This document was received together with the press release appearing in the previous post. Again, reader comments are strongly encouraged. – Ed.
The Human Resource (HR) Design Project has completed the first phase of its process. The initial work team draft recommendations can be found here. Many of the reports are lengthy and discuss very detailed issues related to personnel policies and have a fairly limited effect on the careers of most employees. The most notable exception is the report of the compensation work team. If the recommendations of this work team were enacted, every employee on campus would be affected.
For this reason, the Wisconsin University Union (WUU) prepared this analysis. We are interested in your comments on the work team recommendations and our review. We can be reached at firstname.lastname@example.org, www.wuu.info and on Facebook.
Review of Primary Recommendations
A. Determining Compensation through Analysis of Labor Markets
The compensation work team recommends that labor market analyses form the basis of the UW compensation plan. Defining the labor market of an occupation, however, can for all practical purposes merely give the desired result. Some examples:
A food service worker can be compared to similar positions in a range of other labor markets, such as: 1) Big Ten Schools, 2) other large campuses, 3) local public employers, 4) local large corporations, and 5) all local employers-small and large, public and private. If 4 and 5 are chosen as the labor market for comparison, the analysis would most likely show that the compensation (the sum of pay and benefits) of UW employees is greater than the compensation within that other labor market. This is despite the fact that UW is institutionally different than a small restaurant with different needs, resources, and purposes. Alternately, comparison with employees within labor markets 1, 2 and #3, would likely indicate largely equivalent levels of compensation.
UW faculty had established a set of 11 comparable peer institutions as the basis for determining its relative compensation (Gov. Commission on Faculty Compensation, 1984). However, a recent legislative analysis used a different set of peer institutions that yielded reduced inequities between the UW and other universities (Compensation for faculty and academic staff. LFB, 2011). Similarly, a 2012 UW System report on compensation systems used three universities as a comparison group (Personnel Issues for Reorganization Taskforce, 2012)
A research specialist performing complex laboratory activities can also be compared to the same five labor markets. Unlike in the case of the food service worker, there are more likely to be large differences within each market. For example, local employers may have substantially different pay practices (profit sharing, stock options or alternately high pay and few benefits). However, there may be fewer differences between the groups, such as overall compensation between Big Ten schools and local large corporations.
Summary: There is no standard labor market for any group or individual occupations (with the exception of building trades). There are often valid arguments to be made for or against choosing one group over another. However, choice of a particular labor market as the standard will frequently determine the result.
B. Labor Markets and Discrimination
To better understand some of the problems in simply adopting labor markets as a single guide think back to 40 years ago. Women were largely locked out of the labor market. When they did work, their options were limited: clerical, education, food service, nursing, etc. Women who held jobs that required higher education were paid only a fraction of what men made in similar jobs (nurses vs. engineers). African-American and Hispanic workers were also segregated into very limited occupations, which were largely unskilled and were paid at lower rates than similar occupations held by whites.
While there have been improvements in the labor market in terms of race and sex integration, most occupational fields are still largely segregated. Women make on average about 80% of the salaries of men. Men are overwhelmingly employed in skilled trades, women in health care, and black men are unemployed at three times the rate of the overall population. Discrimination and job segregation always results in lower wages and worse working conditions for the group subject to the bias.
Summary: There is no “natural” or market-based reason why nurses are paid less than plumbers or teachers are paid less than computer operators. These are the effects of long-term discrimination and job segregation. Thus, when the “labor market” is used as the basis for establishing wages and benefits, the discriminatory patterns of compensation in other institutions may be copied into our own.
C. “Segmented” Labor Markets
As noted above there are many forms of labor markets for a specific occupation. One type has large employers, is usually unionized, and has higher pay and benefits and better working conditions. Another type is composed of small employers, is non-union, and tends toward unstable employment. In addition to these two traditional markets, there is a large labor market for self-employment for some occupations.
Many studies have shown that the three labor markets operate fairly separately. Employees who work for small employers with lower pay tend to stay in these jobs and do not move into large employment situation or self-employment. It also means that changes in compensation in one market has little effect on other labor markets located in the same geographic area.
