Copyright 1997 CAUSE. From CAUSE/EFFECT Volume 20, Number 2, Summer 1997, pp. 8-16. Permission to copy or disseminate all or part of this material is granted provided that the copies are not made or distributed for commercial advantage, the CAUSE copyright and its date appear, and notice is given that copying is by permission of CAUSE, the association for managing and using information resources in higher education. To disseminate otherwise, or to republish, requires written permission. For further information, contact Julia Rudy at CAUSE, 4840 Pearl East Circle, Suite 302E, Boulder, CO 80301 USA; 303-939-0308; e-mail: [email protected]
by Celeste M. Giunta
In the spring of 1996, the California State University systemwide office implemented a new classification series and compensation structure for its information technology professionals. Development of this project, which affected more than 1,000 union employees, took five years of discussions and negotiation. This article reviews the process and describes the compensation structure, union negotiations, and lessons learned.
Three years ago the California State University systemwide office (CSU) shared information about a new job design approach for information technology professionals in a CAUSE/EFFECT article.1 The goal of the new approach was to accommodate changing skill requirements, promote ongoing skill development in information technology jobs, and enhance organization effectiveness by optimizing the use of the available skill mix. The CSU faced numerous challenges in implementing the new approach given its size, diversity, and the fact that information technology professionals are represented by a union.2
At the time the article was published, the CSU had completed the classification and compensation design and development phases of the project and was on the verge of negotiating with union representatives the compensation changes necessary to support the new approach. That article presented trends for change that influenced the job design and provided an in-depth discussion of the job design process and outcomes, but did not address the compensation proposals because of their confidential nature at the time.
In the spring of 1996, after five years of development, discussion, and negotiations, the CSU began to implement the new information technology (IT) classification series and supporting compensation structure -- an effort that has affected over 1,000 employees. This article shares the final phases of the project, describing the design and development of the compensation structure, negotiations with the union, implementation, and lessons learned.
A formal study to explore alternative job design approaches for information technology jobs was initiated in 1991. The impetus for the study came from both outside and within the CSU. External pressures for change resulted from shrinking state funds and resources, coupled with expanded demands for educational access and accountability for technology services. Additionally, the CSU was strategically reorganizing and decentralizing decision-making and budget authority to individual campuses. Likewise, information technology departments on individual campuses were becoming increasingly decentralized and diverse in terms of management philosophy, organizational structure, information infrastructure, and resources. This complex diversity alone mandated job design flexibility.
Information technology managers were frustrated by existing job structure limitations, which resulted from the CSU classification structure. The classification and its associated salary structure were seen as barriers to effective work and utilization and development of human resources. Employees also expressed frustration at the limited opportunities for growth and performance recognition. As is common in most public entities, CSU classifications were narrowly defined, task driven, and hierarchical. The companion salary ranges were based on seniority and composed of narrowly defined step ranges. Together, these structures could not be readily adapted to the dynamic nature of today's information technology work environment.
As the CSU explored job design alternatives, it identified a fundamental need for an information technology job structure that would give managers flexibility in work assignment and resource management. More specifically, managers needed flexibility to achieve desired goals by assigning work in a way that expedited its completion and optimized the use of the available skill mix. The job structure needed to be supported by an equally flexible, market-competitive compensation structure that provided greater opportunity to promote and reward performance and development on the basis of competencies and skills valued by the organization.3
Negotiating these fundamental compensation changes was the biggest challenge and obstacle to implementation. Fortunately, CSU executive leadership articulated and stood behind a newly developed staff compensation philosophy and set of goals that supported the new approach. This commitment was crucial to the resolution of the negotiation process and to the ultimate implementation of the new IT classification series.
Chief among these goals was a pay-for-performance plan for all employees represented by collective bargaining agreements, including faculty. The commitment to pay-for-performance was viewed as essential for the successful implementation and administration of broader, more flexible salary ranges designed to support the IT job design. The specific goals identified for CSU staff classification and compensation were to:
The project was organized into four key phases:
Data collection. The purpose of this initial phase was to define goals, prioritize issues, gather internal job data, and conduct external research. Information was gathered through position questionnaires and focus groups with constituent groups.
