JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.. Southern Economic Association is collaborating with JSTOR to digitize, preserve and extend access to Southern Economic Journal.

The Junior Reserve Officers' Training Corps (JROTC) is a high school initiative that serves many at-risk students. Its goals range from reducing dropout rates and improving academic achievement to preparing students for military careers. Using data from High School and Beyond (HSB) and the National Educational Longitudinal Study (NELS), we estimate average treatment effects on students attending schools that typically host JROTC. Applying a twostage matching technique, we find that JROTC participants have poorer academic outcomes than other students; although, a large portion of these differences is explained by their at-risk status. In addition, program effects appear to differ by demographic group, with black participants having lower dropout rates than white participants. The program also appears to improve self-esteem scores of females. Although the majority of JROTC participants do not join the military, we find large marginal enlistment effects.

Recent research has analyzed the issue of whether local government services are public or private in nature. In analyzing the determinants of local government spending both Bergstrom and Goodman (1973) and Borcherding and Deacon (1972) developed variants of the median voter model which permitted estimation of 'publicness' parameters in reduced form expenditure equations. In both studies the estimated parameter indicated that municipal services are divisible in consumption, a characteristic of private goods. Subsequent studies have tended to corroborate this result (Gramlich and Rubinfeld, 1982;McMillan, et al., 1981;Pack and Pack, 1978, Pomerehne, 1978; Pommerehne and Frey, 1976; Vehorn, 1979). However, all of these studies have used the same theoretical model, varying only the data set used for empirical estimation. This paper raises some questions about the appropriateness of the median voter model for estimating the divisibility question, and proposes an alternative theoretical approach.A number of fundamental questions have been raised about the accuracy of the median voter model in describing local decisionmakers' behavior. Romer and Rosenthal (1979a; 1979b) argue that empirical studies based on the assumed dominance of the median voter have failed to prove that actual municipal expenditures correspond to those desired by the median voter. The median voter model views local government as setting expenditure levels in a competitive manner in response to the preferences of the voter with median tastes. Two candidates competing for elected office would be expected to propose the median voter's position in order to win an election. Voters are assumed to be well informed and free from fiscal illusion. The electoral process also is assumed to be free of the voting power of public employee and other special interest groups. These and other aspects of the median voter theory have prompted questioning as to whether the electoral process is capable of producing policy outcomes in line with the median * The authors would like to thank Thayer Watkins, Roger Folsom, Geoffrey Nunn and Randy Eberts for valuable comments. We would also like to thank the Earhart Foundation for financial assistance.

Analyzing the provision of local public services in the context of a market for local government services has yielded numerous important insights into local first behavior (Martin and Wagner, 1978;Martin, 1977;Wagner and Weber, 1975). Any change in local government structure -institutional or legal -can be examined for its effect on the competitive conditions in this market and the degree to which it promotes or retards centralization (monopoly) in the provision of government services. The major advantage of decentralized service provision in a metropolitan area is the broader choice of public services offered to voters as compared to a single, centralized provider 1 (Tiebout, 1956). Martin and Wagner (1978) use the model of intergovernmental service competition to analyze the effect of one specific structural change in municipal services markets in California -enactment of the Knox-Nisbet Act in 1963 which created Local Agency Formation Commissions (LAFCO) in 57 of the state's 58 counties. 2 Each LAFCO is composed of appointed representatives from existing local government jurisdictions -the county and cities -and has the authority to accept or reject any proposed boundary change by a city or special district, including incorporation, annexation, consolidation, or dissolution. In effect, existing jurisdictions are given control over entry into the local government services market. Martin and Wagner argued that the LAFCOs have used this authority over boundary changes to severely reduce the number of new incorporations in California and to erect substantial entry barriers to potential competitors in local public service markets. They predicted that the effect would be a rapid growth in government spending due to the enhancement of the monopoly power of existing jurisdictions. Their empirical tests gave some support to this hypothesis -the advent of LAFCO reduced new incorporations by 42 percent and raised the level of spending in California counties 35 percent above what it otherwise would have been.The present study extends the initial work by Martin and Wagner. The LAFCO legislation not only reduced municipal incorporations, it also encouraged municipal boundary changes via annexation. LeGates (1970) Public Choice 36: 53-62 (1981) 0048-5829/81/0361-0053 $01.50.

Little evidence is currently available on the regional growth effects of federal defense spending. In this study, econometric models for state personal income and manufacturing employment between 1976 and 1985 are specified and estimated. Pooled cross-sectional time-series data are used, and the estimation procedure corrects for serial correlation, heteroscedasticity, and contemporaneous cross-sectional correlation. The results indicate that aggregate defense spending has a positive effect on both growth measures. However, when defense expenditures are disaggregated, only investment-type outlays appear to consistently affect state economic growth.