School Efficiency: Commissioner Says SADs Are More Efficient
(from Maine Townsman, October 1999)
by Geoff Herman, MMA Director of State & Federal Relations

The simple hypothesis of State Commissioner of Education Duke Albanese is that School Administrative Districts (SADs) are designed in such a way that they are more efficiently run than the municipal school systems in Maine. The commissioner says that his department’s research clearly shows SADs to be more cost-effective based on per-pupil spending in similarly sized school systems.

Since MMA has long supported greater school efficiency in Maine, we decided to look more closely into the commissioner’s assertion and the data the Department of Education uses to back it up.

THE DIFFERENT SCHOOL STRUCTURES

In terms of structure, there are basically three K-12 public school systems in Maine: the single municipal school system; the school union system, and the school district system.

There are 47 single municipal systems in Maine. A single municipal system is likely to exist when a town or city has a sufficient population to support its own grades K-12 school system. Additionally, some small towns have municipal school systems just for lower grades. Typically, the municipality operates an elementary school and pays tuition for the older students to attend secondary or middle/secondary school systems in the area.

A school union system exists when a number of small municipal school systems, typically at the elementary grade level, band together for the purpose of sharing a school superintendent. The governance of the elementary school doesn’t centralize except for the purpose of sharing the superintendent. There are 32 school unions in Maine, involving over 120 municipalities and plantations.

The school district system includes School Administrative Districts (SADs) and Consolidated School Districts (CSDs). SADs tend to be systems which govern the entire K-12 educational program in a multi-municipal region. There are 73 SADs in Maine that include over 270 municipalities and plantations.

CSD’s tend to be regional secondary school or middle/secondary school systems that sometimes work in concert with a school union operating at the elementary grade levels. There are 13 CSDs in Maine involving 40 municipalities and plantations.

UNABASHED SAD SUPPORTER

Commissioner Albanese is an unabashed supporter of the school district model. At a recent meeting with some municipal and school officials, the commissioner offered the opinion that the creation of the SAD structure was the best public policy decision in the field of education that the state has ever made. Efficiency seems to be one of the reasons the commissioner is so supportive of SADs. Among the benefits of SADs that the commissioner celebrates is the alleged ability of the district school structure to deliver educational services at a lower per pupil cost than municipal school systems of similar size.

Although it is not always clear in discussions with state education officials that low per-pupil spending is considered a positive achievement, Commissioner Albanese will hardly let a conversation go by without mentioning the department’s study that shows that SAD’s consistently deliver at lower per-pupil costs than municipal school systems. The department’s analysis is found in Table I.

As can be seen, the department’s analysis begins by segregating all the school systems into their respective student population categories, beginning with the smallest school systems and running in increments of 500 students all the way to the state’s largest school system, which is the City of Portland with its 8,000 students.

The analysis then identifies the operating costs of those school systems. Although the term "operating costs" when used in the context of the school funding formula doesn’t normally include such "program costs" as special education, vocational education, and transportation, for the purpose of this analysis, the term "operating costs" includes all general fund costs except major capital outlay, debt service and transportation expenditures.

These operating costs are aggregated for all the municipal, SAD and CSD school systems in each population category and those aggregated sums are divided by the respective resident student population counts. The mathematics yields the aggregate per-pupil operating cost for municipal school systems, SADs and CSDs in each population category.

Surprisingly, almost without exception, in each of the population categories for which there are both municipal and district school systems, the school district per-pupil operating costs are less than the per-pupil costs in similarly-sized municipal school systems, when analyzed in the aggregate.

THE COMMISSIONER’S HYPOTHESIS

Commissioner Albanese explains the data with the claim that the SAD organizational structure allows for natural economies of scale. Albanese points out that the SAD is likely to have larger purchasing power and a savvy, professional business manager who is able to squeeze out efficiencies in purchasing. Albanese also says that the SAD doesn’t suffer from the multiple governance groups that can be involved in a school union/CSD structure, for example. When there are elementary schools with different systems of instruction feeding a middle school or high school, the commissioner believes the efficiencies of a unified system are simply unavailable.

Commissioner Albanese also believes that a more effective use of facilities helps explain the lower per-pupil costs delivered by SADs. When school lines are drawn by municipal boundaries, and the student population in one town is sharply declining but in the neighboring town growing quickly, the commissioner believes that parochial, unconsolidated governance structures would be naturally resistant to making the best use of the two facilities, and one school would go partially unused while new construction would be demanded in the other town. This explanation holds that an SAD, because of its consolidated aspect, is less susceptible to facility inefficiencies and the vicissitudes of a student population shifting from one town to another.

To encourage more efficient governance structures and overcome the ‘home rule’ resistance to consolidation, the commissioner believes that policy makers should consider developing financial incentives in the system of state educational subsidy so that the school systems that were willing to adopt new structures that would achieve operative efficiencies would be rewarded for those decisions.

UNPACKING THE DATA

One element of the department’s data that clearly stands out is the methodology of aggregating the information for all the school systems, by type, in each population category. On the theory that something instructive might emerge if the aggregate data was unpacked, MMA sorted the department’s information to show the per-pupil costs of each school system in each of the population categories. Some additional information, also provided by the Department, was included with the per-pupil cost information, specifically the per-pupil valuation figure for each school system, and a percentage of state aid figure. (see Table II)

The per-pupil valuation figure is simply the value of taxable property within the tax jurisdiction of the school system divided by the school system’s student population. Per-pupil valuation is a strong indicator of the capacity of the school district to raise revenue for the schools. If there is a great deal of taxable value relative to the number of students in the school system, the tax burden necessary to raise an appropriate amount of revenue is less painful to bear.

The percentage of state aid figure is a calculation of the state’s share of the "foundation allocation" which is the principal building block of the school funding formula. This percentage does not take into account the debt service allocation or the amount of the school’s "local option" appropriation, but the percentage figure still gives an indication of whether the school system is a "low receiving" or "high receiving" school unit.

IT’S VALUE, NOT STRUCTURE

The first and most obvious observation to be made when reviewing the unpacked data is that the strongest factor related to per-pupil operating costs is per-pupil value, not the structure of the school system. There is a near-perfect correlation between per-pupil value and per-pupil operating cost. When an SAD has a high per pupil value, it is just as likely to have a per-pupil operating cost that competes head-to-head with a similarly situated municipal school system. At the same time, when a municipal system has a low-per-pupil valuation, its per-pupil operating cost falls into the same ranks as its SAD counterparts.

The correlation between per-pupil value and per-pupil operating costs is so strong it is rarely broken in any school population category. For example, when the per-pupil valuations are grouped, it is a rare school system that defies what would otherwise be its completely predictable per-pupil operating costs. For example, in the 1,000-1,499 student population category, four out of five of the communities that spend over $6,000 per-pupil have per-pupil valuations over $500,000. In that same category, six of seven communities that spend less than $5,000 per student have per-pupil valuation under $300,000.

And when those few unpredictable operating costs do occur, there is no link between the aberration and the structure of the school system. Millinocket and Winthrop (both municipal systems) have lower per-pupil valuations than would account for their relatively high operating costs, but for the municipal school systems in Ellsworth, Falmouth and Scarborough the reverse is true, and their per-pupil costs are down despite substantial per-pupil value.

The exceptions aside, the first conclusion drawn from this data is that that strongest predictor of per-pupil operating costs is per-pupil value. It is also clear that with some notable exceptions the SADs in Maine are far more likely to have lower per-pupil values than the municipal school systems. Of all the school systems in Table II (1,000 to 4,000 student population), for example, 16% of the SADs had a per-pupil value under $200,000, but only 3% of the municipal schools fit that category. 50% of the municipal school systems have per-pupil values over $200,000, but only 31% of the SADs have that much per-pupil value to work with. The average per-pupil value among all the SADs in the 1000 to 4000 student population category is $276,925, and for the municipal school systems in the same category the average per-pupil valuation is $361,004.

If per-pupil spending is as strongly related to per-pupil value as it appears to be, and if the SADs, as a class of school systems, have less per-pupil value than the municipal school systems, then the apparent cost efficiency being attributed to the SADs may be a manifestation of their relative valuational poverty in disguise.

WHAT SUPERINTENDENTS SAY

It is obvious that there is more to this data than meets the eye. One obvious factor behind spending is resources, but what else distinguishes SAD and municipal school systems? Is Commissioner Albanese correct when he says that SADs enjoy better economies of scale or more efficient use of their facilities? For insight into the behind-the-scenes factors that drive per-pupil operating costs, MMA sent the applicable data to a number of school superintendents and subsequently connected with those superintendents over the phone.

Winthrop Schools and SAD 36

The Winthrop school system and SAD 36 (Livermore, Livermore Falls) have quite a bit in common. With nearly the same student population, roughly the same per-pupil valuation, and within the range of the same labor markets, the biggest difference between the school systems seems to be the per-pupil operating cost, which is $650 greater in Winthrop.

Winthrop’s Superintendent Terry Despres has an explanation at the ready which defines an essential difference between SADs and municipal school systems that has little to do with governance structure. It’s age.

Winthrop’s municipal school system is a lot older than the newcomer SAD’s, and the state has long been heavily financing new school construction with nary a penny of state money (until recently) contributing to the renovation of aging municipal school infrastructure. Despres points out that Winthrop has been maintaining its aging school buildings and dealing with associated problems like ADA compliance lawsuits, all on its own nickel and heavily on its operational budget.

Because the department’s analysis of per-pupil operating costs under review doesn’t include major capital outlay and debt service, this distinction between aging infrastructure and more modern school facilities is significant. Simply put, the cost of operating the older school facilities, which are more likely to be found in the municipal school systems, is greater. At the same time, the debt service cost associated with more modern facilities, as are likely to be found in SAD’s, is not included in the calculation of a school’s per-pupil operating costs.

Winthrop’s high tax mill effort for its facilities is scheduled to come down, however. After 13 years on its waiting list for new construction, and after one locally-rejected school construction proposal four years ago when the voters appeared to disagree with the proposed location of the new high school, two phases of a badly needed new construction plan have just recently cleared all state and local hurdles. According to Despres, when the new school facilities are built the town’s school maintenance and operation costs are scheduled to drop over $2.00 a square foot, the leased facilities will no longer be necessary, and the extensive costs associated with air quality control should all but disappear.

Waterville Schools and SAD 47

Jim Morse, the superintendent of SAD 47, agrees with Commissioner Albanese that SADs have a greater capacity of efficiency because of the economy of scale they can generate. As an example, Morse points out that the three towns that make up SAD #47 (Oakland, Sidney, and Belgrade) have one high school, one middle school, and only two elementary schools, in Oakland and Belgrade. This partial consolidation of the elementary schools creates an efficiency that would not be available without a regional structure.

Morse did point out that SAD 47, which is next door to a municipal school system in Waterville, is to some degree reversing roles with its urban neighbor. In 1960, the combined population of the three towns in the SAD was just 6,000, and now that population is 13,000 and the portable classrooms piling into SAD 47 are a dramatic manifestation of that population growth. In some defense of the per-pupil cost rates in Waterville, which the data shows to be over $700 higher than SAD 47’s, Morse points out that when a student population is sharply increasing, the cost escalation is not necessarily dramatic at first, as the unactivated capacity of that school system becomes utilized. By the same token, when a school population is in decline, as is the case in many of Maine’s service center communities, the per-pupil operating costs are unable to drop precipitously with that declining population, because the same basic critical mass of infrastructure and staff must still be maintained.

Just down the road in Waterville, Superintendent Ed Leblanc watches over a high school, a junior high, and two elementary schools, just like his SAD 47 counterpart. Waterville’s student population is dropping and is now approximately 370 shy of the 2,500 students that go to SAD 47 schools. Leblanc points out that the "service center" municipal school systems like Waterville tend to have significantly higher student populations with special needs, which certainly has a direct impact on the bottom line per-pupil operating costs. A vivid example of that phenomenon is found in the Portland school system, which has developed an English-as-Second-Language program which, by Maine standards, is enormous.

Several of the municipal-system superintendents mentioned a phenomenon related to special programs that is difficult to prove or quantify but which they feel is taking place. As the theory goes, the service center community for demographic reasons tends to have a higher student population with special educational needs. When the programs are put into place to address those needs, the programs become a magnet for parents who are seeking quality special education services, and the concentration of special education services intensifies.

Auburn School System and SAD 17

Cathy Fanjoy, the business manager of SAD 17, says that the area labor market plays a role in per-pupil operating costs. SAD 17 includes Norway, Paris, Oxford, and five surrounding towns that are 20 miles northwest of Lewiston-Auburn. With respect to SAD 17 and the Auburn school system, almost all the statistical factors are the same. The two systems have approximately the same number of students and the same amount of taxable value behind each student. The rate of school subsidy from the state is roughly the same, as well. Structurally, SAD 17 can’t point to a consolidation of its elementary schools to explain the numbers. There is an elementary school in each of the eight communities, the largest with 650 students and the smallest with 44.

