Back to: Finance
A Review of Operating and Capital Grant
Formulas For Education In Saskatchewan
By J. A. Volk (1990)
SSTA Research Centre Report #90-12: 45 pages, $11.
| The Operating Grant Formula | The Foundation Grant Program review was commissioned by the
Saskatchewan School Trustees Association Research Centre to assess the effectiveness of
the operating and capital grant formulas currently used in Saskatchewan. A large number of
concerns appeared to result from unrealistic expectations and misunderstandings,
therefore, part of the report is devoted to an explanation of the operation and purpose of
the formulas. It was found that the operating grants formula continues to meet the objectives set out when it was introduced in 1972. Changes in the delivery of education services and modifications that have been made to the formula suggest that there are certain changes that should be made. |
Back to: Finance
The SSTA Research Centre grants permission to reproduce up to three
copies of each report for personal use. Each copy must acknowledge the author and the SSTA
Research Centre as the source. A complete and authorized copy of each report is available
from the SSTA Research Centre.
The opinions and recommendations expressed in this report are those of the author and may
not be in agreement with SSTA officers or trustees, but are offered as being worthy of
consideration by those responsible for making decisions.
A grant formula is a mechanism designed specifically to distribute monies allocated to third parties such as school boards by government to assist with the costs of social programs. The Foundation Grant Formula and the Capital Grant Formula are intended to assist boards with meeting operating and capital expenditures. It is important to note that these two formulas do not generate grant monies hut merely distribute the money that has been allocated to boards by government through a political process that is independent of the grant formula components. This study, therefore, will confine itself to a study of the equity of the grant formulas. The amount of money that should be available for distribution by the grant formulas is indeed an important consideration but it falls outside the purview of this study. It will become necessary, of course, to make reference to the amount of money available to boards but it is not the purpose of this study to comment on the adequacy or inadequacy of monies allocated to education by government. Thus, the question to be answered is simply this: De the two grant formulas distribute the monies allocated to boards, fairly? An examination of the more important components of the formula will serve as a useful first step in attempting to answer this question.
I. THE OPERATING GRANT FORMULA
The Saskatchewan Foundation Grant Formula, as other foundation grant formulas, has several basic objectives:
(1) To bring about a greater degree of equality of education opportunity.
(2) To provide for greater education opportunities with a "reasonable" local tax rate.
(3) To retain for boards a high degree of local autonomy.
(4) To hold boards accountable for the consequences of local decision-making and local priority setting.
(5) To provide for special adjustment factors in the formula which will compensate boards for circumstances over which they have little or no control. These circumstances would include a marked loss or increase in students, the need for small schools and costs associated with the conveyance of students.
It should be noted that these objectives embody the principles of equity, autonomy and efficiency which are central to any discussion of education finance.
The principle of equity holds that the quality of educational services in a community should not be dependent upon the wealth. of that community. There is, however, some disagreement about what constitutes a wealthy community.
The principle of autonomy holds that local authorities (school boards) are best able to establish local priorities and address local needs.
The principle of efficiency holds that there must be an acceptable relationship between educational inputs and educational outputs. Efficiency is closely related to the concept of accountability which calls for responsibility in the exercise of local autonomy.
The 1990-91 Estimates of the Saskatchewan provincial government indicate that $360,278,300 was allocated to school boards as Grants to Schools-Operating. Most of this allocation is distributed to the 110 school jurisdictions in this province through the grant formula. A good grant formula will distribute the approximately $360 million equitably to 110 school boards. Let us examine but a few possible, if not practical, ways in which this money could be distributed.
(1) A simple but hardly a practical method would be to divide the $360 million by 110 to arrive at an equal amount for each jurisdiction.
(2) The money could be distributed on a par-pupil basis. On the basis of approximately 200,000 students the provincial allocation would provide $1800 per student. Jurisdictions would simply receive $1800 for every student enrolled in the system. It will be noted later that this method would not be as equitable as one might think.
(3) The money could also be distributed on the basis of need. The less wealthy Jurisdictions are in greater need of provincial money since their ability to raise local revenue is less. The ability-to-pay principle is indeed desirable and acceptable to use in allocating monies but its implementation is difficult. Wealthier jurisdictions have a greater ability-to-pay while the less wealthy jurisdictions have a lesser ability-to-pay. But what makes a jurisdiction mare wealthy or less wealthy? This question needs to be answered for the Foundation Grant Formula is based on the ability-to-pay principle. The indicator of wealth as used in the formula is the assessment of the jurisdiction. For two similar Jurisdictions the one with the higher assessment is considered to be the wealthier jurisdiction.
The present grant formula was first used in 1972. Several modifications have been made to the formula since then hut its basic structure has remained the same. Originally the formula vas expressed as the following algebraic equation:
A - B = C
In the equation A represented the "recognized costs" of a jurisdiction, B represented the "local input" and C represented the grant paid to the jurisdiction. Since the letters A, B and C are used to represent other constants and variables in education and since C is the dependent variable in the formula at the jurisdictional level it may be advisable to rewrite the formula in the following form:
Grant = Recognized Costs - Local Input.
As the formula indicates, the grant is the variable and will be -be difference between the recognized costs and the local input. The constants used to calculate both the "recognized costs" and the "local input" are determined by policy and are not subject to negotiations. The recognized costs are sometimes called assigned costs and represent a defined need for a board.
