Government Regulation of Business

By Thomas, C.R., Maurice, S.C.

Edited by Paul Ducham


As you learned in your first course in economics, every society must decide how best to use its scarce labor, capital, natural and Environment Resources for production of goods and services. To get the most from society’s scarce resources, production and consumption must be organized in ways that avoid inefficient use of resources and inefficient consumption of goods and services. In this chapter, we examine efficiency from the larger perspective of society as a whole, rather than a single firm. Social economic efficiency exists when the goods and services that society desires are produced and consumed with no waste from inefficiency in either production or consumption. To reach this goal, two efficiency conditions must be fulfilled: productive efficiency and allocative efficiency. We will examine both of these conditions shortly.

Under ideal circumstances, a perfectly competitive market reaches an equilibrium that is both productively efficient and allocatively efficient. Unfortunately, market conditions are not always ideal (or even approximately so), and perfectly competitive markets then fail to bring about social economic efficiency. In this chapter, we will discuss and analyze the six reasons why competitive markets can fail to perform in a socially efficient fashion. You will see that imperfectly competitive market structures—monopoly, MONOPOLISTIC COMPETITION, and oligopoly—are never expected to achieve social economic efficiency because these market structures always lead to allocative inefficiency and, in some instances, may also fail to accomplish efficient production. The prospect that social economic efficiency can be reached through perfect competition is so compelling it serves as the foundation for antitrust policy in the United States, as well as competition policy in Canada, the European Union, and elsewhere around the world.

Efficiency Conditions for Society

As stated above, two efficiency conditions must be met to avoid waste and thus ensure that society enjoys the greatest gain possible from its limited resources. Markets must operate with productive efficiency so that society gets the most output from its resources. Productive efficiency exists when suppliers produce goods and services at the lowest possible total cost to society. Should markets fail to achieve productive efficiency for any reason, then resources will be wasted, diminishing the amounts of goods and services that can be produced in every industry. Managers produce at the lowest possible total cost by choosing the combination of inputs on the firm’s expansion path. Thus productive efficiency happens whenever managers operate along their firms’ expansion paths in both the short-run and long-run periods.

Allocative efficiency, the second condition needed for social economic efficiency, requires businesses to supply the optimal amounts of all goods and services demanded by society, and these units must be rationed to individuals who place the highest value on consuming them. Because productive resources are scarce, the resources must be allocated to various Industries in just the right amounts, otherwise too much or too little output gets produced. The optimal level of output is reached when the marginal benefit of another unit to consumers just equals the marginal cost to society of producing another unit. The price on the market demand gives the marginal benefit buyers place on consuming that extra unit of the good. Thus allocative efficiency requires production up to the point where the maximum price consumers are willing to pay for the last unit produced just equals its marginal cost of production. Since this is the point on demand where P = MC, economists frequently refer to this condition for efficiency as marginal-cost-pricing.

Let’s consider an example now. Suppose a market currently operates at the point on market demand where price is $100. At this output level, the value or marginal benefit of the last unit consumed is $100. At this output, suppose suppliers only need $60 worth of resources to make this last unit. Using the logic of marginal analysis, you can see that the current output level is too little, because an additional unit of output would add more total value (MB = $100) than it adds to total cost (MC = $60), thereby generating a net increase of $40 in social well-being. Now suppose that the market is operating at an output level for which consumers value the last unit at $35 and producers use $55 worth of scarce resources to make it. Now, too much output is produced, because it is inefficient to use $55 of resources to produce a good worth only $35 to consumers. Thus the optimal level of output is the unit for which demand price equals marginal cost.

Once society’s scarce productive resources are allocated efficiently among competing industries, the resulting output must then be rationed or distributed to the individuals in society who get the most value from consuming them. This is exactly what happens when buyers engage in voluntary market exchange. At the current market price, consumers whose demand prices (i.e., marginal valuations) equal or exceed market price will choose to buy the good. Consumers whose valuations are less than the current price will not buy any of the good, leaving more for those who value consumption of the good more highly. This process by which prices serve to ration goods to their highest-valued users through voluntary exchange is generally referred to as the rationing function of prices. We now summarize the conditions required for social economic efficiency in a principle.

Principle Social economic efficiency occurs when two efficiency conditions are met: (1) industry output is produced at the lowest possible total cost to society (productive efficiency), and (2) every industry produces the socially optimal amount of a good or service and these units are rationed or distributed to the individuals in society who value them most (allocative efficiency).

Social Economic Efficiency under Perfect Competition

Now we will show you that markets in perfectly competitive equilibrium achieve both productive and allocative efficiency.

Figure 16.1 shows demand and supply curves for a perfectly competitive industry. Supply curve S represents either short-run or long-run industry supply. In competitive equilibrium, 800 units are bought and sold at the market-clearing price of $60 per unit (point A). We will now explain why the competitive equilibrium at point A is both productively efficient and allocatively efficient.

Productive efficiency under perfect competition

As we explained, productive efficiency occurs when firms operate on their expansion paths, because input combinations on expansion paths—both short-run and long-run expansion paths—minimize the total cost of producing any particular level of output. Probably the most compelling reason to believe managers will actually operate on their expansion paths is rather simple, yet powerful: Economic profit cannot be maximized unless total cost is minimized for the profit-maximizing output level. Managers wishing to maximize profit and the value of their firms must operate on their expansion paths. This reasoning applies in both short- and long-run periods of production. In contrast to the short run, however, firms that fail to produce efficiently in the long run must exit the industry. These inefficient firms suffer losses at the long-run competitive equilibrium price—minimum longrun average cost—and will be forced either to produce efficiently or exit. As a further benefit of productive efficiency, consumers pay the lowest price possible for the good.

At point A in Figure 16.1, each firm in the industry produces its portion of total industry output by using the combination of inputs that minimizes its total cost of production. If S is a short-run industry supply curve, then the firms in the industry may earn positive, negative, or zero profit at the market price, $60. Alternatively, if S represents a long-run industry supply curve, then all firms earn zero economic profit and produce at their minimum average costs, $60. In either case, 800 units cannot be supplied at a lower total cost since all suppliers are producing on their expansion paths.

Allocative efficiency under perfect competition

Recall that demand prices along a market demand curve are exactly equal to the marginal benefits buyers receive by consuming another unit of a good. For this reason, market demand, D, in Figure 16.1 is also labeled “MB.” You learned that supply prices along both short-run and long-run industry supply curves equal the industry’s marginal cost of producing additional units of output. Thus industry supply S is also labeled “MC.”

Let’s suppose the industry produces 400 units of output. At this level of output, buyers place a value of $100 on getting an additional unit to consume, while sellers require $40 worth of extra variable inputs to produce an additional unit. Since MB (= $100) exceeds MC (= $40), the 400th unit should be produced since it adds $60 to net benefit. This reasoning applies to every unit up to 800 units. Any output beyond 800 units, however, is too much output since MC exceeds MB beyond 800 units. For example, at 1,000 units, variable inputs worth $70 are used to produce the 1,000th unit of output, but the extra good is only worth $40 to consumers. Clearly, production and consumption beyond 800 units is inefficient.

Allocative efficiency in a perfectly competitive industry must occur at the equilibrium output level determined by the intersection of demand and supply. At the point of competitive equilibrium, demand price (MB) equals supply price (MC). In Figure 16.1, you can verify that the marginal benefit of the 800th unit is $60, which equals the marginal cost of the 800th unit. Competitive equilibrium always establishes the optimal level of output for society, as long as the demand curve correctly and fully measures marginal benefits of consumption and the supply curve correctly and fully measures marginal costs of production. As we will show you in this chapter, demand and supply curves may not always correctly measure marginal benefits and costs. When this happens, competitive markets may produce either too much or too little output.

Finally, notice that the equilibrium price, $60, successfully rations 800 units of output to the consumers who value these goods most. Consumers with relatively low values, represented by the segment of demand below point A, do not voluntarily pay $60 to purchase output beyond the 800th unit because these units are worth less to them than the market price. These potential buyers voluntarily choose not to consume the good, leaving the 800 units for those buyers who place a value on the good of at least $60. Thus perfectly competitive markets achieve allocative efficiency because the optimal amount of the good is produced, and this amount is rationed or allocated to the highest-valued users.

