Loading field map
Banking theory is the branch of monetary economics that asks why banks exist, what they do, and how their distinctive features—maturity transformation, liquidity creation, and the issuance of demandable debt—shape the financial system and the broader economy. It is not primarily a theory of how individual banks maximize profit, though that enters, nor a theory of financial markets in general, though banks interact with them. The field's central puzzle is that banks are simultaneously ordinary firms and extraordinary institutions: they take deposits repayable on demand and lend them out for longer terms, creating liquidity for depositors and funding for borrowers while exposing themselves to the risk of runs.
Three questions organize the field. First, why do banks exist at all? In a world of perfect information and frictionless markets, borrowers and lenders could deal directly through securities markets, and intermediaries would be redundant. Banking theory therefore explains banks as responses to specific frictions: information asymmetries, transaction costs, and the need for liquidity insurance.
Second, what makes banks fragile? Banks fund long-term, illiquid assets with short-term, demandable liabilities. This maturity transformation is socially valuable—it lets savers hold liquid claims while borrowers get stable long-term funding—but it creates the possibility of self-fulfilling runs. If enough depositors believe others will withdraw, it is rational for each to withdraw early, even if the bank's assets are fundamentally sound. Understanding this fragility, and whether it is an inevitable cost or a correctable flaw, is a central theoretical concern.
Third, what is the social function of banks, and how should they be regulated? Because banks create money-like claims and are central to the payment system, their failures have externalities beyond the bank itself. Banking theory thus connects to questions of monetary policy transmission, financial stability, and the design of regulation—capital requirements, deposit insurance, lender-of-last-resort facilities—that aim to preserve banks' benefits while limiting their risks.
Modern banking theory emerged in the twentieth century, but its questions have older roots. Nineteenth-century debates over free banking, the gold standard, and the Bank Charter Act of 1844 wrestled with whether banks' note issuance and credit creation were inherently destabilizing. The British Currency School argued that banks should be constrained to hold reserves fully backing their notes; the Banking School countered that the demand for credit naturally limits issuance. These debates concerned the macroeconomy more than the microeconomics of the bank, and they did not produce a systematic theory of why banks exist.
The Great Depression of the 1930s, with its wave of bank failures, made the fragility of banks a pressing practical problem. But the theoretical tools to analyze it did not yet exist. The field as a distinct body of theory took shape in the 1960s and 1970s, when economists began applying the new information economics—the study of situations where one party knows more than another—to financial institutions.
A crucial early contribution came from the economist George Akerlof's 1970 demonstration that asymmetric information can cause markets to fail entirely. If sellers know the quality of their goods and buyers do not, buyers will only pay the average price, driving high-quality sellers out of the market. Applied to banking, this suggested that banks might exist to overcome information problems that prevent direct lending. A borrower knows more about her own project than a lender does; a bank that specializes in screening and monitoring borrowers can solve this problem more cheaply than dispersed individual lenders.
The modern theory of bank runs was established in the early 1980s by Douglas Diamond and Philip Dybvig, whose model remains the field's canonical reference point. In their framework, depositors face uncertain consumption needs: some will need their money early, others later. A bank offers demand deposits that let each depositor withdraw when needed, while investing in long-term projects that pay higher returns. This improves welfare for everyone. But the same contract creates a coordination problem: if depositors who do not actually need early withdrawal believe that others will withdraw, they too will withdraw, and the bank—which cannot liquidate its long-term assets at full value—will fail. The model showed that runs are not caused by bad assets or irrational panic; they are an equilibrium outcome of a contract that is otherwise socially beneficial.
This insight reframed the policy debate. If runs are a rational response to the structure of demandable debt, then deposit insurance—which guarantees depositors they will be paid regardless of others' behavior—can eliminate the run equilibrium while preserving the bank's liquidity-creating function. The model also clarified why central banks act as lenders of last resort: by standing ready to provide liquidity to solvent but illiquid banks, they can reassure depositors and prevent runs.
Banking theory is not divided into sharply opposed schools in the way that, say, macroeconomics is divided between Keynesians and monetarists. Instead, it is organized around complementary research programs that emphasize different frictions and functions. These approaches coexist and often combine.
The dominant tradition treats banks as delegated monitors. The core problem is that lending requires evaluating borrowers and ensuring they repay. Individual savers lack the expertise and the incentive to do this; a bank that pools many deposits and lends to many borrowers can specialize in screening loan applicants and monitoring borrowers after the loan is made. Because the bank's own capital is at stake, it has an incentive to monitor carefully. This approach explains why banks make loans that are hard to sell on markets: a loan's value depends on the bank's private information about the borrower, which cannot be easily transferred to a third party.
A related strand emphasizes banks' role in relationship lending. Over time, a bank learns about a borrower's character, business prospects, and reliability—information that is not publicly available and cannot be captured in a credit score. This gives banks a comparative advantage in lending to small businesses and other opaque borrowers. It also creates a form of lock-in: a borrower who has built a relationship with one bank cannot easily switch, giving the bank some market power.
The information approach has limits. It explains why banks exist but says less about their distinctive liability structure—why banks fund themselves with demand deposits rather than equity or long-term debt. It also struggles to explain the modern growth of securitization, where loans are pooled and sold to investors, which seems to reduce the importance of bank monitoring.
The Diamond-Dybvig tradition treats banks as liquidity providers. The problem here is not information but time: depositors do not know when they will need their money, and long-term investments are more productive than short-term ones. A bank offers a contract that gives each depositor the right to withdraw on demand while investing in long-term assets. This is a form of insurance: depositors who need money early are subsidized by those who do not, and everyone is better off because the bank can invest productively.