Design and development. Based on information collected in phase one, a conceptual design was developed. To build consensus and ensure an operationally successful approach, the conceptual design was refined by partnering with two committees. The subject expert team helped troubleshoot job and skill requirements from a technical perspective. The human resources team worked to troubleshoot the approach from a classification perspective.
Negotiations. Once the classification and compensation structures were developed and management approval obtained, negotiations with union representatives began. Negotiations broke down and ended when the parties reached an impasse largely as the result of overall compensation goals and initiatives, not specifically the IT series. After exhausting the collective bargaining process, including mediation and fact finding, management made the decision to unilaterally implement the new IT series along with other compensation changes.
Implementation. The implementation process took a full year. It involved systemwide joint training of human resources and information technology managers on the new classification and compensation structures, as well as the individual campus processes for making determinations that moved employees into the new structure.
The job design model developed by the CSU evolved based on the two key goals identified in the data collection phase: flexibility and skill development. The key design strategy to addressing both goals was to focus on broader job functions and common skill sets, rather than specific job tasks. This strategy was consistent with human resources trends identified in the external research and provided the opportunity to expand job recognition to include the contribution of the individual.
Frustrated managers can probably count the times they have heard a human resources representative say "we pay for the job, not the person" when trying to recognize a top performer. The reality is each individual brings a unique contribution to his or her job and disregarding this contribution has always been problematic. Jobs don't define people, but people often define jobs because people, not jobs, accomplish work. As work continues to shift away from a set of discrete, ongoing tasks to "whatever it takes to get the work done" to meet workplace demands, the role of the individual and his or her competencies becomes even more central to the accomplishment of work and determination of pay.4
Focusing on functional differences allowed the CSU to collapse the information technology community from over twenty discrete classifications into six broad classifications inclusive of all job levels (see Table 1 for a brief description of each).
| Analyst/Programmer-Performs analysis and development of systems and technology-based solutions to meet user needs, including applications, databases, and related systems. |
| Operating Systems Analyst-Performs operating systems analysis and maintenance, including network and database systems, and their interfaces to all other systems. |
| Information Technology Consultant-Provides a broad range of consultative support to students, faculty, and staff to enhance the use of technology systems and information access. |
| Network Analyst-Engineers, analyzes, and supports all networks carrying voice, data, video, and/or broadcast transmissions. |
| Equipment/Systems Specialist-Installs, modifies, and maintains equipment and systems, with a hardware and systems configuration focus. |
| Operations Specialist-Operates, monitors, and controls multi-system information systems in data, voice, and/or video processing. |
Figure 1 illustrates how the old classifications were collapsed into the broader classifications with skill levels.
Creating broad classifications inclusive of all job levels, rather than traditional discrete, incremental classifications, provides for the essential flexibility and skill development. First, the approach is more suited to information technology knowledge workers because it broadens the focus from job content and scope to include the knowledge, skills, and contributions of the individual doing the work. The term knowledge worker refers to the fact that much of the work in information technology is performed in the mind using a varied, abstract knowledge base. Often work cycles are long and complex, making it difficult to get a snapshot of the full depth and diversity of the work. As a result, traditional, static job definitions do not capture the full essence of information technology work.5 Second, as a position evolves or individual skills develop, the approach provides the opportunity for more natural, fluid growth and development without artificial job-level barriers. Finally, changes in work assignments are readily accommodated without an immediate need to review classification or level assignments.
Based on the analysis of existing work through the data collection process, only three distinct levels were clearly distinguishable within each of the six classifications:
While the CSU program utilizes skills and competencies to define levels within the broadly defined classifications, it is not a true skill- or competency-based program.6 Skill- and competency-based programs are "person-based," rather than job-based. This means that pay is based strictly on a repertoire of skills or competencies (see definitions) that are performed by the person, and increases to base pay are based on the addition of skills sets or competencies. True skill/competency-based plans do not measure job tasks or content for pay determination.7
In the CSU model, skill requirements and development are key factors for progression within a broad classification, but job content is still the main determinant of pay. The CSU program is more of a job-based, skill/competency-influenced approach designed for information technology knowledge workers.8
To support the administration of the three broad skill-based level definitions, skill-level guidelines were developed. These include three core skill/competency dimensions that were identified as critical to successful performance in information technology work: (1) technical know-how, which encompasses depth, breadth, and integration of knowledge, (2) critical thinking skills, which cover problem solving, future thinking skills, and organizational, self, and project management, and (3) interactive capabilities, which include listening, communication, team, and leadership skills. (See Table 2 for excerpts from skill-level guidelines for professional level positions.)
