Exploring the relationship between performance management and program impact: A case study of the job training partnership act

Programs that involve multiple levels of government may suffer from a principal‐agent problem: Lower levels of government may wish to pursue different objectives than the higher level of government that provides funding. One strategy for dealing with this problem is to establish a performance management system where units operating the program are accountable for meeting performance standards and are rewarded or sanctioned depending on how well they perform. Title II‐A of the Job Training Partnership Act provides training for economically disadvantaged adults and has operated under a performance management system since 1983 when the program was established. The federal government's goal for the program is to maximize impact on the employment and earnings of participants, but because control groups are not available for the 640 local programs, proxy measures of performance must be used. In this paper, data from an experiment in 16 sites are used to determine how closely measured performance corresponds to program impact. It is concluded that there is only a weak correspondence between the two measures and that the Department of Labor should avoid making significant rewards or sanctions based on the current performance management system. © 2000 by the Association for Public Policy Analysis and Management.     

Getting Performance from Performance Management: A Framework for Strategic Management Choices

The choices for optimal performance management depend upon the circumstances – there is no single ideal management approach. The principals have three basic tools available to encourage high performance from subordinate entities: they can delegate authority, impose rewards or sanctions, and/or develop (or enhance) a performance measurement system. These tools allow the principals to motivate, delegate, educate, reevaluate, and/or reassign the agent in order to generate increased learning and improved outcomes from the program. Thus, depending on the nature of the agency culture and the quality of the performance measures, different managerial responses will have different results.     

Thirty Years of Changing Federal, State, and Local Relationships in Employment and Training Programs

There have been three major training programs in the UnitedStates in the past thirty years: the Manpower Development andTraining Act (MDTA)from 1962 to 1973, the Comprehensive Employmentand Training Act (CETA)from 1973 to 1982, and the Job TrainingPartnership Act (JTPA)from 1982 to the present. MDTA was a categoricalprogram, with service providers funded directly by the federalgovernment. CETA was a hybrid block grant program that gavelocal units of government substantial autonomy in administeringthe basic training component, but CETA also includedcategoricalprogramsfor specific target groups and for public service employment.Over time, CETA was increasingly regulated. JTPA is regulatedmore by the states and the private sector, and in 1992 amendmentstargeted the program more sharply and restricted activitiesthat could be undertaken. Federalism in employment and trainingprograms has followed a course similar to other areas, withcooperative federalism ending in 1978 being replaced by coercivefederalism. In recent years, stales have started a number ofinnovative programs.     

Expenditures on children and child support guidelines

This article examines the literature that estimates parental expenditures on children; describes the types of child support guidelines that are being used by the states; and compares how the child support awards that emerge from each state's guidelines compare to the empirical estimates of expenditure patterns on children. The findings indicate that the states' guidelines appear to be more or less consistent with the estimates of expenditures on children. In a few cases, however, the guidelines require less in support from the noncustodial parent than the parent would have spent on the child(ren) in an average intact family. In many other cases, the guidelines require child support payments that are very close to the lower bound of the estimates of expenditures. This article concludes with a discussion of the difficult value judgments that ultimately must be made in setting child support awards.     

Demonstration evaluations and cost neutrality: Using caseload models to determine the federal cost neutrality of New Jersey's reach demonstration

The Reagan Administration's Interagency Low Income Opportunity Advisory Board (ILIOAB) requirement that states conduct impact evaluations of their welfare reform demonstrations no doubt led some states to conduct evaluations when they otherwise might not have. The ILIOAB requirement that demonstrations be cost neutral to the federal government, on the other hand, limited the ability of states to test promising approaches that require more resources. The New Jersey experience with REACH clearly demonstrates the limitations of caseload modeling for assessing cost neutrality. In retrospect, too much might have been asked of a fairly crude forecasting technique. If carefully designed, experiments can provide better information on participant impacts and cost neutrality, although different experimental designs are preferred for estimating participant impacts than for estimating cost neutrality or overall impacts.     

One Standard Fits All? The Pros and Cons of Performance Standard Adjustments

Despite the wide‐ranging use of performance measures within the public sector and the growing use of performance bonuses to recognize performance achievements, the use of formal performance standards adjustment procedures in public performance measurement systems is relatively rare. Burt S. Barnow of Johns Hopkins University and Carolyn J. Heinrich of the University of Wisconsin–Madison set forth the basic arguments in favor of and against the use of formal or informal procedures for adjusting performance expectations. They describe how performance standards adjustments processes are currently (or have been) in use, review the evidence of their effectiveness or problems encountered in their application, and explore the consequences of the inadequacy of performance standards adjustments. The authors offer recommendations for the improvement of public sector performance measurement systems and conclude why this area remains fruitful for future research experimentation. The nice thing about standards is that there are so many of them to choose from. —Andrew S. Tannenbaum