Lecture Notes: Fundamentals of the Neoclassical Framework
1. The Efficiency of the Market
In The Wealth of Nations, Adam Smith argued that self-interest would lead people to act in the public interest, as if guided by an invisible hand, more surely than if they intended to (or professed to) do so. Food was placed on your table by the butcher, farmer, baker and brewer not out of charity, but instead because it benefitted them to provide it. Thus, Smith argued that a policy of laissez faire, in which government left the market alone, would best increase the wealth of nations. Following this tradition, in the early 19th Century Ricardo demonstrated that mutually beneficial gains from trade resulted from specialization and trade according to comparative advantage. However, it wasn't until much later that economists were able to mathematically prove Adam Smith's argument, at least for a particular set of well-defined perfect market conditions.
The Italian economist Vilfredo Pareto systematically defined the social interest as efficiency, in which scarce resources are best allocated to competing uses in order to maximize the net value individuals in society derive from them. If a trade benefits at least one party, without hurting another, then such a trade is a Pareto-improvement in the social welfare (meant in the sense of well-being, not of government transfer programs). When all possible trades have already been made, so that no gains from trade remain to be exploited, then we may say that the economy is Pareto Optimal (sometimes called Pareto-efficient), and no individual may be made better off without making someone else worse off.
Later, Nicholas Kaldor argued that Pareto's definition was insufficient. If a reallocation of resources could benefit one person only by harming another, but it made the former better off more than it made the latter worse off, then such a reallocation would not improve efficiency according to Pareto's definition, but it would still increase the overall value that society received from the use of scarce resources. For example, suppose that the government had granted some producer a monopoly, whereby he could increase his profits at the expense of consumers, but is now considering revoking the monopoly and allowing competition. Pareto's definition is not very useful in comparing the two cases, since the monopolist will be harmed. Because modern economists are loathe to make interpersonal comparisons of individual utility, Kaldor proposed that Pareto Optimality be modified by the compensation criterion: if someone could be made better off enough to compensate the loser and still have something left, then such a change would be become Pareto-improving. So the monopoly could be dismantled, and the consumers would gain more from lower prices than it would cost them to compensate the monopolist. Kaldor's point is that his criterion allows us to say more about the comparison of efficiency and Pareto Optimality. However, compensation needs only to be possible, not actual.
Before we can describe the proof of the efficiency of perfect markets, we need to understand more about competitive equilibrium. Mathematicians describe equilibrium as a steady state, where systems come to rest, and in the late 19th Century Alfred Marshall, the father of marginal analysis, applied this concept to what we now know as the supply and demand model of the market. In one market, partial equilibrium is found at the intersection of a rising marginal cost (supply curve) and a falling marginal value (demand curve). Of course, cost is really another term for the foregone value of other goods that could have been produced, but weren't. The price which equates the quantity-supplied and the quantity-demanded is a stable equilibrium, since any other price would result in market forces that push it towards the equilibrium. In Marshallian economics, many buyers and sellers of the same homogeneous good would lead the market to a competitive equilibrium, with the result that the net value of a good, its total value less its total cost, would be maximized.
However, marginal value and marginal cost only make sense in a relative sense, because costs depend on other prices and the price we are willing to pay depends on what other goods costs. Furthermore, the production or consumption of one good may have spillover effects (A.C. Pigou called them externalities) on other goods or other parties than the buyer and the seller. All markets are thus interdependent, and so how can partial equilibrium analysis be helpful in examining the efficiency of the overall economy? Leon Walras first formulated the conditions for general equilibrium, where all markets cleared simultaneously (technically, he defined it by saying that excess demand for each and every had to be equal to zero if its price were nonzero, but if its price were zero, it is not a violation if the good were in excess supply). He also recognized that because only relative prices mattered, we need only solve for n-1 of n markets.
Kenneth Arrow and Gerard Debreu finally provided the mathematical proofs economists had been seeking. First, they were able to show that the Walrasian general equilibrium under perfect competition, the competitive general equilibrium, must exist for at least one set of n-1 prices equating marginal costs and marginal values. It is thus theoretically possible for a system of competitive markets to all clear simultaneously even though they are all interdependent. Of course, for the actual economy to find such a set of prices would require both the conditions of perfect competition and of perfect information, or incredible luck.