| Skill Level | Foundation | Career | Expert |
|---|---|---|---|
Technical Know-How |
Basic knowledge of specialty area with limited ability to integrate elements within the specialty. | Functional, working knowledge of specialty area. Capable of intergrating skills and knowledge from other specialties | Advanced and comprehensive knowledge of specialty area. Capable of substatial integration from other specialties to achieve innovative results. |
Critical Thinking Skills |
Capable of solving problems where precedents exist. Refers others appropriately. | Applies theories an principles and uses reasoning and logic to analyze problems, explore alternatives, and implement the appropriate solution. | Understands problems from a broad, interactive perspective. Is able to develop and implement solutions that combine information in new ways. |
Interactive Skills |
Able to present ideas clearly in writing and orally. | Competent at interpreting and communicating information, ideas, and instructions. | Demonstrated expertise using persuasion and negotiation to build cooperation to expedite projects. |
Position skill requirements, as defined by management, are the primary determinant of skill level within an IT classification. Specific skill requirements are compared to the skill-level definitions and guidelines to determine the best level fit. The "person in level" concept (see Figure 2) is applied to determine the incumbent's eligibility for growth within and between skill levels. Movement to a higher skill level is based first on the need for a position at a higher level, and second on an incumbent's abilities. In the public sector, this is seen as a critical budgetary control mechanism.
The foundation of traditional compensation structures is the job -- a set of discrete tasks and skills. As the nature of jobs and work continues to change and expand in the workplace, compensation structures need to evolve and adapt. Traditional, rigid salary structures designed to limit discretion are outmoded and are being replaced by more flexible, innovative programs, which place the art of managing back into the hands of the managers.9
As the CSU became clearer in its job design approach, the need to overhaul its compensation structure became paramount. Broader classification and skill levels did not fit with the existing structure of narrow ranges with a 20 percent spread from the minimum to the maximum rate and defined steps. Additionally, salary progression through this step structure was essentially automatic, provided satisfactory performance was maintained. This left little opportunity to recognize or reward differences in contribution or promote development. Changes to this compensation structure had to be negotiated with the union.
To support broader jobs and recognize a greater variability in individual skills and competencies, a broadbanded compensation structure was the logical alternative. Most organizations introducing skill/competency-based or -influenced programs rely on some form of broadbanding. It is a compensation strategy designed to provide a broader view of work and more readily accommodate variable levels of skill, performance, and progress by significantly expanding the spread between the minimum and maximum rates of a salary range. Generally, broadbands are implemented when an organization seeks to decentralize and streamline pay administration and promote lateral career growth.
Typically, organizations that implement true broadbands define four to six salary bands for the entire organization. Each band encompasses major occupational groupings and the full spectrum of job levels. For example, the first band may include all non-exempt office support jobs; the second all non-exempt technical jobs; the third exempt professionals; the fourth management; and the fifth executives.
Broadbands can have a spread from the minimum to maximum rate of the band of 70 percent to 150 percent. Usually, to be considered a true broadband, the spread is at least 100 percent. This compares to traditional salary ranges that typically have spreads of 4060 percent. The concept of defined steps is antithetical to the basic premise of broadbands, which are designed to accommodate variability. Within a broadband, however, an organization may define pay zones based on various skill or market factors to assist in pay administration.
The broadbands utilized by the CSU are more accurately described as career bands because they are based on a job family. Each IT classification is essentially a job family inclusive of all job levels within that family. A key consideration in the job design and compensation structure was to provide for lateral growth and development -- a characteristic often associated with career bands.10
The individual career bands for each classification were developed based on a market study. The starting point for developing each career band was the midpoint. It was targeted to the statewide market rate for the fully proficient level within each classification. Statewide rates were used because the same structure applies to all twenty-two campuses and the chancellor's office location.
