Econometric cost functions have begun to appear in education adequacy cases with greater frequency. Cost functions are superficially attractive because they give the impression of objectivity, holding out the promise of scientifically estimating the cost of achieving specified levels of performance from actual data on spending. By contrast, the opinions of educators form the basis of the most common approach to estimating the cost of adequacy, the professional judgment method. The problem is that education cost functions do not in fact tell us the cost of achieving any specified level of performance. Instead, they provide estimates of average spending for districts of given characteristics and current performance. It is a huge and unwarranted stretch to go from this interpretation of regression results to the claim that they provide estimates of the minimum cost of achieving current performance levels, and it is even more problematic to extrapolate the cost of achieving at higher levels. In this article we review the cost-function technique and provide evidence that draws into question the usefulness of the cost-function approach for estimating the cost of an adequate education.

School districts are confronting difficult choices in the aftermath of the financial crisis. In prior recessions, districts often muddled through by imposing a combination of tax increases and expenditure cuts that avoided involuntary personnel reductions. Today, the financial imbalance in many school districts is so large that there is no alternative to teacher layoffs. In nearly all school districts, layoffs are currently determined by some version of teacher seniority. Yet, alternative approaches to personnel reductions may substantially reduce the harm to students from staff reductions relative to layoffs based on seniority. First, because salaries of novice teachers are often much lower than those of veteran teachers, seniority-based layoffs lead to more teachers being laid off to meet any given budget deficit, with the associated implications for class size. Second, because teachers vary substantially in their effectiveness, staff-reduction polices that do not consider effectiveness likely will allow some ineffective teachers to continue teaching while some more effective teachers lose their jobs. Third, because many districts have redesigned human resource policies to place greater emphasis on the recruitment and retention of effective teachers, they may have hired disproportionately more effective teachers over the last several years than in prior years. In such cases, seniority-based layoffs will be even more detrimental for quality. Finally, if, as in many districts, novice teachers are concentrated in schools serving low-achieving students and students in poverty, then a seniority-based layoff approach will disproportionately affect the students in those schools. As a result, many school district leaders and other policymakers are raising important questions about whether other criteria, such as measures of teacher effectiveness, should inform layoffs. This policy brief, a quick look at some aspects of the debate, illustrates the differences in New York City public schools that would result when layoffs are determined by seniority in comparison to a measure of teacher effectiveness. Due to data limitations and an interest in simplicity, our analysis employs the value-added of teachers using the 4th and 5th grade math and ELA achievement of their students. Unsurprisingly, we find that layoffs determined by a measure of teacher effectiveness result in a more effective workforce than would be the case with seniority-based layoffs. However, we were surprised by facets of the empirical results. First, assuming readily available measures of teacher effectiveness actually measure true teacher effectiveness, an assumption to which we return below, the differences between seniority and effectiveness-based layoffs are larger and more persistent than we anticipated. Second, even though seniority-based layoffs imply laying off more teachers, the differential effect on class size is very small in our simulations, though it would be larger for larger budget reductions. Third, there is somewhat greater school-level concentration of layoffs in a seniority-based system, though with a few notable exceptions, both methods result in fairly dispersed layoffs, with the vast majority of schools having no more than one layoff in grades four and five combined. So where does this leave us? As a result of the limited applicability of teacher value-added measures to the full population of teachers as well as concerns about potential mis-measurement of effectiveness associated with using value-added measures even when available, neither seniority nor measures of value-added to student achievement should be the sole criterion determining layoffs. However, ignoring effectiveness measures completely as seniority-based systems do, is also problematic. Instead, the use of multiple measures of effectiveness for layoff decisions holds promise for softening the detrimental effect of layoffs.

From a nationwide perspective, the number of teachers leaving schools in the United States – frequently referred to as teacher turnover – is not very large. Between 2003-2004 and 2004-2005, for instance, 83.5% of teachers stayed in the same school, while only 8.1% transferred between schools and 8.4% left teaching (Marvel et al.,2007). However , averages rates hide the fact that some schools lose teachers frequently, particularly schools serving black and low-achieving students (Hanushek et al., 2004; Boyd et al., 2007). High rates of teacher turnover can destabilize the learning environment in schools, disrupting instructional continuity in classrooms and jeopardizing the educational experience of students, many of whom are already at a disadvantage (Shields et al., 1999; Loeb et al.,2005). Administrators may find it difficult to implement policies or make changes if the teaching force in a school is in constant flux.