The above existence theorem was only the first piece of the proof. Next, they demonstrated the first fundamental theorem of welfare economics. It says that if markets are complete, then any competitive general equilibrium will be Pareto-optimal even under Kaldor's compensation criterion. By complete markets, it is meant that there is a market for every possible good or service anyone might wish to buy or sell (no thin or illegal markets), that there are no externalities or spillover effects (since these would be captured in a special market), and that there are no public goods (since a market won't work without the ability to exclude free riders, nor can a positive price equal marginal cost if marginal cost is zero). We can therefore say with more than a little confidence that a system of completely free markets would be perfectly efficient if the three conditions of perfect markets (perfect competition, perfect information, and complete markets) were met.
The second fundamental theorem of welfare economics is about the distribution of income in an efficient market economy. Economists argue that income is determined by what you own and what value the market places upon it. The proof states that there is a competitive general equilibrium for each and possible distribution of ownership of economic resources, and we know from the first theorem that such an equilibrium would be perfectly efficient (under the condition of complete markets). This means that an unequal distribution of the goods and services produced by a market economy results not from markets per se, but rather from the initially unequal ownership of the inputs to production themselves. The second theorem might thus justify the redistribution of property as a preferred policy to interference in markets or redistribution of income.
The first theorem is often dismissed as too unrealistic, since perfect market conditions are probably never met in their mathematical purity. One interpretation of it, however, is that the three conditions provide a starting point for improving the efficiency of market economies: government policies might solve market failures, by increasing competition, improving information, and fixing incomplete markets, to improve the workings of the economy. Significant intellectual assistance in one aspect of this strategy was provided by Ronald Coase's theorem. This states that if property rights are well-defined and enforceable (regardless of who actually holds them), and transactions costs are sufficiently low, then all externalities could be efficiently internalized into a market. Government needs only to clarify property rights, and the market will right itself without intervention. If producers pollute a river that is owned by somebody with clear property rights, then they would have to pay the costs of that pollution, either to clean it up or reimburse the property owner, whichever is more efficient. If the producers held the property rights, then those affected could pay them to stop polluting if it were efficient to do so, and the outcome does not depend on who actually owns the rights. What this does suggest is that pollution, overgrazing, overfishing, and the myriad of other externalities result from the inability to either define or enforce property rights, and thus results in what Garrett Hardin calls the tragedy of the commons.
However, any certainty that the strategy of slowly fixing those parts of a market economy that fail to meet perfect market conditions is eroded by the theorem of the second best. In a nutshell, this theorem says that even though we can compare an economy under perfect market conditions (the first best) to all other possibilities, we cannot prove which of the imperfect varieties is second best. That is, if we solve a market failure, we cannot be confident that the overall social welfare has improved; one market failure often offsets another, and solving one can possibly make things worse. Taxes and subsidies are inefficient because they interfere with market prices, for example, but what about taxes on pollution or subsidies for education or research? Monopolies are inefficient and patents create monopolies, but wouldn't the elimination of patents slow down the rate of technological progress?
Friedrich A. Hayek dismissed the relevance of the first fundamental
theorem, not because he believed that markets weren't efficient, but rather
because he believed that it focused on the wrong things. He argued that
market economies may not be perfectly efficient, but they are relatively
efficient, in large part because of how they use information. He built
on the work of Ludwig von Mises, who argued that resources and products
could not be accurately valued without markets, and thus resources could
never be allocated to their competing uses efficiently in a non-market
economy. Hayek argued that market prices did not only serve to equate marginal
values with marginal costs, they conveyed information to market participants.
The costs of centralizing the information necessary to make efficient economic
decisions would be infinitely expensive, but markets are decentralized,
and prices tell potential buyers and sellers much they could not know otherwise.
While perfect information may not exist, markets do not require that their
participants know that much. Furthermore, since motivation is inherently
individual, market economies are better at encouraging harder and smarter
work, innovation and risk-taking because they reward individual initiative
so well. So a decentralized laissez faire market economy may not
be perfectly efficient, but it is more adaptable (in the evolutionary sense)
and dynamically efficient than other possible economic systems.
