In the Matter of the Arbitration
of a Dispute Between
MADISON AREA TECHNICAL COLLEGE
TEACHERS' UNION, LOCAL 243,
AFT, WFT, AFL-CIO
AREA BOARD OF VOCATIONAL, TECHNICAL
AND ADULT EDUCATION DISTRICT NO. 4
Mr. Steve Kowalsky, Staff Representative, Wisconsin Federation of Teachers, AFT, WFT, AFL-CIO, 2021 Atwood Avenue, Madison, Wisconsin 53704, appearing on behalf of Madison Area Technical College Teachers' Union, Local 243, AFT, WFT, AFL-CIO, referred to below as the Union.
Mr. Jon E. Anderson, Godfrey & Kahn, S.C., Attorneys at Law, 131 East Wilson Street, P.O. Box 1110, Madison, Wisconsin 53701-1110, appearing on behalf of Area Board of Vocational, Technical and Adult Education District No. 4, referred to below as the Board.
The Union and the Board are parties to a collective bargaining agreement which was in effect at all times relevant to this proceeding and which provides for the final and binding arbitration of certain disputes. The Union requested, and the Board agreed, that the Wisconsin Employment Relations Commission appoint an Arbitrator to resolve a dispute reflected in a grievance filed by "(t)he Union on its own behalf". The Commission appointed Richard B. McLaughlin, a member of its staff. Hearing on the matter was held on January 11, and February 12, 1991, in Madison, Wisconsin. Each day of hearing was transcribed, and the parties filed briefs and waived the filing of reply briefs by October 31, 1991.
The parties stipulated the following issue for decision:
When the college assigns an employee to a load level over 100% in one semester, is an adjustment computed in the succeeding semester computed from a 100% load level or a 90% load level?
RELEVANT CONTRACT PROVISIONS
. . . This agreement that is entered into shall supersede and cancel all previous agreements, verbal or written or based on alleged practices between the parties. Any amendment or agreement supplemental thereto shall not be binding upon either party unless executed in writing by both parties.
Section F-Teaching Load and Class Size
A 100% teaching load is defined as a load range of 90-100%. Any load above 100% is considered an overload. If this condition is unavoidable, adjustment shall be made in the teaching load during the succeeding semester. Any load under 90% is considered an underload . . .
The request for arbitration filed with the Commission in this matter stated the background to this grievance thus:
While we are filing as the employe organization, the parties have jointly agreed a ruling on the language referred to in the grievance . . . is required since our respective interpretations vary so greatly.
This view was borne out at the hearing, where the parties reached the following stipulations:
(T)he issue of the college's authority to assign or not to assign an overload is not before the arbitrator in this case. (1)
(T)he language at issue here is Article VI, Section F. The language in question has not substantively changed since 1969. (2)
The parties were unable to stipulate the factual background to the grievance. Evidence centered on bargaining history and "past practice".
The Evidence On Bargaining History
Paul De Rose is presently employed by the Board as a Physics teacher in its Science Department. In 1968 he was a member of the Union's bargaining team. He testified that at some point during those negotiations, the Board made the following proposal:
A 100% teaching load is defined as a load range of 95-105%. Any load above 105% is considered an overload. If this condition is unavoidable, adjustment shall be made in the teaching load during the succeeding semester. Any load under 95% is considered an underload . . .
The parties ultimately agreed to state the load range as "90-100%", with "(a)ny load above 100%" defining "an overload" and "(a)ny load under 90%" defining "an underload". De Rose acknowledged that the 1968 negotiations occurred after the Board's budget had been set, but he stated that monetary compensation was not the Union's focus in opposing the 95-105% range. Rather, "we were interested in not being overloaded". (3) De Rose stated the parties specifically addressed compensating an overload in one semester with an underload the following semester:
Well, the adjustment was to be an underload . . . An adjustment would be an underload, which would be a way underload, and that was the exact intent of the language, if you give us an overload, you'll give us an underload for adjustment. (4)
De Rose noted the compromise ultimately reached by the parties was subject to some criticism from both teachers and administrators, but represented the best result the parties could fashion in light of "the A, B, C, D categories" then being advanced by the State.
