Founded in 1932, the Citizens Budget Commission is a nonpartisan, nonprofit civic organization devoted to influencing constructive change in the finances and services of New York City and New York State governments. Because employee compensation represents approximately one-half the City's budget, the Commission has always paid careful attention to employment practices and wage policy. And because 2000 is a year in which contracts with virtually all the City's employee unions expire and are being renegotiated, the Commission decided for this year to devote much of its research and other resources to informing and influencing this process in a constructive manner.

Our efforts began in December of 1999 with a conference on "The Citizens' Stakes in Collective Bargaining." With the participation of union leaders and management representatives, we discussed the citizens' interest in a municipal workforce that is better paid, better managed, and more productive. When the first of the expiring contracts reached its ending date, March 31, 2000, we reminded the Mayor that timely settlement of the negotiations was important to citizens. While workers can receive raises retroactively, it is virtually impossible for the citizens' benefits in the form of better managed and lower cost services to be implemented retroactively.

In June of this year, the Commission released a report, Developing a Public Workforce for the New Millennium, prepared by The Delaney Policy Group. This report examined examples of how collective bargaining has been used in other cities to promote better and more efficient public services. It pointed to three strategies that New York City should follow in order to match and exceed the gains in competing urban areas-performance pay, better training to prepare the workforce, and building competition into the structure of service delivery through competitive bidding.

This report on the contract between the Board of Education and the United Federation of Teachers is the next step in the Commission's ongoing efforts to provide suggestions to improve public services through collective bargaining. Public education is a vital service that needs reform, and we seek to make a positive contribution toward that effort with this report. The Commission plans additional activities for later this year to present ideas for the contracts between the City and other municipal employee unions.

The Commission's activities relating to collective bargaining are overseen jointly by two of its standing research committees-the Budget Policy Committee and the Technology and Public Service Committee. I co-chair the former Committee with Richard A. Levine, and Bud Gibbs and Frances Milberg co-chair the latter committee. The other members of these committees are Alan M. Berman, Jeremiah Blitzer, Mark Brossman, Deborah A. Buresh, Lawrence B. Buttenwieser, Denis Curtin, Evan A. Davis, Stephen DeGroat, Cheryl Cohen Effron, Morton Egol, Linda Fan, Kenneth D. Gibbs, Daniel J. Gross, James F. Haddon, Peter C. Hein, H. Dale Hemmerdinger, Paul M. Hopkins, Brian T. Horey, Lawrence S. Huntington, Jerome E. Hyman, Peter C. Kornman, Robert Kurtter, Hugh Lamle, James L. Lipscomb, Stanley Litow, William F. McCarthy, Frank McLoughlin, John R. Miller, Philip L. Milstein, Lionel I. Pincus, Steven M. Polan, Robert E. Poll, Carol Raphael, Edward L. Sadowsky, Lee S. Saltzman, Barbara Shattuck, Larry A. Silverstein, Robert W. Strickler, Barry F. Sullivan, Merryl H. Tisch, Robert V. Tishman, W. James Tozer, Jr., Ronald G. Weiner, Stephen H. Weiss, and David I. Weprin.

This report was prepared by Charles Brecher, Director of Research, Elizabeth Lynam, Senior Research Associate, and Cynthia S. Searcy, Research Associate. They gratefully acknowledge assistance and cooperation in data collection from several members of the staff of the Board of Education. Nicolette Macdonald, the Commission's Publications Coordinator, prepared this report for publication.

EXECUTIVE SUMMARY *

INTRODUCTION *

CURRENT POLICIES AND PROBLEMS *

THE CHALLENGE AND OPPORTUNITY OF INCREASED RETIREMENTS *

RECOMMENDATIONS *

CONCLUSION *

Improving the quality of the New York City public schools is a complex problem, without a single, simple solution. Political and educational leaders should pursue long-run strategies that have the potential eventually to yield positive results. In 1996, as part of its Budget 2000 Project, the Citizens Budget Commission recommended two basic strategies: (1) Harness the innovative energy of competitive processes through charter schools, competitive bidding for some instructional services, and other measures; (2) Reform the governance and management of public schools to improve accountability and to provide financial incentives for high performance by principals and teachers.1

Since then considerable progress has been made in pursuing these strategies, but much remains to be accomplished. A major opportunity to promote this reform agenda is the negotiation of a new contract for public school teachers by the Board of Education, the Mayor, and the leadership of the United Federation of Teachers (UFT). In 1995 these parties agreed to a five-year contract that expires on November 15, 2000. In September, formal negotiations will begin over terms of the next contract.

The purpose of this report is to identify priority changes to the contract that would help make the public schools more effective. Provisions of the current contract relate to compensation, working conditions, the criteria for assigning teachers to schools and positions, and the mechanisms for developing and disciplining teachers who perform unsatisfactorily.

The context in which these negotiations will take place includes the potential loss of many teachers due to retirement. This situation offers both a challenge and an opportunity.

The challenge arises from the need to replace a large number of teachers. It is possible that as many as 17,000 teachers will retire in the next four years.

The opportunity arises because the large-scale replacement of more senior teachers with new hires will free substantial resources that can be used to improve educational effectiveness. Because current policies give large rewards for seniority, the retiring teachers are also the highest paid teachers. In contrast, the newly recruited teachers are in the lowest rungs of the current pay scale. The replacement in this manner of 17,000 teachers over four years would lead to annual savings in teacher compensation at the end of the period of about $490 million.

Policies established in the current contract are directly related to four significant problems: (1) Lack of performance incentives for teachers. (2) Shortages of qualified teachers in selected categories and at selected schools. (3) Insufficient managerial authority for principals. (4) Inefficient deployment of teachers that reduces time spent in instructional activities.

If organizations are to perform well, then their employees should be paid in relation to their contribution to that performance. Historically, this was the logic for establishing pay differentials for teachers with more experience and education. However, over time the pay structure has given more emphasis to these factors than is justified by their contribution to teacher performance. Under the current contract teachers can increase their pay by up to $8,270 annually by earning graduate credits and by up to $29,820 annually due to longevity.

Numerous studies have failed to show a systematic relationship between graduate school credits and teacher performance, and experience significantly adds to teacher effectiveness only during the initial part of a teacher's career. Thus, most of the current differences in pay among teachers established by the contract bear little relationship to how well the individuals function as teachers, nor do they serve as an incentive to high performance.

Innovative school districts in several areas have established compensation systems that reward other factors. These systems share the common purpose of seeking to link performance and pay more closely. The approaches being tried differ in two key aspects. First, some provide rewards to individual teachers, while others reward larger work units such as an entire school. Second, some base the reward on the performance of students using standardized tests or other measures as the yardstick, while others provide the reward based on specific accomplishments (such as submitting a portfolio or passing a national teachers examination) of staff members themselves rather than their students.

A new plan ought to have three basic elements. First, base salaries ought to represent a large majority of total compensation to teachers, but should leave a significant portion of compensation to be determined based on performance. The base salaries should vary with a few increments based on experience and with a significant increment for widely accepted professional development milestones such as National Board Certification.

The second component of pay should be based on the accomplishments of individual teachers during the year. This should follow the Douglas County (in Colorado) approach described in this report, with pay available for taking on additional responsibilities, for achieving certain developmental goals, or for special service to the school or district.

The third component of pay should be given to schools or other work groups whose students have educational gains above those that could normally be expected. The approach used in Dallas and described in this report can be a model.

The Board of Education does not have a single job classification known as "teacher." Rather it hires people in 225 separate titles that are specific types of teachers.

Determining when there is a "shortage" for a particular job title is conceptually difficult. Shortages, like beauty, are to some extent in the eye of the beholder. That is, employers identify situations when they must change their recruitment practices to adapt to changes in the nature of the labor supply as a "shortage."

The Board has been criticized for failing to adapt its recruitment and hiring practices to current conditions. The Board recruits and hires close to the start of the school year, relies primarily on City University and other New York City institutions for new teachers and does less extensive recruiting at upstate State University campuses and at out-of-state colleges than do many smaller districts, and fails to identify ways to improve its recruiting efforts by collecting data on why recruits do not accept jobs or where they do accept jobs.

How, then, does the Board determine that a shortage exists? It classifies job titles as in a shortage (or not) annually based on its recruiting experience. In the 1999-2000 school year, 33,633 teachers or about 43 percent of the total were in 142 shortage job titles.

The shortage conditions impact most acutely on the schools that teachers consider the least desirable. Because the deployment of teachers is governed by their seniority, the more experienced teachers are more likely to be in schools they desire, while the new and often uncertified teachers must accept the least desired locations. Based on their inability to attract certified teachers, about 40 percent of the public schools are in a shortage situation in the sense that they have serious recruiting problems. The combined share of schools with recruiting problems (40 percent) and job titles defined as in shortage (43 percent) indicates that about 17 percent of the system's total positions are in serious shortage.

The new contract should provide differential pay for teachers qualified for and serving in specific titles that are suitably deemed in a shortage condition, and who also agree to work in those schools facing the greatest difficulty in securing qualified teachers. In this way, added funds can be targeted to recruit those teachers needed most to those places with the greatest need.

As previously indicated, about 17 percent of the Board's positions are in an acute shortage condition. Differentials in the range of 20-25 percent ought to be offered for the jobs. Based on current average teacher salaries, providing 20-25 percent differentials to this group would cost between $133 million and $162 million annually.