Despite all the similarity, the data shows SAD 17 to have a per-pupil operating cost over $1,100 less than the Auburn school system. Fanjoy thought one factor was the area labor market. Where the salaries and benefits of the school teachers, administration, and other staff make up 80% of a school’s operating budget, a comparison of per-pupil costs has to include a comparison of the area wage rates in the respective labor markets.

Auburn School Superintendent Barbara Eretzian reiterates the observations about the financial impact of special education. In addition to the programs Auburn is operating, the city school is paying for 20 out-of-district placements, which can be extraordinarily expensive. Eretzian wanted to point out, however, that there is Medicaid reimbursement now for some of the services provided to qualifying students, which the city puts right back into the school’s special education account, so even if there are high per-pupil costs related to special need services, as reflected in the data, at least some of that cost is being borne by the federal government.

With respect to the efficient use of facilities, Eretzian said that Auburn’s facility system is very similar to SAD 17’s. Auburn has a high school and a middle school and seven neighborhood elementary schools, the continued life of each of which is fiercely protected by the respective neighborhood communities. Unlike SAD 17, however, each of the Auburn elementary schools has student populations in the 250-500 range.

Eretzian also pointed out that there is a limit to the "economy of scale" theory. In her experience, the maximum purchasing economies can often be obtained in systems the size of Auburn’s. When larger purchasing systems are created, the price of the commodity doesn’t go down any further, but the process of administering the purchase becomes significantly more complicated. Bigger is not always better, and Eretzian says she has to keep an eye on consolidated purchases, whether they be raw commodities or professional development programs, to make sure that the consolidation is actually cost effective.

Caribou School System and SAD 1

Two other school systems where a comparison is suggested are the Caribou school system and SAD 1 (Presque Isle, Mapleton, Chapman, Castle Hill and Westfield). In this comparison, the only major statistical difference is student population, with Caribou having 700 less students than are enrolled in SAD 1. Otherwise, the two systems have similar per-pupil valuation, state subsidy levels, share the same labor market, and have per-pupil operating costs that are only separated by $76.

Caribou’s School Superintendent Art Benner suggested two additional factors that should be taken into account before comparing school systems "apples to apples."

One of those factors is vocational education. According to Superintendent Benner, the school systems in the several regions around the state that have developed vocational education programs for the larger region are, at least in some cases, taking on operational costs that are not borne by the other school systems.

Benner also said that the financial implications of tuition students should also be carefully reviewed. According to Benner, the emerging phenomenon of negotiating tuition rates rather than sticking with the tuition rate structure established by law (the school’s actual per-pupil operating cost or the state average, whichever is less) is forcing some school systems to subsidize the costs of educating the nonresident students.

In fact, the department’s data that is the subject of this article highlights the tuition factor in the following way.

"There are many factors that influence the size of a unit’s per-pupil operating cost. One of the factors that affects the amount of the per-pupil operating cost…is a state-mandated tuition rate that may be charged by any administrative unit for students received from other school units. Consequently, the proportion of students that are tuitioned to other school administrative units is related to the per-pupil operating costs of the school the school administrative units. In particular, units that tuition all of their pupils have, in general, lower per-pupil operating costs than do other units…In contrast, units which tuition few (or none) of their pupils to other units have, in general, higher per-pupil operating costs that do other units."

CONCLUSION

There is probably not an organizational system that involves human beings that cannot be criticized by the observer for being inefficient to some degree. It’s an easy claim.

With respect to the alleged efficiencies of the SAD structure compared with the municipal school structure, the claim hasn’t been verified. It appears that regardless of the governance structure, at least with respect to the school systems of any size, the efficiencies of purchasing and facility use are obtainable. Regardless of the governance structure, neighborhoods and communities are going to fight to retain smaller schools closer to the community, especially for the younger children. Not surprisingly, the most significant factor that predicts per-pupil spending is per-pupil capacity to spend. And the older municipal school systems have at least two factors working against them when asked to compete with younger regional systems to achieve reduced per-pupil costs: the age of their facilities and the demographic trends that yields a student population with a higher level of special educational needs.

"Economies of scale" is too facile an argument. There is more to this data than meets the eye.