This article examines the interaction between school accountability and local control over revenue raising and resource allocation. In particular, it asks whether accountability policies are more or less effective at improving student outcomes in states with stronger local control. Local control is operationalized with multiple measures, including the percentage of education funding from categorical grants, state supreme court rulings overturning school finance systems, local entities' taxing authority, and principals' self-assessments of their ability to control school resources. Using the National Center for Education Statistics' Schools and Staffing Survey and National Assessment of Educational Progress, the article finds that accountability policies are more effective when there is greater local control.

This paper presents suggested matches for the geographical coding (geocoding) of metropolitan areas in the 1970, 1980, and 1990 Censuses. The Census Bureau used different definitions and taxonomies to describe the geography of metropolitan areas in these three Census years. As a result, the geographical areas referred to by the standard Census Bureau definitions differ among the three Census data sets. The geographic matching scheme explained in this paper attempts to maximize consistency over time for metropolitan areas in the U.S.

The assertion that choice-driven competition between schools will improve school quality rests on several largely unexamined assumptions. One is that choice increases the competitive pressure experienced by school leaders. A second is that schools will seek to become more effective in response to competitive pressure. In this article, we use responses from a survey of Milwaukee public school principals to examine these assumptions. Our results suggest that there is a substantial amount of variation in how principals experience competitive pressure. Somewhat surprisingly, the extent to which principals perceive competition for students is not related to geographic factors such as the number of nearby schools. However, perceptions of competition are related to student achievement as well as to transfer rates out of a school. Although some schools respond to competition by trying to improve through curricular or instructional changes, a more common approach is to use outreach or advertisement.

Even as research has begun to document that teachers matter, there is less certainty about what attributes of teachers make the most difference in raising student achievement. Numerous studies have estimated the relationship between teachers' characteristics, such as work experience and academic performance, and their value-added to student achievement; but, few have explored whether instructional practices predict student test score gains. In this study, we ask what classroom practices, if any, differentiate teachers with high impact on student achievement in middle school English Language Arts from those with lower impact. In so doing, the study also explores to what extent value-added measures signal differences in instructional quality. Even with the small sample used in our analysis, we find consistent evidence that high value-added teachers have a different profile of instructional practices than do low value-added teachers. Teachers in the fourth (top) quartile according to value-added scores score higher than second-quartile teachers on all 16 elements of instruction that we measured, and the differences are statistically significant for a subset of practices including explicit strategy instruction.

Michigan's Proposal A fits within the broad school finance equalization movement. First implemented in 1995, the reform stemmed from concerns about inequities in property tax burdens and expenditures and radically changed the financing of public schools. The primary source of funding shifted from local property taxes to state sales taxes and finance became highly centralized at the state level. A modified foundation system replaced district power equalization as the mechanism for distributing basic aid to school districts. Under the foundation system, spending per pupil was sharply increased in previously low spending districts and was essentially frozen for higher spending districts. This paper describes the nature of these changes, the short-term results, and the tensions that are likely to foster future change. We also briefly touch on the school choice and accountability reforms that Michigan implemented at approximately the same time as school finance reform.

Making schools accountable through state testing was the preeminent educational reform of the 1990s. Thirty-nine states now administer some form of performance-based assessment; twenty-four states attach stakes to their tests; and forty states use tests scores for school accountability purposes (Stecher and Barron 1999). Proponents argue that using student scores on curriculum-based tests as a measure of school effectiveness encourages teachers to teach the curriculum. It sets a minimum standard against which schools can be judged; and it quantifies school equality in a way that parents and politicians can easily understand. By setting student improvement goals for schools, the state can motivate school personnel to reach continuously higher, while also identifying those schools unwilling or unable to meet the prescribed goals.

In "Public School Finance and Urban School Policy," Thomas Nechyba summarizes his research using computational models to estimate the impact of policies on schooling and residential location decisions. Perhaps because of the fixed costs of developing sufficiently complex and accurate models, computational models are rarely used for educational policy analysis, even though they have the potential to greatly enrich our understanding of policy dynamics and our ability to predict the impact of policy choices. Nechyba's work stands out because it applies such models to estimate conditions under a variety of policy alternatives. His framework assumes that families' utility is a function of house quality, private consumption, and school quality, which is itself a function of per pupil spending and average peer inputs.