The spread from the minimum to maximum rate for each career band varies from 90 percent to 145 percent, based on the statewide market rates identified for the job family encompassed within the broad IT classification. The four professional level classifications (analyst/programmer, operating systems analyst, network analyst, and information technology consultant) have the same minimum and maximum rates. The small differences in proficiency rates identified for the different classifications in the market study were not meaningful in the context of broadbanding.
Additionally, sub-ranges are defined within the career bands for each of the three skill levels. The sub-ranges were developed by matching the criteria defined at each skill level to job levels with comparable knowledge and skill requirements in the market. The sub-ranges associated with each skill level are comparable to the approach of developing pay/skill zones to assist in managing pay within broadbands.11 In the CSU structure, each skill level sub-range overlaps with the adjacent sub-range by 1020 percent. This is primarily to account for varying entry qualifications and to allow for continuing recognition of performance and development within a skill level without inappropriate skill level advancements.
Figure 3 illustrates the career bands and sub-ranges that comprise the IT compensation structure. The career bands have spreads ranging from 90145 percent, based on market rates. Likewise the sub-range spreads vary and are noted in parentheses.
Currently, as a result of negotiations, movement through the bands occurs as the result of three types of salary increases.
(1) Performance-based salary increases. These are individual salary increases from a pool negotiated with the union. They are granted based on performance and each campus has the discretion in the distribution of this pool. Campuses must use their allocated performance pool, but may also supplement this pool at any time in any amount to recognize an individual.
The performance-based salary increases are the only variable component in the current compensation package. They allow for the recognition of contribution, which includes skill development and acquisition, taking on of new assignments, as well as traditional performance factors. The flexibility of this compensation component is viewed as essential to the success of the new classification and compensation structure.
(2) Service-based salary increases. These are salary increases that recognize service and are of a specific percentage negotiated with the union. They are granted to eligible employees with satisfactory performance on their anniversary date. As Figure 3 shows, employees are eligible for service-based salary increases only up to the midpoint, or fully proficient rate, within the overall classification range. The rationale behind this is that employees at the midpoint and above are being paid at a fully proficient level, and future salary increases should be based solely on demonstrable performance and development.
(3) In-classification progressions. When an employee advances to a higher skill level, this is referred to as an in-classification progression. At that time, an employee must receive an increase to the minimum of the new sub-range or of at least 5 percent. The criteria for advancement to a higher skill level are discussed under the job design.
(4) General salary increases. In addition, general salary increases are used to maintain market competitive salary ranges and pay. Because uniform percentages are applied to the salary ranges and individual salaries, there is essentially no movement within the salary band as the result of a general salary increase.
Managing costs is a central concern when broadbands are implemented. Within the CSU program there are several built-in control mechanisms. First, employees are only eligible for service-based salary increases to the midpoint of the career band. Second, skill-level sub-ranges identify market-appropriate pay zones for various job levels. Finally, and most critical, management has budget accountability for the distribution of pay-for-performance funds.
Gaining union acceptance of a new approach to classification and compensation is challenging; the CSU process took a total of three years. Initially, the union representing employees in IT classifications refused to bargain regarding the impact of the new series. By the time the parties reached the negotiation table, the process coincided with full contract negotiations for the bargaining unit covering IT professionals. This meant the new IT series was only one of several key management initiatives for change proposed to the union. Foremost among these initiatives was pay-for-performance. As mentioned, this initiative was viewed as critical to the successful implementation and administration of the new series, but was also a major stumbling block in the negotiation process.
Once negotiations began, considerable energy was devoted to educating union representatives on why the CSU wanted to make these changes and how the new structure would work. Through this process, the union expressed some positive reaction to the new opportunities available to employees, but were guarded about the concept of management discretion and differential treatment of employees -- both key elements of a pay-for-performance and competency-influenced program.
After negotiating for several months and exchanging numerous proposals related to the compensation structure and other unrelated management initiatives, negotiations broke down and the parties requested to be certified for mediation. The mediation process was not successful either, and the negotiation process was taken to the final step of fact finding. This process involved a fact-finding hearing with representatives from the CSU, the union, and an impartial third party. Unfortunately, the parties were still unable to reach an agreement through this final step.