Social economic efficiency and maximization of social surplus

We now wish to demonstrate that Social Surplus is maximized in competitive equilibrium. Consider again the 400th unit of output in Figure 16.1. The marginal value of this unit is $100. The buyer who purchases the 400th unit must pay the market price of $60 to get it, and so enjoys Consumer Surplus on this unit equal to $40 (= $100 - $60). The producer who supplies the 400th unit is willing to do so for a price as low as $40. The supplier of the 400th unit gains Producer Surplus equal to $20 (= $60 - $40). (Recall that the $20 producer surplus is economic rent if S in the figure represents long-run industry supply.) Thus, the consumption and production of the 400th unit generates $60 of net gain for society. By this reasoning, every unit for which demand price exceeds supply price contributes positively to society’s total surplus. As you can see from the figure, production and consumption of all units between 0 and 800 must be undertaken to maximize social surplus.

For consumption and production levels greater than competitive equilibrium— beyond point A in the figure—social surplus will be smaller than at point A. Consider the 1,000th unit, for which demand price, $40, is less than supply price, $70. Production and consumption of this unit are inefficient because $70 worth of scarce resources is transformed through production into a good worth only $40. Obviously, this would be wasteful for society. Fortunately, there is no price for this good that would stimulate buyers and sellers voluntarily to make such a wasteful transaction. Therefore, competitive market forces lead to the exact level of consumption and production that maximizes social surplus, and hence maximizes the value of this free market to society.

We must emphasize that, while competition maximizes social surplus, this does not imply that either consumer or producer surplus is maximized individually— it is the sum of the two that is maximized. Moving away from competitive equilibrium (point A) can result in one kind of surplus rising while the other surplus falls, but it must always reduce total social surplus. We can now summarize the Results of this section in a principle.

Principle Markets in perfectly competitive equilibrium achieve social economic efficiency because, at the intersection of demand and supply curves, conditions for both productive efficiency and allocative efficiency are met. At the competitive market-clearing price, buyers and sellers engage in voluntarily exchange that maximizes social surplus.



Competitive markets can do a number of desirable things for society. Under perfect competition, producers supply the right amount of goods and services, charge the right price, and the right consumers get the goods produced. The “right” amount to produce is the allocatively efficient amount. All units of output are produced for which consumers value those units by more than society values the resources required to produce them. No units are produced that cost more to supply than they are worth to buyers. Competitive suppliers cannot control the price of the product, because there are many producers and the products they sell are virtually identical. Consequently, prices in competitive markets are determined by the impersonal forces of market demand and supply. In the long run, consumers get the very lowest price possible, consistent with firms remaining financially viable since market forces drive prices down to minimum long-run average cost. Even in the short run, however, market prices are determined by the cost structure of firms that operate on their short-run expansion paths, so costs are minimized given the fixed amount of productive capacity in the short run. Finally, voluntary exchange at the market-determined price insures that industry output is purchased by the consumers who place the highest value on consuming the goods and services.

Unfortunately, not all markets are competitive, and even competitive markets can sometimes fail to achieve maximum social surplus. Market failure occurs when a market fails to achieve social economic efficiency and, consequently, fails to maximize social surplus. The causes of market failure can stem from either productive or allocative shortcomings, or both. Six forms of market failure can undermine economic efficiency: MARKET POWER, natural monopoly, negative (and positive) externalities, common property resources, public goods, and information problems. We will briefly explore each of these market failures in this chapter and offer a short discussion of some of the most important policies government can adopt to improve matters.

In the absence of market failure, no efficiency argument can be made for government intervention in competitive markets. As long as competitive equilibrium works to maximize social surplus, any government intervention that moves the market away from competitive equilibrium will reduce social surplus. However, when market failure does create inefficiency, government intervention in the market can, at least theoretically, improve market performance. This motivation for government intervention to overcome market failure and increase social surplus supplies a public interest rationale for policymakers to justify government Government Regulation. Most politicians understand that “fixing a market failure” provides a politically attractive reason for intervening in the marketplace. For this reason, politicians frequently dress up their pet projects as solutions to “market failures”—real or imaginary.

Market failure creates an important and genuine opportunity for government to improve market performance in ways that will add to social surplus. As history demonstrates more often than we wish, even the best-designed government policies fail to be effectively and successfully implemented. The reasons for government failure—government intervention that reduces social surplus—are not difficult to understand. The politics of special interest groups frequently generates government rules and regulations designed to promote the welfare of one group at the expense of the rest of society. And even the best intentions cannot guarantee successful implementation of policies since bureaucrats face challenges of their own. Government bureaucrats’ best efforts can be partially thwarted or entirely compromised by incomplete or obsolete information about the industries they regulate. For example, the Environmental Protection Agency (EPA) cannot effectively regulate pollution emissions unless it knows and understands both the technology of producing goods and the technology of pollution control for dozens of pollutants and thousands of industries. Grappling with such an enormous amount of information and knowledge, which is constantly changing, makes it impossible for regulatory agents to always set pollution standards at the optimal level and perfectly enforce these standards. Of course, we should remind you that perfection is not usually optimal. Expecting zero episodes of government failure is not only unrealistic, but it is also not optimal since avoiding every instance of government failure would itself impose huge costs on society.

Many bureaucrats labor with little incentive to do the best job possible. Government agencies that implement cost-saving measures may find their budgets cut by Congress in the following fiscal year. In spite of the limitations facing government agents who administer and enforce government regulatory and antitrust policies, the pursuit of solutions to market failure problems constitutes one of the most valuable roles for government to play in any society. Business leaders who understand the strengths and weaknesses of various government policy options can play especially important roles in shaping rules and regulations under which their firms must operate. We will now examine the nature of market failure and discuss some of the more important and effective ways government can attempt to make markets generate greater social surplus.


The achievement of productive and allocative efficiency in perfectly competitive markets ensures that the market-determined prices and quantities will maximize social surplus. However, only perfectly competitive markets can meet the necessary condition for allocative efficiency—marginal-cost-pricing. As we have demonstrated in previous chapters, price will always exceed marginal cost under monopoly, monopolistic competition, and oligopoly, that is, under imperfect competition. The reason allocative efficiency is lost under imperfect competition can be directly attributed to the market power that all imperfectly competitive firms possess. Market power always leads to allocative inefficiency and lost social surplus. In this section, we will show you why market power reduces social surplus. We will also examine monopoly markets and briefly discuss the role of antitrust policy in promoting competition.

Market Power and Allocative Inefficiency

Market power is something that competitive firms don’t have—the power to raise product price without losing all sales. Firms with market power, however, do not sell standardized commodities in competition with many other firms. Firms with market power can set prices anywhere they wish along their downward-sloping demand curves. The value, of course, of possessing market power comes from the opportunity to raise price above costs and earn economic profit, something competitive firms cannot do. All imperfect competitors possess some degree of market power. The problem with market power, from society’s point of view, is the loss of allocative efficiency that comes about when imperfectly competitive firms set their prices to maximize profits.

For all firms with market power (not just pure monopolies), the marginal revenue curve lies below the firm’s demand curve. For this reason, prices charged by firms with market power always exceed marginal revenue: P > MR. Since profit maximization requires producing at the output level for which MR = MC, it follows directly that firms with market power must price above marginal cost (P > MC) to maximize profit. As a consequence, all firms possessing market power fail to achieve allocative efficiency, and thus market power diminishes social surplus.

Any degree of market power reduces social surplus, but a high degree of market power may do so much damage to social surplus that a government remedy is warranted. When the degree of market power grows high enough, antitrust officials refer to it legally as “monopoly power.” No clear legal threshold has been established by antitrust authorities to determine when “market power” crosses the line to become “monopoly power.” Nevertheless, antitrust agencies aggressively seek to prevent firms from acquiring monopoly power and can severely punish firms that acquire or maintain monopoly power in illegal ways. Before we discuss antitrust policy toward monopoly practices, we must present the case against monopoly.

Market Power and Deadweight Loss

The best way to understand the problem caused by market power is to compare an industry under two different equilibrium situations: perfect competition and pure monopoly. We will do this using the Louisiana white shrimp industry as our example. Initially, let’s suppose the industry is perfectly competitive. Figure 16.2 shows market demand and supply conditions in the market for Louisiana white shrimp. Industry demand, D, is downward-sloping and, also measures the marginal benefit to shrimp consumers. The longrun competitive industry supply curve for Louisiana shrimp, SLR, is horizontal because the shrimp industry is a constant-cost industry. The flat supply curve indicates that long-run marginal and average costs are constant and equal to $100 per 10-pound basket (LMC = LAC = $100).

The competitive, market-clearing price for 10-pound baskets of shrimp is $100, and the market quantity is 20,000 baskets per month. The competitive equilibrium at point C generates social surplus each month equal to $2 million (= 0.5 * 20,000 * $200). No greater amount of social surplus can be generated than under perfect competition at point C.