Investments should also be made in bringing new people into the teaching profession and helping them obtain certification. The Board should greatly expand existing efforts and aggressively pursue new options for uncertified teachers to gain their credentials. The Board could contract with local colleges to design programs, perhaps delivered on site at schools or district offices, which offer appropriate graduate credits and help prepare for the certification exams. One-time bonuses could be paid to uncertified teachers when they meet all the requirements. An investment of $3,000 in each of these recruits to help them obtain certification, and a $1,000 bonus for achieving it, would cost $20 million annually, if all those eligible successfully participated.

If a sudden upsurge in retirements causes a sharp, but temporary, increase in hiring needs in the next year or two, then the Board could also anticipate temporarily drawing on retired teachers as a source of qualified personnel. Retired teachers could be paid to work for a defined period without losing all, or even some, of their pension benefits. In order to avoid abuses, the arrangement should be limited to those who have been retired for at least one year and limited in duration to just one year of paid employment.

Problem: Insufficient managerial authority.

Two significant limitations on principals' authority are the contractual provisions relating to seniority and the cumbersome disciplinary process. In the 1998-99 school year only 523 of 75,209 or 0.7 percent of teachers reviewed were rated unsatisfactory. Some principals are reluctant to give such ratings or seek to discipline teachers for other reasons because of the time and effort required for the arbitration process that follows. In February 2000, 301 teachers were awaiting completion of their disciplinary proceedings. The average time for these proceedings was over 18 months, and one case had been pending for almost seven years.

Perhaps the leading constraint on principals' ability to use their judgment in deploying staff is the contractual requirement that seniority determine most assignments. Principals should be given more discretion in determining which teachers may transfer to their school, and what assignments each teacher is given within their school. An appropriate step would be to modify the seniority rules to allow principals to select teachers for given posts from among those three (or five) candidates with the most seniority.

Another important boost to principals' managerial authority is to make it easier for them to discipline bad performance or behavior. A principal should have the ability to terminate a teacher on grounds of misconduct or repeated unsatisfactory performance. While an appeal mechanism is an appropriate safeguard, it should be a streamlined process.

Overall about one-third of teacher time is spent in duties other than classroom instruction. About 17 percent of all available teacher time is required to cover classes while their colleagues are on preparation periods or administrative duties. An almost equal amount of teacher time is diverted from regular classroom assignments because teachers are on sabbatical or other leaves, are serving as union representatives, or have other non-instructional assignments.

This misallocation of the scarce resources of available teachers should be rectified by: reducing combined preparation and administrative periods for junior high and high school teachers from 10 to five per week; reducing preparation periods for elementary school teachers from five to three per week; eliminating sabbatical leaves for travel and study; eliminating subsidized time for union activities. Reallocating teacher time in this way would be the equivalent of adding more than 4,250 teachers to the payroll.

The previous recommendations would be significant steps in the necessary, broader long-run effort to improve the performance of New York City's public schools. Moreover, these changes require less additional funding than is likely to be available due to the replacement of more-senior with less-senior teachers.

Of the eventual annual savings of nearly $500 million, less than half would be needed to fund the pay differentials for teachers in shortage schools and job titles ($162 million) and to provide the financial incentives to uncertified teachers to become permanently qualified ($20 million). The recommendation to restructure compensation to de-emphasize longevity and reward performance is "budget neutral" and need not require additional funding, and the recommendation to increase classroom time actually yields savings.

The likely savings in the Board's payroll in coming years is a major opportunity for educational and political leaders to set clear priorities for the use of additional funds to improve the schools. The most desirable strategy for new money is to increase the quantity, rather than the price, of the labor and capital resources made available to children. With respect to quantity of labor, the Board has larger average class sizes than the rest of New York State. A priority claim on new resources should be hiring sufficient additional teachers to achieve significantly reduced class sizes in the early grades.

With respect to capital inputs, modernization and added capacity of school buildings are another high priority. The Citizens Budget Commission previously has recommended allocating capital budget funds to create modernized, air-conditioned schools equipped with computers to be used on a year-round basis. Maintaining the current 180 days per year of schooling for students, but spreading the time over the full calendar year rather than only the current school year, is an accepted practice in other states and would increase the capacity of existing schools by as much as one-third. Using $250 million of potential annual savings to support debt service on bonds would permit about $3 billion to be added to the Board's $7 billion, five-year capital plan.

Finally, the scope of this year's negotiations should not be confined to the pay and working conditions of public school teachers. Stimulating competition to public schools, as well as improved staffing and management of the public schools, is the second element of the long-term strategy for educational reform. A leading policy instrument for promoting innovation is charter schools. State law sets the framework for these schools, but it must be recognized that the teachers' union is a powerful lobbying force in the State legislature that helps shape these laws.

Just as collective bargaining sometimes involves agreements for the Mayor, the Board and the union to support pension law changes, the negotiations can also seek joint support of measures to facilitate alternatives to badly performing schools. For example, the 1998 charter school law requires that a charter school with more than 250 students have its teachers covered by the collective bargaining agreement that applies to public schools. In New York City, this cap is unrealistically low and should be raised or abolished in order to permit established school management firms to launch charter schools. Part of the bargaining should be obtaining union support for appropriate legislative changes.

After five years of living with a contract that made little progress in improving public education, the Mayor and the Chancellor now have an opportunity to achieve major gains. The current round of negotiations should not just raise the income of teachers; it also should raise the quality of public education.

Most New Yorkers are justifiably concerned about the quality of their city's public schools. Too few elementary school students become proficient in reading and mathematics, and too many high school students drop out or graduate without acquiring the skills necessary for success in college or in the contemporary labor market. An inadequate education often condemns these youth to dead-end jobs and leaves the city with an uncompetitive labor force. The future well-being of New York's economy and residents is dependent on an improved school system.

Multiple factors underlie the public schools' poor performance. Students' families are sometimes unable to provide the support that aids effective learning; many school buildings are crowded and in disrepair; curricula are not always well-designed or sufficiently rigorous; and the governance and management of the system are not structured to provide clear accountability.

Given the complexity of the problem, there is no single, simple solution. Instead, public officials should steadily and aggressively pursue long-run strategies that have the potential eventually to yield positive results. As part of its Budget 2000 Project, the Citizens Budget Commission recommended two basic strategies: (1) Harness the innovative energy of competitive processes through charter schools, competitive bidding for some instructional services, and other measures; (2) Reform the governance and management of public schools to improve accountability and to provide financial incentives for high performance by principals and teachers.2

Since the Commission released its report nearly four years ago, considerable progress has been made in pursuing these strategies. The State passed legislation authorizing charter schools in 1998 and 97 charter schools have been established; the governance of the Board of Education and its community school districts was revised in 1997 to make district superintendents more accountable to the Chancellor; earlier this year a landmark agreement with principals eliminated their tenure and linked their pay to school performance.

Much remains to be accomplished, but the two strategies remain sound. A major opportunity to promote this reform agenda is the negotiation of a new contract for public school teachers. Improvement in the quality of teaching cannot be achieved by administrative fiat. It requires new ways of recruiting, deploying, and compensating teachers. And these policies are determined largely through collective bargaining.

The processes and standards for assigning and paying teachers are set in negotiations among representatives of the Board of Education, which governs the schools, the Mayor, who funds the system through the City's budget, and the leadership of the United Federation of Teachers (UFT), which represents the Board's pedagogical employees. The outcome of these negotiations is embodied in the contract agreed to by the Board and the UFT.

In 1995 these parties agreed to a five-year contract that expires on November 15, 2000. In September, formal negotiations will begin over terms of the next contract. These negotiations are not just about the level of pay for teachers; they represent a major opportunity for the Mayor, the Board of Education, and the UFT to agree upon changes that can improve the quality of the public schools.

The purpose of this report is to identify priority changes to the UFT contract that would make the public schools more effective. The first section describes the policies set forth in the current contract and the problems associated with them. The second section considers the challenge and opportunity presented by the potential loss in coming years of a relatively large number of current teachers due to retirement. The final section presents recommendations for changes in the next contract that can improve the performance of the public schools.

Among the important provisions of the current contract are those relating to compensation, working conditions, the criteria for assigning teachers to schools and positions, and the mechanisms for professional development and for disciplining teachers who perform unsatisfactorily. It is important to understand these complex provisions in order to identify how they can be constructively changed.

Compensation. Teachers' compensation has five elements-a basic annual salary, optional additions to that salary, health insurance, fringe benefits provided through a union welfare fund, and pension benefits provided through the New York City Teachers' Retirement System.

The annual salary of a teacher is based on two factors-length of time served at the Board of Education and level of graduate education achieved. Longevity increases are granted in multiple increments or "steps," with adjustments every six months or annually during the first eight years and then at 10, 13, 15, 18, 20, and 22 years of service. Educational achievement is categorized in terms of the number of graduate credits or degrees earned, with increases granted when teachers earn 30 credits or a masters degree and 60 credits or a doctorate.

The multiple salary levels available based on combinations of education and length of service are shown in Table 1, which is extracted from the current contract. The rewards are much greater for length of service than for educational achievement. The maximum gain from graduate education is $8,270, the difference between the annual salary of a teacher with a bachelor's degree and one with a doctorate. The maximum gain from longevity is more than three times greater or $29,820, the difference between the salary of a new teacher and one with 22 or more years of service.

