Russia is now drafting soldiers. Below is a new economic discussion of in-kind taxes, of which the military draft is a good example.
In-kind taxes – obligations of citizens owed in goods or services, rather than money – are some of the most important taxes in human history, and even in recent times are used in significant ways. For example, according to The Military Balance, 59 percent of countries in the world in 1995 obtained military manpower by conscription: forcing citizens into military service. More than six dozen countries, including the United States, have constitutional provisions for taking private land using the power of eminent domain. Eminent domain is also used to redistribute intellectual property, and may be employed more frequently in the future as that type of property becomes more valuable. In-kind “public service” labor payments to local governments can be a major part of the tax burden faced by the poor in developing countries. Another example: two dozen countries compel their citizens to participate in civic elections.
Although public finance deals extensively with the question of cash versus in-kind transfers, in-kind taxes are almost completely neglected. For example, neither of the public finance textbooks by Stiglitz and Rosen mentions in-kind taxes in general, or labor conscription in particular. The purpose of this paper is to examine the efficiency properties of in-kind taxation (hereafter, IKT), with special emphasis on avoidance behaviors.
Behaviors for avoiding the Vietnam War draft are still famous today. They include entry into protected occupations, obtaining political favors, moving to Canada, or embellishing medical conditions. Quantitative work on avoidance behaviors has shown that the Vietnam War draft induced a significant increase in college enrollments (Baskir and Strauss 1978). Desertion by conscripts can be rampant, and draftees are less likely than volunteers to reenlist when their required tour is finished. Leon Friedman (1969, pp. 1545-6) describes evasion of the Union Army’s draft, “enrolling officers … were frequently lied to, avoided, and even physically attacked… new towns sprang up just across the northern borders in Canada … some men maimed themselves in order to fail the physical requirements for the army.” Landowners are known to modify their property in order to avoid being targeted for condemnation or other limits on land rights. Stroup (1997, p. 57) describes how landowners have reacted to the Endangered Species Act by “managing their land … in a way that almost assures that it will not be suitable for listed species.”
Avoidance activities like these serve to restrict the quality and quantity of supply of resources to be taken, yet the small literature on the economics of eminent domain and military conscription typically takes the supply as given, emphasizing instead the purported tendency of in-kind taxation to distort comparative advantage (i.e., that the public project fails to be supplied by those most suited to do it) while it economizes on treasury revenue. This paper treats IKT as a price-regulation phenomenon – the public recruiter is a buyer who has coercive power over its suppliers – and explicitly models the social costs, and private benefits, of suppliers’ avoidance behaviors.
In-kind taxation has been implemented in a variety of ways. Supplier compensation is one variable. Another is the granting of an option (if any) for suppliers to substitute a monetary tax payment for their obligation to supply under the IKT. The price-control framework readily addresses various implementation options, offers new conclusions, and changes some old ones. Under some conditions, “fairer” IKTs – those that do not accept monetary payments and thereby widely distribute the IKT obligation – are more efficient than less fair IKTs. These results may help explain where and when various implementation options are exercised.
The IKT implementation options present a tradeoff between avoidance costs and opportunity costs. IKT’s are capable of realizing many of the gains from comparative advantage, even while controlling prices. This implication appears to match the reality in which the resources obtained through IKT are far from random. Even among men born 1950-53 whose military draft eligibility was chosen by a random draft lottery, military enlistment was by no means random. Angrist (1990, p. 315) refers to “the fact that armed forces selection criteria were not random ...” and reports that more than three-quarters of draft-eligible men in these cohorts did not serve in the military. At the same time, roughly ten percent of those not draft eligible did serve. Although IKTs distort comparative advantage to some degree, the price control framework suggests that comparative advantage distortions might enhance efficiency because they can alleviate incentives to avoid the tax.
Previous work on conscription has debated the nature of the relationship between monetary taxes and in-kind taxes. Lee and McKenzie (1992), Ross (1994), and Warner and Asch (1996) claim that economizing on treasury revenue, as in-kind taxation is supposed to do, is socially valuable because of the deadweight costs of collecting monetary taxes. However, these papers do not model the social costs of effort to avoid IKTs; real-world IKTs interact with income taxes, and the former may generate more deadweight costs because they are more concentrated. Birchenall and Koch (2014) use a mechanism-design framework to look at the two types of taxes simultaneously, and conclude that IKTs may reduce overall efficiency by raising the marginal deadweight cost of income taxes. This paper does not contribute to the money-versus-IKT debate, and just assumes that in-kind taxation reacts not only to the efficiency considerations noted above, but also to social or political preferences to regulate prices paid to suppliers.
Section II begins with this paper’s conceptual point of departure: a model of avoidance behavior as a Tullock-style rent-seeking contest. A simple, but critical, result is that the social costs of avoidance are convex in the amount of avoidance. Section III lays out the other dimensions of IKT policies, relates them to empirical observations of military recruitment schemes, and interprets them in an equilibrium context. Section IV overturns some of the conclusions from the literature. Section V looks at policy options for the mix of social costs, and suggests that comparative advantage distortions may not be the primary social cost of IKTs. Section V concludes with some ideas of why the incidence and design of IKT tax policies vary over time, by country, and across policy domains.