Now suppose that a consortium of investors form a company called Shrimp Depot, and they buy every one of the Louisiana shrimp suppliers, transforming the industry into a pure monopoly. As we will explain shortly, antitrust laws in the U.S. would certainly prevent such a consolidation, but let’s compare the monopoly equilibrium (point M) to the competitive equilibrium (point C) anyway. Once Shrimp Depot owns all of the productive capacity, the competitive industry supply curve, SLR, now represents the long-run marginal cost and average cost curve for the monopoly shrimp firm. Shrimp Depot possesses market power and finds its profit-maximizing production level by setting marginal revenue (MR) equal to marginal cost (LMC). As you can see in Figure 16.2, Shrimp Depot maximizes profit by producing 10,000 baskets of shrimp per month and charging a monopoly price of $200 per 10-pound basket.

Under monopoly, consumer surplus shrinks by $1.5 million, which is the area fbCM. A portion of this lost consumer surplus—the area of the red-shaded rectangle fbeM (= $1 million)—is transformed into producer surplus or monopoly profit in this example. Shrimp Depot makes $100 (= P - LAC = $200 - $100) profit on each one of the 10,000 baskets of shrimp it sells, which amounts to $1 million of economic profit.

Allocative efficiency is lost under monopoly pricing since monopolists—and any other firm with market power—do not practice marginal-cost-pricing. At point M, the marginal benefit of the 10,000th basket of shrimp is $200, which exceeds its marginal cost of production, just $100. Since price exceeds marginal cost, resources are underallocated to the shrimp industry, and too little (shrimp) is produced. As a result of this allocative inefficiency, social surplus falls by the amount of the gray-shaded triangle eCM. This lost surplus, which equals $0.5 million (= 0.5 * 10,000 * $100), is called deadweight loss, because the entire surplus lost on units not produced represents a complete loss of surplus to society. We must stress that deadweight loss, which is attributable to the lack of marginal-cost-pricing, arises not only in monopoly markets but also in markets with any degree of market power.

Promoting Competition through Antitrust Policy

To reduce the cost of market failure caused by market power, most industrialized nations rely on antitrust laws, as they are known in the U.S. Canada and the European Union countries refer to their antitrust policies as competition policies. Since the purposes are quite similar, we will focus our rather brief discussion in this textbook on antitrust policy and enforcement in the U.S. Antitrust laws seek to prohibit business practices and actions that reduce or restrain competition, based on the fundamental acceptance of competition as a powerful force for achieving large social surpluses and protecting consumers and competitors from firms with substantial market power. We can only take a brief look here at this fascinating area of law and policy. To gain an understanding of the theory and practice of antitrust policy, you must take at least one course in antitrust law.

Monopoly power, or a high degree of market power, can arise primarily in three ways: (1) actual or attempted monopolization, (2) price-fixing cartels, and (3) mergers among horizontal competitors. Firms may engage in Behavior designed specifically for the purpose of driving out existing rivals or deterring the entry of new rivals, thereby reducing competition, perhaps to the point of gaining a monopoly position. Firms may be found guilty of actual monopolization only if both of the following conditions are met: (1) the behavior is judged to be undertaken solely for the purpose of creating monopoly power, and (2) the firm successfully achieves a high degree of market power. Businesses can also be guilty of attempted monopolization if they engage in conduct intended to create a monopoly and there is a “dangerous probability of success.”

When businesses are unable to drive out their rivals, or choose not to do so for fear of antitrust penalties, they may instead seek to reduce competition by colluding to raise prices. Collusive price-setting is absolutely forbidden, and guilty parties incur substantial financial penalties and even jail time. Nonetheless, the enormous profit potential from successful price-fixing continues to attract plenty of practitioners. Fortunately, cooperation among rival firms can be very difficult to establish and maintain. Antitrust enforcement agents at the Department of Justice make no secret about their willingness to grant clemency from prosecution to the first cartel member to expose and plead guilty to a pricefixing deal. The prisoners’ dilemma really works!

Horizontal merger policy represents the third important area of antitrust doctrine designed to prevent monopoly power from arising through the process of merger or acquisition of direct rival firms. Horizontal mergers or acquisitions happen when two or more firms that are head-to-head competitors—selling the same product in the same geographic markets—decide to join operations into a single firm. Such mergers can obviously lead to an increase in market power, although many horizontal mergers are too small to damage competition. Horizontal mergers can have beneficial effects on social surplus when the merged firm enjoys substantial economies of scale that could not be realized by separate production operations. Of course, consumers only benefit directly from such scale economies if the merged firm passes the cost saving along through reduced prices. Antitrust agencies require firms planning large mergers to notify antitrust authorities of their intentions prior to merging. Antitrust officials typically have 90 days after notification to study a proposed merger and to let the prospective merging firms know whether antitrust agencies intend to challenge the merger or to approve it. In most cases, businesses will abandon a merger if antitrust agencies plan to challenge the merger in court.

Antitrust policy and enforcement in the U.S. is the responsibility of the Department of Justice (DOJ) and the Federal Trade Commission (FTC). While the language of the law itself is not particularly complex, the legal application can be tremendously complicated. Antitrust litigation can also impose heavy costs on businesses, whether they are defendants or plaintiffs. In 1993, American Airlines reportedly spent $25 million defending itself in a predatory pricing trial that took the jury less than one hour to decide the airline was not guilty.

Before leaving our examination of monopoly, we must consider a special kind of monopoly, called natural monopoly, which would be harmful to break up through the application of antitrust laws. We explain in the following section that natural monopoly arises when one firm can produce the amount of good desired by society at a lower total cost than having two or more firms share in the production of industry output. As it turns out, antitrust remedies to break up a natural monopoly would increase total costs and create productive inefficiency. Natural monopoly requires other methods of government regulation to reach an efficient outcome that maximizes social surplus.

Natural Monopoly and Market Failure

Sometimes monopoly can have desirable consequences for society. One such occasion arises when a single firm can produce the total consumer demand for a product or service at a lower long-run total cost than if two or more firms produce the total industry output. This situation is called natural monopoly, and it causes market failure. In natural monopoly, productive efficiency requires a single monopoly producer, which, as you now know, results in allocative inefficiency and deadweight loss to society. Many public utilities, such as electricity, water, natural gas, local telephone, and cable television, are commonly thought to be natural monopolies. If two or more local phone companies serve a community, then each phone company must string its own set of telephone wires. By having a single phone company, the community must pay for just one set of telephone lines. The same logic has been applied to other municipal services that require costly distribution infrastructure. One way to avoid the needless duplication of distribution lines is to give one company a monopoly franchise in return for the right to let public regulators set the price of service. We will examine some of the complexities of regulating price under natural monopoly shortly. First, we must explain more carefully the conditions on long-run cost that lead to natural monopoly.

Natural monopoly is another way of saying that LONG-RUN COSTS are subadditive at the level of output demanded by consumers. Long-run costs are subadditive at a particular output level Qbarif any division of Qbar among two or more firms is more costly than letting a monopoly produce all Qbar units. Thus the terms “natural monopoly” and “subadditive costs” mean exactly the same thing. One way for natural monopoly to develop is for long-run average cost to fall continuously, so that economies of scale extend to all output levels. With continuous economies of scale, cost subadditivity and natural monopoly exist at all levels of output. This can happen for public utilities when large quasi-fixed costs of distribution lines (pipelines, telephone lines, fiber-optic cable, electricity power lines, water lines, and sewage lines) are spread over more units of output.

Figure 16.3 illustrates the nature of cost subadditivity for a water utility in a small town. The water plant and underground distribution lines cost $12 million and are quasi-fixed inputs because they are employed in a fixed amount for any positive level of output. Municipal bonds are sold to pay for these inputs, and the debt payment on the bonds is $60,000 per month. The average quasi-fixed cost for the water plant and distribution lines, AQFC, declines continuously, as shown in Figure 16.3. (Note that water consumption is measured in 1,000-gallon units of consumption.) The long-run marginal cost of water is constant and equal to $2.50 per 1,000-gallon unit. Thus LMC is a flat line in Figure 16.3 equal to $2.50 for all output levels. Long-run average cost, LAC, is the sum of AQFC and LMC. LAC declines continuously as the $60,000 quasi-fixed cost is spread over more units of output, and LAC approaches LMC as water consumption gets very large. You can verify that costs are subadditive in this example by comparing the total cost if one firm produces 40,000 units, which is $160,000 (= $4 * 40,000), to the total cost if two equal-size firms produce 20,000 units each to reach 40,000 units, which is $220,000 [=($5.50 * 20,000) + ($5.50 * 20,000)]. When 40,000 units per month are demanded, monopoly water service saves the community $60,000 per month. With a monopoly water utility, the community pays for just one water plant and distribution network. With two water utilities, the community must pay for two water plants and two distribution networks. Incurring the quasi-fixed capital cost one time, instead of two or more times, and then spreading the cost over all of market demand creates large cost savings, and a natural monopoly arises.

