Fractional Reserve

The Reserve Requirement Ratio (RRR) is the legally mandated minimum percentage of customer deposits a bank must hold as vault cash or deposits with the central bank. This ratio, set by a central bank (e.g., the Federal Reserve), directly controls the money multiplier effect. For example, a 10% RRR means a bank can lend out 90% of its deposits, theoretically allowing the initial deposit to expand the money supply by up to 10x.

This is the core mechanism of fractional reserve banking. When a bank lends money, it creates new bank deposits for the borrower. These new deposits can be re-deposited and lent out again, creating a chain reaction. The theoretical maximum expansion of the money supply is calculated by the money multiplier formula: 1 / Reserve Requirement. This process demonstrates that most modern money is credit money, created by commercial banks, not physical currency issued by the state.

Banks face two primary risks under this system:

• Liquidity Risk: The risk that a bank cannot meet immediate withdrawal demands because its assets (loans) are illiquid. This can trigger a bank run.
• Solvency Risk: The risk that the total value of a bank's assets falls below its liabilities (deposits), meaning it is fundamentally insolvent. Central banks act as lenders of last resort (e.g., via the discount window) to provide emergency liquidity and prevent solvent banks from failing due to temporary liquidity shortages.

Fractional reserve banking is a key transmission channel for monetary policy. By adjusting the reserve requirement, discount rate, and conducting open market operations, a central bank influences how much credit commercial banks can create. Lowering reserve requirements or interest rates encourages lending and expands the money supply (expansionary policy), while raising them has the opposite (contractionary) effect, making it a primary tool for managing inflation and economic growth.

Fractional reserve is often contrasted with full-reserve banking (or 100% reserve banking), a system where banks must hold reserves equal to 100% of their demand deposits. In this model, banks cannot create new money through lending; they can only lend from time deposits (savings accounts). Proponents argue it eliminates bank runs, while critics contend it would severely restrict credit availability and economic growth. Narrow banking is a related, modern proposal for a safer banking model.

Critics, following economists like Hyman Minsky, argue that fractional reserve banking is inherently pro-cyclical and contributes to financial instability. During economic booms, asset prices rise, collateral values increase, and banks extend more credit, fueling a speculative bubble. During a downturn, the reverse occurs, leading to a credit crunch and reinforcing the recession. This Minsky Moment describes the point when overleveraged institutions are forced to sell assets, causing a sudden collapse in market value.

Some algorithmic stablecoin designs attempt a pure fractional reserve without direct collateral backing. They use on-chain logic and secondary governance tokens to regulate supply, acting as a "share" in the system's reserve. This model is highly sensitive to reflexivity and loss of confidence, as seen in historical de-pegs, because the "fraction" of backing can rapidly approach zero during a bank run scenario.

Automated Market Makers (AMMs) like Uniswap implement a form of reserve accounting. Each pool holds two assets in reserve according to the constant product formula (x * y = k). Impermanent loss occurs when the external market price diverges, effectively depleting the pool's reserve of one asset relative to simply holding. LPs provide the fractional reserve that enables continuous trading.

The critical difference from traditional finance is that DeFi's "fraction" is almost always greater than 1.

• TradFi Bank: $100 in deposits → $90 in loans (10% reserve).
• DeFi Protocol: $150 in collateral → $100 in minted assets (150% collateralization). This over-collateralization replaces trust in a central entity with cryptographic enforcement of the reserve ratio.

A fundamental advantage of DeFi is the verifiability of reserves. Unlike opaque bank ledgers, protocols like MakerDAO publish real-time, on-chain proof of all collateral assets backing issued stablecoins. This transparency allows any user to audit the actual reserve fraction instantly, moving from trusted audits to cryptographically verified proof of solvency.

A fractional reserve system operates on the premise that not all depositors will withdraw funds simultaneously. This allows banks—or crypto lending platforms—to lend out a majority of deposited assets. The primary risk is a bank run, where a loss of confidence triggers mass withdrawals, potentially exceeding available reserves and causing insolvency. This is a fundamental counterparty risk where users' claims on their full deposit are not backed 1:1 by liquid assets.

Centralized (CeFi) and some decentralized (DeFi) lending platforms employ fractional reserve models. They accept user deposits (e.g., USDC, ETH) and re-lend them to borrowers or use them in other yield-generating strategies. Key risks include:

• Protocol Insolvency: If a large portion of loans default or collateral value crashes, the platform cannot cover withdrawals.
• Liquidity Mismatch: Assets may be locked in longer-term strategies while liabilities (user deposits) are demandable instantly.
• Transparency Gap: Unlike on-chain DeFi, CeFi platforms often lack verifiable proof of reserves.

