Complexity, governance and constitutional craftsmanship - Constitutional Economics

American Journal of Economics and Sociology, The, Jan, 2002 by Richard E. Wagner

I

Allocative Outcomes, Constitutive Rules, and Economic Policy

FOR THE MOST PART, DISCUSSIONS OF ECONOMIC POLICY treat the network of economic relationships that constitute an economy as a relatively simple phenomenon. Economic policy is often treated as a form of mechanics, with policy being much like tinkering with the engine of an automobile. It is thus an intervention into the engine's operation, the purpose of which is to change the performance of the engine in some fashion. Modern legislatures are like large garages with many engines being worked on at the same time. It is common, for instance, to find analyses of economic policy that describe policy as an instrument for selecting among positions along a production frontier.

This presumption of inherent simplicity is revealed again and again in textbook illustrations of the effects of government interventions into particular markets. Rent control, for instance, is portrayed as creating a shortage in the market for rental housing and reducing the return earned by the owners of those units. These effects can be illustrated with reference to Figure 1. Without rent control, the market-clearing price and quantity are [P.sub.0] and [X.sub.0] respectively. If rents are controlled at [P.sub.2] suppliers are willing to supply only [X.sub.1] units of rental housing while tenants want [X.sub.2] units. Tenants therefore want a greater number of units or a larger amount of space than the owners of rental property are willing to provide. Rent control changes the market outcome ([P.sub.0], [X.sub.0]) to the controlled outcome ([P.sub.2], [X.sub.2]), thus changing resource allocation within a society.

This portrait of rent control as simply changing allocative outcomes seems correct at first glance but turns out upon further consideration to be inadequate, if not wrong. This portrait is one that captures only the direct, immediately visible effects at the point of policy injection. It does not account for all of the subsequent indirect effects that emerge throughout the economy as people respond to the myriad changes in profit opportunities that the policy generates. The very complexity of economic life, moreover, precludes any effort to do so in anything but a formal and highly stylized manner. These changes in profit opportunities all emerge through the intensified competition that rent control sets in motion among tenants. At the market outcome ([P.sub.0], [X.sub.0]), there is no unresolved competition among tenants. This is not to deny that tenants might wish that rents were lower. But rents are what they are by virtue of their having been generated through a process of open competition that has been f ramed by the rules of private property and freedom of contract. At the market equilibrium, there is no scope for further competition.

The enactment of rent control generates new opportunities for competition among tenants. At the controlled price of [P.sub.2], there are [X.sub.2] people competing for [X.sub.1] places. (1) A full examination of rent control cannot ignore this competition among tenants. Yet the course of this competition cannot be predicted in exact detail, which means that rent control cannot truly choose allocative outcomes, but can only modify the framework of rules within which outcomes emerge through interaction among market participants.

Setting a rent control at [P.sub.2] does not mean that rents actually will be [P.sub.2] and not [P.sub.1]. For one thing, the mere enactment of a law does not mean that it will be effective. There are many drugs whose sale is illegal that nonetheless can be obtained easily. Rents might be controlled at $600 and yet housing might still rent for $800, because people would rather pay $800 than do without. A genuine census of rental contracts might even find instances in which those contracts carried prices of $800 despite the rent control. How frequently this might occur would depend on such things as the penalties for violation of the control and the budget of investigative and enforcement agencies. Far more likely than rental contracts being made above the controlled price is a change in the structure of contracts and contractual relationships that would have much the same effect. There are an indefinitely large number of ways this can be accomplished. The initial impact of rent control is to prevent market pa rticipants from making transactions they wish to make. It is easy to understand why those participants would be interested in pursuing other contractual formats to allow them to exploit the gains from trade that rent control otherwise would prevent them from exploiting.

Rent control can change the form that competition among tenants takes, but it cannot eliminate that competition. An open and direct competition will give way to some indirect, more complex, and costlier form of competition. There are an indefinitely large number of particular forms that this more complex competition can take, varying in their degree of complexity and indirectness. One simple illustration is a one-year lease at $600 per month, with that lease requiring also the payment of $2,400 in cash as a condition for securing the lease, therefore bringing the effective monthly rent to $800.2 Besides an under-the-table payment, there are numerous ways this payment could be given contractual standing. It could be as a deposit against damage, with the owner or his agent serving as the final arbiter of damages. It could also be as a $24,000 deposit that is refundable without interest, provided the interest rate was 10 percent.