Regulating Price Under Natural Monopoly

When costs are subadditive at the level of output demanded by society, a monopoly produces the desired output at the lowest-possible total cost to society. Amonopolist, however, maximizes profit by pricing above both marginal and average cost. Under monopoly, as we explained previously, consumers not only get too little output, they also can end up paying more than average cost for each unit purchased. Breaking up a natural monopoly is undesirable because increasing the number of firms in the industry drives up total cost and undermines productive efficiency. State regulators of public utilities—known as “public utility commissions” (PUCs) or “public service commissions” (PSCs)—face a challenging task in regulating the price that natural monopolies can charge. Regulators would like to force a pricing structure on natural monopolies that creates social economic efficiency. Under natural monopoly, as we will now show you, no one price can establish social economic efficiency.

We can best explain the pricing dilemma facing regulators by returning to the previous example of the water utility. Figure 16.4 reproduces the long-run average and marginal cost curves facing the water utility. Recall that LAC falls continuously as the quasi-fixed cost is spread over more units of water output. Clearly, then, regulators wish to maintain a monopoly in water service to fully exploit the cost subadditivity that extends over the entire range of water production, because economies of scale exist at all levels of output in this case. The city’s demand for water and corresponding marginal revenue are shown in Figure 16.4 as D and MR, respectively. While utility regulators do not wish to encourage (or force) competition by increasing the number of water utilities, they also recognize that if the monopoly water utility is unregulated, it will operate at point M on demand, which maximizes the water utility’s profit. When water price is $6.50 per 1,000-gallon unit, only 20,000 units are demanded, and the city faces a deadweight loss due to monopoly of $40,000, which is the area of the (unshaded) triangle Mws (= 0.5 * 20,000 * $4). At point M, the unregulated monopolist’s economic profit is $20,000 per month [=20,000 * ($6.50 - $5.50)].

To solve the problem of allocative inefficiency created by monopoly pricing at point M (P is greater than MC at point M), the PUC might set the legal price of water at the point on demand where P = LMC, which is $2.50 per unit (point s). At this price, 40,000 units of water are consumed each month and social surplus is maximized. Unfortunately, when a utility experiences economies of scale at the socially optimal level of consumption, LMC is less than LAC, and marginal-costpricing always creates economic losses for utilities operating in a region of economies of scale. As you can see in this example, the owners of the water utility face unit costs of $4 per unit (point t) and lose $1.50 (= $4 - $2.50) on each of the 40,000 units, creating a $60,000 monthly loss. While marginal-cost-pricing succeeds in maximizing social surplus, it is not a viable pricing scheme because no utility would continue providing water at a loss. Indeed, investors will not build the water plant and distribution pipelines in the first place if they believe regulatory authorities plan to implement marginal-cost-pricing in the face of continuous economies of scale.

A second-best pricing solution in the face of continuously declining LAC involves setting price as close to marginal cost as possible, yet high enough to allow the utility to break even by earning zero economic profit. With this method, the PUC regulates price at the level that minimizes the loss of social surplus but allows investors to earn a normal rate of return. As it turns out, the second-best pricing solution simply requires setting price equal to LAC. In this example, $4.50 is the closest price to LMC that allows the water utility to remain economically viable. When paying a price of $4.50 per 1,000-gallon unit, buyers can be happy that they are paying the lowest possible price that will keep water-utility investors from shutting down operations and moving their capital to its best alternative use. The problem with average-cost pricing is that price exceeds marginal cost, resulting in a deadweight loss of $10,000 (= 0.5 * 10,000 * $2)—the area of the gray-shaded triangle zxs in Figure 16.4.

As long as economies of scale extend over the entire range of demand for utility services, no single, uniform pricing of output can achieve social efficiency and create financially viable firms. A pricing solution does exist that can satisfy social efficiency conditions and maximize social surplus: two-part pricing of utility services. A two-part pricing plan charges utility customers a fixed access charge plus a usage fee based on the number of units purchased. By wisely setting the access charge and usage fee, regulatory authorities can solve the utility pricing dilemma. The solution is to set the usage fee equal to marginal cost and set the access charge to spread the loss caused by marginal-cost-pricing across all utility customers. The total monthly water bill for Q units of water is computed as follows:

Total bill for Q units = LMC * Q + L / N

where L is the total loss generated by marginal-cost-pricing, and N is the number of households served by the utility. In this example, suppose the water utility in Figure 16.4 serves 4,000 households. The ideal two-part pricing plan is to set the usage fee at $2.50 per 1,000-gallon unit of water consumed and to set the monthly access charge for each household at $15 ( $60,000/4,000). The total monthly water bill for Q units of water is computed as follows:

Total water bill for Q units = 2.50 * Q + $60,000/4,000

= 2.50Q + 15

Under this pricing plan, a household using 12,000 gallons of water per month (Q = 12) pays $45 (= $2.50 * 12 + $15) per month for water. Notice that the allocatively efficient amount of water is consumed (40,000 units per month), and productive efficiency is achieved because economies of scale are fully exploited by letting one firm produce the entire industry demand.

We should not leave you with the impression that regulating natural monopoly is an easy matter. Many difficulties arise, and the brief treatment here serves only to introduce the fundamental nature of the problem. In a full course on regulation of natural monopoly, you will gain a more complete understanding of the financial complexities of rate-of-return regulation (a widely used average-cost-pricing practice), multiproduct cost subadditivity that can arise for multiproduct utilities, and incentive-compatible regulations designed to motivate utilities to invest in cost-saving technologies. Many economists have made lucrative careers for themselves by specializing in public utility regulation practices. Both utilities and regulatory agencies hire these experts.



Another important cause of market failure in competitive markets arises when the actions taken by market participants create either benefits or costs that spill over to other members of society. When these spillover effects are beneficial to society, economists call them positive externalities. Flu vaccination provides an example of a spillover or external benefit creating a positive externality. When one person at the office chooses to get a flu vaccine, everyone who works with that person benefits from a reduced probability of catching the flu at work. Alternatively, when spillover effects are costly to society, economists call them negative externalities. Pollution is a particularly important example of negative externality. If an upstream business chooses to dump polluted wastewater into a nearby river, parties downstream who use the river for recreational or productive purposes— swimmers, boaters, and fishing companies, for example—will bear the spillover or external cost of this pollution through reduced enjoyment and productivity of the river.

External or spillover benefits and costs undermine allocative efficiency because market participants, when making consumption and production decisions, rationally choose to ignore the benefits and costs of their actions that spill over to other parties. Consequently, competitive market prices do not include the social benefits or costs that spill over to other members of society. Equilibrium price must equal both marginal social benefit and marginal social cost to provide buyers and sellers with the correct incentive to make allocatively efficient decisions. In competitive markets experiencing either positive or negative externalities, the equilibrium price sends the wrong signal to buyers and sellers, causing them to consume and produce either too much or too little output. Since the wrong amount of output is produced, both types of externality create a deadweight loss reflecting the lost social surplus of allocative inefficiency. For many businesses, the externality of greatest consequence for their profits is the negative externality created by pollution since government environmental agencies usually attempt to impose remedies on business polluters. Thus we focus our analysis here on pollution generated by businesses in the process of making goods and services for society.

As we stated previously, managers rationally ignore spillover or external costs when making their profit-maximizing production decisions. Profit maximization only concerns private costs of production, that is, costs incurred by firm owners to use productive resources. Because external costs do not affect profits, managers will likely ignore these costs that spill over to others in society. External costs, nonetheless, are real costs borne by society for the production of goods and services. The social cost of production is the sum of the private cost incurred by producers and any external or spillover cost imposed on other members of society:

Social cost = Private cost + External cost

Economists sometimes say that a negative externality drives a “wedge” between social and private costs of production:

Social cost - Private cost = External cost

The larger the external costs of a negative externality, the greater the difference between social and private costs of production, and the greater will be the resulting deadweight loss.

