The Unbroken Window

George Akerlof’s canonical example of the market for lemons (information asymmetries leading to unraveling in free markets with subsequent welfare losses) seems to be a perfect illustration of the problem of asymmetric information. Ironically, just as what has happened in the canonical descriptions of public goods problems (lighthouses and honey bees), market power (oil cartels), information problems in money markets (private currency provision) and other examples, the problem seems to have been solved by the markets. In other words, the very strongest examples used by proponents of state intervention into free-markets (i.e. that markets fail) have turned out to illustrate precisely the opposite! Markets work extremely well even when conditions are not favorable to them doing so. (Do not interpret this to mean that markets always work – but even in cases where they do not, the relevant question ought to be, “as compared to what?”) In other words, describe why Akerlof’s lemons story was likely to be true in the early 1970s while empirically it has been shown not really to hold in 2009? What factors does the canonical story overlook?

Two major economic factors are overlooked in Akerlof’s story. Remember, his story was that “bad quality will drive out good quality” if sellers possess more information than buyers do (think about what happens when buyers possess more information than sellers do). Think back to our Harley example in class. The canonical description assumes that buyers will do nothing at all to try to figure out what products are lemons and what products are good ones. In fact there are enormous incentives for buyers to find this information out. Furthermore, Akerlof seems to allow lots of cash to be left on the table by sellers of high quality Harleys. When the market unravels, owners of high quality Harleys would like to sell them, and they know that there are buyers who would like to buy them. Thus, there is a huge private incentive for suppliers to overcome this information problem by setting up third-party evaluation and guarantee services, or discovering other signals that are costly to fake for the owners of low-quality motorcycles. Both of these factors have been important in the market for used cars today.

As a side note – this process is illustrative of how capitalism works generally. Voluntary exchange is not a smooth and frictionless system – quite the opposite. The “feature” of capitalism is that whenever there are unexploited profit opportunities, people are free to try to exploit them. They may fail. There may be structural problems that get in their way, but they are free to fail – and others are free to learn from the mistakes and successes of others. The constant trial and error inherent in a truly capitalist system help actors overcome all manner of frictions.

EVERY economic problem is one of asymmetric information. In fact, there is virtually no transaction (potential or actual) in human history that is categorized by actors having perfect information. Yet, we have grown richer as people have become freer. If the asymmetric information story was endemic to capitalist economies, then why do we see a growth in the quantity and quality of goods in capitalist countries and a lack of goods and quality in countries that are less free? Before the first manufactured goods were ever sold, before the first fruits and vegetables were sold in early farmers’ markets, from the time the first cottage was sold, etc. each transaction suffered from the information asymmetry that Akerlof describes. Yet powerful forces overcome these asymmetries. Good houses get sold. Tasty and healthy food gets sold. Strong and comfortable clothing gets sold. And all of this happens without intrusion of the state. In fact, it is the intrusion of the state that allows information asymmetries to persist by entrenching incumbent producers and providing advantages to special groups of consumers. Those who like to point out that “the individual health insurance market in California” is unraveling conveniently overlook the political interruptions that have led to this outcome. It would be a useful exercise to study the health insurance “market” today to see what factors are preventing healthy consumers from signaling to insurers that they are in fact healthy, and what factors are preventing insurance companies from doing things to figure out who the healthy consumers are. If you believe that insurance companies and healthy customers are too stupid to attempt to do these sorts of things … well, then, you might as well argue that no individual or business is smart enough to do anything in this world.

Again, this is not to say that markets always work. Even if we assume that markets suffer from intractable information problems, what is the relevant policy question? “Standard” textbooks will offer up governmental involvement in markets as a solution. Governments should force people to reveal quality truthfully, or governments should make it costly for people to fake being good quality sellers. But think about this for a moment. If the fundamental problem is that “outsiders” cannot have access to the same information as “insiders” then what leads you to believe that another group of outsides with no direct interest in the transactions (what reward does government get for solving the problems, or what penalty does it face for NOT solving the problems?) will have any better information than anyone else? Furthermore, even if we want to make heroic assumptions about governments having more information than market participants do, then you are going to also have to assume that the government acts properly on this information. I am not so sure that other factors are not at work when the government decides to provide information to markets, or to regulate product quality and safety.

