3 Public Policy Aspects of Finance

Garry Schinasi
Published Date:
December 2005
  • ShareShare
Show Summary Details

As examined in previous sections—based solely on the observation that finance intrinsically embodies uncertainty about human trust—finance can have both positive (or efficiency-enhancing) and negative effects on economic processes and systems. Both originate in the element of trust embodied in finance, and the balance between the two depends on the ways in which finance, as a process, facilitates the assessment, transformation, pricing, and allocation of the uncertainty and risks associated with this trust. The leaps of faith by members of society to engage in trust relationships so as to obtain private benefits have produced both private and social efficiency (and welfare) gains throughout history. However, history has also revealed that confidence in trust is a fragile feature of human interaction in all social endeavors, including economic and financial activities. When confidence in the trust underlying economic and financial transactions breaks down, so too can the ability of markets and financial institutions to perform their basic pricing, allocative, and intermediary functions. This breakdown can threaten financial stability. In effect, the element of human trust, and whether that trust is strong or weak, helps to determine the strength and stability of the economic and financial mechanisms that provide order to everyday economic life.

Can and will unfettered market forces lead to the right balance of these positive and negative aspects of finance? Is individual rationality (utility-and profit-maximizing behavior) sufficient to achieve collective rationality (or socially optimal outcomes) in finance?24 If not, what steps can be taken to move society closer to the right balance?25

Box 3.1.Prisoner’s Dilemma

The prisoner’s dilemma is a gaming situation often experienced in life and in financial markets. It illustrates how the pursuit of self-interest rather than cooperation can lead to an outcome for the collective that is inferior to other feasible outcomes.1

The original “game” was created by Albert Tucker, a Princeton mathematician. Tucker’s dilemma was a more intuitive representation of a game discussed in a paper by a Rand mathematician, Merrill M. Flood; Flood’s paper was followed by additional experiments at Rand with his colleague Melvin Dresher.2 The original dilemma was posed as follows: Two men, charged with a joint violation of law, are held separately by the police. Each is told that

  • if one confesses and the other does not, the former will be given a reward and the latter will be fined;
  • if both confess, each will be fined.

At the same time, each has good reason to believe that if neither confesses, both will go free.

The incentives implicit in this situation are for each individual to follow the dominant strategy of confessing, which gives a smaller sentence to each prisoner for each of the choices the other prisoner might make. However, when each prisoner follows his dominant strategy, each receives a sentence greater than if neither confessed. In this situation, when both parties pursue their own interests exclusively and do not cooperate with each other, the outcome is worse for both of them, even though in any given situation, either party is better off not cooperating.3

In finance, it is not difficult to imagine situations like this classic dilemma. For example, in the midst of market turbulence in, say, the bond market, the five market makers accounting for most of the trading might encounter a situation in which each of them would strongly prefer to withdraw from market-making services so as to use inventory and liquidity to protect their own portfolios, including in other markets that might later be affected if the turbulence turns to a widespread crisis. Thus, each has the incentive to withdraw, and if they all do so, the worst case scenario—a dysfunctional market—has a high probability of occurring. However, if they cooperate and all jointly decide to continue to supply market-making services, the turbulence would probably be limited and subside, leaving all of them better off.

An elaboration on the classic but simple and concrete prisoner’s dilemma will make the problem clearer. A more modern adaptation might unfold as follows: Consider a situation involving two prisoners (although it can be generalized to any number participants). Two prisoners arrested on a minor charge are suspected of committing a major crime as well. Conviction carries a jail sentence of one year for the minor crime for each prisoner and a total of 10 years for the major crime. The police are having difficulty investigating the major crime and are under some pressure to resolve it. Each prisoner is questioned separately and is presented with the following choice: testify that the co-suspect committed the major crime (noncooperation) and go free, provided the co-suspect does not also testify, or remain silent and receive either a one-year sentence if the co-suspect remains silent or serve the full 10-year sentence if the co-suspect testifies. In the event both testify to the other’s guilt, they split the prison sentence for the major crime (five years each).

The possible outcomes to these separate decisions follow:

  • Both suspects remain silent. This is the cooperative solution, and they each receive a year in jail for the minor offense (R = –1, R = –1; R is the reward for cooperation). This is in fact the best outcome for each prisoner, as they would share the highest total payoff (–2).
  • One suspect testifies against the other, while the other remains silent. Each is tempted to testify and go free (and receive T = 0; T is the temptation to confess). If one prisoner testifies against the other, and the other does not, the former goes free, while the latter receives the full 10-year sentence (T = 0, S = –10; S is the sucker payoff for remaining silent).
  • Both testify against each other. If they both testify against each other, they will each receive a five-year sentence (P = –5, P = –5; P is the penalty payoff for confessing guilt).

While the best outcome is for both prisoners to remain silent, each of them has little incentive to choose this option, knowing that the other prisoner understands the risks, and also understands that his choices do not affect his co-suspect’s choices: each must choose independently, and without prior consultation. They also both have the strong temptation to squeal on the other prisoner and go free. Given these conditions, each prisoner has the strong incentive to make the choice that minimizes his prison term regardless of what the other prisoner chooses. Thus, it is rational for each prisoner to testify against the other and obtain a five-year sentence for the major crime, because remaining silent poses the risk of the worst outcome, a 10-year sentence. In this case, they take an equal share of the worst outcome of the game, rather than taking an equal share of the best outcome of the game.

These options can be represented in a payoff matrix as follows:

B SilentB Testifies
A Silent(–1, –1)(–10, 0)
A Testifies(0, –10)(–5,–5)

In general, for a two-person game to pose the Prisoner’s Dilemma, the payoff matrix, and the structure of the various payoffs must be as follows: T > R > P > S, and 2 R > T + S.

B TestifiesB Testifies
A Silent(R, R)(S, T)
A Testifies(T, S)(P, P)
1For a readable and entertaining explanation, see Prisoner’s Dilemma (Poundstone, 1992).2See “Some Experimental Games” (Flood, 1952).3See A Mathematician Plays the Stock Market (Paulos, 2003).