Figure 16.5 shows why the “wedge” of negative externality makes allocative efficiency impossible to achieve in competitive markets. The demand curve for the competitively supplied good correctly measures the marginal social benefit of the good: D = MSB. The marginal private costs of production are given by the competitive supply curve: S = MPC. Competitive market equilibrium is established at the intersection of demand and supply (point C), where QC units are produced and consumed at price PC. Production of this good by competitive suppliers creates an external cost that spills over to society. The marginal external cost, shown as MEC in Figure 16.5, increases with the level of output. At every output level, the marginal social cost curve is the vertical sum of the marginal private cost and marginal external cost: MSC = MPC + MEC. As you can see, in competitive equilibrium, too much output is produced because MSC exceeds MSB at point C. Allocative efficiency occurs at QE (point E), where MSC equals MSB. By producing the units from QE to Qc, the competitive industry creates a deadweight loss on each unit that costs more to produce than it is worth to society. The area of the gray-shaded triangle DWL is the amount by which social surplus is reduced by overproducing and overconsuming the good.

Perhaps you are now thinking that a “good” manager should consider all external costs that spill over to society to “do good for society.” The issue of “doing good for society” often surfaces in class discussions when we analyze the loss of social surplus caused by pollution. Perhaps now is a good time to stress how little economists have to say about “doing good.” As you know from your course in business ethics, any debate about how managers should handle spillover costs to society raises complicated subjective and ethical issues concerning the appropriate level of social responsibility of business enterprises. Although ethical issues fall outside the realm of objective (positive) economic analysis, we can offer you two objective reasons to ignore external costs in decision making. First, managers who choose not to ignore external costs will produce less output than would maximize profit, which, of course, reduces profits and wealth of the firms’ owners. If your firm operates in a competitive industry, you will be forced to exit in the long run. You can safely predict that other managers in your industry, wishing to survive in the long run, will make profit-maximizing decisions.

A second important consideration concerns the possible legal consequences for you if your shareholders believe your practice of including social costs in decisions conflicts with your legal responsibilities to protect the value of the firm. The legal standards that apply to this area of executive responsibility are not clear, but we advise you, as we have throughout this book, to make decisions that will increase the value of the firm. Government authorities certainly don’t count on individual firms to sacrifice profits for the good of society. If, for example, you do decide to undertake a costly investment in “green” production technology for your firm, prepare to show owners of the firm that “green” production methods are indeed economically efficient (i.e., “green” production lies on your firm’s expansion path) or that buyers will substantially increase their demand for your product when they hear about your sensitivity to the environment. In the absence of a clear profit justification for internalizing the costs of negative externalities, you may find yourself in legal trouble—while you search for a new job!

As we have stressed throughout this chapter, government intervention is warranted only when there is market failure that government policy can fix at lower cost to society than the market failure itself. In the case of negative externalities, public policymakers can eliminate allocative inefficiency by devising methods of forcing firms to internalize external costs they would otherwise rationally choose to ignore. Once external costs are internalized, firm owners face the full social cost of producing goods and services, and allocative efficiency is restored. Taxation and assignment of property rights provide two of the most effective methods government can employ for internalizing costs. We can now summarize our discussion of negative externality in the following principle.

Principle A producer creates a negative externality by imposing an external cost on other members of society without making a compensating payment for the harm caused. Negative externality drives a wedge between social and private costs of production, which causes producers in competitive equilibrium to overproduce the good or service. The loss of allocative efficiency due to negative externality creates a deadweight loss to society.

Pollution: Market Failure and Regulation

Let’s consider again the example we mentioned previously: An upstream competitive industry chooses to dump polluted wastewater into a nearby river. Parties downstream, who use the river for recreational or productive purposes— swimmers, boaters, and fishing companies, for example—are burdened by the external cost of this pollution through reduced enjoyment and productivity of the river. First we will analyze the market failure of this competitive industry caused by the pollution it creates. Then we will turn our attention to the role government environmental policymakers can play in solving the externality problem. We will show you how environmental economists identify the optimal level of pollution emissions and demonstrate that a properly set charge or tax on emissions can motivate profit-maximizing firms to reduce their pollution levels to the socially optimal level.

Allocative inefficiency and market failure

In Figure 16.6, the marginal external cost (MEC) caused by the pollution externality is increasing as industry output rises. MEC for the 6,000th unit of output is $2.

The marginal private cost (MPC) incurred by competitive firms to produce the 6,000th unit is $3. MPC represents the competitive industry supply curve, S, because suppliers ignore the external cost of pollution. Marginal social cost to produce the 6,000th unit is $5 (= $2 + $3). MSC is constructed by repeating the vertical summation at every output level.

As usual the competitive equilibrium price, $3, is found at the intersection of demand and industry supply (point C in the figure). Competitive suppliers behave inefficiently, because managers in this market increase production up to the point where price equals marginal private cost. Since the marginal social cost of the 6,000th unit is $5, industry output in competitive equilibrium exceeds the level that would maximize social surplus. Allocative efficiency happens when the industry produces 4,800 units at point E in Figure 16.6. The deadweight loss of social surplus caused by overproduction and overconsumption equals the area of the gray-shaded triangle ECu, which is $1,200 (= 0.5 * 1,200 * $2).

Regulators can restore allocative efficiency by taxing producers on their pollution emissions, causing them to internalize the external pollution costs. Environmental policymakers and enforcement authorities have employed numerous taxation methods with varying degrees of success. We will present one of the more widely used taxation methods here: emissions taxes (or charges). This method employs market incentives to encourage firms to choose the optimal level of emissions and control activity (called “pollution abatement”). To set the proper tax rate on pollution emissions, environmental authorities must first determine the socially optimal rate of emissions, which typically applies to a specific geographic area that regularly experiences “excessive” levels of pollution.

The optimal level of pollution (and abatement)

Throughout this text, we have applied the reasoning of marginal analysis to solve a variety of optimization problems. Finding the socially optimal level of pollution provides one more opportunity to demonstrate the power of marginal analysis. To find the optimal pollution level, policymakers must be able to measure with a reasonable Degree of Accuracy the benefits and costs to society for different levels of pollution emissions. As you will see, finding the optimal level of emissions also determines the optimal level of effort for firms to expend reducing, preventing, or controlling pollution emissions from their production facilities. Such activity is called pollution control or pollution abatement.

The benefit accruing to society from reducing pollution is equal to the dollar value of damages from pollution avoided by pollution reduction or abatement. The measure of pollution damages to society includes all costs attributable to pollution, such as costs of illness to humans, value of lost productive and recreational use of environmental resources, cost to society of lost biological habitat, and so on. Measuring damages is a controversial “science” involving multidisciplinary analytical methods. You can learn more about this important area of research and methodology by taking a course in environmental economics or reading the textbook for that course. While the scientific and environmental cost data required to estimate accurately pollution damages are generally substantial and costly to obtain, the task must nonetheless be undertaken by government environmental agencies if they wish to make optimal policy decisions.

To find the optimal level of pollution, marginal damage caused by pollution must be estimated, that is, the addition to total damages attributable to discharging one more unit of pollution into the environment must be known. In Figure 16.7, the curve labeled “MD” shows marginal damage for various rates of pollution emission. As you can see by looking at MD, emissions cause no damage below a threshold level of 400 tons per year, after which each additional ton causes evergreater marginal damage to society. The damage caused by the 600th ton discharged is $20. We can add marginal damages for all units discharged and obtain the total damage caused by any specific level of pollution emission. In this case, total damage caused by 600 tons of emissions is the area under MDfrom 400 units to 600 units, which is $2,000 (= 0.5 * 200 * $20). From this computation, it follows that the benefit to society of abating these 200 tons of pollution—reducing emissions from 600 to 400 tons—is the avoided damages of $2,000 per year. For this reason, the MD curve measures the marginal benefit of pollution control.

Reducing pollution emissions almost always requires an expenditure of resources on pollution control activities. Since business owners are also members of society, their costs to abate pollution must be considered when policymakers determine the optimal level of pollution. The marginal cost of abatement, shown as MAC in Figure 16.7, gives the addition to total abatement cost of reducing emissions by one more ton per year. We must stress here that to interpret properly the MAC curve, you must recognize that abatement activity (the number of tons of pollution saved) increases by moving leftward along the MAC curve. The MAC curve intersects the emissions axis at the uncontrolled level of pollution, which is the amount of pollution that is discharged if producers make no effort at all to control pollution. In this example, the uncontrolled level of pollution is 1,600 tons per year. Since the cost of abating pollution usually starts off at a relatively low incremental cost and rises for higher levels of abatement effort, MAC rises as emissions fall (moving leftward from 1,600 tons per year to lower levels of emissions in Figure 16.7).