A great illustration of my skepticism is what happened when government regulators initiated vehicle emissions standards in the 1970s in the name of consumer safety. Because cars emitted all kinds of noxious fumes like SO2 and NO2, regulators did not think consumers would have enough incentive or information to ensure than cars they were buying were clean and safe (among other reasons). So rather than simply setting an output standard that said something like, “hey car companies, make sure your cars have fewer emissions of poison,” what they did instead was to tell car companies EXACTLY how to regulate their emissions. How so? The government mandated the installation of catalytic convertors on every car. What is the big deal? Well, installing catalytic convertors is very expensive – in fact it is thought to add up to $1,000 or more to the price of every car sold. There happen to be myriad ways to reduce emissions without the use of catalytic convertors. So what?

1. At the time the law passed, several car manufacturers had emissions standards that EXCEEDED the standards set by the new laws.
2. But the law required that even these companies be required to put catalytic convertors on their cars. Do you have any idea who these companies were? Yes, it was foreign manufacturers who were in competition with their dirtier American counterparts.
3. And do you know who MAKES the catalytic convertors? Is it any coincidence that this was the type of emissions standard that government regulators chose?

To restate the issue, to argue that government as 3^rd party should be responsible for overcoming the information asymmetries “inherent” in market processes – you would be arguing that, for example, regulators that work in the Consumer Product Safety Commission have a stronger incentive to know if a toy is safe than a parent of a two year old child who is purchasing a toy – I’ll let you answer that when you become parents yourselves. In addition to the issues raised above, I don’t think that is such a strong case.

But that is not all! There are two additional problems governments create when they get involved in markets in the name of “fixing” information imperfections.

First, how can governments credibly signal to consumers and producers that they have in fact fixed the problem? Isn’t there also an information asymmetry between the sellers of government services (i.e. the government itself) and the “buyers” (ok, stop laughing) or government services (i.e. the taxpayers)? If information problems are inherent in market transactions, are they not also inherent in government transactions with its citizenry? What makes the government so perfect? When officials walk into marbled domed buildings, do they suddenly reveal all information truthfully? Do they suddenly acquire information they would be unable to obtain as private citizens?

And what is the canonical “solution” to information problems? Have government intervene. But if there is an information problem in the provision of government services to American citizens, what is the canonical solution? More … government?  Houston, we have a problem. In Latin it is referred to as quis custodiet ipsos custodes.

The second problem is perhaps more serious. Generally efforts by third parties (governments) to intervene in information markets result in exacerbations of problems they are trying to fix. In the case of information markets – governments can create or exacerbate the moral hazard problem. Take the banking industry as an example. In traditional banking the information asymmetry is that the banks themselves know more about the riskiness of their lending activities than do the depositors of banks. As a result what tended to happen (in regulated systems, but that is for another class) was that the moment depositors thought that their bank was making bad loans, they would pull their money out of the bank. The problem is, even if the bank itself was NOT making bad loans, it could go out of business and depositors could lose lots of money because of the panic (due to the information asymmetry).

In the old days, banks did all sorts of things to overcome this problem (these things are now illegal or unattractive for banks to do because of new national and state bank regulations). However, FDR in the 1930s instituted “deposit insurance” which basically guarantees that depositors will not lose their deposits if their bank goes out of business. This deposit insurance comes from the pocket books of taxpayers. What is the problem with deposit insurance? Well, it EXACERBATES the information problem from both the bank AND depositor perspectives.

How?

From the perspective of the bank – having the government insure the deposits of your customers encourages the bank to take on more risk than it would if it were acting on its own without the FDIC cushion. This may work out in terms of higher returns, but it also turns out to cause more bad investments and MORE bank failures. From the perspective of the depositors – if you know that your deposits will be insured, what efforts do you take to monitor your bank? How many of you studied the lending activities of the banks you had your checking accounts at? My bet is that exactly ZERO of you would do this. And this therefore encourages more deposits, and more deposits at unsound banks than would prevail in a pure market system. In other words, the existence of deposit insurance creates a moral hazard – and this moral hazard was created by government efforts to overcome the information problem in banking.  This is generalizeable to a host of other activities the government does in the name of providing fixes to information problems.

Tags: information