Market Failure

When Markets Get It Wrong

Sometimes free markets get the right answer completely and demonstrably wrong – this is a Market Failure. For the most part, markets are good at dividing up limited resources. Prices tell us things, people willingly exchange goods and are usually both better off as a result, and competition encourages efficiency in a way that central planning never has. But there are certain scenarios where a free market results in a situation that's noticeably worse for society than it could be, and these aren't unusual occurrences; they appear everywhere.

Economists use the term "market failure" for when a market left to its own devices produces a result which isn't the best for society. This could mean too much of something damaging is produced, too little of something helpful, or a product people want isn't made at all.

Most of this comes from four things:
- Externalities (where costs or benefits impact people who weren't involved in the original exchange – think of a steel plant causing asthma across a town)
- Public goods (goods the market won't provide enough of because you can't make people pay for using them)
- Information asymmetry (where one party in an agreement knows more than the other, a concept George Akerlof explained in a 1970 paper that won him the 2001 Nobel Prize)
- Market power (when a company or a small group of companies have a very strong hold on a market allowing them to reduce how much is available and raise prices)

Each of these violates a key idea within the competitive model. And when those ideas aren't true, the price and amount available at the equilibrium point no longer maximize society's wellbeing.

Externalities: Costs and Benefits That Spill Over

Consider externalities, costs and benefits that spread outwards. A steelworks in Gary, Indiana, fires up its furnaces and releases sulphur dioxide into the air. The owners pay for the coal, the workforce, and the machinery. They don't pay for the asthma treatment in the nearby neighbourhoods, the deteriorating buildings, or the harm to local woods. Those costs are borne by people who didn't agree to anything and haven't received any of the profits. The difference between what the company pays (the private cost) and what society actually pays (the social cost) is a negative externality.

When there's a negative externality, the market produces too much. Steel ends up being too inexpensive because the price doesn't account for the total amount of harm being done. The people buying and selling are pleased at the market's equilibrium, but society would be better with less steel and less pollution; the amount that would be best for society is lower than the amount the market will make without intervention.

Positive externalities operate in the opposite way. If you have a flu jab in October, you're obviously better off, avoiding a week of feeling awful. But your colleague benefits, as does your elderly neighbour, and the person next to you on the train, because you're less likely to give them the virus. This extra benefit is a positive externality.

With a positive externality, the market produces too little. Not enough people are vaccinated because each person only considers their own benefit versus the expense and doesn't think about the benefit they are giving to everyone else. Social benefit is higher than private benefit, so the amount that would be best for society is higher than the amount the market will produce on its own.

This is a consistent pattern and is often examined in the AP exam. Negative externalities mean overproduction, and positive externalities mean underproduction. If a free-response question describes a polluting factory and you say the market 'underproduces', you will lose marks.

Public Goods and the Free-Rider Problem

On the Fourth of July, fireworks over a city are for everyone to see, whether or not they helped pay for them. You can't stop one person from seeing them without ruining it for everyone else, and your enjoyment doesn't diminish the experience of the person next to you.

A public good has two key characteristics:
- Non-excludable (meaning it's not really possible to prevent people who don't pay from benefitting from it)
- Non-rivalrous (one person using it doesn't reduce the amount available for anyone else)

National defense is the example everyone uses in textbooks. Once a country has an army, all its citizens are protected, regardless of their taxes. A lighthouse's warning is available to all ships that pass, clean air is available to all, public radio signals reach anyone with a receiver, and basic research funded by the National Science Foundation (like the work that led to GPS, which the military created in the 1970s and then made public in 2000) benefits everyone.

The issue is pretty straightforward: why would you pay if you're going to get the benefit anyway? This is the "free-rider problem". People who are being sensible will wait for others to pay for it. But if a lot of people take advantage of the situation and don't pay, then the good will either not be provided at all or will only be provided in a much smaller amount than society actually thinks it's worth.

Private companies aren't able to fix this. A fireworks display company can't send a bill to everyone who watches because it can't prevent people from watching if they haven't paid. The market doesn't work, even though the good is clearly more valuable than the expense of creating it. Governments get involved and pay for public goods with taxes, and this isn't necessarily because of a belief in government itself, but because it solves a very specific and identifiable flaw in the market.

Information Asymmetry: The Market for Lemons

With information asymmetry, the seller of something has much more knowledge than the buyer. Think about a used car: the seller will know if it spent two days underwater, if the gearbox is noisy when cold, or if the engine uses a lot of oil. You won't. This big difference in information can destroy a whole market.

George Akerlof explained this in a famous 1970 article. Imagine that half of all used cars are good ("peaches", worth $10,000) and half are terrible ("lemons", worth $5,000). Because a buyer can't tell the difference, they'll offer an average price of $7,500. Owners of the peaches will realize their cars are worth $10,000 and won't sell at $7,500, so they'll remove them from sale. This means that what's left for sale will have a higher proportion of lemons. Buyers will see this, and lower their offer. More peach owners will stop selling. This cycle will continue until only lemons are left, or the market disappears completely. This is called adverse selection - the poor quality items push the good ones out.

