This paper develops a model of private school choice that incorporates measures of the availability and costs of private schooling, as well as the quality of existing public schools, in the local market for secondary education. Using data from the National Longitudinal Survey of Youth and the 1980 Private High Schools File, I estimate this model on all individuals enrolled in high school in 1979 and all Catholics enrolled in high school in 1979. I find that the availability of a private high school of one's religious affiliation in one's county of residence significantly increases the likelihood of private school attendance. Further, per capita educational expenditures in one's county of residence are inversely related to the likelihood of attending a private high school. Finally, private school enrollment decisions are relatively insensitive to tuition and fees levels. These results provide a link between aggregate models of private school enrollment within educational markets and microdata models of private school choice.

This paper develops a model of private school choice that incorporates measures of the availability and costs of private schooling, as well as the quality of existing public schools, in the local market for secondary education. Using data from the National Longitudinal Survey of Youth and the 1980 Private High Schools File, I estimate this model on all individuals enrolled in high school in 1979 and all Catholics enrolled in high school in 1979. I find that the availability of a private high school of one's religious affiliation in one's county of residence significantly increases the likelihood of private school attendance. Further, per capita educational expenditures in one's county of residence are inversely related to the likelihood of attending a private high school. Finally, private school enrollment decisions are relatively insensitive to tuition and fees levels. These results provide a link between aggregate models of private school enrollment within educational markets and microdata models of private school choice.

This paper explores the relevance of the principal-agent model for analyzing development projects using data from World Bank-funded projects. After demonstrating that World Bank loan agreements can be viewed as principal-agent contracts, the paper explores the importance of the agency problem in determining project performance. Predictions from an adversarial model contrast with those of a cooperative model. The importance of information in the adversarial model links World Bank supervision to project performance. Data support the relevance of the agency problem and the role of supervision as monitoring. The paper concludes with suggestions for modifying project selection and implementation to reduce agency problems.

A model incorporating a discrete full-time/part-time/do not work labor choice and decomposing labor supply into its full-time and part-time components was used to examine the increase in labor supplied by nurses (the increase per nurse on average) over the 1977-88 period. Both full-time and part-time supply increased. Full-time supply increased because the percentage of nurses who chose to work full-time increased (full-time hours worked, conditional on working full-time, stayed the same). Conversely, part-time supply increased because part-time hours worked, conditional on working part-time, increased (the percentage of nurses who chose to work part-time stated the same).

This study examines the income redistributional consequences of the public financing of higher education in Pakistan. The results show that the tax incidence and the subsidy to higher education entail a redistribution from the middle and upper to the lower income groups. Earlier studies of other less-developed countries indicated a redistribution from the lower to the upper income groups.

This paper reexamines the Baumol-Wolff (1988) hypothesis that there exists an exclusive international Convergence Club of rich nations whose members will move nearer each other with the passage of time. We argue that the Baumol-Wolff analysis does not really justify such a claim except in a relatively weak sense. Even in this weak sense we find that it is the poorer nations and not the richer who show convergence tendencies. We also show that the Diffusion hypothesis of Gomulka (1971) offers better hopes of identifying such a Convergence Club in a more meaningful sense. Using a generalisation of the Gomulka hypothesis, we find that there are two mutually exclusive Convergence Clubs - one for the richer nations and one for the poorer ones.

It is commonly believed that firms prefer layoffs to worksharing, in part, because layoffs economize on fringe benefit costs. We find that when labor markets are characterized by optimal implicit contracts, layoffs will never occur in equilibrium, regardless of the level of fringe benefits.