Two additional points should be made about the salary policies incorporated in the current contract. First, it enhanced the emphasis given to longevity. Effective in December of 1999 a new "step" was added to the schedule for 22 years of service; prior to that, the last step was at 20 years of service. This disproportionately benefited teachers with 22 or more years of service. Second, continuing the prior policy of rewarding longevity meant that the pay increases received by teachers over the term of a contract exceed the amounts by which the salary at any given step is increased. To illustrate, consider that all the specific pay amounts shown in Table 1 are 11 percent greater than they were at the start of the contract in 1995, reflecting the increases negotiated in that contract.3 However, all the teachers working over the period of the contract will have earned pay increases greater than 11 percent, because they also will move up the seniority steps. Thus, for example, a teacher starting in 1995 would receive the 11 percent increase granted to teachers in step 1a and also would have moved from step 1a to step 6a to gain an additional increase of about 9 percent, bringing the total pay gain over the contract to more than 20 percent.

As shown in Table 2, most teachers have earned the differential payments associated with advanced graduate education. More than 34,000 or 43 percent of the 78,468 teachers have a doctorate or equivalent graduate credits, and another 27 percent have earned a master's degree. In terms of longevity, more than 14,000 teachers or 18 percent have at least 22 years of service and over 21 percent have at least 20 years. About 10 percent of the teachers are in their first year, and about 35 percent of the teachers have less than five years of service.

Teachers can receive additional pay above their base salary by volunteering to work extra hours. This additional work typically takes place after the regular school day, when teachers conduct extracurricular activities, or during the summer when special programs are conducted. Pay for this work is at a fixed hourly rate (currently $33.18) and does not vary with education or longevity. In fiscal year 2000 the Board budgeted approximately $291 million for these additional services; this represents an average of about $3,700 per teacher.4 Since not all teachers participate in these programs, those who do may earn substantially more than $3,700.

It should also be noted that in June 1999, the Board and the UFT agreed to special compensation arrangements beginning in the 1999-2000 school year for teachers working in selected schools having unusual problems. These 40 schools, designated Extended Time Schools (ETS), have been on the State Education Department's list of Schools Under Registration Review (SURR) because they have been performing badly for several years. Teachers must make a three-year commitment to work in these schools. These teachers receive salaries 15 percent higher than what they would otherwise have been paid under the contract. In exchange, the teachers agree to participate in a week-long professional development program before the start of the school year and to work an additional 40 minutes each school day, usually before the start of the regular school day. This extra 40 minutes is devoted to professional development during September and for two days per week in other months; for the remaining three days per week during the school year, the extra 40 minutes is spent in small group (less than five students) instruction.5

Through the extra compensation for summer and extracurricular activities and for the ETS plan, many teachers are able to achieve annual earnings well above the base salaries shown in Table 1. For example, a highly experienced teacher with the maximum education who works at an ETS school would earn $80,500 rather than the $70,000 maximum shown in the contract schedule; and a newly hired teacher with a master's degree who volunteered for after-school or summer programs might earn $6,000 in supplemental pay to raise the base pay of $36,045 to $42,045.

In addition to setting salaries, the contact provides that teachers receive health insurance, welfare fund benefits, and a retirement benefit. With respect to health insurance, the Board must make available a choice of plans and pay the full cost of a comprehensive plan. This parallels the benefits provided to City employees, and it is a more generous benefit than is typically provided by private sector employers.6 In fiscal year 2000 the average cost per teacher for this coverage was $3,505 annually.7

The contract also requires an annual contribution by the Board to the UFT's welfare fund of $1,320 per teacher. This amount was increased in the current contract from $1,145 in 1995. The union administers this fund and provides supplemental benefits to the basic health insurance plan. These supplemental benefits include relatively generous prescription drug coverage and modest reimbursement for dental services and for optical services. In fiscal year 1998, the latest audited year, the UFT welfare fund received contributions totaling $164 million and made benefit payments of $154 million, with the remainder covering administrative costs.8

Table 3 summarizes the changes in each type of compensation (excluding pensions) in the 1990s for five representative categories of teachers. The "minimum" amount shown is for a teacher starting in each year with minimum qualifications; this entry-level compensation increased approximately 31 percent over the decade, slightly more than the 30 percent increase in local consumer prices. However, for a teacher starting at the entry level in 1990 and remaining employed for ten years, compensation grew 79 percent because of increases granted for longevity. For a teacher following a typical career path, and receiving increments for graduate credits as well as longevity, the gain was 108 percent. A teacher who had a graduate degree and some experience at the start of the decade received an increase of almost 70 percent. Finally, a teacher at the maximum pay level in 1990 received an increase of 41 percent.

The relationship between collective bargaining and retirement benefits is complex. Teachers participate in the New York City Teachers' Retirement System (NYCTRS). This pension program is governed by State law, and its benefits are protected by provisions of the State constitution that prohibit reductions in employee pension benefits once they are granted. The Board and the UFT sometimes have agreed as part of collective bargaining to seek legislation changing pension benefits, and this agreement has been incorporated into the contract. However, on other occasions, the UFT has unilaterally sought legislation enhancing benefits, and these changes are not part of collective bargaining. In the latest legislative session the UFT, together with other public employee unions, sought and obtained two major types of pension enhancements. The first, opposed by the Mayor and not discussed as a part of collective bargaining, provides for automatic cost of living adjustments to pension benefits. The second, supported by the Mayor, increases benefits for employees with more than 10 years of service and eliminates a previously required 3 percent contribution to the pension fund by employees with less than 10 years of service. The latter legislation, which effectively gives employees at least a 3 percent pay increase, is intended to be a part of the broader package agreed to in the next set of contracts with municipal employees.

Benefits under the NYCTRS vary depending on when a teacher started work, with four "tiers" or cohorts of members. Benefits were made less generous for later cohorts, because the State constitution prohibited reducing benefits for those already hired, but the cost of the benefits was unsustainable within fiscal constraints. While the variations among tiers are complex, benefits generally differ based on a teacher's age and years of service. The basic differences are as follows:

• Tier I is for teachers who joined the system prior to June 30, 1973. Members may retire after 25 years of service at age 55. Benefits are 50 percent of the final year's salary for the first 20 years of service, plus an additional 1.7 percent of the final year's salary for every year served beyond 20.
• Tier II is for teachers who joined between June 30, 1973 and July 27, 1976. Members are eligible to receive benefits at age 55 with 30 years of service or at age 62 with 25 years of service. Benefits are 50 percent of the average of the member's highest three consecutive annual salaries for 20 years of service plus 1.7 percent of the same three-year average for every year served beyond 20.
• Tier III is for teachers who joined between July 27, 1976 and September 1, 1983. This tier was transferred to Tier IV in 1983.
• Tier IV is for teachers who joined on or after September 1, 1983. Benefits are based on the average of the member's three highest consecutive annual salaries ("average base salary"). Members may retire at age 62. For those with less than 20 years of service the benefit is 1.7 percent of the average base salary for each year of service. For those with 20 or more years, the benefit is 2 percent of the average base salary for each of the first 30 years and 1.5 percent additional for each year of service beyond 30. Teachers may also opt to retire at age 55 with a minimum of five years of service, but the benefits are reduced significantly from those available if one retires at age 62.

Working Conditions. The UFT contract sets policies on the length of the school year and of the school day. The teachers' school year begins on the Tuesday after Labor Day (with students beginning attendance on the next day) and ends two weekdays before the last day of June. During this period teachers must receive a one-week vacation in February and other holidays as scheduled by the Board. Generally, school is scheduled for 180 days during the September to June period.

The contract defines the school day as six hours and twenty minutes. This day is divided into eight periods, with a total of 20 minutes allowed in intermediate and high schools for students to move between classrooms between periods. All teachers are given one period free for lunch. Depending on their specific assignment teachers receive at least five preparation periods per week. Preparation periods are a time when the teacher has no assigned duties, but should use the time for professional activities such as preparing lessons. Allowing for lunch and preparation periods, a teacher has a maximum of 6 periods per day available for teaching. However, the contract limits intermediate and high school teachers to being assigned to teaching to 25 periods per week or five periods per day. For these teachers, the remaining time may be assigned to administrative duties.

In an unsuccessful effort to use teacher time more productively, the contract was changed in 1995 to limit the duties to which teachers could be assigned during administrative periods. Principals cannot assign teachers to cafeteria supervision and other specified tasks. However, teachers also cannot be assigned to regular classroom instruction during these periods. The effect was to require the Board to hire additional nonprofessional staff for cafeteria supervision and other duties, while teachers received more time for loosely defined professional activities outside the classroom. Thus, what was presented as a productivity gain actually increased costs. Board officials assert that the added time for teachers to prepare for classes may improve instruction, but the wide range of activities in which teachers may engage make this connection difficult to establish firmly.

The contract determines one other important aspect of working conditions. It limits the number of students who can be assigned to a class. These maximum class sizes are 15 for pre-kindergarten, 25 for kindergarten, 32 for elementary schools, 33 for junior high schools and 34 in high schools (and 50 for high school physical education and music classes.) However, exceptions to these maximum class sizes are permitted routinely due to the lack of available classroom space or other constraints.