2. Market Failure
When the perfect market conditions necessary to prove the perfect efficiency of a free market economy are not present, markets are said to fail. Instances of market failure include: a) natural monopolies, monopolistic competition, oligopoly, monopsony, or other failures of competition; b) adverse selection, moral hazard, productivity wages, principal-agent problems, fraud, simple ignorance, or other failures of information and incentive; or c) tragedies of the common, public goods, thin markets, transactions costs, or other forms of incomplete markets. Often, economists argue that market failure comes from government's failure to define and enforce property rights, as per Coase's Theorem, or from government interventions that make matters worse. The market, it is argued, is self-correcting in the long-run and does not require intervention. However, there are other viewpoints.
Smith's argued that self-interest would work in concert with the social interest, but the Prisoner's Dilemma model explains how self-interest might in fact make everyone worse off. In the model, a simple game, two individuals who do not know each other very well are arrested for a crime they may or may not have committed together, placed in separate cells, and asked to confess and testify against the other. Each is told that if he will be worse off if he does not confess, regardless of what the other prisoner does. For example, he might be told that if neither prisoner confesses, then they will each spend six months in jail for a minor charge, and if both confess, they will receive a light sentence of five years, but if only one confesses, then he will go free while the other will do hard time. Because each prisoner is better off confessing regardless of the other's actions, both will spend five years in jail instead of six months. Self-interest leads to uncooperative behavior, and lack of cooperation makes everyone worse off.
Of course, sometimes society doesn't want people to cooperate with each other. If the prisoners are in fact guilty, we want them to not cooperate with each other, and that is why the Mafia works in the prisoners' interest but against those of society. We want firms in the same industry to not cooperate with each other, but rather to compete, produce more, and drive prices lower; if they colluded with each other their profits would be higher but society would be worse off. But we also don't want competitors setting bombs in each other's stores or poisoning their competitor's food products, we don't want crime and anarchy. In lawless societies like Somalia in the early 1990s, the economy cannot function. A certain level of cooperative behavior is necessary for society to work for the benefit of all, even though it may restrict individuals from following what they perceive to be their own self-interest. Such cooperation, like the clear definition and enforcement of property rights, may be seen as a fundamental public good.
A pure public good is a good with two particular characteristics. It has nonexcludable benefits, so that those who don't purchase it can still receive it (we call these folks free riders), and the cost of producing it is not dependent on the number who receive it, so that marginal cost is zero and price cannot equal marginal cost if the producer is to receive any income from selling it. This type of good cannot be efficiently provided by a free market, but requires some form of public agency (government) to provide the good by requiring everyone to pay (through taxation) regardless of use, since free riders would not voluntarily pay and the efficient price is zero.
Of course, few goods are pure public goods, but many have some elements
of a public good. For example, goods with positive externalities have benefits
which are nonexcludable, and reducing negative externalities may also be
seen as a quasi-public good. Though enforcing property rights may be a
public good, we may also view the right of eminent domain, the government's
right to take property with proper compensation, as a public good for those
times when uncooperative individuals use their property in a way which
may harms the interests of society.
3. Government Intervention
By definition, governments have a monopoly on legal coercion within the jurisdiction recognized by other governments, and they may use this for good or ill, for the interests of the governed or the interests of those doing the governing. Those who view government in the former way often rely on John Locke's theory of the social contract, where government is created by individuals to solve problems that they alone may not solve. Those who view it in the latter way may rely on theories of the predatory state, created (at least initially) by military conquerors as a method of ruling their empires. Mancur Olson, for example, has an interesting model of the self-interested warlord who creates a stable government and growing economy in order to maximize the amount he can steal over time. Because rule by the few requires at least some degree of tacit cooperation by the many being ruled, the truth is likely to be somewhere in between these two extremes.
Governments which desire to act in the economic interest of the governed may find justification for intervention in the provision of public goods and the correction of market failures. At a minimum, they define property rights and enforce a code of basic cooperative conduct through civil and criminal law, and provide for the common defense. They may provide other public goods, such as police and fire protection, road building, and the coinage of money. They may use policies of antitrust or regulation to promote competitive behavior, or they may use taxes or subsidies to affect the provision of goods with spillover effects. Governments may use material or coercive incentives to affect the behavior of private firms, or they may nationalize firms to run them directly.
