Evolution of Bank Capital Requirements from Basel I to Basel III

The evolution of bank capital requirements across three Basel accords fundamentally transformed how the world’s largest banks manage their balance sheets. Each accord tightened minimum buffers, widened the definition of what qualifies as capital, and introduced new ways of measuring risk — forcing banks to rethink asset composition and operational leverage over three decades.

Basel I: The First Global Minimum

In the late 1980s, a string of bank failures and the Latin American debt crisis exposed how lightly capitalized some major institutions had become. The Bank for International Settlements convened regulators from 10 leading nations — the Basel Committee on Banking Supervision — to establish a common floor.

Basel I, agreed in 1988, set a simple rule: banks had to hold capital equal to at least 8% of risk-weighted assets. The elegance was deceptive. Assets were bucketed into rough risk categories — government bonds carried 0% weight (no capital needed), mortgages 50%, corporate loans 100%. A bank holding $100 million in corporate loans had to hold $8 million in capital.

What counted as capital? Primarily Tier 1 — actual shareholder equity, retained earnings — and Tier 2 — subordinated debt, loan-loss reserves. Tier 1 had to be at least half the total.

Basel I worked. It harmonized minimum standards and arrested a race to the bottom where poorly capitalized banks undercut well-capitalized rivals. But it was blunt. A AAA-rated mortgage and a junk-rated corporate loan faced the same 100% weight. No capital charge existed for interest-rate-risk in the trading book or operational failures. Banks learned to game the system — warehousing low-weight assets, securitizing loans to shift them off balance sheet.

Basel II: Risk-Weighting Sophistication and Hidden Leverage

By the early 2000s, derivatives had exploded, the rating industry had matured, and regulators wanted banks to do more precise risk measurement. Basel II, finalized in 2004 and phased in through 2006, kept the 8% minimum but rewired almost everything else.

Under Basel II, capital requirements hinged on three pillars: minimum capital-adequacy-ratio calculations, supervisory review, and market discipline. The minimum itself fragmented. Banks could use:

• Standardized approach: similar to Basel I, but more granular weights based on bond ratings (AAA = 20%, A = 50%, BBB = 100%, below BBB = 150%).
• Internal Ratings-Based (IRB) approach: larger banks could model their own probability of default and loss-given-default for each counterparty, reducing capital charges on loans to low-risk borrowers.

For the first time, capital requirements explicitly covered credit-risk, market-risk, and operational risk — losses from failed processes, fraud, system failures. Operational-risk capital charges often ran 12–15% of income.

The seduction was clear: if you owned a portfolio of low-default-probability loans, you could hold less capital than a bank with higher-risk borrowers. This incentivized internal risk modeling but also created opacity. The models were proprietary, assumptions varied widely, and the benefit of lower capital weights invited herding into similar assets.

Basel II also permitted banks to hold less capital if they hedged their credit-risk with credit-default-swaps. A loan-holding bank could buy protection and reduce its capital charge. The problem: the protection seller (often another bank or shadow-bank) accumulated concentrated risk with no haircut for the correlation of defaults in a systemic crisis. When Lehman Brothers failed in 2008, the credit-default-swap market seized up and capital buffers evaporated.

Basel III: Capital Multiplied, Definitions Narrowed

The 2008 financial crisis revealed that banks’ capital was thinner than stress tests had suggested. Goodwill, intangible assets, and deferred-tax assets had eaten into equity. Hybrid instruments — subordinated bonds with coupon-skipping features — didn’t absorb losses during the crisis the way regulators hoped. By the time Lehman was in freefall, leverage ratios (assets to equity, unweighted) told a more honest story than risk-weighted capital ratios.

Basel III, agreed in late 2010 and phased in through 2019, more than doubled the bar:

• Common Equity Tier 1 (CET1) minimum: 4.5% of risk-weighted assets, up from 2% under Basel II. With regulatory buffers added, the effective floor is 10.5%.
• Tier 1 capital minimum: 6% (including CET1).
• Total capital minimum: 8%, unchanged, but the mix shifted — more had to be highest-quality.
• Unweighted leverage ratio: introduced a 3% floor independent of risk weights. A bank with $3 trillion in assets had to hold $90 billion in Tier 1 capital, period.

Capital definitions narrowed drastically. Goodwill and intangible assets were deducted entirely from CET1. Deferred-tax assets from loss carry-forwards faced caps. Tier 2 subordinated debt had to be subordinated to all other obligations; many hybrid instruments were grandfathered but not replaced.

Banks faced a binary choice: raise equity, shrink risk-weighted assets, or both. European and Japanese banks disproportionately shed assets. U.S. banks — already higher-capitalized in the crisis and facing stronger earnings recovery — raised substantial equity through secondary offerings.

Basel III also introduced liquidity standards. The Liquidity Coverage Ratio (LCR) required banks to hold enough high-quality liquid assets to survive 30 days of cash outflows under stress. The Net Stable Funding Ratio (NSFR) ensured longer-term stability. These created new incentives to hold Treasuries and mortgage-backed securities and to reduce reliance on overnight wholesale funding.

The Mechanics of Risk-Weighted Assets

The heartbeat of all three accords is the risk-weighted asset calculation. Take a bank’s balance sheet:

• $50 billion in Treasuries (0% risk weight): $0 in RWA.
• $20 billion in mortgages, 50% weight: $10 billion in RWA.
• $30 billion in corporate loans, 100% weight: $30 billion in RWA.
• Total RWA: $40 billion.

Under Basel III, this bank needs Tier 1 capital of at least $2.4 billion (6% of $40B). If it holds $3 billion, its Tier 1 ratio is 7.5% — comfortably above the floor but unremarkable for a major bank.

The weighting isn’t static. Regulatory changes, rating downgrades, and accounting shifts move the needle. A mortgage-backed security rated AAA may carry a 20% weight; if downgraded to A, it jumps to 50%. A corporate loan to a AAA-rated counterparty might be 20%; to a BB-rated firm, 100% or more. Banks with large exposures to single countries or sectors see their risk weights shift with political events, credit-rating changes, or regulatory restatements.

Collateral and Netting

Both Basel II and III allowed banks to reduce risk-weighted assets by pledging collateral or using netting agreements. A $10 billion loan secured by $8 billion in Treasuries could be treated as $2 billion of net exposure. This encouraged collateral-heavy transactions and formal master netting agreements.

The flip side: if collateral suddenly becomes illiquid (as happened with mortgage-backed securities in 2008) or if a counterparty fails and collateral has to be recovered from bankruptcy, the protection disappears overnight.

Implementation and Transition Costs

Basel I was implemented swiftly — major banks reached compliance by 1992. Basel II took longer; implementation extended to 2008 in many jurisdictions, and even then, regulators like the SEC declined to fully adopt the internal-models approach for U.S. banks. (The U.S. instead kept a more conservative standardized approach.)

Basel III’s phase-in lasted until 2019, but regulators immediately began negotiating Basel IV — a further tightening of output floors and restrictions on lower risk weights under internal models. Many of those provisions have now been agreed but not yet implemented.

See also