Teacher Assignments. The contract establishes seniority as the primary basis for assigning teachers to schools and specific jobs within schools for which they are qualified. The filling of vacancies is based on seniority, with teachers able to move from one school to another based on their seniority. Among teachers at a school, the assignment of specific roles is based on seniority within the qualified group. More senior teachers are more likely to get the grade level assignment they prefer and to opt for "cluster" or specialized support roles rather than regular classroom assignments. In hard times, seniority is also the basis for determining who is retained rather than laid off.

One notable exception to the seniority provisions is available through the School-Based Management/Shared Decision-Making (SBM/SDM) clause of the contract. It allows schools to operate under alternative rules if they establish a committee to make decisions under the alternative rules. In order for this alternative arrangement to come into effect, at least 75 percent of the UFT members at the school must initially approve the arrangement and 50 percent of the members of the committee must be teachers. The UFT district representative and president must approve any measures adopted by the committee that modify contract provisions. In effect, the SBM/SDM clause allows schools to adopt alternatives to seniority for hiring if a principal is willing to share this discretion with a committee of teachers and others, and if the teachers are willing to do so. This approach has been adopted at about 250 schools, some of which are older schools that have themes (performing arts, aviation, etc.) that require teachers with particular skills or interests and some of which are small schools under the New Visions Schools program. New Visions Schools are newly established public schools, sometimes viewed as a "contracting-in" model of charter schools, and are co-sponsored with community organizations and have an educational specialty or theme.

Professional Development and Discipline. The contract specifies how principals can deal with teachers who are not performing satisfactorily. It contains requirements for how teachers' professional competence shall be improved and how problematic teachers shall be disciplined.

For newly hired teachers who lack formal training in education, the contract requires two types of staff development. First, these teachers must participate in ten days of training designed jointly by the Board and the union. Five of these days are before the school year begins and five during the school year. (When such teachers are hired after the start of the school year, only three days of training are before they begin their assignments.) Second, these teachers are assigned a mentor, who is a teacher with at least five years experience, to work with them during the first year.

Experienced teachers who are judged to be performing unsatisfactorily by the principal in a formal annual review may voluntarily participate in a Peer Intervention Program. The Board and the union appoint a Peer Intervention Panel, who in turn select experienced staff to serve as confidential intervenors. These intervenors help unsatisfactory teachers develop a plan to improve their performance or counsel the teachers on alternative career opportunities. The Board has agreed to assist such teachers in finding alternative employment opportunities within the system.

The disciplinary procedure for teachers receiving repeated unsatisfactory ratings or charged with misconduct is detailed in the contract and in the State Education Law. Generally, teachers can be suspended from classroom duties based on a principal's decision, but the teacher must continue to be paid during the disposition of the principal's charges, usually a lengthy process. The charges are reviewed by arbitrators, and recently efforts have been made to expedite the arbitration process.

The policies established in the current contract are related to four significant problems that inhibit the performance of the public school system:

1. Lack of performance incentives for teachers.
2. Shortages of qualified teachers in selected categories and at selected schools.
3. Insufficient managerial authority for principals.
4. Inefficient deployment of teachers that reduces time spent in classroom instruction.

Each of these issues warrants further discussion.

Lack of Performance Incentives. If organizations are to perform well, then their employees should be paid in relation to their contribution to that performance. Historically, this was the logic for establishing pay differentials for teachers with more experience and education. Graduate training and classroom experience were believed to be linked with better teacher performance. However, over time the pay structure has given more emphasis to these factors than is justified by their contribution to teacher performance. Numerous studies have failed to show a systematic relationship between graduate school credits and teacher performance, and experience significantly adds to teacher effectiveness only during the initial part of a teacher's career.9 Thus, most of the current differences in pay among teachers established by the contract bear little relationship to how well the individuals function as teachers, nor do they serve as an incentive to high performance.

This lack of relationship between pay and performance is demoralizing to good teachers. The best teachers see weaker colleagues receive equal or greater paychecks, and feel unappreciated and unrewarded for their superior service.

The absence of performance pay is also inconsistent with efforts in recent years to improve accountability in the governance and management of the public schools. The Board is insisting on greater accountability from the Chancellor, and the Chancellor in turn has been given greater authority to hold district superintendents accountable. A major breakthrough in negotiations earlier this year with principals abolished their tenure and established a performance pay arrangement for principals that permits district superintendents to hold them more accountable. The contract provides for incentive payments of $15,000 annually for principals in the top 5 percent of performance reviews; $10,000 for those in the top 5-15^th percentile, and $5,500 for those in the 15-25^th percentile. The logical-and essential-next step in creating greater accountability is a performance pay arrangement that permits principals to hold teachers accountable and pay them accordingly.

In recognition of the dysfunctional consequences of basing teachers' pay only on education and longevity, many educational, corporate and political leaders have supported developing new compensation arrangements for teachers. In 1999 the National Education Summit, a meeting of governors and corporate chief executive officers, agreed to work with teacher organizations to negotiate "pay-for-performance" incentive plans into teacher salary structures.10

Innovative school districts in several areas have established compensation systems that reward other factors. These systems, sometimes called "merit pay," vary, but share the common purpose of seeking to link performance and pay more closely. The approaches being tried differ in two key aspects. First, some provide rewards to individual teachers, while others reward larger work units such as an entire school. In the latter case, the principal or a committee in the school determines how to allocate the financial incentive among the individual members of the relevant work unit. Second, some base the reward on the performance of students using standardized tests or other measures as the yardstick, while others provide the reward based on specific accomplishments (such as submitting a portfolio or passing a national teachers examination) of staff members themselves rather than their students.

Because the experience with these performance pay plans is relatively new, definitive evaluations are not available. However, results at some of the districts with longer-running initiatives are positive.11 For example, the Dallas public school district, the tenth largest in the nation, has since 1994 used sophisticated measures of student achievement based on standardized tests to reward schools whose pupils achieve greater than expected gains. About one-fifth of the schools receive the annual incentive payments, which are $1,000 for teachers and $500 for nonprofessional staff. The State of Kentucky has provided incentive payments to schools based on greater than expected gains on its standardized tests in two-year cycles that began in 1992. About one-third of the schools earned the incentive payments, with the award per teacher ranging between $1,300 and $2,600.

The Douglas County school system in Colorado provides an example of an approach that rewards individual teacher behavior as well as group efforts. The county and the teachers union agreed in 1993 on a pay scale that provides no reward for longevity and no annual cost-of-living increases. Instead teachers increase their base pay by earning one or more of four types of incentive payments: (1) A $1,000 bonus for being designated an outstanding teacher based on submission of a portfolio. About one-fifth of the teachers obtained this payment each year. (2) Payments ranging from $250 to $500 for successfully completing training courses in areas such as computer skills. (3) Payments for taking on additional responsibilities at the school or district level. For school level tasks, payments range from $35 to $200 for specific responsibilities such as committee assignments and mentoring colleagues. For district level tasks, payments averaging $800 are given to teachers who participate most actively in these types of activities. (4) Group incentive payments are available to groups of teachers who design and implement an innovative program to improve student performance. In the 1998-99 school year 33 of 39 schools gained these awards with payments per teacher averaging about $400. 12

These are just three examples of performance pay plans that have been in place for several years and are "working" in the sense that the teachers voice satisfaction and there is evidence of improved student performance.13 The attractiveness of these plans is evident is the adoption of similarly motivated efforts in more recent years in large urban districts such as Denver and Cincinnati.14

New York City also has begun to experiment with performance pay for teachers. A pilot program known as "Breakthrough for Learning" was launched in 1998 in Community School Districts 19 and 23. Funding was raised privately by the New York City Partnership and Chamber of Commerce. These funds are made available to the staff of schools whose performance exceeds targets based on the performance of peer group schools. The entire staff of high performing schools receive bonuses, with the annual bonus for teachers being $2,000. As of the end of the 1999-2000 school year, payments to six schools in District 19 have been made.15

Selected teacher shortages. The Board of Education does not have a single job classification known as "teacher." Rather it hires people in 225 separate titles that are specific types of teachers. This includes many titles with relatively few incumbents, notably specialized bilingual instructors such as the four bilingual Cantonese high school social studies teachers (and other subject-specific bilingual instructors in Mandarin, Hebrew, Russian, Creole and other languages), and specialized vocational school teachers such as the instructors in window display, aeronautics, baking and barbering. However, nearly 29,000 of the more than 78,000 teachers are in the single job title "common branches" teacher, those teaching general classes in the elementary schools. Other classifications with more than 500 incumbents are identified in Table 4.

To be eligible for permanent appointment an individual must be "certified" by the State of New York in their job classification. The State's requirements for certification include completion of a master's degree that includes at least 18 credits in education, completion of 36 credits (graduate or undergraduate) in the relevant subject area, and passage of a State administered certification test. Individuals who lack State certification may be given a temporary license, but they are permitted to teach for only four years, during which time they are expected to become certified. After four years, uncertified teachers may continue to work with a temporary license only after a special review by the State Education Department. The State prohibited the hiring of uncertified teachers in SURR schools effective in September 1999 and plans to stop issuing temporary licenses effective in September 2003.

