Fractional Reserve Banking: Definition and How It Works

Fractional reserve banking is a core concept to understand how banks work and how the U.S. financial system has changed over time. Here’s what to know.

» Skip down to how this system relates to the Fed rate

Fractional reserve banking is a system in which banks (and credit unions) keep a portion of their customers’ money in bank accounts — called deposits — and can use the rest to make loans, and to a lesser extent, investments.

To use a classic example: If you deposit $1,000 into a savings account and the bank keeps 10% in reserves, your bank holds onto $100 and lends out $900 to another customer. That customer spends $900 on a car repair and the auto shop deposits the money. The shop’s bank keeps $90 and lends out $810, and so on. Fractional reserve banking allows banks to essentially create money in the economy.

The U.S. central bank, called the Federal Reserve or the Fed, has required banks of a certain size to have a percentage of funds tied up in reserves. Before March 2020, the percentage was 3% or 10% of money held in transaction accounts, such as checking or savings accounts, and the percentage depended on a bank’s size. Banks could satisfy reserve requirements in two ways: holding cash themselves (as vault cash) or putting funds in an account at the Federal Reserve (as a reserve balance).

In March 2020, the Fed lowered the reserve requirement ratio — the minimum percentage of deposits kept as reserves — to 0%, meaning there’s no reserve requirement for banks. The Fed says it doesn’t have plans to change it, but the ratio can be adjusted. Banks still hold reserves despite not having a Fed requirement^[0]

Customer deposits aren’t the only funding source banks can use for loans. Banks can borrow from other banks and the Federal Reserve to manage their short-term business needs, which can include clearing payments and funding consumer loans. When banks borrow from each other overnight, they use the federal funds rate. Banks have to balance the gap between short-term bank deposits (that customers can withdraw from at any time) and the longer-term loans they provide.

Fractional reserve banking only describes part of how U.S. banking works. Reserves and reserve requirements play less of a role in how the Fed steers the economy today than historically. Instead of having reserves as a safety net, banks have capital and liquidity requirements to withstand potential economic crises. In simple terms, capital is a bank’s financial cushion to manage losses, and liquidity is how much cash (or assets that can be quickly converted to cash) a bank has to pay bills and fulfill customers’ withdrawals and other requests^[0]

The Fed’s use of the fractional reserve system has evolved over the past 30 years in parallel with other major free-market economies, say New York Fed researchers. Given the complexity of the U.S. system, the current risk-based approach with bank requirements for capital and liquidity to limit how much leverage banks can take on is more effective than a system primarily focused on reserves or reserve requirements, the Fed researchers say. Leverage, put simply, is the use of borrowed money to invest.

Reserve requirements predate a national currency, and became nationwide through the National Bank Act of 1863. Banks with national charters, or business licenses, originally had to hold 25% in reserves.

Reserve requirements were originally thought to guarantee that deposits could be converted into cash for the whole banking system. However, bank runs and panics in the late 1800s and early 1900s disproved that idea. Bank runs occur when people rush to withdraw their cash in fear of their bank failing.

To manage occasional spikes in public demand for cash from banks, the Federal Reserve Act of 1913 created the Federal Reserve System and the Federal Reserve became the lender of last resort to U.S. banks in trouble. Over time, the Fed took on more powers to help keep the economy stable and growing, including changes to reserve requirements.

Since 1913, the Fed’s reserve ratio has fluctuated many times, as low as 0% and as high as 26%, varying based on bank account types, banks’ size of deposits, and the geographic location of banks. The 0% reserve ratio has remained in place since late March 2020^[0]

The Fed used three monetary policy tools to control the direction of interest rates and the economy: open market operations, reserve requirements and the discount rate. Controlling the size of reserves meant controlling interest rates.

Here’s a quick breakdown:

• Open market operations — buying and selling government securities such as Treasury bonds in an open market — was the Fed’s way of expanding or shrinking the amount of reserve balances. Buying securities adds money to the economy through banks’ reserve balances. Selling them decreases the money supply.

• Reserve requirements helped keep banks’ demand for reserves consistent even as economic conditions changed. To satisfy reserve requirements overnight, banks with excess reserves would lend to banks in need using the federal funds rate, or Fed rate.