These are difficult, and in some ways philosophical questions. By applying certain features of the economics of the public sector, this chapter frames the public policy aspects of finance and financial stability considerations, and suggests that

  • finance entails both positive and negative externalities;
  • finance intrinsically embodies imperfect information (at the very least, uncertainty about human trust);
  • access to finance—the process that provides effective store-of-value services—is a public good, as is access to a universally accepted means of payment (fiat money);
  • overall, some market imperfections produce “too little of a good thing” while others produce “too much of a good thing”;
  • because of market imperfections, maintaining effective finance and financial stability requires a balance between private market forces and both private-collective and public sector action on behalf of members of society.

This chapter examines why fiat money and finance can and should be regarded as public goods and other public policy aspects of finance and financial stability. In order to do so, it first applies the economics of the public sector—particularly the concept of market imperfections or failures—to finance.

Sources of Market Imperfections in Finance

A market imperfection occurs when a market outcome (or market force) deviates from a standard used by economists to define the economically efficient quantity and allocation of goods and services.26 Economic efficiency means an allocation of resources that leads to a combination of production, consumption, and so on in which no individual can be made better off through a reallocation of economic resources without making some other individual worse off. Such outcomes are known as Pareto-efficient or Pareto-optimal. A fundamental theorem in welfare economics is that under well-specified conditions—including perfect competition and information—the “invisible hand” of market forces will lead to economically efficient (Pareto-efficient) outcomes.

Within this context, market imperfections occur because of the existence of five deviations from the definition of a perfectly competitive economy: externalities, public goods, incomplete information, incomplete markets, and a lack of competition.27

When a market imperfection occurs, the price established in the market will not equal the marginal social benefit of a good and it will not equal the marginal social cost of producing the good. As a result, the good in question will be consumed or produced (or both) in quantities that are economically inefficient. This can occur even when the market price equals both the marginal private cost and marginal private benefit that meets the economists’ definition of an equilibrium price. That is, even in what economists consider to be an equilibrium, there can be deviations from a socially optimal outcome. This is the true import of a market failure—a situation in which a sustainable equilibrium is economically inefficient and socially suboptimal.

Note that a market failure is defined in terms of the economic efficiency of a price and resource allocation outcome, not in terms of the stability properties of equilibrium. Thus, a market failure need not pose a threat to market stability. Also note, however, that a market failure—particularly in finance—can lead to imbalances, which in turn could create a potential threat to stability.


An externality is a secondary or an unintended consequence. In finance, externalities arise when a financial activity imposes benefits or costs on third parties or in markets that are not directly involved in the activity. If the externality provides benefits, it is a positive externality; if it increases costs, it is a negative externality.28 If the activity imposes costs on a sufficiently large number of counterparts or markets, it could potentially become systemic, and pose systemic risk. If an externality provides benefits to society at large, it may also be a public good (see next subsection).29

In general, an externality—negative or positive—drives a wedge between private costs and benefits and social costs and benefits. Thus, the market-determined price, quantity, and allocation between counterparts of a financial activity may not be socially optimal, implying that the activity would not be produced and consumed to the point that the social marginal costs and benefits exactly match the private costs and benefits. The less-than-optimal outcome is reached because the external costs or benefits of goods are not factored into individual and market demands and supplies, because individual consumers and producers do not directly bear the external costs or reap the external benefits. Looking at this from the demand side only, goods that convey positive externalities will be underdemanded and, given rising marginal costs of production, underproduced. Similarly, goods that convey negative externalities will be overdemanded and overproduced.

Some aspects of finance are simultaneously associated with the potential for both positive and negative externalities. Banks funded by short-term deposits provide liquidity to potential borrowers and superior opportunities for risk sharing. At the same time, however, in a completely unregulated environment—even in equilibrium situations—individual banks could be exposed to bank runs and panics as a result of imperfect or incomplete information, as banks were in the early part of the twentieth century. While the provision of liquidity and opportunities for risk sharing are positive externalities, bank runs and panics are negative externalities; in some circumstances, there are significant net social costs associated with bank failures and collateral damage. As noted earlier, social arrangements and mechanisms have been implemented to rule out this possibility (deposit insurance, for example) but they themselves are costly to society, thus there are both private and public trade-offs.

Box 3.2.Sources of Market Failure in Finance

Public good

  • Finance provides unit-of-account services to financial balances (ix+)
  • Finance extends universal acceptability benefits of fiat money to financial system (+)


  • Trust in finance enhances efficiency in intertemporal and interspatial allocations (+)
  • Financial system creates network benefits (+)
  • Finance subject to contagion and systemic risks (–)

Incomplete information

  • Incomplete information in finance leads to price misalignment, resource misallocation, and multiple equilibria, possibly resulting in liquidity and credit runs (–)
  • Asymmetric information between borrowers and lenders leads to adverse selection, moral hazard, and credit rationing (–)

Incomplete markets

  • Uninsurable liquidity risks (lender-of-last-resort financing) increase economic uncertainty (–)
  • Nonprice discrimination in provision of finance leads to missed exchange opportunities (–)

Imperfect competition

  • Single money issuer improves services provided by fiat money and economizes on transaction balances (+)
  • Monopoly of money supply generates seignorage revenues with incentives for overissue (–)
  • Economies of scale and too-big-to-fail considerations lead to insufficient or excessive competition between financial institutions and with new entrants (–)
Note: (+) and (–) indicate a positive or negative contribution to market efficiency.