Moral hazard is the opposite. After an agreement is made, someone's behaviour changes because the other party can't supervise them closely enough. For instance, if you have full fire insurance, you might become a bit more relaxed about candles, not bother with an electrical check, or delay replacing an old dryer vent. The insurance company is now paying for any damage, not you. The insurer can't watch you all the time.

We can correct these issues in the real world with things like warranties (where the seller shows they're confident by covering repair costs), vehicle history reports like Carfax (which give you the information you're missing), professional qualifications for professions like medicine and law, and rules that require certain information to be revealed, like the SEC's financial reporting rules for companies that are publicly traded.

Market Power

Market power means that you pay $120 a month for cable in many American cities because in many places, only one company provides broadband. There's no one else to switch to. The company charges far more than it costs to provide the service and your only choices are to pay or not have internet. Market power is when a company can charge prices that are higher than in a normal competitive situation.

In a perfectly competitive market, many companies sell the same thing and no one company can influence the price. Competition drives the price down to the lowest amount it costs to make something, companies don't make a profit in the long term, and customers get the largest amount of the product at the lowest possible price that keeps the businesses going.

A monopoly is the most extreme version of this. One seller dominates, there isn't anything really like what they're selling, and they have total control of how much is made. A monopolist makes less of the product than in a competitive situation, charges more for it, and creates a "deadweight loss" (meaning transactions that would have been good for both buyer and seller don't happen). Deals that would have happened with competition are now lost.

Natural monopolies are a specific case where it's cheaper for one company to supply an entire market than for two or more to do so. Think of water companies, electrical systems, and railway lines. Having two separate sets of pipes running under a city would be a huge waste! Because of the enormous "economies of scale" (the more you make, the cheaper each unit becomes), the monopoly is efficient at production, but without rules, it will exploit its ability to set high prices. It's this conflict between efficient production and the temptation to overcharge that's why natural monopolies are controlled by the government, instead of being broken up into smaller companies.

But a company doesn't have to have complete control to cause problems. Oligopolies, where a small number of big companies are in charge (like the major U.S. airlines, mobile phone companies, or social media sites), create outcomes somewhere between perfect competition and a full monopoly. And, truthfully, they're usually nearer to a monopoly.

What Government Can Do (and Sometimes Makes Worse)

Governments have ways to fix "market failures":

Pigouvian taxes make a company's own costs match the true cost to society by charging a tax on each unit of pollution. This causes the market to shrink to the most efficient size. The EU Emissions Trading System, which began in 2005, works on this idea.

Subsidies lower the price customers pay for things with positive effects on others (like cheap flu shots or Pell Grants for college), and increase how much of them are used to the best amount for society.

Regulation can set limits on how much pollution is allowed, require details to be revealed, or set safety rules. These "command and control" rules directly limit harmful things, although they can be quite basic, forcing all factories to reduce pollution by the same percentage, regardless of the cost to each one.

Tradable permits limit the total amount of pollution and allow companies to buy and sell permission to pollute. A company that can cheaply reduce pollution does so and sells its remaining permits to a company for whom cleaning up is more expensive. The total pollution limit is reached at the lowest cost to the economy as a whole.

The government itself provides the service, paid for with taxes. This is how we get national defense, interstate highways, and basic scientific work through the NIH and NSF.

However, government intervention doesn't always make things better. "Government failure" happens when the government's actions create new problems or make existing ones worse. Those in charge of regulation sometimes don't have all the information about the industries they're looking at. Political pressure can push policy to favor groups with influence instead of the public in general. A badly designed subsidy for farming might encourage too much corn to be grown, and it just sits in storage. A limit on rent can create a shortage of homes that's even worse than the problem of affordability it was meant to solve. In fact, New York's 80 years of experience with rent control has resulted in a huge amount of research on this.

Here's an example: A factory's pollution causes $20 of damage to people's health for every item made. The government puts a Pigouvian tax of $20 on each item. Before the tax, the factory made 1,000 items at its normal cost. After the tax, the cost to make each item goes up by $20, and production falls to 800. This is now the level where the true cost to society is matched by the benefit. Those 200 items that aren't being made anymore were items where the total cost to society was higher than the total benefit. The deadweight loss disappears. The tax brings in $20 x 800 = $16,000, which can be used for cleanup or to help people who have been affected.

The point isn't that government always succeeds and markets always fail. Both markets and the government can get things wrong when allocating resources. Essentially, good policy looks at the imperfect results of a market and the imperfect results of the government doing something, and chooses the one that's closer to being efficient.