Governments may also intervene to address macroeconomic problems. Say's Law argues that supply creates its own demand, since the aggregate price level will adjust when aggregate demand fluctuates, and the economy is inherently stable. This may not be true, however, if producers and consumers have limited information, or suffer from "money illusion", since they might interpret nominal changes in the prices that affect them as real, not illusory. John Maynard Keynes argued that Say's Law is backwards, that demand creates its own supply (up to a point), and aggregate demand is inherently unstable because of investment's volatility and the multiplier effect. From the depths of the Great Depression, Keynes argued for an interventionist countercyclical fiscal policy to offset and stabilize the business cycle, and military spending during World War II seemed to prove his point. New Keynesians tend to focus on price rigidities that prevent the economy from remaining stable when shocks occur, and on internal mechanisms that tend to amplify instability into a persistent phenomenon.
Milton Friedman suggested instead that government was the source
of instability, not the solution. He argued that the Great Depression resulted
primarily from the monetary negligence of the Federal Reserve Bank, the
U.S.'s central bank. The
monetarist school he founded generally
dismissed the desirability and effectiveness of fiscal policy, emphasizing
instead money supply volatility as the source of macroeconomic stability.
The new classical school went further, to argue that government's monetary
or fiscal policy could only work by fooling people, and rational expectations
would lead the economy to respond to any policy change in an unpredictable
way. Government's task, then, is to be both as stable and as predictable
as possible.
4. Public Failure
To say the state is justified in economic intervention does not necessarily mean that the state is able to intervene efficiently. Paying for the provision of public goods requires that government tax its citizens, for example, and since taxes generally are levied on observed activities (buying goods, earning income, owning property, et cetera) they tend to alter people's behavior and discourage productive activities. Like private markets, governments also have both information and incentive problems. In a market economy, the problems that fall into the public sphere are likely to be the most intractable, because markets were not able to solve them in the first place. Information about the degree of inefficiency to be corrected and the effects of intervention are guesses at best.
Incentive problems are especially difficult, however. All large bureaucracies suffer from the principal-agent problem, as the agents put their own interests above that of the principal who pays them. Yet the goals of a private firm (profit maximization) are relatively easy to measure, while governments have many varied and competing goals. When performance is difficult to measure and monitor, motivation to perform better suffers. When success is hard to measure but failure is obvious, bureaucracies tend to become extremely risk averse. Government agencies created for a specific task find it dangerous to complete it, since doing so means that the agency is no longer needed. Agencies who regulate industries for the good of the public often become "captured" by that industry, as they begin to identify with the specific individuals that they regulate rather than the unknown public that they rarely see in a positive light.
Politicians who ostensibly manage the bureaucracy have information and incentive problems too. Even the most idealistic politician finds that success requires getting things done, and getting things done requires getting in power, staying in power, and extending power. Because there are too many issues that are too complex, a politician must rely on the expert advice of the bureaucracies, and they don't necessarily find it in their interest to keep him or her well-informed. Politicians must become deal-makers and vote-traders (this is called logrolling), and they must provide largesse (the pork barrel) for their key constituencies if they wish to stay in power, if only because, as Peltzman suggests, their voters give them credit for spending projects in the home district but blame their higher taxes on the projects they must pay for in everyone else's district. Finally, as the lecture on social choice demonstrated, even if government could somehow solve all of its managerial and incentive problems, the fundamental problem of determining the social interest remains. There is no reliable way to determine what society prefers or what policies are in the social interest, and thus it is impossible to know if even an incredibly effective government is acting efficiently.
In sum, markets may be relatively efficient but the presence of public
goods and other market failures may provide a justification for government
intervention. This intervention is not likely to provide a perfect fix,
however, since governments also have incentive and informational problems,
even if the government somehow knew what intervention was in the social
interest. Social choice theory, however, suggests that there is no possible
method of accurately gauging the preferences of society. Instead, policy
is messy and indeterminate, and it is not really possible for a rational
state policy to emerge to determine what public goods should be provided,
or what market failures should be corrected.