Externalities may also arise in finance in situations in which many individual market participants take independent actions that would benefit them separately and collectively only if a small number were engaged in the activity, and would be harmful to everyone if a large number engaged in the activity simultaneously. Consider the classic bank runs during the global financial panics that occurred in 1931 when many banking systems needed to be closed for several days. Bank depositors began withdrawing cash from particular banks thought by some to be experiencing difficulties. Ultimately, a bank would run out of liquid assets and close its doors, which led to concerns about solvency. Once the word spread, third-party depositors with deposits in other (also third-party) banks started questioning whether their banks would be able to make good on deposits. As the process continued, even good banks were experiencing runs, thus, even good banks ran into solvency problems as their depositor base dwindled. In this case, individually rational decisions—to withdraw deposits from the banking system—collectively created the negative externality of driving the banking system into the ground, which imposed costs on everyone (Diamond and Dybvig, 1983).

Liquidity runs can also occur in markets. Consider a market with many traders, each exposed to most other traders in the market. Even in highly efficient and liquid markets, liquidity problems can arise when traders withdraw from trading because they receive “news” that one of the traders is having difficulties obtaining financing. In reaction, all other traders pursue their self-interest and stop trading, perhaps with all other traders. This creates a chain reaction in which liquidity is reduced in the market. To the extent that liquidity in one market affects liquidity in another, there might be contagion or systemic effects from the initial liquidity pressures in the one market.30

Network externalities can also capture some of the benefits of scale in finance. A network externality exists when a product’s value to the user increases as the number of users of the product grows. Each new user of the product derives private benefits, but also confers external benefits (network externalities) on existing users. Network externalities can cause market failures—for example, a network may not reach its optimal size because users fail to take external benefits into account. Network externalities can also be negative and in the extreme become systemic.

Public Goods

A public good (or common good) the extreme form of a positive externality. A public good has two defining characteristics: (1) the producer of the good is unable to control who benefits from consuming the good (nonexcludability in supply); and (2) consumption of the good by one person does not affect the benefits received in consuming the good by others (nonrivalry in consumption). Nonrivalry in consumption means that the marginal cost of providing the benefit to an additional consumer is zero. Nonexcludability in supply means that no one would be willing voluntarily to help supply the good or to pay for using it. Briefly stated, a pure public good conveys benefits that are both nonexcludable in supply and nonrival in consumption.31

Examples of goods that possess these properties are the provision of national defense against aggression, the maintenance of social law and order, the redistribution of resources to achieve a collectively chosen norm of social justice, and traffic monitoring and enforcement at intersections. Taking the first example, the security of national defense is a nonexcludable good: in providing national defense against aggression, the government (the supplier) cannot exclude citizens from enjoying its benefits and the marginal cost of an additional citizen enjoying the benefit is zero. In fact, it is impossible to exclude any citizen from enjoying these benefits; such a good is defined as a pure public good. In addition, the fact that one person receives the benefit does not diminish the ability of another citizen to also receive the benefit, and the marginal cost of providing the good does not increase as more individuals reap the benefits. Thus, national defense is a good that is nonrival in consumption. Although national security could, in principle, be provided by the private sector, there are incentives to be a free rider, defined as the reluctance of individuals to contribute voluntarily to the production of a public good.32

Table 3.1 presents a typology showing how private and public goods differ in the two characteristics of public goods. (See Sandler, 1992.) Note that even if a good is rival in consumption, it can still have a public-good character if the benefits it provides are nonexcludable. An example is the natural clean air we breathe—it is a pure public good, until there is rivalry in its use, such as when a company pollutes the air in a community. The company’s actions cannot alter the supply of air to the citizens in the community, but the company’s consumption of clean air (its pollution) reduces the ability of the citizens in the community to consume clean air.

Table 3.1.Typology of Benefits from Goods, Based on Characteristics of the Goods
RivalPure private goodsImpure public goods
Private externalities
Common-pool resources
Nonrival2,3Impure public goodsPure public goods
(zero marginal cost

of consumption)
Local public goods

Club goods
Source: Sandler, 1992.

According to Sandler (1992, p. 5), “Benefits of a good, available to all once the good is provided, are called nonexcludable. If the benefits of a good can be withheld costlessly by the owner or provider, then benefits are excludable.”

According to Sandler (1992, p. 6), “A good is nonrival or indivisible when a unit of the good can be consumed by one individual without detracting, in the slightest, from the consumption opportunities still available for others from that same unit.” For nonrival goods, exclusion is possible (by charging fees or club membership) but undesirable because it results in underconsumption; but without exclusion (that is, without charging for the good), there is the problem of undersupply (because there is no incentive to supply it).

Partially rival means one’s consumption of the benefit diminishes the benefits of others but does not eliminate or preclude others from receiving some benefit from consumption—for instance, a crowded park or fishing stream, both of which are nonexcludable.

Source: Sandler, 1992.

According to Sandler (1992, p. 5), “Benefits of a good, available to all once the good is provided, are called nonexcludable. If the benefits of a good can be withheld costlessly by the owner or provider, then benefits are excludable.”

According to Sandler (1992, p. 6), “A good is nonrival or indivisible when a unit of the good can be consumed by one individual without detracting, in the slightest, from the consumption opportunities still available for others from that same unit.” For nonrival goods, exclusion is possible (by charging fees or club membership) but undesirable because it results in underconsumption; but without exclusion (that is, without charging for the good), there is the problem of undersupply (because there is no incentive to supply it).

Partially rival means one’s consumption of the benefit diminishes the benefits of others but does not eliminate or preclude others from receiving some benefit from consumption—for instance, a crowded park or fishing stream, both of which are nonexcludable.

With the development of the joint field of law and economics, a new practical insight emerged: the assignment of property rights can be used to “internalize” (that is, make private) the costs of the negative externality.33 The introduction of property rights, therefore, tends to reduce the need for an outside agent (government) to create the conditions necessary for eliminating or reducing the adverse consequences of an externality. For instance, in the clean air example, if the community was granted property rights over the air in its boundaries, the citizens could collectively produce a private solution by imposing a user fee on the company, or restricting its activities to lower the level of pollution. This can occur in finance. For instance, a government could grant the authority to privately exchange financial instruments to a set of private agents provided they conformed to certain rules and regulations. Stock exchanges are an example.