Abating the 1,000th ton of pollution (resulting in 600 tons of emissions) requires producers to spend an additional $50 on abatement effort. The total abatement cost incurred by producers to abate 1,000 tons per year is the area under MAC from 1,600 tons leftward to 600 tons: $25,000 (= 0.5 * 1,000 * $50). At 600 tons of emissions (and 1,000 tons abated), the total cost to society for this level of annual pollution is the sum of total damage and abatement cost:

Total social cost of 600 tons = Total damage + Total abatement cost

= $2,000 + $25,000

= $27,000 per year

Notice that the optimal level of pollution is not 600 tons per year, because increasing the level of pollution (and reducing the level of abatement) reduces the total cost to society from pollution.

To see why, suppose the industry increases pollution to 601 tons per year, which reduces the number of tons abated to 999. Total damage rises with the higher level of pollution, but only by $20. Total abatement cost falls by $50 due to the reduction in abatement activity. The net effect on society of increasing pollution by one ton is the difference between MAC and MD, which is $30 (= $50 - $20) for the 601st ton of discharge. As you can now see, pollution should be increased to 800 tons per year to minimize the total cost to society from pollution. The minimum possible total cost of pollution is the sum of the areas of the two shaded triangles in Figure 16.7:

Total social cost of 800 tons = Total damage + Total abatement cost

= (0.5 * 400 * $40) + (0.5 * 800 * $40)

= $8,000 + $16,000

= $24,000

To curb pollution to just 800 tons per year in this particular geographic region, environmental authorities may take a “command and control” approach. Let’s suppose currently that no pollution control is happening, so 1,600 tons are discharged. Environmental authorities notify producers in the region that only 800 tons of discharge will be allowed henceforth. All producers may then be commanded to reduce pollution levels equally to accomplish a total reduction of 800 tons (= 1,600 - 800). The cost of compliance for the industry in this example is $16,000, the amount spent by the industry on pollution control efforts. Numerous policy shortcomings undermine the desirability of using this kind of direct command and control approach. Recently, environmental policy makers have turned to regulatory methods that create economic incentives for businesses to not only comply with emissions targets but to also go beyond compliance by investing in more pollution control assets and engaging in research and development to find new economically efficient means of controlling pollution. We limit our discussion to one of these methods: emission taxation.

Optimal emission tax on pollution

Emission taxes levied on each ton of pollutant discharged can create an effective economic incentive for firms to make pollution and abatement decisions efficiently. Figure 16.8 shows how an emission tax works using the same MD and MAC curves that led us to conclude that 800 tons of emissions per year are optimal, because this level of emissions minimizes society’s total cost from pollution (including the cost of pollution abatement to reach 800 tons).

By setting the emission tax rate at $40 per ton, regulatory authorities create an incentive for firms in this industry to reduce pollution from the uncontrolled level of 1,600 tons per year to the optimal level, 800 tons per year. To see why this happens, suppose less abatement is undertaken, and the industry discharges 1,000 tons and abates 600 tons. At this level of pollution, firms can pay $40 to discharge legally one ton of pollutant or firms can spend $30 to abate the 1,000th ton of emissions. Obviously, the firm chooses to abate rather than pollute, and saves $10. In general, then, pollution abatement continues as long as MAC is less than the tax rate, $40 in this case. Firms will increase abatement effort and reduce emissions until the optimal level of pollution is reached at 800 tons of emissions. Notice that the industry will not reduce its pollution below 800 units, because then abatement becomes more costly than simply paying $40 to discharge the pollutant into the environment. So, by allowing firms legally to discharge all the pollutant they wish, as long as they pay $40 for every ton, regulators can motivate industry to pollute and abate at socially optimal levels, and collect $32,000 (= 800 * $40) in emission tax revenue from business in the process!

Emission taxation appeals to environmental authorities for several reasons. When authorities can estimate MD and MAC with reasonable accuracy, setting the efficient emission tax rate is a straightforward task. Once the tax rate is set optimally, regulatory authorities know that they can rely on the cost-minimizing behavior of business managers to choose the optimal levels of emissions and abatement. They also know that firms will have an incentive to invest in R&D for the purpose of finding less costly methods of abating pollution. Aton abated, after all, is a ton not taxed. Businesses may well prefer to have old-fashioned legal restrictions on the amount of pollution that can legally be dumped into the environment. A legal limit of 800 tons imposes only the cost of abatement, $16,000, on the industry, while businesses avoid paying $32,000 in emission fees. Many other interesting issues concerning the regulation of pollution are covered in depth in courses on regulation and environmental economics. If you plan to work in an environmentally sensitive industry subject to a great deal of green regulation, you should plan to take at least one such course.

This concludes our discussion of externality as a source of market failure. We will now examine another form of market failure that arises when people cannot be forced to pay for consuming a good or service or using a natural resource.

I L L U S T R AT I O N 1 6 . 1

Taming Negative Externality with Congestion Pricing

Traffic congestion represents a market failure caused by negative externalities generated by automobiles driving on crowded roadways. Even without congestion, driving a car creates exhaust emissions that pollute the air that everyone must breathe. This negative externality occurs regardless of traffic conditions. However, when congestion arises, each car stuck in traffic creates even more pollution (car engines are running longer) than when traffic flows freely. During periods of the day when traffic is light and no congestion occurs, the only negative externality created is the pollution cost imposed on society. However, once roadways become congested, not only the social cost of pollution rises but the time cost to each driver stuck in traffic rises as well. On a congested highway, each additional driver adds a small additional time cost to every other driver. This external cost on other motorists drives a wedge between private and social costs of using a roadway.

As we explained in our discussion of negative externalities, this spillover cost leads to allocative inefficiency in road use because all drivers ignore the external cost their cars impose on everyone else. Thus, motorists make road use decisions based on private costs that understate the true or full social cost of using a particular roadway at a particular time of day. If motorists incurred the full social cost of road use, they might choose to drive at a less congested time of day or even give up driving private cars and use public transportation. And, of course, if they also had to bear the full social cost of their auto emissions, they might well choose to drive a cleaner car or use public transportation.

Market failure created by traffic congestion provides an opportunity for government intervention to improve resource usage and social well-being. One way to improve traffic congestion is to build more public roadways and expand the number of lanes on existing roads. The cost of building new roads and expanding old ones can be exceedingly high in urban areas where land is scarce and expensive. (Of course, urban areas are typically the locations where congestion is a problem!) Furthermore, adding concrete to reduce congestion may worsen the problem of automobile emissions as more people will decide to drive when adding new roads or widening old ones reduces congestion.

Another way for government to improve matters is to charge a fee or toll for the privilege of driving on roads, and then to raise this toll to a level that will reduce congestion. Nobel laureate economist William Vickrey developed this approach, commonly called “congestion pricing,” in the 1950s. Vickrey believed that charging drivers higher tolls during peak hours and lower tolls at off-peak hours would close the gap between the private and social costs of driving. By raising the toll at rush hour to reflect the higher marginal congestion costs, transportation officials seek to flatten out the “peaks” in demand each day by giving drivers an incentive to switch their time of travel from peak to offpeak periods.a This is precisely the technique that urban planners are experimenting with in Stockholm, Sweden.

According to a recent article in The Wall Street Journal, automobile traffic in Stockholm is such a nightmare, especially during morning and evening rush hours, that the city has undertaken an experimental test of “the world’s most sophisticated traffic-management system.” Traffic engineers and urban planners face a particularly difficult problem in Stockholm, because the metropolitan business district spans a number of small islands that are linked by several bridges. The drive into the city at the peak morning rush hour usually takes three times as long as it does during off-peak hours. The traffic control system charges drivers tolls that vary according to the time of day. To implement this complex pricing scheme, the Swedish government contracted with IBM Corporation to install transponder boxes that attach to windshields for the purpose of deducting tolls from bank accounts. IBM also installed laser detectors to read license tags, and a video camera network capable of tracking every car in Stockholm.

During a six-month test period, the dynamic toll system successfully reduced peak-period travel time by one-third—without building or expanding a single new bridge or road. The figure nearby shows the structure of the congestion-pricing plan employed in Stockholm. The WSJ article also reported that during the trial period of the congestion-pricing system, exhaust emissions and carbon dioxide fell by 14 percent in the inner-city area of Stockholm. And “some of the biggest beneficiaries weren’t the drivers, but cyclists and bus riders.” Bus ridership rose by 9 percent during the trial period.

Now that the experiment is over, Stockholm government officials have scheduled a voter referendum to decide whether to continue using the congestionpricing system. A poll of voters at this time finds 52 percent of the voters back the pricing plan. Since the purpose of government intervention is to remedy market failures for the public good, it must be encouraging to Stockholm officials to see their plan winning broad voter support. Indeed, urban and transportation planners in Bangkok, New York, Dublin, Prague, Copenhagen, and San Francisco are all considering the same sort of congestion-pricing plan for their innercity roadways.