During the 1971 negotiations, each party proposed changing the language of Article VI, Section F. The Board's proposal reads thus:
A 100% teaching load is defined as a load. Any load above 100% is considered an overload. If this condition is unavoidable, adjustment shall be made in the teaching load during the succeeding semester.
The Union's proposal reads thus:
A 100% teaching load is defined as a load range of 90-100%. Any load above 100% is considered an overload. If this condition is unavoidable, the teacher shall be compensated with an underload equal to the previous overload in the succeeding semester. Any load under 90% is considered an underload. If this condition is unavoidable, the teacher shall be compensated through special assignment.
During the 1975 negotiations, the Board made the following proposal:
A 100% teaching load per semester is defined as a load range of 90-110%. The academic year teacher load percentage total shall not exceed 200%. Any load under 90% in either semester is considered an underload. If this condition is unavoidable, adjustment shall be made through special assignment to the instructor.
Edward Hellegers, currently an Economics teacher for the Board, served on the Union's negotiating team during the 1975 negotiations. He testified that the Union regarded this proposal as a vehicle by which the Board could reduce the number of teaching positions. He stated the Board advanced a similar proposal in the 1979 negotiations. In each case, the parties did not reach any agreement to change the language of Article VI, Section F.
The Evidence On "Past Practice"
Past practice evidence is not uncommon in arbitration proceedings, but is placed in quotes above to reflect its unique role in this matter. Initially, the parties stipulated that there was no controlling past practice regarding Article VI, Section F. The Board was permitted to withdraw from this stipulation, and each party submitted evidence on the point. Neither party's arguments advance the evidence as the independent source of a binding obligation.
Abdulcadir Sido is currently the Board's Dean of Health Occupations, and actively participates in the process by which work loads are assigned to teachers. He testified that every adjustment he has made for a teaching overload has been computed from a benchmark of 100%, not 90%. He stated that the Board has compensated teachers for a teaching overload by three different methods: (1) paying them for the teaching load exceeding 100%; (2) reducing their teaching load in a succeeding semester below 100%; and (3) scheduling teachers in two successive semesters for a teaching load which does not total more than 200% for the entire year.
Sido cited the cases of Keith Fleming and Jami Bandt to illustrate the first method.
He also testified that in adjusting teaching loads under the second method, he referred to a memo bearing the signature of "Toni Walski, OTT Program Director". That memo reads thus:
For curriculum design reasons, the Home Economics Division is now requesting that their Self/Group Dynamics course be taught in the Fall instead. To meet this request, OT instructors have agreed to a Fall Semester overload equivalent to the number of students in a CDDA section, if this is balanced with an underload each Spring . . .
There are two ways to handle the loads as this request is handled. The instructors' loads could remain the same, 100% for each semester as they are now, or they could be amended to reflect the overload and corresponding underload on subsequent semesters. If the latter were chosen, Fall loads would be amended as follows:
For each instructor in the Fall, adjust the Self/Group Dynamics from 4B/4C 37.2%, to 6B/6C 55.8%. This would increase their total load to 118.6% each.
For each instructor in the corresponding Spring semester, adjust the Self/Group Dynamics from 4B/4C 37.2%, to 2B/2C 18.6%. This would create a balancing underload of 81.4% in the Spring total loads for each instructor.
This adjustment has been discussed with Dr. Wright, the current instructors-- Carol Holmes and Cathy Wilson, and Union President Dick Swanson. The concept has been approved . . .
Swanson did not testify. Sido testified that he followed this memo in assigning Catherine Wilson a teaching load of 118.6% in the Fall Semester, followed by a teaching load of 81.4% in the Spring Semester of the 1988-89 and 1989-90 school years. He stated that he followed the same procedure in assigning Carol Holmes a 118.6% teaching load in the Fall Semester of the 1988-89 school year, followed by a teaching load of 81.4% in the Spring Semester.
Sido cited the cases of Mary Lynn Jensen and Cynthia Grover to illustrate the final method. Jensen was assigned a 90.24% teaching load in the first semester of the 1988-89 school year, and was assigned a 104.8% load for the spring semester. Since this totalled a 195.04% teaching load for the entire year, no further adjustment was made. Grover taught a 92.8% teaching load for the first semester of the 1988-89 and 1989-90 school years, followed by a 107.3% load for the Spring Semester of each year. She received no further adjustment. He testified he used this type of scheduling with other teachers.
