Lobbying expenditures and government output: the NEA and public education.

• Author: Hoyt, William H.
• Position: National Education Association

1. Introduction

   Primary and secondary education have traditionally been provided by local governments. In recent years, however, higher levels of government have become increasingly involved in the provision of public education. Evidence of this increased involvement includes the greater proportion of funding originating from state or federal sources and the imposition of mandates on local school districts by the state and local governments. These mandates include specifying minimum teacher retirement and other fringe benefits, establishing certain curriculum and textbook requirements, and placing maximum limits on the pupils per teacher. The extent of state government involvement in the provision of elementary and secondary education differs greatly across states.

   Most studies of state intervention in education have been restricted to examining the effects of intergovernmental grants on local spending.(1) There has been little positive analysis of why state governments are involved in education or why this involvement varies across states. In this paper, we build upon previous works that consider the role of interest groups in influencing government spending to provide some insights into the role of state governments in public elementary and secondary education and the effects of the state role on local outcomes.

   Our model focuses on the determination of educational spending when two interest groups, groups of high and low demanders, lobby both state and local governments. We characterize the high demanders of educational expenditures as producers, that is, educators. After developing a model that describes the gains to educators from lobbying at the state level and how the potential gains will differ across states and across localities, we test the model's implications using data on political spending of the state affiliates of the National Education Association (NEA). Using data on salaries aggregated to the state level, we find that NEA political spending increases salaries of primary and secondary educators. With data on county-level educational expenditures we find that NEA spending also increases primary and secondary educational expenditures. As predicted by our model, political spending is higher, ceteris paribus, in those states with lower educational spending and salaries.

   2. A Model of Political Contributions with Mandates

      The normative arguments for state intervention in local government's provision of any good, including education, involve the existence of interjurisdictional externalities or economies of scale that can be captured at higher levels of government. The actual role of state governments in education, however, may not be motivated solely by normative reasons but may also reflect political pressure.

      Gary Becker |2~ argued that interest groups compete against each other to influence government policies and to secure wealth transfers for the members of their group. In this section, we develop a simple framework to illustrate how interest group competition can be used to explain the imposition of state mandates on locally provided education. In particular, the model focuses on interest groups' choice to produce political pressure at the local level or at the state level of government.

      We consider two levels of government, local school districts and the central, or state, government. There is a large number (K) of local school districts but only one state government. Each local district provides a single good, elementary and secondary education, and finances it from taxes imposed on its residents.(2) The state neither provides education directly nor does it finance localities' provision of education. The state influences localities' provision of education, however, by stipulating requirements or mandates that all districts must obey. For purposes of simplicity, we assume that the mandates have the effect of placing a minimum constraint on the per pupil expenditures for education in local districts. These expenditures, then, will measure educational output in the locality.(3)

      Any decisions made at either the local or state levels of government regarding educational expenditures are made by elected representatives. Let g measure the level of education, education expenditures per student, provided by each locality and assume that education is a publicly provided private good. Ng is the cost of providing a level of education equal to g to the children of N residents.(4) To simplify our analysis we assume that each locality's residents are divided into two groups, 1 and 2, with |N.sub.1~ and |N.sub.2~ in the two groups. Group 1 members are low demanders of education and group 2 members are high demanders of education. The latter group could be broadly categorized as producers of education as they derive employment or income directly from government's provision of education in addition to receiving consumption benefits. Membership in the consumer group exceeds that of the producer group, |N.sub.1~ |is greater than~ |N.sub.2~. For the purpose of simplifying the model, we assume that all localities have the same population composition.(5) We also assume that each resident pays a head tax, T, to finance education.

      Consider a representative locality in which a member of either the consumer or producer group receives net benefits, W, from education given by

      |W.sub.i~ = |B.sub.i~(g) - T, i = 1, 2. (1)

      Each locality faces a balanced budget constraint of

      NT - Ng = 0. (2)

      By substituting (2) into (1), the preferred level of education by each group is

      d|W.sub.i~/dg = |B|prime~.sub.i~(|g.sub.i~) - 1 = 0, i = 1, 2. (3)

      Because group 2 receives higher benefits (both marginal and total) from the provision of education than group 1, members of group 2 will prefer a higher level of education.

      Given that the determination of educational expenditures is political, both groups must produce political pressure to influence the level of g. The groups can produce pressure at the local level to directly influence the level of education provided or they can produce pressure at the state level to affect the level of education indirectly through the imposition of mandates on localities.

      Influencing legislation requires spending resources. Individuals within each group must organize and engage in activities to pressure politicians. From either group's perspective, the production of political pressure is a function of the resources spent and the size of the group, or

      |Mathematical Expression Omitted~.

      where |Mathematical Expression Omitted~ is the amount of expenditures per member on political pressure production by group i at the jth level of government. Expenditures are used directly for lobbying as well as for encouraging member participation. We assume that pressure increases with expenditures per member and the membership of the group but increases at a decreasing rate. Because of free riding by group members, we also assume that proportionate increases in per member expenditures by each group and proportionate increases in the size of the groups will have a disproportionate effect on the relative political pressure or

      |p(|e.sub.1~, |N.sub.1~)/p(|e.sub.2~, |N.sub.2~)~ |is greater than~ |p(z|e.sub.1~, x|N.sub.1~)/p(z|e.sub.2~, x|N.sub.2~)~,

      where x and z |is greater than~ 1. (6)

      The functional relationship between political pressure and educational expenditures can be given by a simple but general specification:

      |g.sup.j~ = |g.sup.j~(|p.sub.2~/|p.sub.1~), j = 1, s. (5)

      Because group 2 demands more education than group 1, a relative increase in political pressure by group 2 results in an increase in education. Analogously, a relative increase in pressure by group 1 decreases the provision of education.

      Both groups choose their local lobbying expenditures independently of the other (K - 1) localities but lobbying groups at the state level consist of coalitions of group 1 and group 2 members from all of the K local school districts. Formally, each group chooses the per member lobbying expenditures at each level of government (|Mathematical Expression Omitted~) that will maximize net benefits from education, g, or

      |Mathematical Expression Omitted~.

      For the low demand group, group 1, the first...