Suppliers of most goods and services can exclude people who do not pay the firm’s price from consuming the firm’s output. When access to a good or a scarce resource cannot be excluded, market failure may result. In this section, we examine two kinds of market failure caused by lack of excludability, which also possess some degree of the externality problem discussed in the previous section: common property resources and public goods. In the absence of any government intervention, common property resources are generally overexploited and public goods are underproduced.

Common Property Resources

Property rights to resources establish who owns them and who may rightfully use them. Common property resources are resources for which property rights are completely absent or so poorly defined that no one can effectively be excluded from accessing these resources. Since everyone can use these resources and no one can be excluded, they will be overused and underproduced, which diminishes their contribution to social surplus. Some classic examples of overexploitation caused by nonexcludabilty are overhunting of whales, overhunting of bison, and overharvesting of ivory and forests.

The problem of open access to resources is very similar to the problem of negative externality. When a whaler takes his catch onboard, he establishes his property right to the whale under the “rule of capture” that applies to many nonexcludable resources. The whaler may well understand that leaving this whale to reproduce is necessary to insure a sustainable population of whales, but the whaler also knows that other whalers will take the whale if he does not. The whaler considers only his private share of the total social costs of harvesting the whale. In a fashion similar to the case of negative externality in the production of goods and services, nonexcludabilty drives a wedge between private and social cost of resource use, causing the resource to be overexploited.

You may also see here some similarity to strategic decision making since one person’s decision to take a resource depends on his belief about how others will decide to use the resource. Nonexcludability can create a prisoners’ dilemma for resource exploitation. The gain to society from conservation is lost or severely undermined when weak property rights make “use it or lose it” a dominant strategy. We can best demonstrate this with a fictional example of a commonly owned underground pool of oil. Suppose the pool of oil flows under land owned by two companies, Shell and BP. Each company drills a well on its property into the common underground pool of oil. Geologists at both companies explain to managers that to get the maximum total number of barrels of oil, the oil must be pumped out of the ground slowly, so that pressure is maintained uniformly across the pool. Rapid pumping of the oil by either company will cause total yield to fall sharply. Unfortunately, property rights to the oil are poorly defined, and the “rule of capture” applies. Shell and BP share the right to pump the oil, but they only own the oil that comes out of their own company’s well. This sets up a “use it or lose it” property rights situation.

Table 16.1 shows the payoffs according to the production rate each firm chooses. If both firms choose a “slow” (and equal) rate of production, the maximum total amount of oil that can be pumped from this field is 300 million barrels (approximately a 15-day supply for the United States), as given by cell D in the payoff table. Since the firms pump at equal rates in cell D, each firm gets 150 million barrels of oil. If either firm decides to pump “fast” while the other firm pumps “slow,” the faster-pumping firm gets 175 million tons and the slower firm gets 50 million tons. Of course this diminishes the total amount of oil (225 million barrels is less than 300 million barrels). An even less desirable outcome results when both firms choose “fast” and each firm gets only 75 million barrels, for a total production from this pool of 150 million barrels.

Let’s suppose each firm chooses its production rate simultaneously in a onetime decision, perhaps because all the oil is pumped out before either firm can reconsider its production decision. Using the techniques of solving games, you can see that both firms have dominant strategies: produce “fast.” Each firm knows that the other firm can gain by cheating on any agreement they might make to both pump slowly. In cell B, Shell pumps at a fast rate while BP pumps at a slow rate, and Shell gets 25 million more barrels from the common pool than it would get if it cooperated with BP by pumping slowly. And, as Shell knows, BP has exactly the same incentive to cheat. Both firms see that they will both be better off if they cooperate by pumping slowly, but the prospect that the other will cheat makes pumping “fast” a dominant strategy. As in every prisoners’ dilemma game, both parties play their dominant strategies (cell A) and both are worse off than if they had cooperated (cell D).

While the payoffs in this example are hypothetical, the problem certainly is not. By overpumping this pool, pressure falls, which creates numerous smaller pockets of oil that cannot be reached by any single well, and this effectively dooms recovery until drilling technology or crude prices advance sufficiently to make recovery economically feasible. To protect this mineral resource from this tragic outcome, government policymakers have several options. Government could take the property rights to the oil and assume the role of oil producer. Of course this option destroys the incentive for Shell or BP to search for oil in the first place. A second option is for government to regulate production by using government geologists to determine the efficient rate of production and then monitor Shell and BP to make sure neither firm exceeds this regulated rate of production. Athird option is to fix the property rights problem that created the “use it or lose it” strategy in the first place. Unitization is a method of accomplishing this by defining property rights so that each owner of the oil pool possesses equal rights to the oil, regardless of who pumps it out of the ground. Government officials then simply monitor and enforce the unitized property rights. With unitized property rights, all owners will now wish to cooperate by producing at a rate that maximizes the total output (and value to society) of the oil pool.

Because property rights issues can be complicated to analyze, we cannot offer a simple policy recommendation that applies to all situations of market failure caused by nonexcludable access to resources. Improving property rights, however, can be a powerful tool for saving whales, elephants, bison, fish, forests, minerals, species, and the environment. Many environmentalists now believe that efficient assignment of property rights combined with market incentives can successfully replace many ineffective government regulations.

Public Goods

Public goods are characterized by two properties: they are nonexcludable and they are nondepletable. In the case of public goods, the problem of nonexcludabiltiy is called the free-rider problem: suppliers of a good cannot prevent nonpayers from consuming the good or service. If the free-rider problem is severe, firms cannot collect sufficient revenue to cover costs, and no firm will produce any of the afflicted good. A good is nondepletable (or nonrivalrous) in consumption if one person’s consumption of the good causes no reduction in the amount or quality of the good available for consumption by other members of society. By this definition, most goods that are called public goods simply because government supplies them are not actually true public goods.

Either one of the two properties can, by itself, create a problem for the provision of public goods. Taken together, these two properties can completely eliminate the private provision of such goods, necessitating government provision of pure public goods. Goods that possess one or the other of these two properties, in varying degrees of severity, are not pure public goods and may be reasonably well supplied by profit-maximizing firms. It is best to view “pureness” of public goods as a matter of degree.

Perhaps the best example of a pure public good is national defense. Once national defense is provided for one citizen it cannot be withheld from any other citizen (nonexcludable). It is easy to be a free-rider of national defense. Furthermore, one citizen can consume national defense without depleting the amount available to any other citizen (nondepletable). No profit-maximizing firm will supply national defense, since no one will voluntarily pay for a service if he can get it for free when someone else buys it. Examples of pure public goods are few in number. There are, however, many goods that exhibit varying shades of “public goodness.”

Consider the enormously profitable computer software industry. Microsoft’s current operating system software, Windows XP, can be reproduced at nearly zero marginal cost for any number of users who want it. For all practical purposes, computer software is nondepletable. When a good is nondepletable, the marginal cost of production is zero. And, when goods can be produced at zero marginal cost, social surplus for these unusual goods is maximized by giving the good away—that is, the socially optimal price is zero. When Microsoft charges a price greater than zero for its Windows XP software (currently $199 per copy), the market for this software is underserved since price exceeds marginal cost. Deadweight loss is generated, because everyone who places a value on the software below $199 will choose not to buy it, even though the price he is willing to pay exceeds marginal cost. Of course, some of this deadweight loss is offset by software pirates who freeride by making illegal copies to avoid paying Microsoft $199.

As you can see from the Microsoft example, many goods that are not pure public goods will be privately produced and do not require government provision. It would be foolish to have government provide computer software just to prevent free-riders from consuming this nondepletable good. As a general rule, only pure public goods need to be provided by government agencies since private firms will not supply them at all. We summarize our discussion of public goods in the following principle.

Principle A pure public good is nonexcludable and nondepletable. The inability to exclude nonpayers creates a free-rider problem for the private provision of public goods. Even when private firms do supply public goods, a deadweight loss can be avoided only if the price of the public good is zero.



Market failure may also occur because consumers do not possess perfect knowledge. Perfect knowledge includes knowledge by consumers about product prices and qualities, including the hazards associated with a product. Like all other activities we have studied, search for information is carried out to the point where the marginal benefit of more information equals the marginal cost of gathering it. Consumers choose the optimal level of search based on their individual valuations of benefit and their individual search costs. It is a rare consumer who finds it optimal to be perfectly informed. Even when consumers are optimally (but not perfectly) informed about product prices and qualities, market failure due to imperfect knowledge remains problematic. In this section, we will describe how lack of perfect knowledge may lead to prices greater than marginal cost and possibly too few or too many resources devoted to the production of some goods.

