What Is Basel III? Capital Requirements and Impact on Banks

By Markets & Investing DeskUpdated Nov 10, 2026

Basel III
Illustration · Fintrev

Definition

Basel III is an international set of banking standards that requires banks to hold more and better-quality capital, limit leverage, and keep enough liquid assets and stable funding to survive stress.

Key takeaways

  • Basel III raises the minimum share of a bank's risk-weighted assets that must be funded by common equity, the loss-absorbing form of capital, and adds capital buffers on top of those minimums.
  • It introduced the first international leverage ratio, a non-risk-based limit of Tier 1 capital against total exposures, with a 3% minimum.
  • It added two liquidity standards: the Liquidity Coverage Ratio (a 30-day stress test of liquid assets) and the Net Stable Funding Ratio (a one-year funding stability measure).
  • The final package of reforms agreed in 2017 is often called "Basel III Endgame" in the United States and "Basel IV" by some in the industry. It changes how banks calculate risk-weighted assets.
  • Implementation is national. Core parts of Basel III are in force in major jurisdictions. The final reforms have been delayed and contested, most visibly in the U.S. and the U.K.

Basel III is an international set of banking standards from the Basel Committee on Banking Supervision. It requires banks to fund themselves with more and better-quality capital, to limit total leverage, and to hold enough liquid assets and stable funding to survive periods of stress. The Committee developed it in response to the 2007 to 2009 global financial crisis, when many banks turned out to be overleveraged, thinly capitalized and dependent on short-term funding that disappeared overnight. Basel III is not a law. It is a standard that each country’s regulators write into their own rules, so the details, timing and scope differ by jurisdiction.

The three pillars of Basel III

Basel III keeps the three-pillar architecture introduced by Basel II and strengthens each part.

  1. Pillar 1: minimum requirements. These are the hard numbers: capital ratios, buffers, the leverage ratio and the liquidity ratios. They also cover the rules for calculating credit risk, market risk and operational risk.
  2. Pillar 2: supervisory review. Supervisors assess risks that Pillar 1 formulas miss or understate, such as interest rate risk in the banking book or concentration risk. They can require a bank to hold capital above the Pillar 1 minimums.
  3. Pillar 3: market discipline. Banks must publish standardized disclosures about their capital, risk exposures and risk-weighted assets, so investors, creditors and analysts can compare institutions. The revised Pillar 3 framework moved these disclosures onto common templates so figures line up across banks and countries.

In plain terms, Pillar 1 sets the rules, Pillar 2 lets the referee add penalties for risks the rulebook did not foresee, and Pillar 3 makes the scoreboard public.

How Basel III works

What counts as capital

Bank capital is the funding that can absorb losses without the bank failing or having to stop paying depositors. It sits on the liabilities and equity side of the balance sheet. Basel III sorts it into tiers by how reliably it absorbs losses.

  • Common Equity Tier 1 (CET1): common shares plus retained earnings and certain reserves, after regulatory deductions such as goodwill and most intangible assets. This is the highest-quality capital because it takes losses first and never has to be repaid.
  • Additional Tier 1 (AT1): perpetual instruments that can be written down or converted to equity when a bank runs into trouble, with dividends that the bank can cancel.
  • Tier 1 capital: CET1 plus AT1. It is “going-concern” capital, meant to absorb losses while the bank keeps operating.
  • Tier 2 capital: supplementary, “gone-concern” capital such as qualifying subordinated debt. It protects depositors and senior creditors if the bank fails.

Basel III also removed Tier 3 capital, a short-term debt category Basel II allowed for market risk. It tightened the deductions from common equity, because before the crisis some banks reported healthy headline ratios built on capital that could not actually absorb losses.

Minimum ratios and buffers

Every risk-based requirement is a ratio of capital to risk-weighted assets (RWA). Basel III sets three minimums and stacks buffers on top.

Requirement Level (share of RWA) Capital that qualifies
Minimum CET1 ratio 4.5% CET1
Minimum Tier 1 ratio 6% CET1 + AT1
Minimum total capital ratio 8% Tier 1 + Tier 2
Capital conservation buffer 2.5% CET1 only
Countercyclical capital buffer 0% to 2.5%, set by national authorities CET1 under the Basel standard
G-SIB surcharge 1% to 3.5%, by systemic importance bucket CET1

The capital conservation buffer brings the effective CET1 requirement to 7% and the total capital requirement to 10.5%. A bank that falls into the buffer is not shut down. Instead, it faces automatic limits on dividends, share buybacks and discretionary bonuses, and the limits get tighter the deeper it goes. The point is to stop banks paying cash out to shareholders and staff while their capital is eroding, which happened at several banks in 2007 and 2008.

