Please note that my opinions are my own, and the opinions of the anyone or any institution quoted are theirs. The opinions expressed herein do not reflect the opinion of North Carolina State University, its board of directors, the College of Management or any other college, Student Media Authority, or WKNC Raleigh.

Monday, January 22, 2007

The Agency Problem in Public Education

My first attempt to publish this article ended in me writing a 12-page discussion of the topic as a whole. If anyone wishes to see the work-in-progress, e-mail me. Otherwise, here is about as concise as I could get it for lay readers.


I wish to expose the broken system of public education currently putting America at risk for a long, hard fall from super-powerdom. If selected to continue my research in this project, I will produce a detailed analysis of the agency problem in public education and its effect on the quality of our nation’s education.

I will start by taking a look at how to define the "agency problem" from corporate governance theory. Next, I will form an analogous definition for the problem in education, noting caveats and elaborating on my choices for analogies. I will then discuss the reasons our current system fails to solve these problems. Finally, I will propose hypotheses for research and possible methods.

The agency problem may be defined like this: a principal (
analogous to an employer) employs an agent to act in the principal's stead, that is, to make decisions in the absence of the principal; a conflict of interest arises in that the agent performs and makes decisions in his best interest, which can not be the same interests (though they may produce parallel decisions) when the agent is not also the principal. The agency problem, then, is the conflict of interests that arise from differing personalities (tastes, preferences, valuations, etc) of the principle and agent.

The table below is a brief summary of the agency problem applied to corporations and public education.


Public Education




Respective Interests




Board of Directors

School Administration

Home Consumption



Job Consumption




d(perks)/d(profit) is neg

d(doctrine)/d(knw) is neg4

Agency Problem

Above optimal perks

Above optimal doctrine


  1. If a CEO is in charge of maximizing profit, the agent who is directly charged with maximizing knowledge is the teacher. The school, then, as an institution, never enters the picture and can be thought of as the institutional "market" for knowledge transfer, where the principal’s role is as Grasso’s was to the NYSE.

  2. Just as a manager is in charge of maximizing profits, a teacher is charged with maximizing knowledge. Likewise, as a manager is in charge of maximizing firm value, a teacher is charged with maximizing the net-present-value of that child's future earnings (which will vary considerably across individuals and time).

  3. Whether it is to correct some past injustice or to ensure the survival of an epistemology, a teacher derives clear utility from this practice of indoctrination. These teachers value indoctrination in a way managers value their perks consumed.

  4. As knowledge is product of classroom activity, the marginal benefit to the student is increased analytical thinking skills. These skills make a pupil more competitive in labor markets and more productive as a worker, thus increasing his potential future earnings. “Affective teaching” (Sowell), or indoctrination reduces the development of independent thought and analytical skills, as consumption of perks decreases profit in a corporation.

The agency problem can be solved, or eased, in numerous ways by changing the agent's incentive structure directly, rewarding (or punishing) good (or bad) performance, and external incentives. Whereas an optimal combination of all these solutions can align the owners’ and manager’s interests, An optimal combination cannot be reached in public education. Direct incentives are only possible if the teacher teaches only her own children, becoming strictly a parent-teacher, and hence eliminating the agent problem. Performance-based incentives are currently restricted by teacher unions. External incentives are not implemented, save external constraint of behavior through curricula.

Why does the agency problem continue to plague our public schools? I propose two main reasons for the continuance of the status quo, the quality of educators and the unionization of teaching, and I argue that these two factors must necessarily change before methods of solving the agency problem may be applied.

I propose two main hypotheses. Due to the inability to build standard direct and performance-based incentives, the only option left is to restrict the agent's behavior; the imposition of curricula, standardized tests, and teacher accountability occur in the public sector, where the other solutions to the agency problem are restricted.

  1. I argue that the private sector, where performance-based pay and real threat of dismissal, will provide a higher quality education or at least deflate salaries whilst maintaining status quo, freeing up resources for more efficient (or profitable) uses.

  2. I argue that the constraint of teacher behavior in the public sector improves education, but not to the same degree that performance-based pay and dismissal incentive would--that is to say, the alternate method of constricting behavior does not fully compensate for the lack of standard solutions to the agency problem. I expect to observe this result because the constraint of behavior prevents other methods of compensation and incentive-building to be chosen optimally, given they could be chosen.

A simple means comparison between a control group and experimental group may reveal the most information. The closest to a natural experiment we find in education is the marked differences between public schools (unionized, bureaucratic, and monopolistic) and private schools (competitive labor market, competitive market for students).

I wish to test the differences between historically “indoctrinable” subjects and “pure” subjects both in public and private schools. I expect the stringency of curricula in history, literature, and philosophy, for example, to be higher than that in curricula in mathematics and science in the public schools. I also expect the constraints on teacher behavior to be absolutely smaller in private schools in all subjects.

Further research would allow me to develop more robust methods for testing my hypothesis, and may allow me to develop a theme more consistent and relevant to competition and choice in education.

No comments: