Recognizing High Cost Factors in The Financing of Public Education: A Regents Report

Background

Perhaps the most critical policy goal that has guided Regents State Aid proposals over the years is the notion of student equity – in ensuring that a fair level of resources is provided to each student. In many states, the fairness principle has been interpreted by the courts as equality in spending per pupil. In recent years, the traditional policy goal of equity-in-funding is being replaced in many states because of several shortcomings:

• First, its focus is entirely upon the distribution of resource inputs rather than the level of resources required to achieve desired outcomes. Thus one can easily imagine a hypothetical public financing system which passes muster from the standpoint of equity-in-funding because every child – without exception - is receiving $5,000 per pupil regardless of the wealth of the community they live in, no more and no less. However, if student performance within these hypothetical school fell below a desired standard of achievement and the resources required to achieve superior outcomes were $10,000 per pupil, we would immediately conclude that these resources, while equitably distributed, were hardly adequate or sufficient.
• Secondly, an unstated assumption of financing models designed to promote equal spending per pupil is that all students are equal and should be treated equally. When in fact students have equal educational needs, the traditional "equal treatment of equals" financing concept is quite appropriate. The reality of course is that students differ in their educational needs. Pupils with Limited English Proficiency, with disabilities, or from economically disadvantaged households pose a greater educational challenge than do students without these disadvantages. Many states, like New York, have long recognized the importance of providing additional funding for students with high needs (i.e., who are unequal in some sense and must be treated unequally). The proliferation of categorical aids and additional student weightings as components of state school finance systems is evidence of this recognition.
• Both the additional aid categories and specialized weightings reflect a recognition that some students need greater resources than others to achieve at appropriate levels. Thus, the failure to recognize that some students require greater resources to achieve at desired performance levels is a serious flaw in older equity-in-funding approaches. A truly adequate financing system is achieved when every student, regardless of his/her individual differences, has access to the resources required to meet the new academic standards.
• Finally, any effort to cost out the types of educational goods and services required to ensure that a given student has adequate resources to achieve high standards, will consistently yield one other important conclusion. Fairness requires a recognition that the cost of purchasing a standard, adequate, market basket of educational goods and services will often vary dramatically from one labor market area to another. Thus, to cite one important example already recognized in State Aid legislation, the personnel labor costs associated with building an elementary school in New York City (assuming identically experienced personnel in the construction trades) will cost approximately 80 percent more than it would cost in the Capital Region. Once again, a truly fair and adequate financing system must recognize regional cost differentials.

Steps Required to Strengthen the Current Fairness of the State Aid System: An Adequacy Perspective:

In order to advance recommendations designed to improve both the adequacy and therefore the fairness of the current funding system, a variety of interventions must be considered. Two broad classes of interventions merit support if we are to ensure that all students have access to the resources they require to achieve high standards. Each of these interventions addresses a different dimension of the same high cost problem. Each is based upon a fairness principle which assumes that State Aid resources must be equalized not just with respect to differences in district fiscal capacity (as they are now) but with respect to differences in district exposure to high cost factors as well.

In short, these recommendations recognize and correct the fundamental unfairness of allocating $3,000 in State Aid per pupil to two districts which are identical in fiscal capacity, when one is located in a high cost area of the state where this $3,000 has a purchasing power of only $2,250 and a student body 80% of whose households fall below poverty and the other, in a low cost area of the state where the purchasing power equivalent of this $3,000 is $3,500 per pupil and only 10% of its student body is poor. The recommendations include formula changes designed to better recognize that:

• the cost of providing educational and any other professional services to our children is much higher in certain areas of the State than others; and,
• pupils from low income backgrounds who attend schools with high concentrations of students from similar poverty backgrounds, require substantially greater resources than their more advantaged counterparts to achieve the same results.

Gaps In The Current State Aid Financing System - the Two Sides of the High Cost Problem:

An analysis of current aid formulas reveals two major gaps in the recognition of "high cost" factors for aid distribution:

Gap 1: Failure to Fully Recognize Geographic Wage and Compensation Cost Differences

The first gap reflects the absence of any cost adjustment in operating aids that recognize major disparities in the cost of doing business in different geographic areas of the State. This oversight is particularly noteworthy in view of 1997-98 legislative enactments which did incorporate a regional cost adjustment in the Building Aid formula. The Building Aid cost adjustment enacted was based upon wage and compensation data for construction-trade occupations drawn from the Department of Labor’s Occupational Employment Survey. SED research staff have built upon this earlier work by analyzing Occupational Employment Survey wage compensation data in New York State for 78 different professional job titles.

The results of these preliminary analyses reveal, for example, that the professional service costs of doing business in the New York City area run about 34 percent higher than equivalent professional service wage costs in the Capital Region. Conversely, professional service costs in school districts within the North Country labor service area run only about 85 percent of the same professional service costs within the Capital Region. The failure to explicitly recognize geographic cost differences within the major operating aid formulas has led to formula allocations which are inequitable. They ignore long standing recommendations of the both the 1988 Salerno Temporary Commission and the 1982 Rubin Task Force concerning this type of cost adjustment.

The single most important aid formula, Basic Operating Aid, illustrates the problem nicely. Like most other aid formulas (with the most notable exception of the Building Aid formula as noted) it contains no adjustments to reflect differential professional personnel costs in different areas. The end result? A school district of precisely average wealth in the North Country (enjoying a 36% State sharing in the Basic Operating Aid grant of $3,900 per pupil), receives $1,404 per pupil in State Aid ($3,900 x .36), as does a school district of identical wealth in a much higher cost area of the state, for example, the Hudson Valley Region. However, that $1,400 buys roughly $1,640 worth of educational goods and services in a North Country school district, but buys only $1,130 worth of such services in a Hudson Valley school district.

In short, since current aid formulas are wealth equalized, both districts receive precisely the same allocation award per pupil due to their identical fiscal capacity. However, due solely to geographic differences in the cost of doing business, the Hudson Valley school district in this illustration actually receives $470 per pupil less in dollars of equivalent purchasing power than does its North Country counterpart. The same education dollar awarded in one area buys a different amount of educational inputs than the same dollar buys elsewhere. In effect, the equalization goals legislatively designed to promote fairness by explicitly recognizing school district wealth differences are compromised when no recognition is given that different labor regions face markedly different professional labor costs.

Conclusion: This particular shortcoming should be addressed. Recent legislative enactment of cost adjustments in the Building Aid formula should be extended to other, general operating aid type formulas. The justification for such cost adjustments has been extensively described in the finance literature.In addition, both the Rubin Task Force on Equity and Excellence in 1982, and the Salerno Temporary State Commission in 1988 advanced recommendations concerning the need for both a regional cost type of adjustment as well as for added weightings to more appropriately reflect the higher costs of students from disadvantaged backgrounds². As noted, the Legislature also enacted legislation this past year which provided for the use of a building-cost index in the Building aid formula designed to recognize major geographic differences in construction costs throughout the state. Appendix A describes in some detail the method used to construct this index and a preliminary validity test conducted to evaluate its validity. Only the tabled findings are shown below:

Gap 2: Failure To Fully Recognize the Higher Cost of Providing Services to Disadvantaged Students.

The second gap is the insufficient recognition given in current aid formulas of the higher educational costs required to ensure that students from economically disadvantaged backgrounds achieve higher standards. Existing research, previous New York State blue-ribbon Commission studies over two decades, and long standing Federal policies with respect to Title I compensatory education (allocated to states based upon student poverty) provide overwhelming support for public finance policies of equalizing aid with respect to the numbers of pupils in poverty in a district. Indeed, research conducted over twenty years ago by Decision Resources Corporation on Title I programs demonstrated rather convincingly that the effect of a child’s poverty status upon achievement is not only negative, but is intensified as the concentration of children from such poverty backgrounds is concentrated. Chart I illustrates these findings. You will note that as the concentration effect is intensified, the negative effects upon student achievement are experience by both poor and non-poor students alike.

It is important to emphasize that New York’s aid formulas do provide some, albeit limited recognition of the need to equalize aid with respect to such high cost students. Under Present Law, for example, approximately $665 million in aid is driven to districts in the form of Extraordinary Needs Aid – based principally upon the concentration of students in poverty, and Limited English Proficient students. However, weightings currently within the Basic Operating Aid formula provide no such recognition. Since Operating Aid accounts for $5.92 billion, the specialized weighting accorded poverty pupils in ENA effectively represents a roughly 10% pupil weighting when these two aids are considered together.

As a direct result of the limited, but inadequate weighting of pupils from disadvantaged backgrounds, the effects are particularly disequalizing for selected districts with unusually high concentrations of such pupils. The New York City example is especially illustrative. If one arrays the districts of the state from the lowest to the highest wealth districts in terms of property wealth, for example, the 69 districts within the 6^th wealth decile (with property values/pupil above $201,400 per pupil and below $243,500), received on average $1,620 per pupil in Basic

Operating Aid in 1996-97 (vs. $1,616 per pupil for NYC) and $3,122 in total State Aid (vs. $2909 per pupil for NYC). In terms of "horizontal equity" (or "equal treatment of equals"), it seems clear that if only district wealth is recognized for equalization purposes, the treatment of New York City appears to be reasonably equitable i.e., it is treated like other districts of similar wealth.

However, if we array the major districts on the basis of their need/resource status (a measure which takes into account both their wealth as well as the concentration of children in poverty and Limited English Proficient students), New York City is then found to be in the 8^th N/R decile. The 69 districts in this decile receive on average $2,183 per pupil in Basic Operating Aid (vs. $1,616 for NYC) and $ 4,206 per pupil in total State Aid (vs. $2,909 per pupil for NYC). In short, when aid is equalized with respect to both wealth and pupil need, it is clear that some high need districts, like New York City, experience significant equalization disparities when compared with districts of similar wealth and pupil need.

• First, the current state Aid distribution is highly equalized with respect to district wealth. Districts in the poorest wealth decile, received about $6,000 per pupil (CAADM); those in the wealthiest decile, about $1,220 per pupil – a 4.9 to 1 ratio.
• However, if each district’s current State aid per pupil is simply recalculated using an additional .25 weighting for "high cost" (ENA pupils), the current distribution is then judged to be equalizing but less so. Thus the same districts in the poorest wealth decile now receive $5,127 per poverty-weighted pupil (an $871 per pupil "reduction"), while districts in the wealthiest decile receive $1,192 per pupil (a mere $31 per weighted pupil reduction). These aid amounts in the 1^st and 10^th deciles, represent a 4.3 to 1 ratio, indicative of the diminished equalization in the current aid distribution when high cost pupils in poverty are taken into account.
• While the failure to more appropriately recognize high cost weightings such as student poverty diminishes the desired equalizing effects of State Aid as a district's high cost students increase, this problem is especially serious in a moderate-wealth district like New York City (whose wealth is recognized but) whose concentration of high cost students is dramatically at variance with such concentrations in other, similar wealth districts. Thus, while 93 percent of New York City’s enrollment are high cost pupils, in districts of precisely the same wealth, this high cost pupil percentage is only 32 percent.

Conclusion: This particular shortcoming should be addressed. One important type of formula modification that should be explored in order to accomplish this goal would introduce poverty weightings into the pupil count for Basic Operating Aid. This might be done as a single additional formula modification, or in concert with the removal of the PSEN weighting which has been frozen at the 1985-86 values for well over a decade. One of the key issues which must be resolved concerns the type of weighting factor to be given to such pupils. Preliminary analysis of such weightings (both in use in other states or revealed in the research literature), is the lack of consensus that characterizes recommended weightings for this type of high cost factor. In Appendix C, selected findings concerning such weight factors is described. And, in Appendix D, we evaluate the impact of the failure to more fully incorporate these types of high cost pupil weightings into aid formulas.

Appendix A

A Method For Addressing Interarea Differences in Professional Service Cost: Based upon both recommendations of recent State Aid Commissions and legislative initiatives to use cost indexes for Building aid purposes, a regional professional service cost index was developed. This index was designed to improve the fairness of existing funding formulas by recognizing geographical differences in the relative costs of the provision of educational services. In order to accurately capture regional or geographic compensation differences over which a district has no control, wage compensation for a standard set of professional occupations must first be determined and an index created which places this wage compensation in a comparison framework (using the state or a selected Labor Force area such as the Capital Region as a reference norm).

There are two possible methods for developing an index to adjust for geographic variations in the cost of labor. Both are designed to do the same thing - namely, to control for specific attributes of the goods and services being purchased, so that interarea cost comparisons can be validly made. A brief description of each method follows:

Hedonic Wage Index

Chambers and Fowler (1995) suggest a teacher cost index that is based on a hedonic wage model which considers relevant conditions that may attract workers to a certain geographic area or certain positions. This cost adjustment utilizes a comprehensive statistical analysis (typically a series of hedonic equations) which predict the market price or wage compensation that would occur if certain personal characteristics of teachers, variations in local amenities, and job environment factors were presumed to be the same in each local geographic area.

Statewide Wage Index

Rothstein and Smith (1997) argue that unlike the past, when teachers were predominantly female, and underpaid, salaries have increased and the teacher labor market has become a part of a broader labor pool of college educated, professional workers. Other professionals in this group include accountants, health professionals, and managers.

Rothstein suggests that insight into the regional variation in teacher cost can be gained by examining the regional variation in salary of these other professionals within the broader competitive labor market of similarly educated and salaried individuals. Indeed, he argues that in many communities, the overwhelming supply of teachers (and their associated costs) are controlled by a single agency or a very few agencies - in effect, dominating the market for purchasing educational components. In this type of market situation, one supplier can heavily affect the price of professional compensation; therefore, it is more appropriate to use professional salaries of non-teachers (which are subject to a free market condition) as the basis for price estimation. Since, school districts have no control over the salary scales of other professions in their geographic area, a cost adjustment based on regional variations in these salaries is more likely to represent an accurate reflection of true variations in cost as opposed to factors under the discretion of school districts.

Along this same line, analysts at the U.S. Department of Labor note that "establishment" surveys such as the Occupational Employment Survey of employers provide an excellent choice of data sets because they permit the analyst to control for geographic variation in wage costs due to differences in labor quality and quantity. By narrowly defining the occupational titles used for interarea comparison, and by imputing a common percentage distribution of employees across the selected titles in every labor force area, cost differentials due to job mix, and job quality can be controlled, or diminished.

Construction of the Index

For the purposes of adjusting for geographic variations in the cost of educational resources, a regional cost index was generated following a methodology similar to one developed by Rothstein and Smith for the state of Oregon. This involved the use of a statewide index based on median salaries of professional occupations similarly credentialed to individuals in the education field.

Teacher salaries were specifically excluded from the construction of the index because according to Rothstein, "the task of the cost-of-education theory is to imagine what the cost of education in a community would be if the education market were fully competitive." The field of education is clearly not a competitive market. Since districts have a tendency to purchase a large amount of college educated labor in a community, they gain power as buyers, to exert influence over the price they pay. Teacher's unions, publicly elected school boards and the dual role of teachers as tax-payers, also inhibit the ability of the education market to behave competitively.

Data Selection

The list of professional occupations selected for use in this index came from a list of 94 occupational titles used by the state of Oregon, which represented categories for which employment at the entry level typically required a bachelor's degree, and excluded teachers and categories that tended to be restricted to federal and state government. This wage data was provided by the Bureau of Labor Statistics and came out of the 1998 Occupational Employment Statistics (OES) Survey. OES data was provided for all 661 occupational titles for ten regions of New York State, as well as a statewide total for all titles. As noted by U.S. Department of Labor analysts, "Establishment surveys have little information on the demographics of their employees, but …wages and earnings tend to be more accurately reported in establishment surveys as they are based upon administrative records rather than recall by respondents…These factors make establishment data the natural choice …"

The entire 1997-1998 Occupational Employment Survey data file for New York State was made available to staff through the NYS Department of Labor. This large data file of 661 occupational titles and salaries, was reduced to include only the 94 titles used in the index. This list of occupations was then narrowed further to eliminate occupational titles for which data in New York State was missing (3); were educational in nature (5); or were lacking data in five or more regions of the state (8). The final list of 78 titles as well as a summary of those that were eliminated is included.

Statewide Median Wage

The first step in generating a regional cost adjustment was to establish a statewide figure for indexing purposes. This was calculated by taking the NYS total number of employees in each of the 78 titles, and multiplying by the median hourly wage for that title. The result was then summed for all titles, and this result was divided by the total number of employees in all 78 occupations. This produced a weighted statewide median wage of $24.75 per hour for the professional titles making up the index.

Title Weightings

It was also important to avoid the possibility that the index would be skewed due to compositional differences in the percentage distribution or mix of the individuals occupying the 78 selected titles. In other words, if professional wages in the titles selected were found to be identical in two labor force regions, but 60 percent of the employees in region A occupied the 10 lowest salaried titles (vs. a 10 percent employee representation in these lower salary titles in region B), a simple summation of wages could lead to the erroneous conclusion that professional service costs were far higher in region A than in region B. In short, such "apparent" cost differences would be due totally to differences in the title composition of the workforce, not to true wage differences in those titles.

This problem is avoided by weighting the wage for each title based on the relative importance of that title in the group of 78 titles as a whole statewide. These title weights are then applied to every region, therefore, making the distribution or service "mix" of titles in each region a constant. For example, if computer programmers made up 10% of the employees in the 78 titles, then a 0.10 weighting was assigned to computer programmers in every region. This title weighting procedure thus imputes to every labor force region precisely the same mix of employees across the 78 titles in every region.

Title weights were generated by dividing the statewide number of employees in a given title by the total number of employees in the 78 titles of the index. Therefore, the sum of all title weightings equals one.

Final Calculation of the Regional Index

Once the title weights were determined, they were incorporated into the data set for each of the ten regions. The median hourly wage for each title was multiplied by the title weight. This result was summed for all 78 titles and divided by the weighted median professional service wage in the Capital Region ($20.43 per hour) to yield the final professional service wage index for each region. In short, the Capital Region Labor Service Area was used as the reference norm for indexing purposes. For labor force areas that were missing median wages for a given title, two alternative methods were explored for "plugging" these holes. One method involved a simple substitution of the state median wage for the missing wage information for a given title in a particular Labor Force Region; however, it was quickly recognized that this method of wage attribution was biased "upward" (toward the median) in low cost areas of the state, and biased downward in high cost areas of the state. An alternative method developed was based on the creation of a similar regional cost index, using a smaller, but complete set of occupational titles (those titles in which no missing data appeared, n=40). That index was then used, in conjunction with the statewide median salary information for any occupational title which was missing salary information in a specific region, to estimate the missing regional salary item.

The following chart summarizes the cost index for each of the ten Labor Force Regions of the State :

Validity of Method:

Since the delivery of a market basket of professional educational goods and services (assuming agreement on these things) can be presumed to vary in cost depending upon the area of the state in which they are purchased, a wage-cost index of some type had to be constructed. Unlike the Chambers cost index based upon teacher salaries, the cost index developed here was based upon professional salaries of titles over which districts have no control, and which are further assumed - unlike teacher salaries - to be governed by open-market labor forces. In effect, such an index assumes that in communities in which the entry-level cost of such professional titles is relatively high, teacher salaries will be proportionally higher as well.

The titles selected, as noted, were derived from work conducted in creating such a professional service index for the Oregon Education Department. Specifically, these titles were drawn from the Oregon Employment Department's reports and reflected job categories for which a bachelor's degree was typical of entry workers (excluding teachers and categories restricted to state and federal government).

An important preliminary test of construct validity was conducted in order to determine how well the index performed. As a first step, each district's pupil teacher ratio was correlated with each district's total expenditure per pupil. The correlation between the district’s unadjusted expenditure per pupil and the staff ratio was found to be - .474, i.e., the larger the pupil teacher ratio, the lower the overall expenditure per pupil - precisely as one might expect. If one assumes that the true educational purchasing power of these district expenditures is more accurately measured by cost adjusted dollars based on this index rather than raw dollars per pupil – it follows that the cost adjusted expenditure measure should be more highly correlated with a staffing measure like the pupil teacher ratio than the raw unadjusted measure. When this correlational test was conducted using cost adjusted dollars per pupil, the correlation increased to -.532, consistent with expectation.

return to FARU home page