Judith Olson Sutton is the Board's Assistant Dean of the Business Division, and testified that she has never made any adjustment for a teaching overload which did not use 100% as the benchmark. She cited the example of Beverly Klein, who worked a 101.2% teaching load in the first semester of the 1988-89 school year, followed by a 92.9% load in the second semester.
Janie E. Wimberly is the Assistant Dean of the General Studies Division. She testified that she has always used 100% as the benchmark for adjusting for a teaching overload, and was unaware of any case in which her department adjusted an overload using 90% as the benchmark. In her department, the adjustment for an overload has been either monetary or a reduction in teaching load from 100% in the semester following the overload.
Jerry Butler is the Chairman of the Art Department. He testified he has afforded teachers monetary compensation for an overload, but prefers to reduce a teacher's load from 100% in the semester following the overload. He stated he has never used 90% as the appropriate benchmark for an overload adjustment.
Spencer Artman is a Lead Teacher for the Science Department of the General Studies Division. Artman testified that he was aware of five instances in which a teacher who had an overload in one semester received an adjustment in the following semester based on a 90% benchmark. He stated that in two of those five instances the teacher was paid for the overload. In the remaining three, the teacher received an underload in the semester following the overload. The Union introduced the schedules of two of those teachers -- Mark Kern and Bill Huntsman. Kern worked a teaching load of 101.9% in the first semester of the 1989-90 school year, followed by a teaching load of 89.6% in the second semester. Huntsman worked a teaching load of 107.3% in the first semester of the 1990-91 school year, followed by a teaching load of 82.6% in the second semester.
Thurman Hesse is a Lead Teacher in the Technical and Industrial Division. He testified that in the first semester of the 1973-74 school year he worked a 107.5% teaching load, which was adjusted to a teaching load of 81.6% in the second semester. He also stated that in the second semester of the 1978-79 school year he worked a 105.2% teaching load which was adjusted in the first semester of the 1979-80 school year to a teaching load of 83.9%.
Bob Johnson teaches Electronics in the Technical and Industrial Division, and testified that on about six occasions, he worked a teaching load of about 102%, which was adjusted in the following semester to a load of about 88%. He stated that either he or his lead teacher would inform his supervisor of the overload in each instance. One of his supervisors eventually became a member of the Board's negotiating team.
Further facts will be set forth in the DISCUSSION section below.
THE UNION'S POSITION
After a review of the record, the Union argues that the language of Article VI, Section F, clearly and unambiguously establishes that a "load range of 90-100%" constitutes a "full time teaching load." This definition, according to the Union, recognizes the practical impossibility of assigning teachers a 100% teaching load, and has been mutually recognized by the parties since 1969. The Union characterizes the "heart of the issue" here as the fact that the Board's view of the Article VI, Section F, adjustment "would most likely fall in the 90-100% load range and, therefore, remain a 100% or full time teaching load."
The Union asserts that "(l)ogic, reason, and just common sense" mandate a conclusion that "an adjustment for a work overload must be a work underload." This conclusion is assured only by the Union's use of 90% as the bench mark for the adjustment. Using the Board's 100% bench mark, according to the Union, "leads to the outcome of an adjustment for a work overload being a work normal load."
Any other conclusion, the Union argues, leads to paradoxical results, one of which the Union summarizes thus:
Teacher A and Teacher B are in the same department, teach similar classes, and share an office. For the past four semesters, they both have had course loads of 96. In the fifth semester, Teacher B receives a load of 104. Teacher A continues at 96. In semester six, both are again at 96. Of course, the District would argue that Teacher B's load has been adjusted for the 104 overload, but how does Teacher B vis-a-vis Teacher A come to grips with that logic?
The Union concludes that this demonstrates that the Board's interpretation "has done away with the load range concept." Such a result, the Union contends, violates the clear language of Article VI, Section F, as well as "the intent of the language when it was originally negotiated in 1968."
Beyond this, the Union contends that the Board's interpretation seeks to gain in arbitration what the Board attempted, but failed, to gain in negotiations in 1971, 1975 and 1979. Nor does the change proposed by the Union in 1971 warrant any weight in this proceeding, according to the Union. The Union asserts that the 1971 proposal "goes to the issue of adjustment and not benchmarks."
