Optimal provision of education in developing countries: A theoretical analysis

Provision of universal public education has been justified in the economics literature on the grounds of income redistribution from the rich to the poor. This thesis demonstrates how the presence of private tutoring in developing countries compromises this argument. The teachers by providing private tutoring to the students at a cost, divert the benefits of free public education towards themselves. We develop our arguments in the following three models.;In the first paper, we develop a new model of public school education in the presence of private tutoring. The teachers have the incentive to shirk at school in order to create a demand for tutoring. The students buy the service to acquire "full" education. The main question is one of efficient provision of education in the presence of imperfect and costly monitoring given the technology of monitoring available in the economy.;In the second model, we analyze the first model in a club framework. The students are assumed to demand tutoring like a club service. They decide whether to go to a club (tutorial class) and pay the club fee, and the teacher makes a decision on the optimal timing of tutoring the optimal number of students in a club, the number of clubs, and the price of tutoring.;The above two models are generalized to consider students who are heterogeneous with respect to income and ability. It is found that it pays for the teacher to discriminate between the types. The students who pay a high price subsidize the education of the students who pay a low price. The analysis is able to suggest merit-cum-means principle which supports a number of similar policies adopted in many developing countries to encourage the high ability but poor students to pursue their studies. The analysis from the social planner's problem suggests two policies, incentives and regulations, to control the behavior and the productivity of the teachers. The optimal salary of the teachers and the optimal level of monitoring in the economy have also been characterized in both the models.;Finally, the third model analyzes demand for education under different institutional arrangements such as public and private schools. This model suggest a school choice where the students income is the key determining factor. It is demonstrated that the full privatization of education is always to the disadvantage of the poor. Provision of public education is essential to guarantee a minimum level of education to everybody, and redistribution can be achieved through financing it from general taxes. The social desirability of private tutoring is also examined. To analyze the net effect on welfare, we characterize and compare the optimal taxes and the social welfare frontier of the economies with and without tutoring. It is demonstrated that tutoring can be welfare improving.