A pure public good is produced in optimal quantities when the marginal cost of producing an extra unit of that good equals the marginal social benefit from the consumption of one more unit of that good. Public goods can be produced by both private and public sectors.

Public goods create market imperfections because in a completely unregulated market, public goods would be either underconsumed or underproduced, or even not produced at all. Underproduction occurs because the good would be produced only up to the point that the private marginal cost to the producer would exactly match the private marginal benefit to the producer (which is the portion of the social benefit that the producer is able to internalize by producing). Thus, it is difficult to provide incentives for private individuals to produce public goods in sufficient quantities. Goods that are nonrival in consumption but excludable in supply will be underconsumed, but when they are also nonexcludable they will be underproduced. It is worth noting again that the underproduction or underconsumption of a good is defined by the economic efficiency of an equilibrium and not by its stability properties.

As will be discussed in further detail in the next two sections, both fiat money and finance can and should be seen as having significant positive externalities and can also be seen as public goods. This suggests that both private-collective and public sector actions could enhance the private and social benefits of finance beyond what market forces alone would attain.

Incomplete Information

When counterparts in financial transactions (in formal and informal markets, and in bilateral or multilateral over-the-counter exchanges) are not well informed, free-market outcomes will tend to allocate resources inefficiently. For example, because of imperfect information about a company within a particular industry, a local bank may underestimate or overestimate the risks associated with lending to firms in that industry. Moreover, the industry might be particularly sensitive to macroeconomic conditions in a neighboring town or state, which might be about to experience a boom or a bust. As a result of not having perfect, or even sufficient, information on firms and the industry, the bank might tend to take on a suboptimal amount of risk, either too much or too little. The outcome might be, therefore, that the industry would receive more or less capital than it could reasonably efficiently utilize, and the bank and its depositors would end up owning more or less credit risk than is optimal for the bank and its financial stakeholders.

When information is incomplete, adverse selection and moral hazard can lead to situations in which economically desirable goods are driven out of the market by economically undesirable goods. Consider the classic case of the “market for lemons”: the case of the sale of a used car (or a loan originated years ago by a bank) in which the owner of the used car (loan) knows almost everything about the performance of the car (loan) and the potential buyer knows next to nothing (Akerlof, 1970). Because there is a risk of buying a dysfunctional automobile in this secondary market—that is, of buying a lemon—buyers tend to price the risk of a car being a lemon into their offer prices, which tends to lower all prices, even for good cars. The result is that fewer suppliers of cars are willing to sell, especially suppliers of good cars. This also means that more lemons are sold than is beneficial, which further reduces welfare.

Adverse selection in the credit derivatives market, which is a market for buying and selling insurance protection against the risk of a default on a loan, illustrates the same effect.34 Adverse selection results from the fact that lenders who issued credits to borrowers with a higher risk of default are more likely to want to be insured than those who issued credit to borrowers with lower risks. This skewed demand encourages the insurer to raise premiums above the socially efficient price, which reduces the overall amount of insurance provided to a point below the social optimum. A situation can arise in which low-risk insurees are priced out of the market, so there would be a large number of high-risk insurees in the insured pool. In the extreme case, only owners of high-risk credits will be able to obtain insurance, which is highly inefficient. A similar situation can occur in credit markets in which riskier borrowers crowd out less risky borrowers.

One solution to this kind of adverse selection is to monitor. For example, insurance providers—both private and public—have a strong incentive to require inspections. Another solution is to offer co-insurance rather than full insurance. Still another is to insure large groups to capture the diversity of risks. In the extreme, this would mean insuring the entire population of depositors, which would appear to be optimal if it could be properly priced and monitored; but then there is the moral hazard.35

Incomplete Markets

Market forces often fail to provide a good for which the private cost of production is less than what private individuals are willing to pay. This market imperfection is referred to as incomplete markets. Economists have suggested several reasons for incomplete markets, including high transactions costs in running markets, enforcing contracts, and introducing new products; asymmetries in information concerning financial risks; enforcement costs on defaulted contracts; and adverse selection. Thus, incomplete markets often are the result of the existence of other market imperfections, such as incomplete information or insufficient competition.

Finance is thought to be an area where this market imperfection is prevalent, particularly with regard to loans, insurance contracts, and capital-market instruments. An example in lending and capital markets is the market for college student loans. In principle, lending to students is no more risky than lending to small and medium-size businesses. It can even be argued that lending to college students, whose incomes are likely to be higher than average, would be less risky. Until government guarantee programs were put in place, no market developed to offer loans to college students who were willing to pay even market rates for them. Now many major banks have student loan portfolios. A similar analysis can be made for U.S. home-mortgage loans: once quasi-government agencies stepped in to underwrite them, the market for home mortgages expanded at a more rapid pace. Another example is deposit insurance. Even though banks would have been willing to pay for insurance to keep a steady stream of deposits flowing, no such market was created until the government stepped in during the Great Depression to provide deposit insurance.

Fiat Money As a Public Good?

In private exchanges, the use of specific units of currency is rival in consumption and thus conveys private benefits. It is rival in consumption because in bilateral or multilateral trades and exchanges, only the parties to the exchanges benefit from the value obtained in them. To the extent that the supply of fiat money (or another form of legal tender) is fixed at any time, it is also excludable in supply, even though the issuer does not specifically restrict its use to one set of individuals.

