looks more profitable on ROE. It also has half the equity cushion, all else equal.

Risk-adjusted return asks whether the profit paid for the capital used

Banks use measures such as risk-adjusted return on capital, economic profit or hurdle-rate frameworks to compare business lines.

The underlying question is:

did this activity earn enough after expected loss and capital cost to justify the risk we took?

A hurdle rate is not the regulatory minimum capital ratio

A hurdle rate is the return a bank expects a business to exceed after considering capital and risk.

A regulatory capital ratio is a prudential constraint.

One measures required return; the other measures loss-bearing capacity.

Cheap-looking business can be expensive in capital

A loan with a low interest margin can still be attractive if it is low risk and uses little capital. A high-margin loan can be unattractive if expected loss and capital consumption are very high.

The spread alone cannot rank the economics.

Fee businesses can also consume capital

Guarantees, commitments, derivatives and trading services can require capital even when they are reported largely through non-interest income.

Article 82’s fee line therefore connects directly back to capital economics.

Capital protects creditors because it takes the first hit

Suppose a bank has S$100 billion of assets, S$92 billion of liabilities and S$8 billion of equity.

If asset values fall by S$2 billion, equity falls to S$6 billion before the nominal liabilities need to change.

If losses exceed the equity cushion and other loss-bearing layers, creditor claims can become threatened.

Capital makes ordinary failure less contagious

Banking requires taking credit and market risk. Some loans will fail.

A well-capitalised bank can absorb ordinary losses without immediately imposing them on depositors or disrupting payment services.

Capital also supports confidence

Depositors and wholesale funders care not only about current cash but whether the bank can survive future losses.

A strong capital position can therefore lower the probability of confidence loss and expensive emergency funding.

More capital can make debt cheaper

If creditors believe a larger equity cushion protects them, they can demand a lower risk premium.

This is why the economic cost of equity funding is partly offset by safer debt funding.

But equity holders still expect competitive returns

A bank that consistently earns below its cost of equity can struggle to attract capital and can trade below book value for long periods.

That can make future capital raising harder or more dilutive.

The bank therefore manages two survival tests

• Prudential survival: enough capital to absorb loss and meet regulatory requirements.
• Economic survival: enough risk-adjusted return to keep capital providers willing to fund the bank.

A bank can be safe but commercially weak. It can also be profitable in the short run while undercapitalised. Healthy banking needs both.

Capital planning looks forward, not only backward

The bank asks how capital behaves under:

• loan growth;
• recession;
• property-price declines;
• market losses;
• operational losses;
• dividends and buybacks;
• acquisitions;
• regulatory change;
• stress-test scenarios.

A bank that meets today’s ratio but cannot survive tomorrow’s plausible stress is not well capitalised in the practical sense.

Loan growth consumes capital before the full risk is visible

New loans expand assets and risk-weighted assets. If capital does not grow at a compatible pace, ratios can fall.

Rapid growth can therefore create a race between asset expansion and loss-bearing capacity.

Article 84 owns that timing problem: Why Rapid Loan Growth Can Look Better Before the Risk Arrives.

Dividends and buybacks distribute capital out of the bank

When profits are paid to shareholders, they no longer remain inside the bank as retained earnings.

Distributions can be appropriate when capital is comfortably above needs. They can weaken resilience if paid too aggressively before losses emerge.

Stress buffers are designed to be used

Some prudential buffers are intended to absorb stress and allow banks to continue supporting the economy.

A framework that makes banks afraid to use any buffer can produce procyclical behaviour—cutting lending exactly when the economy is weakest.

Capital is therefore both protection and productive capacity

A strong capital position lets a bank take new risk after old risk has produced losses.

Capital is not only a brake on lending. It is also what allows lending to continue through uncertainty.

A worked pricing example

Two loans each earn S$50,000 of annual interest margin after funding cost.

Loan A requires S$100,000 of internal capital and has S$5,000 of expected loss. Loan B requires S$300,000 of internal capital and has S$20,000 of expected loss.

The nominal margin is identical. Their risk-adjusted returns are not.

A worked growth example

A bank grows loans by 20 per cent while retaining only enough profit to grow common equity by 5 per cent.

Unless the new assets carry materially lower risk weights or another capital action occurs, capital ratios can come under pressure.

A worked loss example

A bank begins with S$10 billion of common equity. A severe credit cycle produces S$2 billion of net losses after provisions and recoveries.

Equity falls toward S$8 billion before any new capital or earnings. The bank may remain solvent, but its ability to support the same asset base has weakened.

The durability test: strip away the regulatory acronym

Capital standards will evolve. The enduring questions remain:

• Who takes the first loss?
• How much loss can the bank absorb before fixed claims are threatened?
• What return do equity investors require?
• How much capital does each activity consume?
• Does pricing cover that capital cost?
• How much headroom remains after plausible stress?
• Can the bank keep lending after losses?
• Are distributions weakening future resilience?

The World Return: expensive capital buys continuity through bad futures

Equity investors are paid for bearing uncertainty that depositors and many creditors should not be forced to bear first.

profit → retained earnings or new equity → loss-bearing capacity → continued lending and payments during stress → future profit and resilience.

The bank fails the return test when it treats capital only as an expensive regulatory inconvenience. The purpose of capital is to keep a risky institution capable of serving the economy after risk becomes loss.

Ten misconceptions to remove

Misconception	Better model
“Bank capital is cash in a vault.”	Capital is a loss-absorbing funding layer; liquidity is a separate question.
“Equity is expensive only because shareholders are greedy.”	Equity bears residual loss and therefore requires a higher expected return.
“More equity makes banks safer at zero cost.”	Higher capital can raise lending costs, though safer debt funding partly offsets this.
“The Basel minimum is the right target for every bank.”	Banks operate with buffers and jurisdiction-specific requirements above minimums.
“High ROE always means a better bank.”	ROE can rise because leverage is higher.
“Regulatory capital and economic capital are identical.”	One follows prudential rules; the other reflects internal risk assessment.
“Fee businesses do not consume capital.”	Guarantees, commitments and trading can require substantial capital.
“Profits automatically strengthen capital.”	Only retained earnings remain after dividends, buybacks and other distributions.
“Capital only restricts lending.”	Strong capital also permits continued lending after losses.
“A well-capitalised bank cannot fail.”	Liquidity, operational, governance and asset-quality failures can still overwhelm an institution.

Observable mastery

1. Why does equity normally require a higher return than deposits?
2. Why is capital different from liquidity?
3. How does equity protect creditors?
4. Why can more equity make debt funding cheaper?
5. What is the difference between regulatory and economic capital?
6. Why can high ROE reflect leverage rather than franchise quality?
7. How does capital cost enter loan pricing?
8. Why do retained earnings matter?
9. How can rapid loan growth pressure capital ratios?
10. What useful capability should strong capital preserve during stress?

If those answers connect, bank capital becomes visible as the price of keeping risk honest: shareholders demand more because they stand first in the loss path, and banking needs them precisely because someone must stand there before the bank can promise everyone else that money will still be there tomorrow.

---