The countercyclical capital buffer lets regulators raise requirements when credit is growing too fast and release the buffer in a downturn, so banks can keep lending instead of cutting loans to protect their ratios. The G-SIB surcharge applies to global systemically important banks. These are institutions whose failure would threaten the wider system, and the surcharge grows with their size, interconnectedness and complexity.

The 8% total capital figure is the same number Basel I set in 1988. What changed is the composition behind it. Far more of the stack must now be common equity, and the buffers and surcharges sit on top of the 8%. The ratio of total capital to risk-weighted assets is also known as the capital adequacy ratio.

Risk-weighted assets: the denominator

Risk weighting reflects the idea that a dollar of cash and a dollar of unsecured corporate loans do not carry the same risk. Each exposure is multiplied by a risk weight, and the results are summed:

RWA = Σ (exposure amount × risk weight)

Under the standardized approach, weights come from regulatory tables that depend on the type of borrower, external ratings and, for real estate, the loan’s loan-to-value ratio. Larger banks may get supervisory approval to use internal models that estimate the probability of default and the loss given default.

Banks also need capital for risks other than credit. Market risk covers losses from moving prices in the trading book. The final reforms overhauled this area through the Fundamental Review of the Trading Book (FRTB), which is part of the Basel III package rather than a separate accord. Operational risk covers losses from fraud, system failures or legal events.

A major part of the 2017 reforms is the output floor. A bank’s RWA calculated with internal models cannot fall below 72.5% of the RWA it would report under the standardized approaches, phased in over several years. The floor responds to evidence that model-based risk weights varied widely between banks holding similar assets.

The leverage ratio

Because risk weights can be wrong, Basel III added a simple backstop:

Leverage ratio = Tier 1 capital ÷ total exposure measure ≥ 3%

The exposure measure counts on-balance-sheet assets without risk weighting, plus derivatives exposures, securities financing transactions and off-balance-sheet items such as loan commitments. Under the final standard, G-SIBs carry an additional leverage ratio buffer equal to half of their risk-based G-SIB surcharge. National rules can go further. The U.S., for example, sets an enhanced supplementary leverage ratio for its largest banks.

The liquidity standards

Capital addresses solvency: whether assets are worth more than liabilities. Liquidity addresses whether a bank can meet cash demands as they arrive. Basel III treats the two separately because a solvent bank can still fail in a run.

Liquidity Coverage Ratio (LCR)

LCR = stock of high-quality liquid assets ÷ total net cash outflows over 30 days ≥ 100%

High-quality liquid assets (HQLA) are assets that can be turned into cash quickly with little loss of value: cash, central bank reserves and certain government securities, with lower-quality categories subject to haircuts and caps. Net outflows are estimated under a prescribed stress scenario with assumed run-off rates on deposits and wholesale funding.

Net Stable Funding Ratio (NSFR)

NSFR = available stable funding ÷ required stable funding ≥ 100%

Available stable funding is capital and liabilities expected to stay with the bank beyond one year, weighted by how stable they are. Required stable funding depends on how liquid and long-dated the bank’s assets are. The NSFR discourages funding long-term, illiquid loans with cheap short-term wholesale borrowing, the mismatch that broke several institutions in 2008.

Worked example: a hypothetical bank

The figures below are illustrative. The risk weights are simplified for teaching and are not any jurisdiction’s actual table.

A bank has $100 billion of on-balance-sheet assets:

Asset Amount Illustrative risk weight RWA
Cash and central bank reserves $10 bn 0% $0 bn
Government bonds $15 bn 0% $0 bn
Residential mortgages $40 bn 35% $14 bn
Corporate loans $35 bn 100% $35 bn
Total $100 bn $49 bn

Add an assumed $6 billion of RWA for market and operational risk, and total RWA is $55 billion.

The bank holds $5.5 billion of CET1, $0.8 billion of AT1 and $1.2 billion of Tier 2.

  • CET1 ratio: 5.5 ÷ 55 = 10.0%
  • Tier 1 ratio: 6.3 ÷ 55 = 11.5%
  • Total capital ratio: 7.5 ÷ 55 = 13.6%

Against the minimums plus the conservation buffer (7% CET1, 8.5% Tier 1, 10.5% total), the bank has headroom on every measure. Assume it is not a G-SIB and the countercyclical buffer is set at 0%.

Now the leverage ratio. Suppose off-balance-sheet commitments add $8 billion to the exposure measure, for $108 billion in total. Tier 1 capital of $6.3 billion gives 6.3 ÷ 108 = 5.8%, above the 3% minimum.

The example shows why the two measures exist side by side. If the bank moved $20 billion from corporate loans into government bonds, its RWA would fall by $20 billion and its CET1 ratio would rise to about 15.7% without raising a dollar of new capital. Its leverage ratio would not change, because total exposures stay the same. Risk-based ratios reward lower-risk assets. The leverage ratio caps how far a bank can expand its balance sheet regardless of how safe its risk weights say the assets are.