Recognizing that the collective bargaining process had been exhausted without resolution, CSU executive leadership elected to unilaterally implement the proposed contract provisions in the spring of 1996. These provisions included a new compensation structure for all classifications, pay-for-performance salary increases for all classifications, and the new IT classification and compensation structure. Subsequently, through another round of negotiations, the union agreed to the compensation structure and salary increases were successfully negotiated in the summer of 1996.
Chief among the union's concerns with the new IT structures was management discretion and potential favoritism (exactly what traditional structures are designed to minimize) and adverse impact on employees. The union had a very broad interpretation of adverse impact. It sought guarantees that no employee would be placed at what was viewed as a lower level, and assurance that employees would not experience reduced pay opportunity as a result of a skill-level placement.
In the implementation process, CSU limited adverse impact to no individual employee's salary being reduced. If a campus found through the implementation process that an employee's salary was at a higher level than the skill level determined, they were encouraged to work with the employee on a development plan to bring performance and pay into alignment.
Working through negotiations to address these concerns and maintain some safeguards in the contract (such as service-based salary increases and seniority rights) was key to the ultimate success of reaching an agreement. However, the chief factor leading to an agreement was executive leadership's commitment to their position. Their willingness to move to unilateral implementation for the first time in the history of the CSU had a dramatic impact. They were committed to making compensation changes they believed were critical to the future of the University and effective management.
The new classification and compensation approach introduced a new way of thinking about people, jobs, and pay that was in conflict with traditional, comfortable practices. Just as ongoing communication was critical during the negotiation process, it became even more critical at implementation. Managers who had wanted flexibility suddenly had it, but had few tools and little experience to make it work. The new approach dramatically changed the roles in the classification and compensation process: human resources shifted from a directive to a more consultative role; line managers had to become more active and accountable in making classification and pay decisions; and employees had to become more self-directed in supporting their own development and career growth.
Because of the significance of these changes, many felt that implementation should be delayed until a complete new infrastructure was in place. However, it is clear now that if the CSU had waited, implementation would never have occurred. The classification and compensation structural changes have supported strategic initiatives and are proving to be a driver for further change.
Campuses were given complete flexibility in the implementation process, but were invited to training sessions sponsored by the chancellor's office. Because of the technical nature of IT jobs and impact of the change, it was deemed essential to jointly train human resources and information technology managers on the new structures. Campus representatives were encouraged to attend the training in teams. The training effort was substantial and the first of its kind within the CSU system, but still only a first step.
The training sessions were a combination of lecture, campus work sessions, and roundtable discussions. The key goal was to introduce the new approaches to classification and compensation and provide examples of best practices to assist campuses in developing their own administrative processes and procedures. Tools were provided for developing new position descriptions and making skill-level determinations. Other topics included communicating the changes to employees and supporting performance management processes.
Another key goal of the training was to open dialogues within and among campuses. Through the training sessions, campuses began to develop their own implementation strategies and plans. Many campuses chose to use implementation committees made up of central computing managers, decentralized computing staff, and human resource representatives to ensure equity and develop new campus processes. Feedback on the committee process from most campuses was positive. Many noted that implementation had gone more smoothly than anticipated. A couple of campuses reported committees having difficulty effectively reaching decisions or consensus on placement of employees within the new structure. Each campus is unique and needed the flexibility and opportunity to develop its own solution.
The author's experience in working with one team is illustrative. Initially, managers on the implementation team struggled with the new concepts. Significant time had to be devoted to discussing the concepts and working definitions so that they made sense in their environment. Ultimately, it was the practical applications that made the concepts come to life. Managers were almost surprised when they realized they had finally made the decisions necessary to place employees in the new structure. The process of gaining a comfort level was difficult, and at times unsettling, but once achieved, most managers commented on the relative ease of the new structure.
The need to continue the training and support process is apparent, particularly in the areas of compensation and performance management. As discussed in the previous article, these are essential supporting systems for longer term success and to achieve the original goals of the program to improve organization effectiveness.12
Looking back, the lessons learned at the CSU focus more on the change process itself than on the actual design features. This is similar to the findings of Hewitt & Associates in their study of organizations implementing broadbands, in which they noted that the design is not as critical as the organizational readiness for change.13
Key lessons learned to date in the change process include:
When entering into a change process of this magnitude, it is important to be willing to expect and tolerate problems. It is nearly impossible to reach perfection before implementation of broad change initiatives. The important thing is to learn from challenges and continue to move ahead. Being nimble is more important than being perfect.