This approach directly explains the bank's balance-sheet structure—demandable deposits funding illiquid loans—and its fragility. It also generates a distinctive policy conclusion: because runs are a coordination failure, not a sign of insolvency, they can be prevented by deposit insurance or lender-of-last-resort support. The approach has been extended to explain other forms of short-term funding, such as commercial paper and repurchase agreements, which play a similar liquidity-providing role in modern financial systems.
The liquidity-insurance approach has been criticized for treating the bank's assets as a black box. It does not explain why the bank holds illiquid loans rather than marketable securities, nor does it address the information problems that make loans illiquid in the first place. Recent work has tried to integrate the two traditions, showing that banks combine information-based lending with liquidity provision because the two functions reinforce each other.
A third tradition, with roots in the older debates over banking and the business cycle, treats banks as creators of money. When a bank makes a loan, it credits the borrower's deposit account, creating new money. This is not a metaphor: bank deposits are part of the money supply, and banks' lending decisions directly affect the quantity of money in circulation. This approach emphasizes that banks are special because their liabilities serve as a medium of exchange, and that this is what makes their failures so disruptive.
This tradition connects banking theory to monetary economics more directly than the other approaches. It asks how bank lending transmits monetary policy, how the money supply responds to changes in bank reserves, and whether banks' ability to create money is a source of instability. It also raises questions about the boundary between banks and other financial institutions: if a money-market fund offers checkable accounts, is it a bank? The modern regulatory answer is often yes, but the theory has not fully settled the question.
The money-and-payments approach has been less central to the microeconomic theory of banking than the other two, but it has gained renewed attention since the 2008 financial crisis, which showed how the shadow banking system—non-bank institutions performing bank-like functions—could create the same fragility as traditional banks.
A fourth approach treats banks as firms in a market, analyzing competition, market structure, and pricing. This tradition asks how the number of banks, their market power, and their cost structures affect the interest rates they charge on loans and pay on deposits, and how these in turn affect the volume of lending and the stability of the system. It draws on standard industrial-organization economics rather than on the distinctive frictions emphasized by the other approaches.
This approach has produced important results about the trade-off between competition and stability. More competition can reduce banks' profits, making them more fragile because they have less capital to absorb losses; but less competition gives banks market power that can be exploited at the expense of borrowers and depositors. The empirical evidence on this trade-off is mixed, and the theory has not produced a clear consensus.
These approaches are not rivals in the sense of offering mutually exclusive explanations. They emphasize different features of the same institution, and a complete theory of banking needs all of them. The information approach explains why banks hold hard-to-value assets; the liquidity-insurance approach explains why they fund those assets with demandable deposits; the money-and-payments approach explains why those deposits are special; and the industrial-organization approach explains how banks compete with each other and with non-bank financial institutions.
The relationship between the first two is the most developed. A bank that monitors borrowers acquires private information that makes its loans illiquid—they cannot be sold without revealing what the bank knows. This illiquidity is precisely what makes the bank's demandable deposits valuable as a source of liquidity insurance. The two functions are complementary: monitoring creates the illiquid assets that make liquidity provision necessary, and demandable deposits give the bank a stable funding base that lets it hold illiquid loans.
The money-and-payments approach cuts across the others. It is less a separate theory of why banks exist than a reminder that banks' liabilities are money, and that this has macroeconomic consequences. A bank that creates deposits when it lends is doing something different from a mutual fund that pools investors' money; the former expands the money supply, the latter does not. This distinction was blurred in the decades before 2008, when shadow banks created money-like claims without being regulated as banks, and the crisis showed the costs of that blurring.
The 2008 financial crisis reshaped banking theory in several ways. First, it demonstrated that the fragility identified by the liquidity-insurance approach extends beyond traditional banks to the shadow banking system—money-market funds, investment banks, and other institutions that fund long-term assets with short-term, runnable liabilities. The theory of runs now applies to a broader class of institutions than the deposit-taking bank.
Second, the crisis revived interest in the money-and-payments approach. The expansion of credit by shadow banks, and the subsequent contraction when their short-term funding dried up, showed that the money supply is not determined solely by central bank policy and traditional banks. This has led to renewed work on how the financial system as a whole creates money and credit, and on how regulation should respond.
Third, the crisis prompted a reassessment of the costs of bank fragility. The Diamond-Dybvig model suggests that runs are a correctable market failure, but the crisis showed that even with deposit insurance and lender-of-last-resort support, the failure of large, interconnected financial institutions can impose enormous costs on the real economy. This has led to a focus on systemic risk—the risk that the failure of one institution can cascade through the financial system—and on the special problems posed by banks that are "too big to fail."
Current research in banking theory is characterized by several active frontiers. One is the integration of the information and liquidity approaches into a unified framework that can address both the asset and liability sides of the bank's balance sheet. Another is the extension of run theory to modern funding markets, including repurchase agreements and money-market funds. A third is the analysis of macroprudential regulation—rules that aim to protect the financial system as a whole rather than individual institutions—using the tools of banking theory. And a fourth is the question of how digital currencies, both central-bank-issued and private, might change the role of banks in the payment system and the money supply.
Banking theory remains a field in which the central questions are settled enough to provide a framework but open enough to generate active research. The canonical models of the 1980s still structure the debate, but they have been extended, qualified, and in some respects challenged by subsequent work. The field's enduring contribution is to show that banks are not merely financial intermediaries among others, but institutions whose distinctive combination of functions—monitoring, liquidity provision, and money creation—makes them both socially valuable and inherently fragile.