From Basel I to Basel III: what happened

The Basel Committee was formed by central bank governors of the Group of Ten countries in 1974, after the failure of Germany’s Bankhaus Herstatt disrupted international payments. It is hosted by the Bank for International Settlements in Basel, Switzerland, and its membership has since grown to cover 28 jurisdictions.

Basel I (1988) followed the Latin American debt crisis of the 1980s. It set a minimum of 8% capital against risk-weighted assets, using a handful of broad risk buckets. That 8% is the “target standard ratio” often cited for Basel I. Basel I mainly addressed credit risk. Market risk was added through a 1996 amendment.

Basel II (2004) refined risk sensitivity. It introduced the three pillars, added an explicit charge for operational risk, and allowed qualifying banks to use internal ratings and models to set their own risk weights. Its focus remained the amount of capital relative to measured risk.

Basel III (2010 onward) came after the 2007 to 2009 crisis showed the earlier framework’s gaps. Banks had met Basel II ratios with capital of poor quality, risk weights that understated trading and securitization risk, very high leverage and heavy reliance on short-term funding. The Committee agreed on the main package in 2010, with phase-in through the following decade. The package raised the quality and quantity of capital, added buffers, introduced the leverage ratio and created the two liquidity ratios. It also addressed systemic risk through the G-SIB framework. Basel I and II had looked at banks one at a time.

Basel III Endgame and implementation status

In December 2017 the Committee finalized the remaining reforms, mostly about the denominator: revised standardized approaches for credit risk, limits on internal models, a new operational risk method, the output floor and the revised market risk framework. The internationally agreed start date was pushed back by the COVID-19 pandemic to January 1, 2023, with the output floor phased in toward 2028.

Jurisdictions have moved at different speeds. In the United States, the Federal Reserve, the Federal Deposit Insurance Corporation and the Office of the Comptroller of the Currency proposed their version, called Basel III Endgame, in July 2023. It applied to banks with $100 billion or more in assets. The proposal required banks to calculate RWA under more than one approach and use whichever produced the higher requirement. It also extended the countercyclical buffer and certain other requirements to all banks above that threshold. The agencies’ plan pointed to a July 1, 2025 start with a three-year phase-in.

The proposal met heavy opposition from banks and business groups. Industry estimates of the capital increase ran higher than regulators’ own, with figures of up to about 20% for the largest banks circulating in the debate. After the comment period closed in early 2024, Federal Reserve officials signaled broad changes to the proposal, and analysts expected the U.S. timeline to slip. In January 2025, the Bank of England said it would delay its own implementation of the final reforms to January 2027, citing uncertainty over the U.S. approach. Any start date still has to be confirmed in final national rules.

National versions can be stricter than the Basel minimums. The Reserve Bank of India, for example, began implementing Basel III in 2013 and set CET1 and total capital minimums above the Basel floor for Indian banks.

How Basel III affects banks, borrowers and investors

Bank profitability. Equity is generally a more expensive funding source than deposits or debt. When a bank must fund more of its balance sheet with equity, the same profit is spread over a larger capital base, which lowers return on equity unless the bank raises prices, cuts costs or shifts its business mix. The conservation buffer also links capital levels directly to dividends and buybacks, so investors in bank shares watch the distance between a bank’s CET1 ratio and its requirement. Supporters of higher capital point to research suggesting that better-capitalized banks fund themselves more cheaply and have kept lending more steadily through downturns.

Lending and the economy. Critics argue higher requirements push banks to cut lending or raise loan prices, especially for borrowers that carry high risk weights. Supporters respond that banks with thicker cushions can keep lending when losses hit, rather than pulling credit to protect thin ratios. Academic work points both ways, and the effect depends on how far above requirements a bank already operates. Risk weights also shape what banks prefer to hold. Exposures with low weights consume less capital per dollar, which affects pricing across products.

Small and medium-sized banks. The Basel standards are designed for internationally active banks, and the U.S. Endgame proposal applied only above $100 billion in assets. Opponents argued smaller banks would feel indirect costs through their dealings with larger institutions. Regulators maintained that smaller banks fall under separate, simpler capital rules suited to their business models.

Depositors and markets. Pillar 3 disclosures give the public standardized information on bank capital and risk. Stronger capital and liquidity reduce the chance that a large bank’s failure spills into payment systems and broader markets, the systemic channel behind the bank rescues of 2008.