Imperfect Information about Prices

As we just mentioned, consumers will not gather every piece of information about prices and product characteristics as long as information is costly to obtain. We know already from a number of discussions in different contexts that the optimal amount of information for consumers occurs when the marginal benefit of its use is just equal to its marginal cost of collection. As long as marginal benefit is greater than marginal cost, more information will be gathered, but because marginal cost is positive, consumers will never collect information until marginal benefit is zero. This means that they will be unaware of higher-quality products or the lower prices some sellers charge for exactly the same product. Furthermore, the optimal level of information will not be the same across consumers. For some consumers the marginal cost of collecting information will be relatively high. Age, a handicap, high transportation costs, and the opportunity cost of a person’s time all have an effect on the marginal cost of getting information. Marginal benefit will also be different across buyers.

The fact that consumers do not know everything about prices and Product Attributes creates an opportunity for product prices to vary from seller to seller. Recall that in the model of perfect competition with perfectly informed consumers, every seller charges the same price, because every seller’s product is identical to every other seller’s and consumers know this. In reality, as we are now arguing, consumers do not know the price that every seller is charging, even for homogeneous products. Consider even a particular product—like a 150-count box of white Kleenex tissue. Prices will vary from seller to seller, because all buyers will not go to the seller with the lowest price. The marginal cost of Gathering Information about the prices all sellers charge is simply too high relative to marginal benefit. Some sellers will, therefore, survive in the market charging relatively high prices, and consumers will not all be paying the same prices for the same goods.

In addition, firms will not be charging prices equal to marginal cost, because they know information is costly. As long as their prices are not outrageously high, customers will not find it optimal to continue searching for a lower-priced seller. The lack of information gives firms some degree of market power. As we demonstrated in this chapter, market power creates market failure and deadweight loss. In this instance of imperfect information, market power arises not because of product differentiation, or even a lack of perfect substitutes. Market power emerges in competitive markets because imperfectly informed consumers do not possess knowledge of all producers and prices. Even though consumers are optimally informed, they are also (optimally) ignorant about the availability of substitutes. Their ignorance of substitutes creates market power for sellers of homogenous goods, something that did not happen in the model of perfect competition. Thus, imperfect information about sellers and prices can cause market failure in competitive markets.

Imperfect Information about Product Quality

Even when consumers have information, they may not be able to evaluate it correctly. Evaluating information about sellers and prices is not as challenging for consumers as successfully utilizing product quality information. Buyers are frequently unaware of undesirable side effects of chemicals in hair spray or new carpet. Foods may contain harmful substances that are listed on the label, yet the information means nothing to the shopper. And automobiles may have faulty designs that only an engineer can evaluate. We also know that producers sometimes provide false or misleading information to make consumers believe a product is better for the buyer than it actually is. Thus, possessing information does not guarantee that consumers will benefit from the information.

To illustrate the problem, suppose consumers of a product that is competitively supplied misjudge the quality of the product, either because the industry misinforms them or because they all mistakenly evaluate the quality information they possess. In Figure 16.9, market demand curve D is the demand when consumers evaluate product quality to be higher than the true level of product quality. Market equilibrium occurs at point C. Because the true quality is lower than the perceived quality, the marginal social benefit curve, MSB, lies below demand. The allocatively efficient level of consumption and production is found at point E where supply intersects MSB. Since the market price, PC, does not equal marginal social benefit, a deadweight loss due to allocative inefficiency reduces social surplus by the area of the shaded triangle. Of course, if consumers underestimate the quality of a product, demand lies below MSB, and too little of the good is consumed.

The deadweight loss due to imperfect information about product prices and qualities opens the door for profit-seeking entrepreneurs to supply Information Services. Unfortunately, information is very much like a pure public good, so government provision of information about prices and product quality may be warranted if government agencies can provide information services at a lower cost than the deadweight loss consumers bear with their own individual search efforts.

Information as a Public Good

Recall from our previous discussion that pure public goods are both nondepletable and nonexcludable. When one buyer consumes information about product quality, by reading an article in Consumer Reports, for example, that information remains fully available to other buyers in society. After the information is produced once, the marginal cost of providing additional consumers with product pricing and quality information is very close to zero. In the current digital age, it can be difficult for information suppliers to prevent nonpayers from receiving the information for free. It is certainly not impossible, however. Most city newspapers provide many of their news articles, sports scores, and weather forecasts online at no charge. By placing advertisements on their Web pages, the newspapers can generate revenues from their Internet provision of information. Agreat deal of information now flows over the Internet, so the free-rider problem is apparently manageable.

In some cases, specific governmental bodies such as the Food and Drug Administration (FDA) and the Consumer Product Safety Commission (CPSC) may be necessary for the provision of information that is much costlier to produce than weather reports and sport scores. For instance, the CPSC annually inspects children’s toys and alerts consumers to potentially dangerous features. In some cases, it may even set standards that eliminate the danger. Usually, the danger involved is reduced, but the cost is a more expensive product to consumers. For example, the Commission once determined that baby cribs were unsafe because infants could slip through the crib bars. It then set a maximum legal distance between the bars of cribs. Manufacturers as a group had to place bars closer together; this took more bars and cribs became more expensive.

Such a change in product quality involves a trade-off to consumers. While standards usually relieve buyers of evaluating the hazards of products, they also make manufacturers conform to designs that restrict product variety or make products more expensive—and sometimes both of these things result. In the specific case of baby cribs, safety standards prevented consumers from buying less expensive cribs that were undoubtedly not as safe as those that conformed to the CPSC guidelines, but the choice was eliminated and the less expensive models may have suited some consumers’ purposes and budgets.

Whether information problems justify government intervention is a controversial topic. Many economists argue that market failure stemming from information problems requires only additional information, not regulation. On the other hand, more is involved than simply acquiring a publication or reading a more informative description of a product. Once information is acquired, it must be studied, and if it is complicated or technically sophisticated, the costs associated with digesting the information can be high. Under these circumstances, many economists argue that safety and quality regulation are beneficial functions of government.

I L L U S T R AT I O N 1 6 . 2

Comparison Pricing in Health Care Off to a Slow Start

Consumers benefit from having better information about both prices and qualities because such information improves their ability to find the lowest price for the desired level of quality. However, as we explained in this chapter, obtaining information is costly and buyers will not usually find it optimal to gain complete knowledge about all prices and qualities. As a consequence of being imperfectly informed, consumers will make purchasing “mistakes,” that is, with more and better information they would have made a different consumption decision. While it is usually too costly to eliminate all “mistakes,” consumers should be willing to adopt new tools or methods for gathering information and comparing prices or qualities (or both)—especially when the new search tools can be employed for free or at a very low cost. Health care shoppers, however, have been slow to start using the many new search tools recently launched in the U.S.

For the first time, consumers shopping for health care services are now gaining access to doctor and hospital prices for medical procedures, as well as software that makes comparison shopping across doctors and hospitals much easier. A number of major health care insurers—Aetna, Cigna, Humana, and UnitedHealth Group—are developing and expanding their online price search services to reveal rates negotiated with local physicians for various medical procedures and prices paid to local hospitals for health services. Several state governments are also getting into the information business by providing Web-based services listing hospital fees. The primary purpose of these efforts to disseminate pricing information about health care services is to stimulate consumer search and competition among providers, thereby lowering costs to health insurers and state governments. Unfortunately, patients don’t seem to be using these new information services. In a recent Wall Street Journal article reporting on Aetna’s experience with a new online pricing information program, almost all of the doctors in a survey done by Aetna said “their patients hadn’t asked questions about their rates after the program was launched . . . There really hasn’t been any discussion (of prices).”

For all the effort and expense undertaken to create more transparent health care pricing, the usefulness of current pricing data still suffers from serious limitations. Currently, these Web services cover a relatively small number of procedures—no more than 75 common medical services at any one Web site. And, in many instances, the price search software provides a range of prices, rather than specific prices, for each doctor or hospital. This creates uncertainty about the actual price patients will end up paying. None of the online price programs provides any information about service quality, so patients might worry that low price signals low service quality.

Perhaps the most important reason for the lackluster demand for pricing information can be attributed to the low insurance deductibles that many consumers still enjoy. Since they pay only a small fraction of the total doctor or hospital bill, patients don’t have as much incentive to shop for low prices as they do when they are buying a new refrigerator for which they pay the entire price. According to Aetna, “as more consumers have plans with high deductibles, prices will become more important to them.” We suspect Aetna is correct. However, until health care shoppers can access accurate information about a wide range of medical services by most of the suppliers in their local areas, market failure due to imperfect information will continue to keep health care prices higher than the competitive level.