The Union's next major line of argument is that there is no binding past practice governing this point. More specifically, the Union asserts that the Preamble of the labor agreement precludes any recourse to any alleged practice preceding July 1, 1989. Beyond this, the Union contends that the evidence shows that the Board has used either the 90% or the 100% benchmark. It follows, according to the Union, that "(t)he guidepost of past practice does not exist in this case."
The Union concludes that "the Arbitrator should declare that an adjustment for an overload should be based on the 90% level of the load range."
THE BOARD'S POSITION
After a review of the record, the Board asserts that the labor agreement defines not "what rights management has reserved to itself" but "what rights it has ceded away or agreed to share with the employees." With this as background, and after noting that Article VI, Section F, does not define the specific adjustment which must follow an overload, the Board concludes that "(a)s the contract does not define the nature of this adjustment, the determination of an appropriate adjustment must be left to management and the exercise of its permitted discretion."
While acknowledging that the load range concept responds to the practical impossibility of scheduling each teacher at an exactly 100% load, the Board asserts this "safety net" "should not be construed to negate legitimate employer goals of maximizing the assignments of staff." More specifically, the Board asserts that "(t)he load range creates a fiction which equates a 90% assignment with a full load." This fiction can not mask, the Board argues, that the Board "could still assign a full ten more load points to the teacher and still be within the contractual definition of full load." Whether or not the Board chooses to use this excess capacity is a management prerogative, according to the Board.
The Board specifically rejects the assertion that an underload must follow an overload, reasoning thus:
The issue is not whether the overload is followed by a normal load. The union's point of reference is awry. The point of reference must be the extent to which an adjustment following an overload deviates from the assignment which management could lawfully impose under its broad assignment authority.
The union argues that to have a normal load follow an overload is illogical. (This) argument, however, belies logic in light of the revelations of its own witnesses that a 105% assignment is not a 15% overload, but rather is a 5% overload.
The Board's next major line of argument is that the language of Article VI, Section F, is clear and unambiguous "and supports the position of the employer." That the section does not specifically state that an underload be the adjustment which follows an overload is "an obvious and glaring omission (which) must be recognized by the Arbitrator." The section requires, according to the Board, only that an adjustment be made. The specific adjustment, the Board concludes, is determined by the Board.
Beyond this, the Board asserts that evidence regarding the original intent of the drafters' of Article VI, Section F, is irrelevant in light of the clarity of the contract and is factually suspect since "(t)he union . . . is content to rely on the memory of its own witnesses in a situation where the District's primary proponent of workload is deceased." In any event, the Board concludes that the lack of any writing on the point dooms such evidence in light of the Preamble to the labor agreement.
The Board contends that the Union's 1971 proposal on workload unmistakably establishes that the Union is seeking to gain in arbitration what it failed to achieve during collective bargaining. To accept the Union's interpretation here would, according to the Board, impermissibly alter the parties' agreement.
Contending that the evidence establishes that the Board has used 100% as the benchmark in determining the adjustment following an overload, the Board concludes "the application of the adjustment over the past many years of the relationship between the parties (is) further evidence in support of the Employer's interpretation of the clause at issue." Such evidence is not, according to the Board, evidence of past practice, since "the Employer views past practice as a limitation on managerial prerogative." That certain teachers may have received an underload following an overload does not rebut this point, since such testimony establishes only that the Board chose to schedule such teachers "less than 90% following an overload."
The Board then asserts that the Union's interpretation would lead to a harsh and absurd result. The Board puts the point thus:
If the union were to prevail, the Employer would have its authority to assign significantly curtailed. A 102% assignment, an overload, in one semester would be required to be followed by an assignment not exceeding 88% . . . A 102% assignment is not a 12% overload, but rather a 2% overload. Thus, the Employer, under the union's construction of the language at issue, has been divested of its authority to assign 10 load points. This is absurd.
The Board concludes that "the Arbitrator (should) support the interpretation . . . that adjustments in light of an overload are computed from a 100% load level."