This contrast is critical for understanding asset-backed stablecoins.

• Fractional-Reserve Stablecoin: Backed by a mix of cash, commercial paper, and other assets (e.g., early models of Tether/USDT). Carries custodial and credit risk from the backing assets.
• Full-Reserve Stablecoin: Backed 1:1 by a designated reserve (e.g., USDC with cash and Treasuries, or DAI with on-chain crypto overcollateralization). Aims to eliminate the classic bank run risk by ensuring redeemability at all times.

A key innovation in crypto is the ability to cryptographically prove reserves and liabilities. Proof of Reserves (PoR) uses Merkle trees to allow users to verify their funds are included in the total custodial assets. Proof of Liabilities shows all customer obligations. Together, they enable proof of solvency, demonstrating that reserves ≥ liabilities. This transparency is a direct countermeasure to the opacity of traditional fractional reserve banking, though it requires regular, auditable attestations.

In traditional finance, fractional reserve banking is heavily regulated through capital adequacy rules (e.g., Basel III). These mandate that banks maintain a minimum ratio of capital (equity) to risk-weighted assets to absorb losses. In crypto, analogous frameworks are emerging. Regulations may require crypto asset service providers (CASPs) to hold minimum liquidity coverage ratios (high-quality liquid assets to cover net outflows) to mitigate the systemic risk inherent in fractional reserve models.

The interconnectedness of fractional reserve institutions creates systemic risk. The failure of one major lender or stablecoin issuer can trigger:

• Contagion: Losses spreading to other platforms due to interlinked exposures.
• Liquidity Crunch: A fire sale of assets across the ecosystem to meet withdrawal demands, depressing prices further.
• Loss of Confidence: A crisis in one entity can lead to generalized withdrawals (a 'run') on other similar entities, regardless of their individual health. This was observed during the collapses of Terra/LUNA, Celsius, and FTX.

A banking system where financial institutions hold 100% of customer deposits as reserves. This is the direct conceptual opposite of fractional reserve banking, eliminating the practice of lending out deposited funds to create new money. While historically rare in modern economies, it is a model often cited in discussions about monetary stability and is analogous to how many non-custodial blockchain protocols operate, where user assets are not rehypothecated.

The theoretical maximum amount of broad money (like M2) that can be created by the banking system from a given unit of central bank money (the monetary base). It is calculated as 1 / Reserve Requirement Ratio. For example, a 10% reserve requirement implies a money multiplier of 10. This concept quantitatively describes the credit creation power inherent in a fractional reserve system, where a single deposit can theoretically generate multiple loans.

A crisis of confidence where a large number of a bank's depositors attempt to withdraw their funds simultaneously, fearing the institution's insolvency. This is a critical systemic risk of fractional reserve banking, as banks only hold a fraction of deposits as liquid reserves. If withdrawals exceed available reserves, the bank fails. This contrasts with on-chain over-collateralization, where protocols like MakerDAO are designed to remain solvent even during mass withdrawals, as all loans are backed by excess collateral.

The practice where a financial intermediary (like a prime broker) re-uses collateral posted by its clients for its own purposes, such as securing its own borrowing. This creates a chain of leveraged claims on the same underlying asset, amplifying risk in a manner analogous to fractional reserve lending. In DeFi, similar effects can occur through recursive lending on money market protocols, though they are typically transparent and governed by smart contract logic.

The process of creating digital assets pegged to a flat currency. Fiat-collateralized stablecoins (e.g., USDC) often operate on a fractional reserve model where the issuer holds a mix of cash and short-term securities against the tokens in circulation. This contrasts with crypto-collateralized (e.g., DAI) and algorithmic stablecoin models, which use on-chain collateral or algorithmic mechanisms, respectively, to maintain the peg without traditional fractional reserves.

A smart contract in decentralized finance (DeFi) that locks cryptocurrency tokens to facilitate trading, lending, or other financial services. While not a direct analog, it shares a functional similarity: pooled user funds are utilized to provide a service (liquidity) to others. However, it differs fundamentally as user ownership is represented by pool tokens (LP tokens), and the system's solvency is governed by constant-product formulas and public blockchain state, not by a fractional reserve of a centralized liability.