Capital relief under Basel III

“Capital relief” means reducing the capital a bank must hold against an exposure, normally by lowering the exposure’s RWA. Basel III allows this only through recognized techniques, grouped under credit risk mitigation (CRM):

  • Collateral: eligible financial collateral, such as cash or high-grade bonds, reduces the net exposure after supervisory haircuts.
  • Guarantees and credit derivatives: a qualifying guarantee can let the bank apply the guarantor’s risk weight to the covered portion instead of the borrower’s.
  • Netting: legally enforceable netting agreements let a bank measure exposure to a counterparty on a net rather than gross basis.
  • Securitization and risk transfer: moving loan risk to outside investors can reduce RWA, but only if the transfer meets the securitization framework’s tests for genuine risk transfer.

Each technique has strict legal, operational and eligibility conditions. The tighter securitization rules and the output floor both respond to the pre-crisis experience, when banks reported large capital savings from transactions that left much of the risk with them. As a hypothetical example, a $10 million corporate loan at a 100% weight carries $10 million of RWA. If $6 million is covered by a qualifying guarantee from an entity with a 20% weight, RWA falls to $4 million plus $1.2 million, or $5.2 million.

Limitations of Basel III

  • Risk weights remain estimates. Standardized tables are blunt, and internal models can differ widely between banks for similar portfolios. The output floor narrows that gap without closing it, and the leverage ratio is a backstop rather than a precise risk measure.
  • Complexity. The full framework runs to thousands of pages. Complexity raises compliance costs and can make it harder for outsiders to judge whether a bank is actually safe.
  • Uneven national adoption. Since each jurisdiction writes its own rules, scope, thresholds and timing differ. A bank’s reported ratio under one regime may not be directly comparable with a competitor’s under another.
  • Gaps in coverage. Requirements apply to banks. Lending and maturity transformation that move outside the regulated banking sector are not captured.
  • Interest rate and funding risks. The 2023 U.S. bank failures, including Silicon Valley Bank, involved losses on securities as rates rose and rapid deposit flight. U.S. rules at the time did not apply the full Basel liquidity standards to banks of that size, and the episode prompted debate over how interest rate risk and fast deposit runs are treated.
  • Procyclicality is reduced, not removed. The countercyclical buffer only works if authorities raise it in good times and release it promptly in bad times. That depends on judgment, not a formula.

Distinctions readers often confuse

Capital is not cash in the vault. Headlines often describe capital requirements as money banks must “keep in reserve.” Capital is a source of funding, mainly shareholders’ equity, that can absorb losses. A bank can lend out funds raised through equity. Cash and liquid securities are covered by the liquidity rules, not the capital rules.

Capital ratios versus liquidity ratios. A bank can be well capitalized yet short of cash, or flush with liquid assets yet insolvent. Basel III measures solvency with the capital and leverage ratios and measures liquidity with the LCR and NSFR.

Minimum versus requirement in practice. The 8% total capital minimum, the target ratio carried over from Basel I, is not what banks actually need to hold. Once the conservation buffer, any countercyclical buffer, a G-SIB surcharge and Pillar 2 add-ons are stacked on top, a bank’s effective requirement is considerably higher. Banks usually run an extra management buffer above that to avoid payout restrictions.

Basel III versus “Basel IV.” The 2017 final reforms are formally part of Basel III. Some practitioners call them Basel IV because of how much they change RWA calculations. U.S. regulators call their implementation Basel III Endgame.

Basel III versus national law. The Basel Committee has no legal authority. Its standards bind banks only through rules issued by national or regional authorities, such as U.S. federal banking agencies or the EU’s capital requirements legislation.

FAQs

What problem was Basel III designed to solve?

It was designed to fix the weaknesses the 2007 to 2009 financial crisis exposed: capital that was too thin and of low quality, excessive leverage, risk weights that understated risk, short-term funding that could vanish, and the danger that one large bank's failure could bring down others. Each major Basel III tool targets one of those failures.

Is Basel III the same as the Dodd-Frank Act?

No. Dodd-Frank is a 2010 U.S. law covering a wide range of financial regulation, including stress testing, resolution planning and consumer protection. Basel III is an international standard. U.S. regulators implement parts of it through their own capital and liquidity rules, which operate alongside Dodd-Frank requirements.

Does Basel III apply to every bank?

Not uniformly. The Basel standards target internationally active banks. Each country decides which institutions must follow the full framework and may apply simpler rules to smaller banks. In the U.S. Endgame proposal, for instance, the new requirements applied to banks with $100 billion or more in assets.

The Bottom Line

Basel III matters to bank investors, borrowers and anyone judging a bank's safety, since it shapes dividends, lending costs and disclosures. The practical point is that a bank's real requirement sits well above the 8% minimum once buffers, surcharges and Pillar 2 add-ons are stacked. Because each country writes its own rules and the final reforms keep slipping, readers should compare banks' ratios within the same regime.

About this page

Written by Markets & Investing Desk

The Markets & Investing Desk covers investing guides, the financial dictionary, companies, people, economics and market data.

Published
Last updated

Spotted an error? Tell the editors. This page is general information, not financial advice.