The issue for decision has been stipulated. Under any view of the record which poses it, that issue poses a difficult interpretive point. The closeness of the issue dictates that it be resolved as narrowly as possible. It is thus necessary to review the parties' stipulations, to hone the stipulated issue.
The parties have agreed that Article VI, Section F, is the provision requiring interpretation. Beyond this, the parties have agreed that the benchmark for the "adjustment" referred to in that section is the interpretive point posed, and that the "adjustment" itself is not at issue. Finally, the parties' arguments establish that the definition of the benchmark is not in issue. Rather, the issue is solely whether the benchmark for an adjustment in teaching load is 90% or 100%.
Each party contends Article VI, Section F, is clear and unambiguous, and can be interpreted only one way. Each party, however, plausibly advances a different way. Because each interpretation has support in the language of the provision, it can not be said to clearly and unambiguously support either interpretation.
Contractual ambiguity is best resolved by recourse to past practice or bargaining history, since these factors focus directly on the conduct of the bargaining parties, whose agreement is the basis and the goal of contract interpretation.
Stating the desirability of these guides in the abstract, however, offers limited guidance in the resolution of this case. The bargaining history and past practice evidence, with limited exception, will not support any definitive conclusion regarding whether 90% or 100% is the appropriate benchmark.
As preface to an examination of the evidence on past practice, it is necessary to touch upon the parties' citation of the Preamble to the labor agreement. That section "supercede(s) and cancel(s) all previous agreements . . . based on alleged practices". This provision is not relevant to the evidence posed here. Both parties acknowledge that Article VI, Section F, governs the issue, and neither party asserts a past practice which constitutes an obligation independent of the written labor agreement. Rather, each party points to past practice to clarify the language of the current labor agreement. There is, then, no "previous agreement" to be superceded or cancelled by the Preamble.
Although the Preamble does not preclude recourse to past practice in this case, the evidence will not permit any reliable conclusion regarding the appropriate benchmark. The persuasive force of a past practice is rooted in the agreement manifested by the parties' conduct. In this case, there is no conclusive evidence that the parties, by conduct, ever mutually acknowledged that either 90% or 100% was the appropriate benchmark.
The Walski memo, as relied upon by Sido in adjusting for overloads, is the most persuasive evidence offered by the Board that the Union understood that 100% is the governing benchmark. That the agreement the memo references was not corroborated by any party to the agreement does undercut its persuasive force regarding what specifically was agreed to between the Union and the Board. However, even if the agreement is treated as having been reached in the terms noted in the memo, Hesse and Artman, as lead teachers, testified that they did not view 100% as the relevant benchmark. Nor did Johnson, as an individual bargaining unit member. Each of these witnesses testified they shared their view with their supervisors. Whatever agreement was reached regarding the Self/Group Dynamics Instructors had, then, less than unit-wide impact. The issue posed for decision here does, and there is no persuasive evidentiary basis to conclude that the Union ever agreed to the 100% benchmark as the basis for an Article VI, Section F, adjustment in teaching load.
That the asserted practice is not uniform also undercuts its persuasive force. The Board has contended that the instances of underload cited by Hesse, Artman and Johnson indicate no more than that the Board, in its discretion, chose that load. Those instances may show, the Board asserts, only that the teachers could have been given no greater load without producing another overload. Even making the difficult assumption that the proportional relationship between Huntsman's 107.3 and 82.6 per cent loads and between Kern's 101.9 and 89.6 per cent loads is coincidental to a 90% benchmark leaves unaddressed Artman's testimony that he informed Board administration he expected an underload to follow the overload. Similarly, it is difficult to accept Hesse's receipt, in the 1973-74 school year, of an 8.4% underload following a 7.5% overload, and, in the 1978-79 school year, of a 6.1% underload following a 5.2% overload as anything other than the use of a 90% benchmark. Johnson's recall of a similarly symmetrical adjustment was not documented, but still stands unrebutted, as does his testimony that the underload was specifically requested of his supervisors.
The evidence of bargaining history, with one exception, afforded more controversy than guidance. The Union's 1971 proposal was the most controversial piece of evidence, and illustrates the difficulty of drawing conclusions in this area. That proposal, which was not agreed to, seeks essentially the result sought by the Union in this proceeding. This will support the inference urged by the Board that the Union seeks in arbitration what it failed to achieve in negotiation. However, it also supports, with equal validity, the inference that the Union sought to clarify what it thought it had achieved with the less precise terms agreed to in 1969. As noted above, certain aspects of the past practice evidence will support either inference. The same difficulties attach to the more extensive evidence introduced regarding Board proposals.