While State officials have a long-run goal of eliminating the use of uncertified teachers and are pressing the Board to stop hiring them immediately in SURR schools, it is a common practice among many school districts and many private schools to hire promising, uncertified teachers and allow them some time to meet certification requirements. This permits people to enter the profession even if they did not complete a graduate degree in education by earning the necessary education credits while they are gaining relevant work experience. (The current president of the UFT began her teaching career as an uncertified teacher.) In the areas of New York State outside New York City 7.1 percent of the teachers were uncertified in 1998-1999, a figure that reflects this alternative entry route to the profession and that is not typically viewed as a cause for alarm.16

Determining when there is a "shortage" for a particular job title is conceptually difficult. Shortages, like beauty, are to some extent in the eye of the beholder. That is, labor economists often find that employers identify situations when they must change their recruitment practices to adapt to changes in the nature of the labor supply as a "shortage." If previous sources of candidates no longer yield an abundant labor supply, then new sources may need to be tapped, but this does not necessarily mean there is a shortage.

The Board has been criticized for failing to adapt its recruitment and hiring practices to current conditions. Specifically, the Board recruits and hires very close to the start of the school year (and even after it begins), while other districts begin intensive recruitment in the early spring before a school year begins. In addition, the Board relies primarily on City University and other New York City institutions for new teachers and does less extensive recruiting at upstate State University campuses and at out-of-state colleges than do many smaller districts. The Board also fails to identify ways to improve its recruiting efforts by collecting data on why recruits do not accept jobs or where they do accept jobs.17

How, then, does the Board determine that a shortage exists? It classifies job titles as in a shortage (or not) annually based on its recruiting experience. Each year the Board sends a recruitment letter to every certified teacher who is not working at a public school (based on information supplied by the State Education Department). If this mailing does not yield a sufficient number of applicants for the openings in a job title, then that title is defined as being in a shortage. For such shortage job titles, incumbent certified teachers may be authorized to receive additional payments of $3,961 annually for teaching classes in place of taking five of their preparation periods each week. In the 1999-2000 school year, 33,633 teachers or about 43 percent of the total were in 142 shortage job titles. (Refer to Table 4.)

It is worth noting that there is not a direct correspondence between the Board's shortage classification of a job title and a high proportion of uncertified teachers in that job title. While some shortage titles such as high school math and science teachers have very high shares of uncertified teachers, other shortage titles such as reading teachers have relatively few uncertified incumbents (3.6 percent), and at least two shortage titles (health conservation and emotionally handicapped) have no uncertified incumbents. (Refer to Table 4.)

The shortage conditions impact most acutely on the schools that teachers consider the least desirable. Because the deployment of teachers is governed by their seniority, the more experienced (and certified) teachers are more likely to be in schools they desire, while the new and often uncertified teachers must accept the least desired locations. The schools that have a reputation for difficult students and/or poor management are the ones experienced teachers seek to avoid. Consequently, assignment by seniority in a period of labor shortages results in the neediest students being taught by the least experienced and least qualified teachers.

The variation in staffing problems of schools is illustrated in Table 5.18 It classifies schools based on the share of teaching staff who are licensed and permanently appointed. While it should not be assumed that State certification is necessarily a valid measure of teacher quality, in the absence of better data the share of teachers who are permanently appointed is a reasonable indicator of how much difficulty a school has in attracting qualified teachers. About one of every four schools has at least 92 percent of their staff who are both fully licensed and permanently appointed, and almost half of all public schools have at least 85 percent fully licensed and permanently appointed. If a "shortage" school is defined as one with only 75 percent of its teachers fully licensed and permanently appointed, then about three of every ten schools fall into this category. If the criteria is 80 percent, then about four of every ten schools are in the category.

These data indicate that is reasonable to conclude that "shortages" exist within New York City public schools, but this condition is selective rather than universal. About 40 percent of the schools face serious recruiting problems, and the job titles in which these problems are evident impact about 43 percent of the schools' teachers.

The selective scope of the shortage conditions is also evident in the Board's aggregate hiring experience. (See Table 6.) Between 1991 and 1999, the number of teachers increased by 15,243 or 24 percent from 62,919 to 78,162. The Board was able to make a net addition of more than 12,000 certified teachers to its roster during the past decade. The recent number of certified teachers exceeds the total number of teachers employed in 1991, suggesting that the current shortage is caused in part by the expanded hiring of new teachers.

Insufficient Managerial Authority. While some schools have difficulty attracting good teachers, it is even harder for the principals of all schools to manage their staff effectively. Two significant limitations on their authority are the contractual provisions relating to seniority and the cumbersome disciplinary process.

The contract makes seniority the exclusive criterion for determining staff assignments. Teachers select their class and other assignments based on seniority, limiting a principal's ability to staff classes and programs based on individual performance and capabilities. Similarly, principals running schools with good reputations must accept teachers transferring from other schools based on seniority, sometimes limiting their ability to select those most able or best suited to performing well in the school. The only exception to this is the previously described SBM/SDM system for sharing authority with the teachers employed at the school.

Another obstacle to effective management is a time-consuming disciplinary process. Principals are reluctant to rate teachers as unsatisfactory or to charge them with misconduct because of the slow and cumbersome arbitration process that follows. In the 1998-99 school year only 523 of 75,209 or 0.7 percent of teachers reviewed were rated unsatisfactory.19 Some principals are reluctant to give such ratings or seek to discipline teachers for other reasons because of the time and effort required for the arbitration process that follows. In February 2000 the Chancellor reported that 301 teachers were awaiting completion of their disciplinary proceedings. The average time for these proceedings was over 18 months, and one case had been pending for almost seven years. During this period the affected teachers received their salaries and were assigned to "rubber rooms" at the Board headquarters where they must report but are not given any duties and are permitted to read newspapers. The Chancellor asserted: "Having employees paid to read newspapers undermines teacher morale. It also deters managers from bringing disciplinary actions even when obviously merited…. I regret to report that our current disciplinary system appears oblivious to the serious consequences of its institutionalized indolence."20 The Chancellor also noted that the Peer Intervention Program had resulted in only 90 teachers leaving the system in the ten years since it was established.

Limited Time Teaching. New York City lags the rest of the state in the number, as well as the qualifications, of its teachers. In the city's public schools, there are 14.6 pupils for each teacher, a figure about 7 percent higher than the statewide total. (See Table 7.) However, the disparity in average class size between New York City and the rest of the state is far greater; classes are between 14 percent and 20 percent larger-or between double and triple the gap in the pupil-teacher ratio. This means that teachers in New York City spend less time teaching than their counterparts in the rest of the state.

Aggregate staffing levels are based on Board of Education and mayoral decisions about how much to spend on the public schools. But the inefficient use of teacher time that exacerbates the disparities and leads to far greater discrepancies in average class size can be related to contractual provisions.

What are the policies that cause about one-third of all teacher time to be assigned to duties other than regular classroom instruction? As noted earlier, all teachers are entitled to regular preparation periods, and in the junior high and high schools teachers are assigned to administrative tasks for part of the day. Covering classes during these periods requires additional staff. About 17 percent of all available teacher time is required to cover classes while their colleagues are on preparation periods or administrative duties. An almost equal amount of teacher time is diverted from regular classroom assignments because teachers are on sabbatical or other leaves (about 3 percent of all teachers), are serving as union representatives (the equivalent of about 200 full-time teachers), or have other non-instructional assignments.21

A similar finding was reported in an audit by the City Comptroller of the use of teaching staff in a sample of high schools in October 1998. The audit found that teachers in regular high schools spent 64 percent of their time on instruction, and that those in alternative high schools spent 66 percent on instruction. The proportion in the regular high schools was unchanged from a similar audit in the spring of 1996, but the proportion in the alternative high schools had increased from 64 percent.22

A related, but different, point emerges from comparative studies of the length of the school day worked by teachers. The latest such study found that New York City teachers worked the shortest day among large school districts nationwide. Teachers in Chicago, the district with the next-shortest day, worked 30 minutes longer than teachers in New York City, and teachers in Los Angeles worked an hour per day longer. However, this additional time is not spent entirely on instructional activities. Typically, the additional time is required before and after the school day begins for students.23

Cherry Picking. Another frequently discussed, but actually not severe, problem for the Board is the recruitment of some of its teachers by suburban school districts. This so-called "cherry picking" is not a major problem because it takes place only on a very limited scale.

Clearly, many suburban districts offer base salaries that exceed those in the city. The median teacher salary in Nassau County was 40 percent above that in New York City in 1997-1998, and the gap is greater in some of the wealthiest districts. (See Table 8.) The current pay for a "typical" teacher with a masters degree and four years experience in Scarsdale was $55,103, or 43 percent greater than for a similarly qualified teacher in New York City. While these comparisons do not take into account the additional pay available to New York City teachers through the ETS arrangement or for after-school activities (nor do they adjust for the typically shorter workday in New York City), it is, nonetheless, true that some teachers may be attracted by the higher base salaries.

While the pay differentials are sometimes large, the number of opportunities for such movement is small. Two recent investigations of the issue each separately conclude that only about 0.5 percent of the Board's teachers (or about 350) move to suburban districts annually. One analyst sought to estimate the size of the problem by calling five high-paying districts in Nassau and Westchester to ask what proportion of their vacancies they typically fill with experienced New York City teachers.24 Based on this sample, he estimated that 15 percent of the approximately 2,300 annual New York suburban vacancies are filled by former New York City teachers. (The remainder are filled with lower paid newly starting teachers and teachers recruited from other locations.) This points to a loss of about 350 teachers annually to the New York suburbs.