• The discount rate — the rate banks could borrow from the Fed — provided a more expensive, emergency option for banks that couldn’t get lending from other banks at cheaper rates. There was a stigma around borrowing from the Fed so banks tried to avoid it. The discount rate was (and still is) the ceiling for a bank’s borrowing costs.

The Fed kept the supply of reserves limited so that small movements could affect the Fed rate when necessary. If the economy needed a boost, the Fed lowered the federal funds rate, which meant adding more to the supply of reserves and lowering the discount rate. A lower Fed rate means borrowing gets cheaper for banks and their customers, which ideally pushes people to buy more, which helps businesses grow.

However, if inflation got too high, as it did in the early 1980s during the Great Inflation, the Fed decreased reserves and raised the discount rate to make lending more expensive and less accessible. A higher Fed rate discourages banks and consumers from borrowing^[0]

The financial crisis of 2007-2009 pushed the Fed to adapt from a “limited” to an “ample” reserves framework. Instead of using reserves, the Fed now controls the Fed rate through three rates (discussed below). The Fed focused on recovery by keeping the Fed rate low and maintaining a high level of reserves, which rendered the reserve requirements no longer relevant as a policy tool^[0]

To control the Fed rate, the central bank turned to a new set of tools, including paying interest on reserve balances. Reserves used to be a financial drag for banks since that money couldn’t earn them any money, but that changed in 2008 when the Fed started paying banks interest on their reserve balances. In fact, banks now have more in reserves than they did before 2008^[0]

.

• Interest on reserves is the primary way the Fed controls the federal funds rate. A bank has an incentive to lend reserves at a rate higher than what the Fed offers on reserves; otherwise, there’s not much incentive to lend. The Fed uses interest on reserves as a minimum rate that banks will lend to each other.

• Overnight reverse repurchase agreement rate is what banks and nonbank financial institutions can earn on deposits from the Fed overnight. The reason this rate exists is to prevent nonbanks from lending money below the Fed rate, since interest on reserves is only for banks. It’s the minimum rate for all institutions with Fed accounts.

• The discount rate still serves as an upper limit on the Fed rate.

Even when reserve requirements are at zero, banks borrow from other banks and the Federal Reserve in the course of managing short-term business needs. Open market operations still exist but mainly to keep the reserve supply well-stocked.

Full reserve banking operates on the idea that banks must hold onto 100% of customer deposits, including checking and savings account funds. So a bank can’t use deposits to make loans or investments. One of the criticisms of fractional reserve banking is that banks are still susceptible to bank runs — such as the recent collapse of Silicon Valley Bank — and a full reserve banking system would mean a bank run couldn’t happen.

As safe as this other system seems, it can create less economic growth for households and businesses since loans — from mortgages to small business loans — can be more scarce and expensive, according to New York Fed researchers. Banks currently pay costs to hold onto customers’ money, such as infrastructure, compliance and insurance premiums to the Federal Deposit Insurance Corp. (For credit unions, the equivalent insurance is through the National Credit Union Administration.) If banks can’t use deposits for loans, they could seek other ways to recoup costs, such as charging customers more for holding bank accounts.

In discussing fractional reserve banking, the concept of the money multiplier may arise to try to explain the link between the Fed and banks. However, the concept has become outdated due to changes the Fed has made to implement monetary policy.

The money multiplier formula has historically been used to describe the maximum potential amount of money a bank can create in the financial system through new loans. The idea is that the Fed’s reserve requirements affect how much money a bank can create, though that’s no longer the case.

The money multiplier formula is one divided by the reserve ratio.

Money multiplier formula = 1 / reserve ratio

For example, a 10% ratio would mean one divided by one-tenth (or 1 / 0.10), which equals 10. So a bank could make 10 times the initial deposit in the form of loans, which turn into deposits, then into loans in smaller and smaller amounts. If a customer deposits $1,000, the bank keeps $100 and lends $900 to a business. That business deposits $900 in another bank, that bank keeps $90 and lends $810, and so on.

However, the money multiplier formula doesn’t factor in the amount a bank has in excess reserves or what happens when the reserve ratio is 0%. Mathematically, 1 divided by 0 is undefined, or infinity. In real life, the money supply can’t expand infinitely.