The testimony of Hellegers and De Rose, who participated in various negotiations is, with one exception, unhelpful in resolving this point. Each noted the Board negotiators were fully aware of the Union's position that an underload must follow an overload. This may be the case, but it is less than apparent that the Board negotiators accepted that view. That the bulk of the past practice evidence indicates a contrary view is unexplained by Hellegers' or De Rose's testimony.
De Rose's testimony does, however, offer some insight into the purpose of Article VI, Section F. He noted that the bargaining context dictated that the original version of Article VI, Section F, did not seek monetary compensation, but sought to deter overloads. This testimony stands unrebutted. More significantly, it is supported by the language agreed to. The section does not specifically reference monetary compensation, but does expressly note that the "condition" of overload should be "unavoidable". This underscores and supports De Rose's testimony on the point.
Viewed as a whole, the evidence of past practice and bargaining history document the extended history of a divisive point which has been resolved on a division by division, case by case, basis. Neither aspect of the evidence, with the one exception noted above, affords significant guidance on the unit-wide interpretive issue posed here.
This extended preface establishes that the language of Article VI, Section F, with the limited exception noted above, must be interpreted on its written terms.
The language of that section affords more support to the Union's than the Board's interpretation. The Union's interpretation fully accounts for the definition of "teaching load" in the first sentence of the section. That definition defines a "100% teaching load" as "a load range of 90-100%." The Board's assertion that this "fiction" was created to provide a safety net has persuasive force, but strains the terms employed. The first sentence does not define the "load range" as a full-time "teaching load" but as a "100% teaching load". As written, the sentence makes 90% = 100%, not 90% = full-time. That the Board can choose to assign anywhere within this range is undisputed. Having exercised its discretion to assign within the range, however, Article VI, Section F, requires that any load within the range be considered 100%.
The second and fourth sentences, which define overload and underload, refer to a "load", not to the "teaching load" defined in the first sentence. These sentences, by defining what "load" constitutes an "overload" or an "underload" establish why a 105% load is a 5% overload and an 85% load is a 5% underload. The "load range" of the first sentence does not, as the Board asserts, define an overload or an underload. The distinction between "load" and "teaching load" becomes significant because the third sentence governs any "adjustment . . . in the teaching load". If 100% is the sole point governing this benchmark, the Board has in effect read the load range defined in the first sentence out of existence. If the definition of the first sentence was of a full-time teaching load, this view would be more persuasive. The definition, however, is of a "100% teaching load."
That the Union's view of these terms is more persuasive is underscored by the effect of its interpretation viewed in light of the intended effect of the section. As noted above, the language of Article VI, Section F, and De Rose's testimony indicate the parties sought to avoid overloads. The Union's interpretation effects this purpose, while the Board's does not. The Union affords an overloaded teacher a reward for the overload, and in so doing, creates a disincentive for the Board to overload. The Board's interpretation affords no disincentive to overload, for an overload (i.e. greater than 100%) in one semester is typically followed by a normal load (i.e anything between 90% and 100%) the succeeding semester. Under the Board's view, a teacher who typically taught a 94% load could be "adjusted" for an atypical 104% teaching load in one semester with a higher than typical load (e.g. 96%) the following semester. While numerical precision in the scheduling area is more elusive in practice, the example does demonstrate that the Board's interpretation affords no disincentive to overload.
Before closing, it should be stressed that the issue posed is narrow, and the conclusion stated above is also narrow. How the Board adjusts an overload has not been put in issue. The conclusion stated above establishes that if the
Board adjusts an overload in one semester with an adjustment to a teacher's teaching load during the succeeding semester, that adjustment must be made from a benchmark of 90%, not 100%.
When the college assigns an employee to a load level over 100% in one semester, an adjustment computed in the succeeding semester is computed from a 90% load level.
Dated at Madison, Wisconsin, this 14th day of January, 1992.
Richard B. McLaughlin, Arbitrator