The second study relied upon the State's file of certified teachers to determine mobility between the New York City schools and New York's suburbs.25 It found that between the 1995-96 and 1996-97 school years 359 teachers left the city for the suburbs and that 74 moved in the opposite direction for a net loss of 285 teachers. For the subsequent school years, the figures were a gross loss of 395 and a net loss of 302. Again, the figures represent only 0.5 percent of the city's staff. While both studies consider only movement to New York suburbs, it is unlikely that expanding the range to other regional suburbs in New Jersey or Connecticut would bring the figure above 1 percent. In sum, cherry-picking related to salary gaps between the city and suburbs is a limited problem and would become even more manageable if the recommendations made below relating to performance pay and shortage differentials were implemented.

While the movement of teachers to the suburbs is a limited problem, the potential loss of many more teachers due to retirement is a more serious issue. It is both a challenge and an opportunity.

The challenge arises from the need to replace a large number of teachers. As is explained more fully below, it is possible as many as 17,000 teachers will retire in the next four years.

The opportunity arises because the large-scale replacement of more senior teachers with new hires can free substantial resources that can be used to improve educational effectiveness. Because current policies give large rewards for seniority, the retiring teachers are also the highest paid teachers. In contrast, under current policies, the newly recruited teachers are likely to be in the lowest rungs of the pay scale. The replacement in this manner of 17,000 teachers over four years would lead to annual savings in teacher compensation at the end of the period of about $490 million.26 These substantial resources are part of the opportunity presented in the current contract negotiations.

Concern about an increased rate of retirement was raised by the UFT earlier this year based on a survey it conducted of its members.27 The union estimated about 21,000 teachers are eligible for retirement in the next two years. This estimate is based on the age of teachers; the number is those aged 54 or older in the 1999-00 school year. Teachers in Tier I of the pension system with at least 25 years of service are eligible to retire at age 55, but it is not clear that all the teachers meeting the age requirement would also meet the other necessary retirement criteria or would want to retire.

The union's survey indicated that about two-thirds (or 14,000) of the age-eligible teachers were considering retiring in the next two years, and that about 10,000 said they were very likely to retire in the next two years. The union argues that the Board will not be able to recruit a sufficient number of qualified replacements and urged higher salaries in order to retain and recruit teachers.

The key questions for the future are how many teachers really will retire, and whether the Board can recruit a sufficient number of qualified teachers to both replace them and meet its other staffing needs. Some perspective on the retirement rate can be gained from the data in Figure 1. Over the decade from 1990-1991 to 1999-2000, the average annual number of retirements was about 2,700. The figure fluctuated widely with peaks in 1991-1992 and the two years 1995-1997 because in those years incentives for early retirement were provided by the Board to help cope with short-run budget difficulties (despite the shortages of certified teachers discussed earlier). Nonetheless, the 2,700 rate appears to be a longer-term average.

The scenario envisioned by the UFT would require the historic rate to nearly double to about 5,200 in 2000-2001, and more than triple to 8,820 in the next year. No explicit reason is given for the greatly increased retirement rate. The implicit explanation is simply the age structure of the teaching workforce, but complete trend data on this are not available. However, some inferences can be made from data on the years of service of current teachers. (Refer to Table 2.) In May 2000 nearly 15,000 teachers had at least 22 years of experience. This suggests 14 percent or about 11,000 teachers have at least 25 years of service and that about 1.5 percent or 1,200 reach that level of service in each of the next few years. Thus, based on years of service, the potential pool of retirees in the next year was a maximum of 11,000 and over the next four years was 14,600.

Given this estimate, the UFT scenario seems extreme. However, after the UFT published its findings and warning, it negotiated with the Mayor incentives for early retirement for its members. In June the unions and the Mayor jointly recommended to the State legislature that senior members of the pension system be credited with additional years of service toward their retirement in exchange for the union agreeing to changes in the actuarial assumptions used in calculating the City's pension fund contributions. The legislation was passed, providing up to 2 years extra credit towards retirement benefits for Tier I and Tier II members.28 Based on the data discussed above, this measure could enable another 2,400 teachers to be eligible to retire, bringing the potential total eligible over the next four years to 17,000. In this way the union, the Mayor and State officials are working to help make the UFT's scenario come true.

If one assumes a significantly increased rate of retirement due to demographics and the recent pension enhancements, then the critical issue becomes how able the Board is to meet its hiring needs. In recent years, the Board has replaced its departing teachers and added between 2,000 and 3,000 teachers annually. (Refer to Table 6.) If staffing levels remained constant, and if the attrition rate due to factors other than retirement remained constant, then the board's past rate of hiring could accommodate an increase in the average annual number of retirements from 2,700 to between 4,700 and 5,700. This is a significant increase from the historical rate, and is in the range of the potential average rate based on years of service of current teachers, but it is below the UFT's peak projection for 2001-2002. Thus, the increased retirement rate will pose a challenge, but not an overwhelming one. If retirements can be spread more evenly than the UFT anticipates, then the Board could sustain and even grow its teaching staff without efforts beyond those already in place. New policies, such as those recommended in the next section should permit the Board to enhance its hiring record and improve the quality of instructional services.

The unfolding challenge of recruiting a larger number of teachers than has ever been required before is also an unprecedented opportunity. If the projected retirement rates are in the right range, and other causes of attrition persist, then in the next four years the Board of Education will be replacing about half its pedagogical staff. At the same time it will be achieving savings that approach $500 million annually.

A new contract should put in place a new compensation system for teachers that moves away from the less relevant criteria of longevity and educational credits and toward on-the-job performance as the basis for determining their pay. The newly recruited workforce can and should be paid in a way that enhances the Board's performance. Four measures would promote this goal.

1. Restructure the pay schedule to make a larger share of compensation conditioned on job performance and to de-emphasize longevity and graduate educational credits.

The massive replacement of teachers expected in the coming years provides an opportunity for fundamental, rather than incremental, changes in the compensation plan for teachers. New teachers should be hired into a new pay plan that consists of base salaries that vary less than is now the case based on education or seniority, and that provide additional compensation based on job performance. Teachers with substantial periods of service under the old plan might be given the option to remain in a modified version of that system, but the new and more recently hired teachers, who will soon comprise the bulk of the workforce, should be part of a new era in teacher compensation.

The details of a new plan will necessarily be worked out as part of collective bargaining, but from a citizen's perspective the new plan ought to have three basic elements. First, base salaries ought to represent a large majority of total compensation to teachers, but should leave a significant portion of compensation to be determined based on performance. The base salaries should vary with a few increments based on experience and with a significant increment for widely recognized professional development milestones such as National Board Certification. This certification involves review by a nationally selected group of professionals over a year-long period, and must be renewed every ten years. In 1999 only seven teachers in New York City obtained this certification, and the Board employed a total of only 25 such certified teachers. Because it does not affect their pay, the vast majority of local teachers do not seek such certification despite its developmental benefits and national recognition. In contrast, North Carolina began in 1997 giving a 12 percent pay raise to teachers who obtained national certification, and it now has more Board certified teachers than any other state.29

The second component of pay should be based on the accomplishments of individual teachers during the year. This should follow the Douglas County approach described earlier, with pay available for taking on additional responsibilities, for achieving certain developmental goals, or for special service to the school or district.

The third component of pay should be given to schools or other work groups whose students have educational gains above those that could normally be expected. The approach used in Dallas and described earlier can be a model. Appropriate statistical measures should be developed to determine when a school's staff has performed better than expected norms, and the staff responsible for these gains should be rewarded.

The changes recommended here are a restructuring of compensation arrangements; they do not necessarily have an incremental cost or savings compared to what otherwise would have been paid. Instead, the money would be allocated differently. Teachers would receive less of their total compensation as base salary and a substantial portion would depend on their individual and group performance. There would still be considerable variation among the pay of individual teachers, but it would be based less on longevity and more on the quality of their work over the past year.

2. Use financial incentives to overcome shortages.

In economic theory, labor shortages are inherently temporary in a market environment. Rational employers increase wages in shortage occupations in order to attract more workers. Union leaders are now arguing that all teachers are in the shortage category, and that the pay of all teachers should be increased significantly. But the shortages are not so widespread; rather they are concentrated in some job titles and are acutely evident only in some schools. In these cases, substantial new economic incentives should be provided.

The new contract should provide differential pay for teachers qualified for and serving in specific titles that are suitably deemed in a shortage condition, and who also agree to work in those schools facing the greatest difficulty in securing qualified teachers. In this way, added funds can be targeted to recruit those teachers needed most to those places with the greatest need. It would be an ineffective use of scarce educational dollars to give all teachers the financial incentives necessary only for a minority of positions.

Based on the data reviewed earlier, about 43 percent of the teachers are in job titles deemed to be in shortage status because of a paucity of certified applicants. (Refer to Table 4.) Of this group, about 40 percent are needed in schools which teachers otherwise tend to avoid and which have unusually large numbers of uncertified teachers. (Refer to Table 5.) Thus, about 17 percent of the positions that the Board must fill are in an acute shortage condition, and additional pay should be targeted to people recruited for those positions.