More generally, the more that people use fiat money in their exchanges, the more efficient multilateral exchange becomes; likewise, the more efficient other economic processes that rely on exchange also become. Once the use of fiat money extends beyond a certain critical social threshold, the widespread use of money as legal tender provides positive externalities to others not necessarily involved in every exchange using specific units of currency. As the universality of use of fiat money grows, there is a greater pool of potential transactors and liquidity in using fiat money. In terms of the characteristics of rivalry and excludability, as universality of use expands, the pool of transactors and liquidity surrounding the fixed supply of fiat money becomes nonexcludable. In the literature on public goods, this is known as a common-pool resource. It is similar to a lake or stream in which there is nonexcludable access to fish but in which the benefit of consumption of the good (catching fish) is rival because once a fish is caught it can no longer be consumed by someone else—unless there is a catch-and-release policy at the lake or stream.36 This common-pool resource is a nonpure public good, as defined in the upper right corner of Table 3.1 (rival but nonexcludable goods). Note that fiat money itself is not the public good, in part because the marginal cost of providing it to a growing population is not zero (in the way the marginal cost of another person benefiting from national defense would be)—the public good is the efficiency-enhancing pool of transactors and liquidity that develops because of the existence of fiat money and its characteristics and services.

However, once fiat money reaches the point of universal acceptability in providing finality-of-payment services, the positive externality extends to all members of society. In this way, the services provided by fiat money are public goods. In fact, the universal acceptability of fiat money as society’s means of payment satisfies the two defining requirements of a pure public good: nonexcludability in supply and nonrivalry in consumption. The first property is satisfied because in a society in which there is a universally accepted means of payment, the fiat issuer cannot exclude anyone from benefiting from one of its most important services, that of universal acceptability. The second property is satisfied because the fact that one agent receives the benefit of having access to a universally accepted means of payment does not reduce the ability of others to enjoy the benefits of universal acceptability.37 Moreover, the universal use of a common means of payment facilitates more efficient multilateral trade and exchange among members of an economy.38

The idea that fiat money is a public good is ancient, and has been exploited by sovereigns throughout human history. According to Kindleberger (1993, p. 22),

The public good character of money was early recognized by rulers who laid down standards for mints within their jurisdictions and tried to see that they were maintained. The Holy Roman Empire, for example, decreed ordinances for regulating the number of mints within its constituent elements and the weight and fineness of the coins struck, and it sent imperial assayers on visits to see that its standards were adhered to. The lesser governments had their own interests in raising funds—for consumption, for building palaces, and, when war broke out, for hiring mercenaries—and these often clashed with the overall public interest. The principalities, duchies, bishoprics, imperial cities, and the like were tempted from time to time to debase the currency issued by their mints, to earn income by minting more coins for a given amount of metal, and then to encourage the taking of debased coins across the border and exchanging them for good coins. If a coin were not too badly worn, sweated (i.e., rubbed), clipped or adulterated, it could pass at its nominal value, especially if it were a subsidiary coin used in retail trade and the payment of wages. If, on the other hand, it was badly deteriorated or was a large silver or a gold coin, it had to be weighed and tested before the recipient would be willing to accept it. The 12 million escudos paid by Francis I of France in 1529 to ransom his two sons, who had been substituted for him as hostages when he was captured in the war between France and Spain, took four months to count and test, and 40,000 coins were rejected by the Spanish as below standard.

Finance and Financial Stability as Public Goods?

As the universally accepted means of payment, fiat money is a public good because it conveys external benefits that are nonrival and nonexcludable, for example, in the way that it facilitates multilateral trade and exchange. Without it, Jevons’ “double coincidence of wants” would constrain the efficiency and amount of trade and exchange, and economic activity more generally. Fiat money accomplishes this because it eliminates the need for human trust, and is the economy’s surrogate for trust, in trade and exchange.

Finance is an exchange of one service for another with a promise to reverse the exchange; it involves a temporary exchange of fiat money for a promissory note. Within a particular exchange, finance provides rival and excludable benefits to the counterparts of the bilateral or multilateral exchange. That is, individual financial transactions occur among private individuals and provide private benefits.

As a critical mass of financial activity is reached, finance both provides positive externalities and conveys public goods. The positive externalities are associated with the efficiency gains that finance provides over fiat money in facilitating the more efficient intertemporal allocation of resources (such as borrowing against future earnings) and in providing greater and more effective opportunities for all members of society to store value and accumulate wealth. In effect, finance enhances, or leverages, the public good function of fiat money; it amplifies the universally accepted finality-of-payment services of fiat money, both spatially and intertemporally. If the services of fiat money are a public good, some of the services of finance are also a public good.

Moreover, when the level and effectiveness of finance as a process or system reaches a critical mass, the process or system itself provides the opportunity for all to have access to effective and superior stores of value—the process that provides these services becomes a public good.39 It does so because it begins providing benefits to society that reach well beyond the aggregation of the benefits of private individual transactions, and because such transactions also become both nonrival in consumption and nonexcludable in supply. Access to the efficiency benefits of a well-functioning financial system is nonrival because one person’s access to (and consumption of) the benefits of the process does not diminish another’s access to the benefits of the process. Indeed, it can be argued that the more everyone accesses these benefits, the greater is the benefit to all. Access to the efficiency benefits is nonexcludable because no one can be excluded from reaping the broader social benefits of finance’s contributions to the efficiency of economic processes. Finance provides other positive externalities as well, because it enhances the services of fiat money by facilitating a greater number of transactions than fiat money alone could support, and does so intertemporally.

Just as the introduction of fiat money allows for the timing of receipt of income to be separated from the timing of of expenditures, finance—borrowing and lending—allows individuals to shift purchasing power forward in time. This intertemporal separation allows societies to achieve a more efficient allocation of resources, greater production from available resources, and, in the end, greater consumption. These services are public goods and so, too, is finance.

Similar externality and public good arguments can be made for preserving financial stability. Everyone would like to see financial stability preserved because there are both private and public costs associated with bank failures, market dysfunctions, and systemic financial problems. However, no one individual or small group of individuals can do much to prevent problems from arising beyond engaging in prudent portfolio and risk management. Moreover, because the private cost of doing something about systemic risk is too high, and the rewards too low, on balance everyone has the incentive to let someone else worry about it. The provision and maintenance of financial stability would provide benefits to all individuals, and the fact that one person incurs these benefits does not prevent others from doing so. Thus, the principles of nonexcludability and nonrivalry would apply to financial stability just as it does to other public goods such as national defense.