Key lessons learned to date in the compensation program design include:
A successful program must have mechanisms for recognizing all the components of contribution -- taking on increased responsibility, applying new skills, and work performance. Without these mechanisms, the basic premise of the program is missing. The CSU must rely on its current pay-for-performance program for rewarding contribution and performance. The campus flexibility to supplement these funds is an essential component foundation for the program.
A more formal evaluation will occur over the next twelve to twenty-four months as the CSU evaluates the longer term viability of the new approaches and structures and makes revisions based on lessons learned. At this point, there is no going back, only moving forward. The CSU's new classification and compensation flexibility are hard-earned achievements, and the hope is that both the CSU and employees will benefit over time: the CSU with a more highly skilled and versatile work force improving organization effectiveness, and employees with improved skills, enhanced opportunities, and increased job satisfaction.16 The next generation of programs is already under way as the approach is fine tuned based on lessons learned. The CSU will be using similar approaches to revise and update other classification series.
DeCaporale, William. "Switching to Broadbanding -- One Company's Experience." ACA News, April 1996, 8-12.
Franklin, Jerry. "Pay for Technical Professionals: Pay for Skills and Pay for Performance." Personnel, May 1988, 20-28.
Schay, Brigitte W. "Broadbanding in the Federal Government: A 16-Year Experiment." ACA Journal (Autumn 1996): 32-43.
Compensation and Benefits Review, a journal published by the American Management Association (see http://www.amanet.org/).
See the Web site of the American Compensation Association at http://www.ahrm.org/aca/htm
The author would like to extend special thanks to Dr. Thomas W. West, Assistant Vice Chancellor for Information Technology at the California State University System, for his original vision and support; to Dr. June Cooper for her leadership in making the new structure a reality; and to all the campus information technology and human resources managers for their valued contributions to the development of a new structure and their energy during implementation.
1 Elsa Swan and Celeste Giunta, "Organizational Effectiveness and Changing Job Design in the Information Technology Community," CAUSE/EFFECT, Summer 1994, 36-44. The article is available online at http://www.cause.org/ir/library/text/cem9426.txt
2 The California State University System is the largest educational system in the nation, with twenty-two campuses ranging in size from 650 to 29,350 students; 36,000 faculty and staff; and approximately 326,000 students. Non-management employees are covered by several systemwide collective bargaining agreements.
4 Charles H. Fay and Howard Risher, "Work Redesign and Reward Systems: What Will We Do Without a Job?" ACA News, March 1996, 5.
5 Gerald E. Ledford, "Paying for the Skills, Knowledge, and Competencies of Knowledge Workers," Compensation & Benefits Review, July-August 1995, 55.
6 While the terms skill and competency are often used interchangeably, it is helpful to distinguish them. A skill is a discrete capability; competency consists of underlying knowledge, skills, and abilities that are essential for successful performance.
7 Graham L. O'Neill and Deidre Lander,
"Linking Employee Skills to Pay: A Framework for Skill-Based Pay Plans,"
ACA Journal, Winter 1993/94, 14.
9 Kenan S. Abosch and Janice S. Hand, "Characteristics and Practices of Organizations with Broadbanding -- A Study of Alternative Approaches," ACA Journal (Autumn 1994): 16.
11 David Hofrichter, "Broadbanding: A 'Second Generation' Approach," Compensation & Benefits Review, September-October 1993, 57.
15 Gerald E. Ledford Jr., "Designing Nimble Reward Systems," Compensation & Benefits Review, July-August 1995, 50.
Celeste Giunta ([email protected])
is a special compensation consultant for the California State University
System. For the past six years, she has been responsible for leading systemwide
compensation change initiatives, beginning with the information technology
job family.
![[Comments]](/ir/library/html/cem/comments.gif){kind=small}
![[Search]](/ir/library/html/cem/search.gif){kind=small}
![[Home]](/ir/library/html/cem/home.gif){kind=small}