It is difficult to estimate how large a differential would be needed to attract teachers to these difficult-to-fill slots. The ETS arrangement described earlier provided a 15 percent differential for teachers in some SURR schools, but also required them to work additional hours. It drew some experienced teachers to the troubled schools, but was not successful in attracting sufficient additional certified teachers to make those schools comply with the State's standards. Thus, differentials in the range of 20-25 percent ought to be tried. Based on current average teacher salaries of about $50,000 annually, providing 20-25 percent differentials to 17 percent of the teachers would cost between $133 million and $162 million annually.30

Investments should also be made in bringing new people into the teaching profession and helping them obtain certification. Last year more than one-half of the Board's new hires were uncertified. If this pattern continues, then each year about 5,000 uncertified individuals will be brought into the teaching field to fill shortage positions in areas like high school mathematics and science. Instead of being treated as welcome recruits to the profession, they are often treated as second class citizens. They are paid in the lowest category of the wage schedule; they must arrange for and finance their own educational course work to meet certification standards; they must pay fees to take certification exams.

In July 2000 the Board of Regents recognized the need for and desirability of people entering the teaching profession in this way. The Regents authorized school districts to hire uncertified teachers provided they have a good undergraduate record and that they acquire the education credits needed for certification in a planned way within two years.31

The Board of Education has taken some modest steps to facilitate alternative routes to certification. It has established two programs, both limited to uncertified teachers working in SURR schools. The first program reflects a special agreement with the State to aid uncertified teachers hired at SURR schools for the 1999-2000 school year in violation of the State's regulations. Eligible teachers are those whom the district wants to retain this year, and who are nominated by the district superintendent. About 350 eligible teachers were nominated. This group attended a specially designed teacher training program in the summer of 2000 and must enroll at City University education courses in the 2000-01 academic year, with the tuition paid by the Board.

The second program, the New York City Teaching Fellows Program, has 335 participants.32 They are newly recruited, uncertified teachers who pledge to teach for at least two years in a SURR school. They attended an intensive training program in the summer of 2000 (with a $2,000 living stipend) and will participate in mentoring by senior teachers during the 2000-01 school year. In addition, the Board will pay the tuition for their participation in a two-year masters degree program available at three City University colleges.

The Board of Education should move beyond these limited efforts and aggressively pursue new options for uncertified teachers to gain their credentials. Funds should be used to help far more of these recruits to meet certification requirements in a planned fashion. The Board could contract with local schools of education to design programs, perhaps delivered on site at schools or district offices, that offer appropriate graduate credits and help prepare for the certification exams. One-time bonuses could be paid to uncertified teachers when they meet all the requirements. An investment of $3,000 in each of these recruits to help them obtain certification, and a $1,000 bonus for achieving it, would cost $20 million annually or $80 million over the next four years if all those eligible successfully participated.

If a sudden upsurge in retirements causes a sharp, but temporary, increase in hiring needs in the next year or two, then the Board could also anticipate temporarily drawing on retired teachers as another source of qualified personnel. Current pension rules appropriately prohibit teachers from simultaneously collecting pensions and pay checks. However, if a short-run need arises, then these rules could be modified. Retired teachers could be paid to work for a defined period without losing all, or even some, of their pension benefits. In order to avoid abuses, the arrangement should be limited to those who have been retired for at least one year and limited in duration to just one year of paid employment.

3. Enhance principals' managerial discretion.

The Board is wisely pursuing a policy making principals more responsible and accountable for the performance of their schools. The principals' pay is now partly based on school performance. But in order to work more effectively, these laudable steps need to be accompanied by changes in the teachers' contract that give principals more authority over how to engage, deploy, discipline and otherwise manage their staff.

The performance pay arrangement recommended above would be one important such measure. It would allow principals to help determine the pay of teachers by approving them for particular assignments in the individual component of the performance pay, and it would provide incentives for teachers to follow a principal's leadership in seeking to earn the group-based performance pay.

However, principals' managerial discretion is limited in other ways that could be changed through collective bargaining. Perhaps the leading constraint on principals' ability to use their judgment in deploying staff is the contractual requirement that seniority determine most assignments. Principals should be given more discretion in determining which teachers may transfer to their school, and what assignments each teacher be given within their school. An appropriate step in enhancing such discretion would be to modify the seniority rules to allow principals to select teachers for given posts from among those three (or five) candidates with the most seniority. A similar "rule of three" is often used in hiring from civil service lists based on test scores (a supervisor can select from among the three highest scorers), and seniority need not be more binding than exams in personnel decisions.

Another important boost to principals' managerial authority is to make it easier for them to discipline bad performance or behavior. The current system of arbitration is demoralizing to principals who want to be effective managers. The contractual and statutory mandates for this process should be eliminated, and a principal should have the ability to terminate a teacher on grounds of misconduct or repeated unsatisfactory performance. An appeal mechanism is appropriate, but the initial decision should rest with principals and all grievances and appeals should follow a prompt, streamlined process.

4. Increase teacher time spent on classroom instruction.

• Reduce combined preparation and administrative periods for junior high and high school teachers from 10 to five per week.
• Reduce preparation periods for elementary school teachers from five to three per week.
• Eliminate sabbatical leaves for travel and study.
• Eliminate subsidized time for union activities.

 

Reallocating teacher time in this way would have the impact of adding the equivalent of more than 4,250 teachers to the payroll. These gains in productive time could be used to offset the need for additional replacement hires projected in the future.

The previous four recommendations would be significant steps in the necessary, broader long-run effort to improve the performance of New York City's public schools. The recommendations would help in securing a well-qualified set of teachers, provide them with new incentives to work effectively with students and with principals who provide leadership, and increase the time they spend in the classroom. The current round of collective bargaining should be used to help institute these important and constructive changes.

Moreover, these changes require less additional funding than is likely to be available due to the replacement of more-senior with less-senior teachers. Of the eventual annual savings of nearly $500 million, less than half would be needed to fund the pay differentials for teachers in shortage schools and job titles ($162 million) and to provide the financial incentives to uncertified teachers to become permanently qualified ($20 million). The recommendation to restructure compensation to de-emphasize the rewards for longevity and provide new rewards for performance is "budget neutral" and need not require additional funding, and the recommendation to increase classroom time actually yields savings.

The likely savings in the Board's payroll in coming years is a major opportunity for educational and political leaders to set clear priorities for the use of additional funds to improve the schools. Three broad strategies are available.

The first, and least desirable, can be described as "across-the-board" raises for teachers. That is, the funds could be used, even under the recommended restructuring, to more richly reward teachers by increasing the base portion of their total compensation.

This approach is advocated on grounds of equity, and to a lesser extent on the grounds of effectiveness. The equity argument typically is that teachers pay should keep pace with inflation to protect their real incomes. However, this is a weak case in the current context. Despite the much publicized two years of "salary freezes," teachers have actually received pay increases during the last contract period that were well in excess of the rate of inflation without any productivity increases. (Refer to Table 3.) Moreover, the enrichments to the teachers' pension benefits enacted in the 2000 session of the State legislature already have provided compensation gains for next year that exceed in their value the increase in the cost of living projected for the coming year. With respect to effectiveness, the argument is that higher pay is needed across-the-board in order to attract sufficient teachers. However, if the recommendations in this report relating to performance pay and differentials for shortages were implemented, across-the-board raises would not be essential to meet staffing needs.

A second strategy, which is sometimes characterized as "time for money," carries great risks of wasting much of the resources devoted to it. This approach would put available funds into higher salaries for teachers on the condition that they work longer hours. The fact that New York City teachers work shorter days than teachers in many other large urban systems has given this approach some appeal, but the merits are illusive.

Three points reveal the risks and potential waste of using new resources in this way. First, the Board already has a compensation system for hiring teachers to staff activities scheduled for after school or during the summer. It gives teachers extra compensation, and has been successful in staffing the extra-curricular programs it has sought to run. As noted earlier, the Board spent over $290 million last fiscal year to pay teachers for work during such extended hours. Second, there is little likelihood that the extended time would be used productively for instructional activities. As noted earlier, other systems with longer days and the Board's experiment in ETS schools allocate all or most of the added time to teacher preparation and tasks other than classroom instruction. The approach typically does not yield a longer school day for children. If teaching children is the highest priority for teachers' time, then extended hours yields a limited payoff. Third, applying extended hours to all teachers is likely to be wasteful in the sense that the Board (and taxpayers) would be obliged to buy more teacher time than can be effectively programmed. It is unlikely that all school staff, including specialized teachers as well as regular classroom teachers, can be used effectively in after school programs; the current system of hiring those needed for priority activities better targets the use of resources. Thus, "time for money" should be pursued only to the extent the added time purchased is devoted to classroom instruction that directly benefits children by extending their school day or school year.

The third, and most desirable, strategy for new money is to increase the quantity, rather than the price, of the labor and capital resources made available to children. With respect to quantity of labor, as noted earlier, the Board has a lower teacher-to-pupil ratio and larger average class sizes than the rest of New York State. (Refer to Table 7.) While steps should be taken to use the current teachers more effectively, both new and current teachers could be deployed in a way that yields high returns in educational gains for students. Specifically, a priority claim on new resources should be hiring sufficient additional teachers to achieve significantly reduced class sizes in the early grades.