Market Imperfections in Practice: Some Produce Too Little of a Good Thing and Others Too Much of a Bad Thing

Some market imperfections in finance are positive in the sense that they do not produce private or social “bads.” Instead, the market failure is that private incentives and market forces alone lead to the underproduction or underconsumption of financial activities with desirable characteristics and potential private and social benefits. Sources of market imperfections that lead to this kind of outcome are positive externalities, public goods, some forms of incomplete information, incomplete markets, and a lack of competition. A specific example is the lack of competition in supplying student loans, which would tend to lead to an overpricing of the risk of lending to students and therefore to an undersupply of loans to this class of borrower. The challenge in these cases is to provide incentives to supply and consume more of the goods that provide positive externalities and public benefits, that tend to open up new markets, and that increase competition unless there are natural monopolies. (See Box 3.2.)

The introduction of fiat money in the mid-seventeenth century (but which did not come into its own until the early part of the twentieth century with the disappearance of commodity monies and standards) may have been the first “socially collective” push to increase the beneficial aspects of financial activity. The advent of banking and other financial arrangements also helped to increase the potential benefits of finance to individuals and society at large. As financial systems continue to evolve, new arrangements are likely to be created, and existing arrangements improved. One aim of financial-system policies is to ensure that this process of innovation and evolution continues to increase overall economic efficiency and therefore the public as well as private benefits of finance.

However, there is another side to finance. Some market imperfections in finance can be seen as negative in the sense that they lead to outcomes in which private and social “bads” are produced. The market failure in these cases is that private incentives and market forces alone lead to the overproduction or overconsumption of financial activities that have undesirable characteristics and potential social costs. Sources of these market imperfections include negative externalities, some kinds of information failures, and excessive competition. Here the challenge is to provide incentives to minimize the production and consumption of financial activities that result from these imperfections. A specific example is excess competition in particular segments of the market for loans—for example, credit card loans—which can lead to underpricing of the risks (and therefore oversupply of this particular class of loans).

Both sides of market failures in finance can be illustrated in many ways. The most direct is to consider the characteristics of bank money, the closest finance substitute for fiat money (or other forms of legal tender). To the extent that bank money (bank notes or demand deposits that promise to pay fiat money on demand) provides superior store-of-value services to society at large and also enhances (by literally leveraging) the real economic benefits of the universally accepted means-of-payment services of fiat money, it both conveys positive externalities and is a public good. However, if completely left to market forces, because of the inherent uncertainty about trust, the future, and information, the demand for bank money would be less than what is considered socially optimal. Individual private demands for bank deposits would focus exclusively on the private benefits and risks, and not incorporate the external benefits of leveraging the finality-of-payment services of money as legal tender.

A potential negative externality can also be associated with bank deposits, however, similar to the negative externalities of traffic jams. Individuals lend their fiat money to banks in return for a promise to get it back on demand. Meanwhile, banks invest these funds in risky assets, which the banks understand better than the depositors. Thus, these bank notes are risky. If banking was left entirely to market forces, when banks become suspected of mismanaging their assets, individuals have the incentive to withdraw their funds. If a large number of depositors withdrew their funds simultaneously, the bank would quickly run out of liquid funds, become insolvent, and most individuals would lose their deposits. That is, when left to market forces alone, bank deposits can be subject to runs from time to time, as uncertainty about trust ebbs and flows. Knowledge of this possibility would tend to reduce the amount of deposits in the financial system below the economically and socially desirable amount. The introduction of deposit insurance—either privately or publicly funded—would eliminate these sources of market failure and help moderate demand around the socially optimal level, provided the deposit insurance scheme was appropriately priced and monitored for abuse and moral hazard.

Policy Implications and Conclusions

Modern finance is a dynamic network of a large number of individual private financial contracts seeking exclusive private gains. Accordingly, the net social benefits of finance are the aggregate of (albeit difficult to measure) individual private net benefits.

However, it would be an illusion to consider the effectiveness of private finance and its enormous real economic benefits either entirely as private or exclusively as the result of individual private actions and unrestrained market forces. Regardless of opinions about the necessity and efficacy of public policies, obtaining the full extent of the private net benefits of modern finance requires, at a minimum, the existence and effectiveness of many private-collective, publicly sanctioned, publicly mandated, and taxpayer-financed social conventions and arrangements.

Many of these important social arrangements are taken for granted. For example, modern private financial contracts are predominantly written in terms of the social convention of a legally sanctioned unit of account. This unit-of-account service has the properties of a pure public good, and is part of every financial transaction. All receive benefits and none pay a private cost, except as taxpayers. The service could be provided and financed privately, but most attempts to do so throughout history have neither succeeded nor endured. A second example is that settlement and delivery of payments for financial transactions typically require a universally accepted legal tender (a fiat money), which, as argued earlier, also has the characteristics of a public good. Third, transactors presume the availability of legal recourse in the absence of financial contract performance, which relies on the effectiveness of a publicly financed and enforced legal system. Other aspects of social-collective action underlie the effectiveness and efficiency of private finance, including well-run and well-supervised financial institutions, effective micro- and macro-financial system policies, and effective micro- and macro-economic policies.40

Although finance would no doubt exist and bestow benefits without these particular social arrangements, there would most likely be significantly less of it, and it would be significantly less efficient and supportive of economic activity, wealth accumulation, growth, and ultimately social prosperity. In short, the enormous and pervasive private benefits of modern finance and financial systems are possible because the existence of an effective financial system has long been understood and supported as a public good. (This is not meant to imply that all public policy involvement in private finance is appropriate, beneficial, or acceptable.)

Box 3.3.Samuelson’s Store of Value As a “Social Contrivance” Providing a Public Good

Although the public-good nature of fiat money has been well understood—both historically and conceptually—and widely accepted, it was not until the beginning of the post-Second World War period that a rigorous economic-theoretic demonstration proved that “money” provided public goods. “Providing public goods” means that the introduction of money allowed a stylized economy to move to a more beneficial outcome for society as a whole. It was not the introduction of fiat money as a means of payment that provided the lubrication to move the economy to the best outcome, but instead money as a store of value.

