Creating failures in the market for tax planning.

If the government could anticipate particular tax planning methods, it could use the ability to patent such methods to its advantage: it could simply patent the methods--easily regulating the extent to which the tax planning method is used (perhaps not at all)--by deciding who may use it, of course in exchange for a fee. Note that the fee would defray the tax revenue losses of use of the method with the income earned from selling the right to use the technique. Of course, if the government could anticipate such methods, it should have considerable success simply prohibiting them ex ante. In any event, the government traditionally has not been successful in anticipating such methods. (17) The government might also be able to buy such a patent from the innovator who obtained the patent, thereby also regulating the extent to which the method is used. However, doing so would likely be more expensive than trying to curtail the method using more traditional sanctions.

The possibility of patenting tax planning methods suggests an interesting strategy. Somebody--the government, someone effectively subsidized by the government, or a group that was opposed to tax planning activity--might be able to patent one component of many strategies and either refuse to license it or license it only at very high fees. (18) In either case, those seeking to develop and market tax planning methods would find doing so more difficult and less lucrative than it otherwise would have been, given the need to either pay for use of the component or of finding ways to structure around it. (19) This would reduce both search and discovery and dissemination.

Another aspect of tax patents needs to be considered. Some tax planning methods will attract more government efforts to shut them down than others. Which ones do so may depend on fairly predictable factors, such as the amount of revenue lost, publicity arising about a particular transaction that yielded political pressure to take action, or serendipity. How might the possibility that a method may have a short shelf life affect the analysis? Patenting may, on the one hand, shorten the shelf life further, as the information disclosed in the patent application gives the government information to disallow the method; however, patenting may also allow the method to yield more revenue as it is sold aggressively.

B. Public Goods: Creating or Exacerbating Free Rider Problems

Another type of market failure involves public goods. When goods are available free of charge, the market forces that normally allocate resources and create production incentives in the economy are absent. In the case of public goods, markets "fail" because they will not supply a sufficient amount of goods. The public goods problem is the effect of the so-called free rider problem. A free rider is a person who receives the benefit of a good but avoids paying for it. When goods are available free of charge and people cannot be excluded from enjoying the benefits of a good to which they have not contributed, individuals may adopt free riding strategies and withhold paying for the good hoping that others will pay for it. (20) Because there is no easy way to induce parties to reveal their valuation of the public good through the price system, markets do not supply sufficiently large amounts of public goods.

One of the market problems commonly associated with the search for new ideas is the free rider problem. New products and ideas are typically expensive to create, but easy to replicate. The legal system generally wants to establish incentives to increase the supply of public goods. Here, we are faced with the opposite concern and policy objective: the government would like to decrease the incentive to develop new tax planning methods.

The incentive to develop tax planning techniques depends on the aggregate benefit available to the developer. An important component of that benefit may be the ability to sell the technique to others. Others will not be willing to pay for it if they can get it for free; if they can get it cheaply they will not be willing to pay much. The less they have to pay, all else being equal, the smaller the benefit there is to searching for and developing the techniques. The legal system may develop ways to condition the use of a tax planning technique on disclosure. For instance, the government could force public disclosure of any tax planning method somebody used; others would therefore be able to use it for free. The result would be increased dissemination of the method, leading to greater lost revenue as more people use the method. But the incentive to search for and develop the methods in the first instance would be far smaller. People could not get a return from selling the method; moreover, they would know that they might be able to use somebody else's method for free. If markets are efficient, the increased lost revenue would be small (perhaps even zero), but the reduction in dissipation costs could be very high.

C. Asymmetric Information

Another form of market failure reflects the fact that information is asymmetric: different people have access to different information. There are two main effects of asymmetric information: adverse selection and moral hazard problems. These effects are generally seen as socially undesirable, inasmuch as they negatively affect the allocative efficiency role, and possibly the existence, of markets. Transactions may be difficult to effectuate because each party believes the other may be hiding self-serving negative information. In the context of tax planning markets, these concerns turn into a hope, inasmuch as both of these effects can help disrupt the market for tax planning.

1. Adverse Selection: The Market for Tax Lemons

Adverse selection involves somebody who knows he has undesirable attributes dealing with others who may not be able to readily determine whether he has those attributes. The classic example is the used car: many buyers are reluctant to buy a used car because they suspect that, if the car is being sold, the seller must know something bad about it. This is also known as the lemons problem. (21) The basic story for the market for lemons is as follows. Suppose there are three quality levels for used cars: high, medium, and low. Car owners know what type of car they have, but they cannot credibly convey this information to buyers--all owner/sellers will want to say their car is high quality. Buyers are willing to pay an amount that reflects what they expect the car's quality to be. If the pool of available cars includes medium and low quality cars, buyers may only be willing to pay less than the amount that high quality car owners will accept. Thus, high quality cars will not be sold; all cars sold will be of medium or lower quality. Buyers, being rational, lower their expectations of the quality of cars available. Again, however, buyers are not able to tell which are of medium quality and which are of low quality; they base their willingness to pay on the expected quality. As before, this willingness to pay might be less than the amount that medium quality car owners need in order to be willing to sell. Thus the used car market would be comprised of only low quality cars, or lemons.

What would a lemons story look like in the context of tax planning? Tax planning methods, like cars, can be of differing quality. If buyers of tax planning methods cannot tell which type of method they are buying, the methods that are available for sale might all be lemons. Since buyers could not be sure that they were purchasing a high quality method, they would only offer a lemons price--a price the seller of the high quality method would be unwilling to accept. The high quality methods would therefore not be sold. Tax revenue would be higher since those methods would not be being used to reduce revenues. Incentives to search for tax planning methods would decline as well, since the rewards to search would be lower.

An essential component to the market for lemons is the uncertainty that the buyer faces at the time of purchase as to the quality of what she is purchasing. In the context of tax planning methods, the requisite uncertainty exists. Some tax planning methods are comparatively easy to appraise. However, even long-standing methods whose workings are well-known may face the risk of being declared illegitimate. Moreover, as methods are shut down, new ones are developed. Finally, the tax code changes frequently, giving rise to the development of new methods that exploit those changes. Frequent changes to the tax code, bemoaned as they are, thus might help to create socially desirable market failures. (22) Of course, the incentive to search decreases in the length of time that a tax policy has been in place, because the number of tax planning methods left to find decreases. Thus the optimal rate of change would be set by the tradeoff between these two considerations.