Relatively small class size in the early grades has been found to increase student performance.33 (The effects of smaller class size are less clear in the later grades.) However, New York City is characterized by relatively large classes in these grades. A careful study by the Independent Budget Office (IBO) found that class size averaged 23.9 in kindergarten, 25.3 in first grade and in second grade, and 25.9 in third grade. These are the grades where research suggests optimum class size is 18 or less. The IBO estimated that 5,075 new classes would have to be established to bring all classes in the kindergarten through third grade to a maximum size of 18. Given the need for coverage during preparation periods and other factors, these new classes would require over 6,000 new teachers. The IBO estimated the total annual operating costs required for this class reduction at $230 million.34 This sum would be affordable with the savings expected from the replacement of more senior teachers by less experienced ones, but there would still remain the problem of finding classroom space for these additional classes.

With respect to capital inputs, modernization of school buildings with added capacity is another high priority. The Citizens Budget Commission previously has recommended allocating capital budget funds to create modernized, air-conditioned schools equipped with computers to be used on a year-round basis.35 Maintaining the current 180 days per year of schooling for students, but spreading the time over the full calendar year rather than only the current school year, is an accepted practice in other states and would increase the capacity of existing schools by as much as one-third. Using $250 million of potential annual savings to support debt service on bonds would permit about $3 billion to be added to the Board's capital budget in the next four years. This would be a major increment to the inadequate $7 billion, five-year plan approved by the Board in 1999. Of course, all the available savings cannot be used for both capital investments and smaller class sizes, but some combination of these added inputs should be sought.

In brief, the priority uses for new resources made available from staffing changes should be some combination of (a) more teachers in front of smaller classes, and (b) modernized school buildings in use year-round. This should be an outcome of the current round of negotiations.

Finally, the scope of this year's negotiations should not be confined to the pay and working conditions of public school teachers. Stimulating competition to public schools, as well as improved staffing and management of the public schools, is the second element of the necessary long-term strategy for educational reform. A leading policy instrument for promoting innovation is charter schools. State law sets the framework for these schools, but it must be recognized that the teachers' union is a powerful lobbying force in the State legislature that helps shape these laws.

Just as collective bargaining sometimes involves agreements for the Mayor, the Board and the union to support pension law changes, the negotiations can also seek joint support of measures to facilitate alternatives to badly performing schools. For example, the 1998 charter school law requires that a charter school with more than 250 students have its teachers covered by the collective bargaining agreement that applies to public schools. In New York City, this cap is unrealistically low and should be raised or abolished in order to permit established school management firms to launch charter schools. Part of the bargaining should be obtaining union support for appropriate legislative changes.

After five years of living with a contract that made little progress in improving public education, the Mayor, the Chancellor, and UFT leaders now have an opportunity to achieve major gains. The focus of the current round of negotiations should be measures, such as those recommended in this report, that will not just raise the income of teachers, but also will raise the quality of public education.

###

FOOTNOTES

1. Citizens Budget Commission, Budget 2000 Project: Public Education (NY: CBC, 1996).

2. Ibid.

3. The sole exception is the pay for those with 22 years of service. This step did not exist in the previous contract.

4. The amount is the sum of the budgeted amount for "summer and evening school" ($165. 6 million) and for "after school and student activities" ($125.0 million) in Board of Education, School Based Budget Reports Fiscal Year 1999-2000 (June 2000) Systematic Summary, p. 6. Per teacher average based on a total of 78,000 teachers.

5. Board of Education of the City of New York, Office of the Chancellor, Memorandum to Members of the Board of Education From Rudolph F. Crew Regarding SURR School Intervention Strategy, June 22, 1999.

6. Citizens Budget Commission, Modernizing the Municipal Employee Health Insurance Program (NY: CBC, April 1995).

7. See Table 3, Note b.

8. New York City Office of the Comptroller, Bureau of Financial Audits, Analysis of the Financial and Operating Practices of Union Administration Benefit Funds Whose Fiscal Year Ended in Calendar Year 1998, June 30, 2000.

9. Linda Darling-Hammond, "Teacher Quality and Student Achievement: A Review of State Policy Evidence," (Center for the Study of Teaching and Policy, August 1999); S. J. Rosenholtz, "The Organizational Context of Teaching," Learning to Teach (IL: University of Illinois at Champaign-Urbana, 1986); Hamilton Lankford and James Wychoff, "The Changing Structure of Teacher Compensation, 1970-94," Economics of Education Review, 1997, Vol. 16, No. 4, pp. 371-384.

10. National Education Summit, "1999 Action Statement," Adopted October 1, 1999.

11. Findings from model-based research completed by Bisa Cunningham, Cynthia Searcy and Kristen Simpson, "Opportunities for Merit Pay in New York City Public Schools," New York University Robert F. Wagner Graduate School of Public Service Capstone Project (April 2000).

12. The Delaney Policy Group, Developing A Public Workforce for the New Millennium, A Report Prepared for the Citizens Budget Commission, June 2000.

13. UFT leaders often assert that merit pay has been tried and that it "does not work." The example cited is a compensation plan initiated in Fairfax County, Virginia in 1986 and terminated in 1991. This plan established 10 percent bonuses for teachers meeting certain criteria. When the county faced fiscal problems in the early 1990s, its political leaders reduced and then ended the bonus arrangements. This action was not in response to concerns about the plan's effectiveness. This shows that when "merit pay" takes the form of bonuses in addition to substantial base salaries then they become politically vulnerable in difficult economic times. See Albert Shanker, "The end of merit pay?," March 28, 1993.

14. "Denver teachers OK merit-pay experiment," Washington Post, September 11, 1999. James Pilcher, "Cincinnati Public Schools approves merit pay plan for teachers," The Cincinnati Enquirer, May 16, 2000.

15. New York City Partnership, Description of Breakthrough for Learning program; Abby Goodnough, "Teacher Incentive Plan Pays Off Despite Fears," The New York Times, February 9, 2000.

16. New York State Education Department, Information Reporting and Technology Services. As shown in Table 4, in New York City's public schools the figure is 13.9 percent.

17. See Attorney General of the State of New York, "Defendants Trial Evidence Volume Accompanying Post Trial Brief," Supreme Court of New York State, Campaign for Fiscal Equity Inc. versus The State of New York, pp. 172-174.

18. The data are for permanently appointed teachers. Systemwide 6.6 percent of all teachers are certified but not yet permanently appointed, so the data understate the percentage of teachers who are certified within the system. Data from New York City Board of Education are for October 1, 1999.

19. Attorney General of the State of New York, op. cit., p.22.

20. Ibid., pp. 171-172.

21. The estimates in this paragraph are based on an analysis of teacher assignments by Richard Delaney for the Citizens Budget Commission based on data from the Board of Education, Division of Human Resources and Labor Relations for the 1994-1995 school year. They were presented to Mayor Giuliani and Chancellor Ramon Cortines in a letter from the CBC's then-chairman Lawrence Buttenweiser, dated September 1, 1995.

22. Bureau of Management Audit, Office of the Comptroller, City of New York, "Audit Report on the Board of Education's High School Teacher Utilization," MG 98-218A, June 29, 1999.

23. Michael Podgursky, "The Relative Pay and Contractual Work Hours of New York City Teachers," unpublished paper, August 3, 2000.

24. Sol Stern, "The Vanishing Teacher and Other UFT Fictions," City Journal, Vol.10, No.2, Spring 2000.

25. See testimony and evidence provided by Professor Michael Podgursky in New York State Attorney General, op cit., p.176.

26. This estimate is based on the following assumptions: (1) 17,000 teachers retire over four years at a rate of 4,250 annually; (2) the average pay of retiring teachers is that for a teacher with a masters degree and 22 years of service or $65,865 annually; (3) the retiring teachers are replaced by new teachers with a masters degree at a starting salary of $36,045; (4) the teachers hired move up and receive annual increments for longevity. The net difference between the fourth year salaries of the new teachers and that of the retiring teachers is $490 million.

27. United Federation of Teachers, "UFT predicts exodus of more than 14,000 senior teachers starting this June," Press release, January 20, 2000.

28. Up to one year extra credit applies to those working through June 30, 2001 and up to two years to those working through June 30, 2002. Eligible teachers earn one month of credit per year of service up to a maximum of 24 months.

29. Jeff Archer, "A Little Something Extra," Education Week on the Web, Part of a series entitled "Quality Counts '99."

30. This calculation does not take into account the current cost of the 15 percent differentials paid in ETS schools. These currently expended funds would partly offset the added costs calculated here.

31. Abby Goodnough, "Regents Create a New Path to Teaching," The New York Times, July 15, 2000, p. B1.

32. Alison Gendar, "Getting in Shape to Teach," Daily News, August 14, 2000.

33. Jeremy D. Finn, Susan B. Gerber, Charles M. Achilles and Jayne Boyd-Zaharias, Short and Long-Term Effects of Small Classes, Paper prepared for conference on the Economics of School Reform, May 23-26, 1999.

34. City of New York, Independent Budget Office, Memorandum on the Cost of Reducing Early Grade Class Sizes: Request from the Public Advocate, August 24, 1998.

35. Citizens Budget Commission, School Buildings for the Next Century: An Affordable Strategy for Repairing and Modernizing New York City's School Facilities (NY: CBC, September 1996).