The largest component of the recognized costs is the sum of the recognized costs for every student in the system. The per-student recognition on average, accounts for approximately 774 of the recognized costs. The constants used to calculate recognized costs are arrived at in a variety of ways. Average expenditures, known or anticipated increases in expenditures, the introduction of new programs, changes in educational priorities and several other factors will determine a set of constants which in turn will determine the recognized costs for a jurisdiction. For example, an anticipated increase in the cost of gasoline would result in an increase in the cost per km. for operating a school bus. The increase would then be applied to the total number of kms. traveled by the board's buses and would, in turn, increase the total recognized costs for the board. In like manner the decision to place a higher priority on special education could result in an increase in the special funding categories which would result in higher recognized costs for those jurisdictions which have students in these categories.
Most of the costs incurred by boards relate to the number of students that need to be educated. If, in fact, there were a direct relationship between education costs and the number of students to be educated one could simply recognize a determined cost per student to arrive at a hoard's recognized costs. Thus, for example, if the cost to educate a student were $4,000 the recognized cost for a jurisdiction would simply be $4,000 multiplied by the number of students in the jurisdiction. Such a simplified calculation, however, is found wanting on two counts.
In the first place it has been assumed that it costs more to educate a student at the secondary level than at the elementary level. When the present grant formula vas introduced in 1972 there vas a significant difference in education costs between the elementary and secondary levels. The differential is probably less today than it was in 1972 but there is no agreement among educators that the costs are now the same. Some will argue that they should be the same but few, if any, will contend that they are the same. Therefore, the differences in recognized rates for different levels of instruction remain an integral part of the formula. There are different rates for kindergarten students, division I and II students, division III students and division IV students. In the light of changes in the delivery of educational services in recent years these rate differentials need to be reviewed to ascertain their rationality and acceptability if, indeed, they are still needed at all.
Another reason for not simply using the product of a cost per pupil and the number of pupils to be educated is that the cost to educate every student at a given level of instruction is not the same throughout the province. It will cost mare, for example, to educate a division III student who lives twenty miles from school than it does to educate a division III student who lives only two blocks from school. Transportation is but one such variable but there are many more. The formula is designed to give recognition to such factors as: the nature of the school attended (comprehensive school), the size of the school attended (small school factor), sparse population of the jurisdiction (sparsity factor), significant unforeseen enrollment changes (enrollment decline factor) and special needs of some students (special education recognition factor).
We see that the total recognized costs for a jurisdiction are made up of two major components. The general per-pupil costs with provision for different recognition rates for different levels of instruction and the calculated recognized costs for specific expenditure items.
In the formula, the first component, or the general per-pupil costs, is called the "basic program" which is used as a base to calculate some of the specific recognized costs. The sum of the general and the specific recognized costs represents the "total recognized costs" for the jurisdiction and should amount to approximately what the jurisdiction spends on its K-12 education program. The constants or multipliers used in arriving at the recognized costs for bath categories are the same for all jurisdictions in most cases. However, the constants or multipliers themselves are not solely the result of objective observations and calculations. A subjective element is always present when decisions need to be made about the actual constants and increases or decreases in the constants. The subjective influence on this component of the grant formula has weakened the formula in that the recognized costs have not kept pace with the actual costs. For several years recognized costs were from 2% to 3% less than boards' actual costs hut that difference increased steadily over the years and in 1989 the estimated difference was in the order of 164. This difference was calculated by using boards' 1989 audited expenditures as they appeared in the Budget Analysis Printout and boards' recognized costs listed in the 1989 Foundation Operating Grants Estimates. Board expenditures were determined by subtracting expenditures for Other Educational Services, Ancillary Services, Provisions for Reserves, Payments to Other Systems and Special Projects from Total Operating Expenditures.
This difference weakens the formula in two ways. The general per-pupil recognized rates, which determine the basic program, are also used as "caps" for the recognition of tuition fees. If, for example, a board pays tuition fees to another jurisdiction on behalf of one of its division IV students the fees paid by the board are considered a specific recognized cost. However, the maximum amount recognized is the same as the recognized per-pupil rate for division IV students in the basic program. The receiving jurisdiction may, for example, charge a tuition fee of $4,000 while the maximum recognized cost may be but $3,500. The difference needs to be paid by the sending board without recognition. For jurisdictions who have many tuition paying students this may create a serious financial inequity. The sending jurisdiction may be forced to consider providing services for such students rather than sending them to another jurisdiction which, for a variety or reasons, may be the better educational setting for the student.
Non-realistic recognized costs reduce the equalization power of the formula. Lower than actual recognized costs allow for the use of a computational mill rate which is lower than the average actual provincial mill rate. The effect that such a situation has on the equalization power of the formula will be discussed more fully later.
It is important to be aware of the sequence in which board revenues are determined. The local input is always the first board revenue to be considered. If the "calculated" local input is not sufficient to cover the "recognized costs", grants will make up the difference. However, if the calculated local input is sufficient to meet the "recognized costs" no grant is paid to the board. If the local input is small when compared to the recognized costs for the jurisdiction, the grant may be significantly higher than the local input. Grants are only used to make up the difference between a jurisdiction's "recognized costs" and the calculated "local input". The determination of recognized costs has already been discussed. A review of local input would now be in order.
Local input consists primarily of local tax revenues. In some jurisdictions tuition fee receipts supplement the tax revenues but these receipts usually account for hut a small percentage of the local input. For discussion purposes, therefore, it will be assumed that local input consists of local tax revenues only.
If the grant formula is to consider ability-to-pay it follows that the local input must reflect a board's ability-to-Pay. Ability-to-pay would not be recognized if every board made the same local input nor would it be recognized if every board made the same local input per-student. But, the ability-to-pay principle would be recognized if the local input were related to the assessment of the jurisdiction since assessment is considered to be an indicator of wealth.
The local input is determined by the application of a mill rate to an assessment. By applying a Uniform mill rate to the different assessments of the various jurisdictions we generate a different local input for different jurisdictions. Jurisdictions with low assessments will generate lower local input than jurisdictions with higher assessments. The question to be answered, therefore, is what uniform mill rate should be applied?
The basic grant formula has three variables.
Grant = Recognized Costs - Local Input.
At the provincial (government) level we can assume that two of the variables are determined primarily on a political basis while the third one is calculated. The "grants" allocated to school boards is a government decision (cabinet, caucus and/or treasury board) and for 1990 the amount allocated was approximately $360 million. The "recognized costs", too, have a strong political input as painted out earlier in this paper and are estimated at $730 million. Now, if the "grant" is known and if the "recognized costs" have been determined it becomes an easy algebraic problem to solve for the third unknown - the required local input.
Grant = Recognized Costs - Local Input
$360,000,000 = $730,000,000 - Local input.
Therefore, local input - $370,000,000
These calculations indicate that boards will have to raise $370 million locally to be added to the $360 million in grants to cover the recognized costs of $730 million. This does not apply to any particular board but to the 110 boards collectively. A mill rate of 56 will need to be applied to the provincial assessment of approximately 56.6 billion to raise the calculated local input of $370 million. This required mill rate is known as the computational mill rata. The computational mill rata is that mill rate which needs to be applied to the provincial assessment to raise that portion of the recognized costs that is not covered by provincial grants or "other recognized revenues".
Thus, if a mill rate of 56 is applied to every board's assessment the levy will cover that portion of the recognized costs not covered by grants. It is considered to be fair if every Jurisdiction Pays at the same mill rate since the assessment Provides for differences in ability-to-pay.
At the jurisdiction level the recognized costs and the local input are calculated in the same manner as they are at the provincial level but in a different order. At the provincial level the grant was fixed and the local input had to be adjusted to cover that portion of the recognized costs not covered by grants. At the jurisdictional level, however, the local input is calculated (cmr x assessment) and the grant is then adjusted to cover that portion of the recognized costs not covered by the local input.
If the calculated recognized costs of a jurisdiction were indeed the actual costs of the jurisdiction the actual mill rate for the jurisdiction would be 56 (the cmr). The local input would differ significantly from board to board but every board would have a mill rate of 56 mills and in keeping with some of our original assumptions We would then consider the educational burden of boards to have been equalized. The wealthier boards would be paying more hut they would have the same mill rate as the less wealthy boards. In order words, we generally accept that boards with equal mill rates are making a fair and acceptable contribution to the education costs.
It may be well, at this time, to be reminded that the discussion thus far has been highly theoretical. "Recognized Costs" ware based on available data and on some subjective assumptions while the computational mill rate was calculated on the assumption that the Recognized Costs were indeed the board's actual costs. Boards, therefore, will not have uniform mill rates equivalent to the computational mill rate. In fact, uniform mill rates are the exception rather than the rule in Saskatchewan. That should not be surprising nor alarming. The computational mill rate is no more than a mill rate used to determine, calculate or compute an average and acceptable local input. Boards, however, use their local autonomy to make program decisions which will have a direct bearing on local mill rates. That is desirable It would be a sad day for Saskatchewan if all boards did indeed have the same mill rate. Such a situation would suggest that boards have either failed to use their local autonomy to address local needs or the provincial government has made all the decisions for the boards. However, the variations in mill rates is sometimes caused by flaws in the application of the formula.
Available data indicate that the recognized costs are approximately 164 less than the actual costs as listed in school board budgets. Also, the computed local input at 56 mills is significantly less than the actual local input. In fact, the cmr would need to be raised by more than 8 mills to make it equal to the average actual mill rate. Table 1 shows the differences in the cmr and actual average mill rates for the period 1980-1990.
An examination of the formula:
Grant = Recognized Costs - Local Input
might suggest that it does not matter if both the Recognized Costs are too low and the Local Input is too low since the constant is the grant. It might be argued that as long as the difference between the Recognized Costs and the Local Input is $360,000,000 the formula will work. Indeed, the formula will work but the equalization power of the formula is thwarted. As an algebraic formula it will work if you increase or decrease both the
Recognized Costs and the Local Input by the same amount. However, the effect that such tampering with the formula has on its equalization powers is alarming. In the recent Board Survey many boards lamented the unrealistic level of "recognized costs" and the ever increasing "computational mill" rate used in the formula.
Quite erroneously some beards feel that either an increase in the level of recognized costs or a decrease in the computational mill rate would result in increased grants to boards. Not so! It must be remembered that the grant to boards is fixed. For 1990, government has allocated approximately $360 million to boards and this amount is not about to be changed. This fixed amount in grants makes it imperative that the difference between the recognized costs and the local input (assessment x cmr) be $360 million. The computational mill rate can be kept low which will result in a lower calculated local input but then the recognized costs must also be low enough to maintain the difference of 5360 million. Conversely, if the recognized costs are increased significantly a corresponding increase will be required in the local input to maintain the required difference of $360 million. You might conclude, therefore, that the level of the recognized costs and the calculated local input could be at any level, set quite arbitrarily, so long as the difference is $360 million. The formula would indeed continue to work if, for example, the recognized costs were reduced by $10,000,000 and if the local input were reduced by the same amount. However, the formula would break down if the recognized costs remained at their current level and if the local input (cmr) were cut in half.
The observation that the formula would continue to work if like reductions were made in both the recognized costs and the local input can not be interpreted to mean that the formula would continue to operate in an equitable manner. In fact, some of the inequities in the distribution of grant money to Saskatchewan boards are the direct result of indiscreet manipulations of these two components of the grant formula.
By way of example, let us examine two similar rural school divisions in Saskatchewan.
These two divisions are similar in all respects except their assessment. A simple calculation would suggest that they would qualify for grants as follows:
Formula: Recognized Costs - Local Input = Grant
Jurisdiction "A": 5,195,148 - 3,746,400 = $1,448,748
Jurisdiction "B": 5,063,480 - 1,892,240 = $3,171,240
Now, if both boards' expenditures were, in fact, the recognized expenditures as calculated in the formula, then both boards could operate at 56 mills, the computational mill rate. In such a case the formula would be equalizing board costs fully. The local input in dollars by the two boards would be very different as shown in the data hut both boards would be contributing at the same mill rate. Contributing at the same mill rate is considered to be fair since it related to a board's ability-to-pay. However, we know that these two boards can not operate at 56 mills. Their actual costs are considerably higher than the recognized casts and therefore a mill rate greater than the computational mill rate of 56 will be required to cover that portion of the recognized costs not covered by the calculated grant.
Let us examine the data again and make some assumptions about actual costs and actual mill rates. Actual costs, on average, are about 16% higher than recognized costs. For ease of calculation let us assume a 10R difference between the recognized costs and actual costs for these two similar jurisdictions. The actual costs would then be some $500,000 higher than the calculated recognized costs for each jurisdiction. Now we can look at the calculations in two steps:
We can make the following observations:
(1) There is no additional grant for the difference between the recognized costs and the actual costs. The grant is calculated on the basis of recognized costs and is final.
(2) The additional $500,000 must be raised by an increase in the board's actual mill rate.
(3) Jurisdiction A with a higher assessment will require an additional mill rate of 7.5 to raise the additional $500,000 while jurisdiction B with the lower assessment will require an additional mill rate of 14.S to raise the additional $500,000.
(4) Jurisdiction B requires almost twice the mill rate to raise that portion of the board's costs over the recognized costs.
(5) If the recognized costs were the same as the actual costs the local input would be covered by the computational mill rate even though the computational mill rate would, of necessity, have to be higher than 56 mills. The actual mill rate would, nevertheless, be the same for both boards.
(6) The computational mill rate is lower than the actual mill rate only if the recognized costs are lower than the actual casts. There will always be some variations between recognized and actual costs and between computational mill rate and the actual mill rates. The differences are the direct result of local decision-making and local priority setting. However, ma]or differences as currently exist seriously affect the equalization power of the formula.
(7) Local input is equalized only to the level generated by the computational mill rate. Mill rates in excess of the computational mill rata have an adverse effect on those jurisdictions who have below average assessments since a higher mill rate will be required to raise an equal amount of revenue from a lover assessment.
(8) Ideally, the recognized costs should closely approximate the actual costs and the computationally mill rate should approximate the average mill rate in the province.
Some of the observations just made might be best illustrated by the following hypothetical case.
Let us assume that the recognized costs were increased by $200 per student in the basic program calculations. That would mean that the per-pupil recognitions for the four different instructional levels would all be increased by $200 per student. Increasing recognized costs would normally not be done by increasing but one set of constants. However, this example keeps the calculations simple and has the same end result. For approximately 200,000 students in the province the $200 per student increase would increase the recognized costs at the provincial level by $40,000,000. With the Provincial Grant fixed at $360 million the local input would also have to be increased by $40,000,000. This would be done by increasing the computational mill rate by 6 mills (6 mills x the provincial assessment). Now, both the recognized costs and the local input have been increased by equal amounts and the provincial grant has not been changed.
If we now return to the data on page 15 and make these adjustments to the calculations we will find the following:
(1) The recognized costs for jurisdiction A will be increased by $240,000 (1,201 x $200) and for jurisdiction B by $225,000 (1,127 x $200)
(2) The local input Will be calculated With a cmr of 62 (56 + 6) -mounting to $4,147,800 for jurisdiction A and 2,094,980 for jurisdiction B.
(3) The actual casts for jurisdiction A and B were assumed to be $500,000 higher than the original recognized costs. With the additional recognized costs now calculated, jurisdiction A and B would still need to raise $260,000 and $274,000 respectively or additional mill rates of 3.9 and 8 mills. The actual mill, therefore, for jurisdiction A and B will be (62 + 3.9 ~ 65.9) and (62 + 8 - 70) respectively. You will note that these rates compare with 63.5 and 70.8 prior to the adjustments to the recognized costs and local input. You will note too that by increasing the computational mill rate by 6 mills the actual mill rate of jurisdiction B was reduced by .8 mills. Thus an increase in the computational mill rate will decrease the actual mill rate of the lower assessed jurisdictions while increasing the actual mill rates of the higher assessed jurisdictions.
The formula is intended to require a greater local input from those jurisdictions which have a greater ability-to-pay. This objective embodies the principle of equalization. The indicator of ability-to-Pay that we have accepted is the assessment of a jurisdiction. Thus, the foregoing example illustrates how full equalization can occur only if the computational mill rata is as close as possible to the actual mill rate.
Early in this study, five major objectives for the foundation grant formula were stated. It may be well to review these objectives in the light of the discussion thus far.
(1) To bring about a greater degree of equality of education opportunity.
This objective calls for an adequate amount of money for education and an equitable distribution of this money to school boards. The survey results indicated that many trustees feel that the local input should not be increased but that more money is required from government. Education costs will continue to rise as demands upon the education system increase and additional monies will have to be made available to education. The current 50-50 sharing of education costs by the government and school boards is viewed as inadequate by boards. In fact, boards are making a determined effort to have the government's share increased to 60% by 1992. Again, the relative contributions of government and school boards is not the function of the formula and will, therefore, not be pursued in this study. The formula does, nevertheless, distribute the education burden in an acceptable, equitable manner. There do not appear to be any serious discrepancies between educational opportunities for Wealthy boards and less wealthy boards.
An analysis of the 10 most wealthy school boards (high assessments) and the 10 least wealthy (low assessments) indicated that the average mill rate for one group differs marginally from the average mill rate for the other. Table 2 shows that the lower assessed boards in this comparison have a mill rate that exceeds the average of the other group by only 2.15 mills.
(2) To provide for greater education opportunity with a "reasonable" local tax rate.
Table 3 lists the mill rates of the 110 school divisions in Saskatchewan. The arithmetical average mill rate for 1990 is 64.39 mills while the weighted average mill rate is 65.70 mills. Whether these mill rates .are "reasonable" is more of a political question than an educational question. There are, of course, significant variations in mill rates for the different boards but an analysis of the mill rates shows that the differences are probably not much greater than one might expect with boards exercising the local autonomy that the grant formula confers upon them. The difference (range) between the lowest and highest mill rata is 62. It must be noted, however, that the low mill rates from 15 to 50 inclusive are low for very special reasons. The boards in this range include boards that are either very small, have substantial surpluses, educate a large number of Alberta students or operate under unusual circumstances in northern Saskatchewan. If these boards are not included in the calculations the range becomes 25 mills which when, such factors as extensive building programs, a large number of tuition paying students and variations in the sophistication of programs offered, are considered the range is understandable.
(3) To retain for boards a high degree of local autonomy.
Unconditional grants to boards enhance local autonomy. Most grants to boards are unconditional, however, in recent years conditional grants such as educational development funds, shared Services funds and core implementation funds have eroded some of the decision-making powers of boards. Funds that are available to boards for specific purposes only are necessary in each instances as high cost special education funding and certain types of transportation costs but the implementation of government policies through conditional funding is a serious threat to the autonomy of boards. Boards need to be aware of the dangers of special funding which is designed to meet a specific need identified not by the board but by government.
(4) To hold boards accountable for consequences of local decision-making and local policy setting.
The formula has little difficulty meeting this objective. The introduction of special programs, innovative delivery systems, richer staffing patterns and other similar board decisions are paid for by increased local taxation. Boards are directly accountable to the electorate for such additional expenditures.
Similarly, the responsibility for the deterioration of school programs due to a lack of financial input also rests with the board when local input is kept unduly low.
(5) To provide for special adjustment factors in the formula which will compensate boards for circumstances over which they have little or no control.
The formula provides these adjustments through the sparsity factor, the small schools factor, the enrollment decline factor, transportation factors and several less significant factors.
The addition of specific adjustment factors, although well-intentioned, often weakens the formula. The most equitable distribution factors to be used are those common to all divisions, such as students. The formula tends to become negatively discriminatory when factors are introduced which apply only to some boards. ?be great differences i~ divisions in the province makes it necessary that factors to address these differences ore introduced. However, the inherent dangers of these factors would suggest that the number of grant determinants be kept to a minimum. The introduction of special needs program funding factor (SNPF) in 1990 is not in keeping with the basic tenets of the foundation grant formula. An examination of this factor follows.
In 1987, the grant formula was modified to include an Established Program Funding factor (E.P.F.). This factor included the learning disabled (L.D.), the emotionally disabled (E.D.) and those who were formerly recognized under the low cost (L.C.) calculations. The change was Prompted by several factors.
With a changing L.D. and E.D. Population it vas difficult to predict the funds required in any given year. Also, there appeared to be a growing discrepancy in the number of students identified by different school divisions and, finally, there was a growing concern about the disproportionate amount of time that was required for identification purposes.
The change to E.P.F. was followed by a freeze on the financial allocations for E.P.F. This action resulted in some serious inequities 'n the recognition of students in this category. School divisions which had not identified all of their L.D. and E.D. students continued to be funded at a low level while those divisions Which Were well along the Way in their identification process when the funding was frozen continued to be funded at a higher level.
In an attempt to address these concerns and inequities, Saskatchewan Education introduced a Special Needs Program Funding factor (S.N.P.F.) for 1990. Essentially, the allocation for this factor is the product of the number of S.N.P.F. units determined for a jurisdiction and the predetermined financial allocation for each such unit. The number of S.N.P.F. units for which a board receives credit will be determined by the nature of the program offered and by the number of qualified special education support staff engaged by the board. One unit for every 200 students will be the maximum number of units allocated to a school division and for the 1990, the financial allocation for every S.N.P.F. unit is $25,000
The introduction of the S.N.P.P. factor, though well-intentioned, violates some of the basic tenets of the grant formula.
In the first Place, the recognition is program based. Not only is it program based hut it recognizes hut one component of the school program. The Saskatchewan Foundation Grant Formula strives to provide adequate funding for a basic education program in an equitable manner. Distributing funds on the basis of programs offered is not in keeping with this basic objective of the formula. Secondly, the use of numbers of staff for calculating recognition units is a throwback to the pre foundation grant formula. The present formula has refrained from using a P.T.R. (Pupil-teacher ratio) for determining recognition units since its introduction in 1972. The use of a P.T.R. for calculating the S.N.P.F. factor is setting a dangerous Precedent. Finally, the S.N.P.F. factor Provides partial recognition for additional staff members in special education but not in other Program areas. This can only be viewed as a strong departmental influence in setting program priorities at the division level.
The maximum number of S.N.P.F. is indeed based on the number of pupils in a jurisdiction hut any number less than the maximum is a function of the number of special education support staff employed by the division. When all boards ultimately reach the maximum recognition the S.N.P.F factor will become redundant. The same result could then be obtained by simply increasing the per-student recognition in the basic program since all students are used to determine the factor. Criticisms of the S.N.P.F. should not be viewed as opposition to special education funding. Indeed, Saskatchewan Education needs to be commended for its leadership role in the provision of services for special education students. However, it is imperative that the funding for such services respect the objectives of the provincial grant formula. It is encouraging to note that Saskatchewan Education is committed to an ongoing review and evaluation of the criteria used in calculating S.N.P.F. Hopefully, the concerns raised in this study will be addressed in such a review.
Of the 49 boards that responded to the survey only 32 answered question 1.1 which asks whether boards were satisfied with the formula. Eleven boards (23%) indicated satisfaction while 15 boards (31%) indicated that they were not satisfied with the formula. Six boards (124) expressed qualified support for the formula while 16 boards (36%) did not respond to the question. Thus, approximately one half of the boards that Survey question 1.1 expressed support or qualified support for the formula while the other half were dissatisfied.
Support for the formula centered around its ability to distribute monies equitably and its use of the "pupil" as a base for distribution. The provision of unconditional grants and its attention to special education students were also lauded. Only two boards, however, listed local autonomy as one of its desirable features.
Perceived weaknesses in the formula were many and are listed here in the order of the frequency in which they were mentioned:
(1) Seventeen boards (354) felt that the formula does not generate sufficient funds for boards. That the generation of funds is not a function of the grant formula even though the concern for more money is valid, has been fully discussed in this study.
(21 Ten boards (204) expressed strong disapproval of either the use of a computational mill rate in the formula or the changing of the computational mill rate every year. Three boards, however, suggested that the cmr be increased. Again, this study has tried to emphasize the importance of the computational mill rate as an equalizer in distributing grant monies.
(3) Eleven boards (22%) felt that the "recognized costs" were too low. Higher "recognized costs" are indeed desirable hut we need to hear in mind that higher recognized costs must, of necessity, be accompanied by a higher local input if the overall grant remains constant.
(4) Nine boards (184) asked for recognition of special programs. Several boards strongly support program funding as an alternative to the present method. It is difficult, however, to reconcile program based funding with local autonomy.
(5) Other weaknesses listed less frequently were the following:
(a) Excessive documentation is required for special education students.
(b) The sparsity factor is no longer required.
(c) The small schools factor needs review. This factor makes it difficult for boards to close uneconomical small schools.
(d) The small schools factor should also apply to urban jurisdictions.
(e) Boards should not receive grant reductions.
(f) The rural transportation formula favors certain types of rural jurisdictions.
(g) Actual tuition fee costs should be recognized.
(h) Discounts, cancellations and tax arrears should be considered for every board when grant calculations are made.
(i) Reduce the number of indexes used in the formula.
(j) Change the name of the computational mill rate to something less confusing.
(k) Change the school year from the calendar year to the academic year.
(1) Grants-in-lieu should be paid on behalf of all provincial government properties.
The following recommendations and observations may appear to be Tao simplistic but it should be noted that they are made in light of the discussions in this study:
(1) Generally, the formula continues to meet the objectives set out for funding education in Saskatchewan.
(2) The formula's ability to equalize local input is severely compromised by the use of a computational mill rate that is far below the average actual mill rate for the province.
(3) The problems surrounding tuition fee payments would be minimized if recognized per-pupil costs were more realistic. This would result, of course, in an increase in the computational mill rata.
(4) The introduction of more increases the probability of funds.
indexes into the formula inequitable distribution of
(5) Significant changes to the formula always results in a shift of financial support from some boards to other boards. Accordingly, major when a significant increase place.
changes are best introduced in provincial funding place
(6) In view of the low computational mill rate currently being used one would expect the least wealthy jurisdictions to have significantly higher mill rates. However, that is not the case. We can only conclude that wealthier boards do, in fact, tend to spend more and less wealthy boards spend at a lover level, as they are expected to. Table 2 shows that the higher assessed boards have higher per-pupil expenditures than the low assessed jurisdictions.
(7) The S.N.P.F. factor, currently used in the formula, needs to be reviewed. In its present form it is not compatible with the objectives of the foundation grant formula.
A good grant formula distributes education dollars as equitably as possible by addressing the common needs of school boards. It cannot and, indeed, should not address the specific needs of individual boards for any attempt to do so would, first of all, undermine the boards' autonomy and, secondly, it would make the formula so complex that it would lose the support of boards. The Saskatchewan foundation formula does a commendable job on both counts.
Prior to 1987 the capital grant formula used in Saskatchewan had three major components.
(1) A Capital Grant
This component was a direct government grant for an approved project. As a percentage of the total cost of the approved project, the capital grant varied from board to board since it was designed to be greater for the less wealthy boards. On average, this component accounted for 204 - 304 of the total cost of the project
(2) The Board Downpayment
On average, boards were expected to raise approximately 104 of the total cost of the project. However, this component vas "capped" at 2.5 mills on the assessment of the jurisdiction.
(3) A Borrowing component
The portion of the cost not covered in (1) and (2), usually from 60 - 704, was amortized by way of debentures or loans. Debentures and loan repayments, along with interest costs, were picked up by the government.
The calculations used to determine the foregoing components were altered slightly during the years that the formula was used but the concept remained intact.
In 1984 a government/SSTA committee reviewed the capital grant formula and after extensive consultation with school boards made the following recommendations:
(1) That the existing capital funding formula not be changed.
(2) That if it is deemed necessary to change the formula, the changes should be limited to an adjustment of the "cap" on downpayment requirements.
(3) That any proposal to change the present level of capital debt recognition in the operating grant formula be strongly resisted.
(4) That a distinct formula for funding roof repairs be developed by the department. However, this formula must honor the principles of neutrality and equity and should include recognition of preventative maintenance programs.
The formula was changed in 1987. The capital grant component of the formula was eliminated and the local input was increased from an average of 104 of the total approved costs of the project to an average of 20% with no provisions for a "cap". Provisions were made; however, for the amortization of the downpayment by the board. For 1988 and 1989 the formula was limited to the calculation of the board's share of the cost of the project as follows:
Board's Share = .4C (.5A.P. + .25) where:
C = approved cost of the project A.P. = ability to pay factor
~ 1 - board's grant/board's recognized costs
For 1990 a utilization factor was added to the formula. This U-factor is calculated as follows:
U-factor = .8 - Total Schedule A usable area/Total Existing Usable area
The Utilization Factor:
The Application of a utilization factor will increase a school board's share of a capital project. The factor will be applied as follows:
(1) In the determination of "usable area", net areas within a school will be used except toilets, gym service, survey, mechanical and janitor spaces and corridor circulation.
(2) It will apply only when it can be shown that the board has excess usable instructional space.
(3)The location of a school will be considered in determining a utilization factor.
For rural divisions the U.F. will be influenced by the sum of all useable areas in all the division's schools within 30 kilometers of the target school. For urban divisions, however, the U.F. will be influenced by all the usable area in all of the schools in the division.
(4) Currently the U.F. is applied only when the utilized area of a school falls below 80%.
The formula now becomes:
Board's share = C(.4 + U.F.) (.5A.P. + .25)
By way of example, let us assume that a rural board has a utilization of 75% of its facilities. Let us also assume that it receives operation grants equivalent to 60% of the recognized operating costs and that it has received approval for a $1 million project.
Before we calculate the board's share of the cost we need to calculate the U-factor and the A.P. factor
U.F. = .8 - .75 = .05
A.P. = 1 - .6 = .4
The formula now generates the following as the board's share:
= $1,000,000 (.4 % .05)[.5(.4) 4 .25]
= $1,000,000 (.45)(.2 + .25)
= $1,000,000 (.45)(.45)
= $1,000,000 (.2025)
= $202,500 = 20.25% of approved project costs.
The constants of .5 and .24 serve no other purpose but to minimize the impact of the formula on those jurisdictions whose A.P. (ability-to-pay factor) deviates from the average of .5.
Thirty-eight of the 49 boards that responded to the SSTA questionnaire commented on the adequacy or inadequacy of the capital funding formula. Twenty boards (41%) were not satisfied with the formula. The 38 boards who responded to question 2.1 dealing with capital funding were almost equally divided in their support or qualified support and dissatisfaction with the funding formula. The weaknesses of the formula, as perceived by the boards, are listed here in the order of frequency in which they were mentioned:
(1) Thirteen boards (27t) felt that an average local input of 204 of the total cost is too high. Strong support vas expressed for a return to the 10% local input that was required before 1987.
(2) Nine boards (184) called for the return of a "cap" on the local input. Six boards (12$) indicated that the present formula could "bankrupt" small jurisdictions without the protection of a "cap" on local input.
(3) Six boards (12%) expressed dissatisfaction with the approval procedures currently being used.
It was suggested that the procedure could be simplified and could be "kept out of the political arena".
(4) The following were listed less frequently:
(a) The capital budget should be increased.
(b) Costs for relocating modular classrooms should be recognized.
(c) The policy on the recognition of site costs needs to be reviewed.
(d) "We do not want a utilization factor".
(e) "Like the idea of a utilization factor".
(f) The regulations should be changed to allow boards to retain 20% of the sale of properties. This would be in keeping with the 204 local input requirement.
(g) More flexibility is required in the timing governing the sale of debentures.
Equitable funding formulas are based on the common elements of all boards' expenditures. Capital expenditures are not a common annual expenditure for all boards, in fact, some boards have no capital expenditures over long Periods of time. Three additional major differences between the operating and capital funding arrangements should be noted.
(1) On average, the required percentage of local input for capital grants is much less than it is for operating costs.
(2) Local decision-making is more restricted for capital projects than it is for a hoard's operating program.
(3) The distribution of available capital funds is not as equitable as it is for operating funds even though an ability-to-pay factor is used in calculating the local input.
The changes made to the capital funding formula in 1987 were designed to bring about a higher degree of board accountability through an increase in local input. Undoubtedly, the scarcity of provincial money had a significant bearing on the government's decision to change the funding formula. However, calling for more local input not only shifted the financial burden for capital construction from the provincial government to school divisions but also augmented the inequity of the formula. That some jurisdictions are required to have building projects every year while others have not had a major building project for many years nor will have one in the foreseeable future results in unfair demands for financial support of capital projects by some boards. It is difficult to develop a funding formula that will address the needs of hut a few school divisions whose enrollment is increasing while for most jurisdictions the enrollment is shoving a steady and significant decline. The added financial burden for these few school divisions, however, becomes unduly heavy when an already continuous local input is increased for 10% to 20%.
Heavy isolatedexpenditures are best addressed by provincial monies.
In view of the foregoing observations and in light of the responses to the Board Survey the following recommendations are proposed for consideration:
(1) That the board's share of capital projects be reduced to 10% on average. If provincial monies are not available for such a change, consideration might be given to a reduction in the number of approvals given in any one year or to the transfer of the required revenues from operating grants to capital grants. The reduction in operating grants would, of course, be an unpopular step but the required additional local input in the operating formula would be considerably more equitable than an increase in the local input for capital projects.
(2) That long term capital planning by government and school boards be encouraged to facilitate the provision of anticipated needs.
(3) That minor renovations and roof repairs be considered an operating and maintenance expenditure.
(4) That major renovations be funded as other capital projects.
(5) That the provisions for minimum board contributions be deleted. Minimum and maximum contributions tend to destroy the equalization mechanisms built into a funding formula. Now that the "caps" have been eliminated there appears to be no rationale for retaining a minimum local input.
The current capital funding formula with its ability-to-pay factor and its utilization factor attempts to distribute building monies equitably. Basically, the formula is simple and easily understood. The recommendations made in this study would not alter the basic concept of the formula hut would make the distribution of available monies more equitable.
The study vas commissioned by the Saskatchewan School Trustees Association to assess the effectiveness of the operating and capital grant formulas currently used in Saskatchewan. Boards' concerns about the formulas were obtained through a Board Survey to which almost 50% of the boards responded and through resolutions passed at conventions. Many of the concerns were well founded and formed the basis for the recommendations in this report. However, a large number of the concerns appeared to result from unrealistic expectations of the formulas and from misunderstandings about the operation of the formulas. A good portion of the study, therefore, is devoted to an explanation of the operation and purpose of the formulas.
It is not surprising that the operating grants formula received more attention by the boards in the Board Survey than did the capital funding formula. The monies distributed through the operating grants formula are significantly larger than those for capital and also, operating grants affect all boards. Therefore, this paper treats operating grant concerns in greater depth than capital funding.
In 1972 when the Saskatchewan Foundation Formula was introduced it was designed to meet the following objectives:
(1) To bring about a greater degree of equality of education opportunity.
(2) To provide for greater education opportunities with a "reasonable" local tax rate.
(3) To retain for boards a high degree of autonomy.
(4) To hold boards accountable for the consequences of local decision-making and local priority setting.
(5) To provide for special adjustment factors in the formula which will compensate boards for circumstances over which they have little or no control. These circumstances would include a marked loss or increase in students, the need f<r small schools and costs associated with the conveyance of students.
It should be noted that these objectives embodied the basic principles of equity, autonomy and efficiency.
An analysis of available data indicated that generally the operating grants formula continues to meet the objectives set out in 1972. However, changes in the delivery of education services and modifications that have been made to the formula over the years suggest that certain changes in the formula be considered and that some changes that have been made be monitored closely since they are not in keeping with the basic tenets of the formula.
The following observations concerning the operating grant formula are, therefore, made in the report:
(1) The use of the foundation grant formula be continued.
(2) The formula's ability to equalize local input is severely compromised by the use of a computational mill rate that is far below the average actual mill rate for the province.
(3) The problems surrounding tuition fee payments would be minimized if recognized per-pupil costs were more realistic. This would result, of course, in an increase in the comPutational mill rate.
(4) The introduction of more indexes into the formula increases the probability of inequitable distribution of funds.
(5) Significant changes to the formula always result in a shift of financial support from some boards to other boards. Accordingly, major changes are best introduced when a significant increase in provincial funding takes place.
(6) In view of the low comPutational mill rate currently being used one would expect the least wealthy jurisdictions to have significantly higher mill rates. However, this is not the case. We can only conclude that wealthier boards do in fact, tend to spend more and less wealthy boards spend at a lover level, as they are expected to.
(7) The S.N.P.F. factor currently used in the formula, needs to be reviewed. In its present form it is not compatible with the objectives of the foundation grant formula.
In addition to these observations the following components of the grant formula should be reviewed.
(1) The differential in recognized rates for different instructional levels.
(2) The method of determining the number of students used in calculating the basic program.
(3) The continued use of the sparsity factor.
(4) The incremental recognition for comprehensive school students.
The capital funding formula has undergone several changes in resent years. The basic changes from the formula that was used for many years and the current formula are ':be following:
(1) The capital grant component has been eliminated.
(2) The required local input has been increased on average, from 10% to 204 of the approved project costs.
(3) The "cap" on the local input has been eliminated.
(4) A utilization factor has been added.
The recommendations in this paper concerning the capital funding formula currently in use do not differ significantly from those made by the government/trustee committee in 1984. They are:
(1) That the board's share of capital projects be reduced to 104 on average.
(2) That long term capital planning by government and school boards be encouraged to facilitate the provision for anticipated needs.
(3) That minor renovations and roof repairs be considered an operating and maintenance expenditure
(4) That major renovations be funded as other capital projects.
(5) That provisions for minimum board contributions be deleted.
The study would indicate that both the operating grants formula and the capital funding formula are basically sound and continue to meet the objectives set out for the formulas. Additions, deletions and modifications to the formulas will need to be made as program delivery systems and education priorities change. The observations and recommendations in this study are designed to provide for these changes while retaining the basic concepts of the formulas.
Back to: Finance