Creating failures in the market for tax planning.

Exacerbating this standard adverse selection mechanism is an interesting feature of tax planning methods. The government is more likely to detect use of a tax planning method if more people are using it, more money is being saved on account of its use, or some combination of the two. For each sale, the probability that the method cannot be sold again will increase, as will the probability that the buyer cannot use the method. Thus, sellers will want to be paid more, and buyers will want to pay less, than would be the case if the expected returns to use of the method stayed constant. Indeed, the better the method, the more divergent its valuation by the seller and buyer may become. A market for lemons dynamic thus should already exist; the government can strengthen it by announcing increased efforts to find the more popular or more effective methods.

2. Moral Hazard and the Principal-Agent Problem

Moral hazard (23) arises when a person whose behavior cannot be monitored has the ability and incentive to engage in behavior that is not in the interest of the individual or firm that will be affected by his or her actions. The moral hazard problem again entails asymmetric information. The basic story is as follows. Two parties enter into a contract. The value of the contract depends on the amount of effort that one of the parties expends. However, the other agent cannot observe how much effort is put forth. For example, an employer may not be able to tell exactly how hard an employee is working, or an insurance company may not be able to tell whether and how hard a customer is actually trying to avoid accidents. If it were possible to see how much effort is being exerted, one would simply reward that effort. Since it is not possible, the person will not expend as much effort as she otherwise would.

This form of market failure suggests other possible disruptions of the market for tax planning methods. There may be ways to limit a seller's or lessor's ability to assure the buyer or lessee of the method that the method is of high quality. The lemons problem discussed above could thus be exacerbated. But, going further, it may be possible to limit incentives for search for planning methods to be sold by limiting individuals' ability to be compensated for their efforts. In many firms engaged in developing tax planning methods, pay is gauged at least in significant part by performance. If tax rules provide sanctions to individuals developing such methods, the individuals will want to minimize the extent to which a method is associated with them personally; this should make infeasible a contract which rewarded the employee for his performance in developing the method, thus creating a moral hazard problem. It also should limit the efforts he spends in developing them.

D. Other Governmentally-Created Market Distortions

There are several other scenarios in which smooth functioning of the market is disrupted. In the following part we will briefly mention some of these scenarios. Once again, the situations that will be discussed are generally seen as socially undesirable insofar as they disrupt market functioning; however, in the context of markets for tax planning methods, disruption is the aim.

1. Hold-Ups: Fostering Strategic Behavior

The hold-up problem concerns the possibility of opportunistic behavior in a transaction stemming from the transaction's timing. (24) Suppose that one person has to complete her side of the bargain before the other. Once the person does so, she is essentially at the mercy of the other, since there is nothing (other than the court action or reputational concerns) forcing the second person to complete her performance.

How might the hold-up problem apply to the market for tax planning methods? Consider our first story about the comparison of a market with patents to a market with no property rights. We assumed that people would not sell their ideas when there were no property rights; we later noted that the assumption might not be realistic. One way that people would try to create property rights in the absence of patents is through contracts. The buyer would agree that if he resells the idea, he would be subject to large penalties. If the government made such contracts unenforceable, a hold-up problem would exist. After the sale of a method, nothing would prevent the buyer from giving the information to friends or reselling the idea at a lower cost. People might therefore not sell their tax planning methods, especially if the probability of a method being shut down were to increase in the number of users.

2. Risk Misallocations: Allocating Risk to an Inferior Risk Bearer

The higher expected sanctions are, and the less those sanctions can be allocated to lower-cost bearers (people who can best diversify the risk that those sanctions impose), the higher the benefits must be to motivate additional searches for tax planning techniques. Thus, the search is likely to stop sooner if higher-cost risk bearers must bear the risk than if the risk could be allocated to lower-cost risk bearers. Higher expected sanctions (the sanction times the probability the sanction will be imposed) are a well-worn weapon in the traditional arsenal; the gloss here is to ensure that the sanction is imposed on somebody particularly ill-suited to bear it.

Contracts typically allocate risk to the cheapest cost avoider. (25) A party who, for instance, can more cheaply acquire insurance on property might assume the risk that the property will be damaged. The same is not infrequently true for risks the parties do not think to allocate: courts sometimes implicitly or even expressly use the principle of allocating risk to the cheapest cost avoider or best situated risk bearer when deciding ex post who should bear a particular risk. (26) The lower the aggregate costs of transacting, the larger the pie the parties will have to divide; thus, allocation of costs to the cheapest cost avoider (and the allocation of risks to the best situated risk bearer) should encourage contracting.

How could the government create risk and then place it on the most risk averse party? How might it be able to prevent allocation of known risks to the cheapest cost avoider? It is generally believed that people are more risk averse than firms. The government would therefore prefer to place risk on individuals rather than firms. What kinds of risk could there be? As we discussed above, for many tax planning methods, there is risk about whether the method will deliver the promised benefits, and if so, how many times it can be used. Suppose that to use a tax method, a person has to identify an individual as the developer of that method and include the individual's certification that the method works. Suppose further that the government prohibits as against public policy indemnities or reimbursement of the individual by his employer for any liabilities incurred to anyone on account of the tax planning method. The individual could be sued by the buyer of the tax planning method or, for that matter, by the government. The effect of targeting the individual rather than the firm should be that fewer sales would take place, reducing both lost revenue costs and search costs.

V. CONTEXTUALIZING THE ANALYSIS: GOVERNMENTAL INTERVENTION AND MARKET DISRUPTIONS IN PRACTICE

A. Allowing an Optimal Amount of Tax Planning

We argued above that it may be optimal for the government to allow some amount of tax planning. We noted that there are different ways for the government to proceed; one way is to simply legitimize a tax planning method. Two examples in which the government has done so follow:

1. Check-the-Box Regulations

One example is the "check-the-box" regulations for business entity designation. Corporations are subject to entity-level taxation--that is, there is tax when the corporation receives income and again when its shareholders receive that income in the form of dividends. Partnerships are pass-through entities: if the partnership earns income, the partnership itself is not taxed. Rather, the tax arises when each partner pays some share of the tax attributable to that income. At different times in history, the Internal Revenue Service (Service) had differing concerns as to which entity it sought to discourage; at a certain point, the Service disfavored the partnership entity, on grounds that partnerships were being used "as tax shelters." The Service originally sought to impose two tests: (1) a formal test pursuant to which an entity seeking to qualify as a partnership had to lack at least two out of four "corporate characteristics" and (2) a more discretionary test focusing on substance. (27) In 1976, a Tax Court decision gave some comfort to tax planners who met the formal test but perhaps ran afoul of the more discretionary test; (28) in 1979, the Service agreed to follow that decision. Lawyers got better and better at meeting the formal test, to the point where partnership treatment eventually became nearly elective. (29) Eventually, the Service adopted "check-the-box" regulations which effectively acknowledged that partnership treatment had become elective: an entity desiring such treatment could simply "check-the-box." (30)

2. Tracing