Samuelson’s Model

The seminal paper that considered this issue was Paul Samuelson’s, “An Exact Consumption-Loan Model of Interest With or Without the Social Contrivance of Money.”1 In that paper, Samuelson considers a model of a highly stylized economy that produces and consumes only a single perishable good, that exists indefinitely into the future, and that has three overlapping generations (OLG) of finite-lived economic actors: the working young, the middle-aged worker facing retirement, and the retired old. In this model, all actors must consume the single perishable good available in the economy to survive. Because retired workers no longer receive income, they have to find some other way—a contract with the young and middle-aged—to obtain the consumption good during their retirement; otherwise they would die after they became unable to produce.

Individuals could, in principle, enter into private contracts to secure promises to receive goods in retirement. However, the various equilibria considered in the model turn out to be socially suboptimal. The equilibria are suboptimal because to avoid starvation in retirement, consumption loans necessary for survival would lose a significant share of their value in one period. That is, in the consumption-loan model, the equilibrium interest rate would either be negative if population growth were zero, or fall well short of society’s biological growth rate, which would be undesirable from a social point of view. As Samuelson emphasized,

It [the model] points up a fundamental and intrinsic deficiency in a free pricing system, namely, that free pricing gets you on the Pareto-efficiency frontier but by itself has no tendency to get you to positions on the frontier that are ethically optimal in terms of a social welfare function; only by social collusion—of tax, expenditure, fiat, or other type—can an ethical observer hope to end up where he wants to be (Samuelson, 1958, p. 479).

To try to overcome the suboptimality, Samuelson introduced the “social contrivance of money,” as a “social compact” to achieve the socially optimal Pareto-efficient equilibrium. In Samuelson’s words,

The present model enables us to see one ‘function’ of money from a new slant—as a social compact that can provide optimal old age social security.... If each man insists on a quid pro quo, we apparently continue until the end of time, with each worse off than in the social optimum.... Yet how easy it is by a simple change in the rules of the game to get to the optimum. Let mankind enter into a Hobbes-Rousseau social contract in which the young are assured of their retirement subsistence if they will today support the aged, such support to be gua\ranteed by a draft on the yet-unborn. Then the social optimum can be achieved within one lifetime (p. 479).

The social compact introduced comprises valueless pieces of paper and the understanding—and more important, the trust—that these pieces of paper could be used to buy goods now and forever in the future. Samuelson’s money constitutes bot]h a universally acceptable means of payment and, more important, a store of value of no intrinsic value other than its exchangeability for goods in the future. Young and middle-aged workers accept this paper in return for goods they produce because the social contract would hold for them in retirement as well. All of this occurs by social decree, trust, and acceptance by all present and future generations.

Samuelson proved that the introduction of this social compact—this store of value—moved the economy to the socially optimal Pareto-efficient equilibrium. In Samuelson’s own words,

Once social coercion or contracting is admitted into the picture, the present problem disappears. The reluctance of the young to give to the old what the old can never themselves directly or indirectly repay is overcome. Yet the young never suffer, since their successors come under the same requirement. Everybody ends better off. It is as simple as that (p. 480).

In effect, the introduction of Samuelson’s money as a “social contrivance” moved the OLG barter economy to a more Pareto-efficient resource allocation.

Samuelson’s “Social Contrivance” Is Not Just Money: It Is Finance

Within the model, Samuelson’s money is not just a means of payment. It represents a store of value that facilitates intertemporal transfers of purchasing power and thereby creates both a private and a social-welfare improvement.

This is finance as it has been discussed in the chapter.2 The social contract works in the model because it removes the element of human trust between generations, which no doubt accounts for the decline in loan value after the first period. In effect, within the OLG model without the social contract, there is a negative externality that prevents the emergence of the socially optimal outcome. The negative externality is that the economy lacks an important market—a financial market. The retired citizens have no way of establishing a contract with the young that would make the young feel comfortable that they themselves would be repaid. While the model did not explicitly model the frailty of human trust, this is the negative externality. Introducing the “social contrivance of money” as the social compact allows the young and old to enter into private contracts using a vehicle that permits them to internalize in private transactions the negative externality.

1See Samuelson (1958) and Allais (1947).2Cass and Yaari (1966) demonstrate that when durable goods are introduced into Samuelson’s framework both government and private finance in the form of promissory notes (debt) can, under different conditions, produce the social optimum.

Thus, more realistically, while modern finance is primarily a private affair, the private net benefits originate in two inseparable sources. First are the direct individual private net benefits derived from private financial transactions. Second are the indirect private and collective net benefits associated with having access to an effective process of finance and sharing in the enormous collective efficiency gains finance creates for the economic system as a whole. All citizens do not benefit equally or have equal access to these private and social benefits, but the social benefits are there for the taking, especially in the more democratic societies with liberalized economies.

Despite the current consensus that many of the social conventions and arrangements developed over time are essential prerequisites for effective finance, this reflects a relatively new and modern understanding of the role of finance. Less than 75 years ago, society’s mismanagement of both money and finance played an important and devastating role in the Great Depression.41 More recently, some of these lessons had to be learned again, in both mature markets and in an increasing number of emerging- and developing-market countries. For example, in the aftermath of recent corporate scandals in some mature markets, improvements are being advocated and implemented for accounting standards and their enforcement and the efficacy of existing corporate governance procedures and accountability. Likewise, as the result of recent financial crises in emerging-market countries, many of these social arrangements are being aggressively advocated for adoption in less advanced as well as least developed economies and financial systems.

Viewed from the perspective of public-sector economics (and as discussed earlier in the chapter), if finance did not produce positive externalities, public goods, and other sources of market imperfections, the best public policy approach would be to leave finance completely to individual actions and market forces. As observed, however, finance inherently embodies uncertainty about trust and several other market imperfections:

  • Some financial services provide positive externalities, and are therefore underproduced or underconsumed, while others provide negative externalities and are therefore overproduced or overconsumed.
  • Some financial services are public goods and are therefore underproduced or underconsumed.
  • There are information failures, which can work both ways.
  • Financial markets are incomplete, which leads to underproduction of financial products and services.
  • Competition is not always perfectly balanced, and when there is not enough of it goods are underproduced; when there is too much of it goods are overproduced.

In practical terms, market imperfections in finance may lead to the underconsumption and underproduction of some socially desirable financial activities, and the overconsumption and overproduction of some socially undesirable ones. This is a compelling practical motivation for private-collective action and public policy.

The overall objective for collective action would be to move the economic and financial systems toward a more economically efficient and socially optimal level, mix, and allocation of finance, measured in part by the ability of finance to facilitate real economic processes. In this way, collective action in finance could enhance both private and social welfare.

It would seem reasonable to think that private-collective and public actions could be designed and implemented to address each source of market imperfection in finance, depending on how significant the efficiency losses associated with each one might be. In considering this, the effectiveness of policies could be improved if they were designed and implemented in a cohesive fashion so that a policy designed to eliminate the negative impact of one kind of market imperfection does not offset the benefits of a policy designed to deal with another. It is not clear to what extent such policy cohesiveness and coordination is actually achieved in practice by countries or across borders.

When a market imperfection inhibits consumption and production of desirable goods, the challenge is to provide incentives to supply and consume more of the goods that provide public benefits and positive externalities, that tend to open up new markets, and that increase competition unless there are natural monopolies. When a market imperfection encourages consumption and production of undesirable goods, the challenge is to minimize the production and consumption of financial activities that result from these market failures. Because both tendencies exist simultaneously, financial-system policies would be more effective if they strove to strike a socially optimal balance between maximizing the net social benefits of the positive externalities and public goods and minimizing the net social costs of the other market imperfections in finance. As noted already, this will most likely encompass a combination of both private-collective and public policy involvement, which should also be striving to achieve some kind of social optimum. Striving to achieve the social optimum will undoubtedly entail difficult choices, including trading off some of the individual private benefits for the greater good, if and when this can be justified.

As noted earlier, not each and every loss of efficiency requires intervention. It is much clearer that when a market imperfection in finance leads to an inefficiency that is potentially destabilizing, either for financial institutions or for markets or for both, that some form of collective action might be desirable or necessary. Unfortunately, in the financial-system policy literature, often no clear distinction is made between sources of market imperfections that tend to lead to instability and those that do not. Likewise, no framework now exists either for measuring the efficiency losses associated with market imperfections in finance or for assessing the risks to financial stability associated with market imperfections. These are some of the challenges in the period ahead for which an analytical framework for financial stability would be useful for policy purposes.


This way of characterizing the issue echoes the seminal work of Olson (1965) and Sandler (1992).


The Prisoner’s Dilemma illustrates how the pursuit of self-interest can lead to an aggregate outcome that is inferior to other feasible outcomes; see Box 3.1.


Some of this discussion is adapted from Sandler (1992) and Stiglitz (2000). In this chapter, the terms “market imperfections” and “market failures” are interchangeable.


A classic example of a negative externality is the production of pollution as a by-product in producing private goods.


In this sense, externalities are a form of impure public goods; alternatively, public goods can be seen as an extreme form of externalities.


Diamond and Rajan (2002) examine the conditions under which a bank run can, through contagion, create aggregate liquidity shortages.


According to Cornes and Sandler (1996), nonexcludability is the crucial factor in determining which goods must be publicly provided.


The free-rider problem arises because a public good can be consumed without paying for it. This originates in the nonexcludability-in-supply characteristic of public goods.


Ronald Coase received the Nobel Prize in economics for this insight and his related seminal work. See Coase (1960).


Adverse selection occurs when two parties in a negotiation have different amounts of information—asymmetric information—and the outcome restricts the quality of the good traded. This typically occurs because the party with more information is able to negotiate a favorable exchange.


Moral hazard is the risk that one party to a contract can change its behavior to the detriment of the other party once the contract has been concluded. It occurs, for example, when an insured agent takes less than the efficient amount of precaution against the insured event. For example, a bank depositor who is fully insured takes no precaution against insolvency of the bank, such as monitoring the bank’s ability to assess and manage credit risk.


See Maier-Rigaud and Apesteguia (2004). They show differences in rivalry for nonexcludable goods lead to differences in aggregate investment behavior in a neighborhood of the Nash equilibrium. More specifically, the authors demonstrate that public goods and common pools are distinct rather than identical Nash-equilibrium games: the public-goods game leads to private investment above the Nash-equilibrium level of investments while the common-pool game leads to private investment above the Nash equilibrium. However, over time, both games converge to the same Nash equilibrium. Thus, “aggregate behavior in both games is surprisingly similar in the sense that it starts in the neighborhood of the Pareto optimum and moves to the respective aggregate Nash equilibrium” (Maier-Rigaud and Apesteguia, 2004, p. 12).


Tobin (1980, p. 83; 1992) writes, “Social institutions like fiat money are public goods.”


Note that these same arguments apply to the unit-of-account service of fiat money.


For another reason to see finance as a public good, see Box 3.3.


Levine (1999) finds that the legal and regulatory environment of financial intermediaries is positively associated with economic growth. More specifically, Leahy and others (2001) show that the transparency and enforcement of these legal and regulatory frameworks, in particular in terms of investor protection, accounting, and auditing requirements, is broadly linked to innovation and investment in new enterprises. Beck, Demirgüc-Kunt, and Levine (2003) establish that countries with better developed national institutions and policies governing issues such as property rights, the rule of law, and competition are less likely to suffer systemic banking crises.


Bordo (2000) discusses some of the connections between, and historical experience of, unsound money and finance.

    Other Resources Citing This Publication