Summary: Owners of an oil change station opening on University Avenue would not ask UW OHR for the pay/benefits for techs as a means of setting up their pay plan. They would ask a similarly-sized business to determine a comparable wage. UW should not base its pay plan on other institutions solely because they both employ similar occupations. The size, resources, and purpose (such as public or private) of organizations are critical factors in determining employee compensation.
D. Performance Evaluation
UW-Madison has an inconsistent track record in regard to the use of job performance evaluations. Some units in the University use it as a means of awarding merit pay and others make only the most pro forma use of it or not at all.
The recommendations propose the use of performance evaluation to “adjust” pay on a periodic basis after an initial rate is established through examination of the labor market. In the use of the word “adjust,” the committee is recommending that pay can be lowered as a result of the performance evaluation. This is a highly unusual practice in most employing organizations; even more so in public institutions.
A number of questions present themselves in regard to this proposal: How would the standards or metrics for evaluation be established? How often would evaluations be conducted that might affect employee pay? What safeguards would be instituted to reduce discrimination, favoritism and bias? What procedures would be in place to dispute an evaluation that affects an employee’s pay status?
Summary: There are neither the trained staff nor job-related evaluation instruments necessary to implement such a complex and potentially, highly contentious operation. Nor is there an institutional tradition that is ready to embrace the use of these processes for the purpose of “adjusting” (potentially lowering) compensation. As such, it is likely to be unpopular both among supervisors and subordinates.
Defining Total Compensation:
What factors should be taken in consideration when calculating compensation? Should we use the start rate or the maximum rate? Or, should we use the average or median wage of all people in a classification? Should we compare health insurance costs by premiums co-pays or the total costs paid by the employer? How do we calculate various levels of deductibles and co-pays? How many vacation days on average do employees in a particular classification earn? Some jobs have more people with fewer years of service than others which often results in fewer benefits.
Summary: There are over 500 titles in the unclassified service. There are hundreds more in for classified employees. Even if a sample of benchmark employee classes are used, the tasks of comparing jobs with similar qualifications and duties and accurately determining total compensation is a very complex undertaking which can result in the exclusion of important factors.
II. Features that should be Included in Recommendations
A. Cost-of-living increase
The compensation work team does not recommend that compensation be adjusted due to changes in the cost-of-living. Until 2008-09, when UW employees stopped receiving pay increases, an annual across-the-board pay increase was, by an overwhelming measure, the means by which employees received pay raises. Without a cost-of-living increase, inflation will erode the real value of the wages of most employees. Inflation increased 4% in 2008, 2% in 2010, 3% in 2011, and about 1% thus far this year. Thus, in addition to the losses due to the changes in WRS and insurance, UW employees lost 10% in real income as a result of the lack of a cost-of-living increase in the last four years.
Summary: A cost-of-living increase in compensation for all employees is a basic principle of compensation management. It is unlikely that the University will allocate as much funds for “merit” and adjustments due to labor market corrections as they would in an across-the-board increase. Failure to use the cost-of-living adjustment as the basis of a pay plan will increase inequities and reduce job satisfaction.
- Collective bargaining:
Until July 2012, the compensation package of most UW and state employees was determined through negotiations between equal parties- the state and employee unions. The compensation of employees that were not directly covered by the labor agreements were heavily affected by the negotiated agreements. These agreements were voluntarily entered into and democratically ratified by both parties. This process is opposite from the top-down, “scientific” method recommended by the Compensation Committee. Compensation plans that do not have the support or “buy-in” from the affected employees will result in job dissatisfaction.
Neither UW-Madison nor UW-System supported the repeal collectively bargain rights. The Work Group Principles adopted as underlying this process stressed that, as much as possible, the existing collective bargaining agreements should be integrated into this process. The work team recommendations do not reflect the position of UW or the precedent advisory groups in understanding the importance of this process for the last forty years.
III. Administrative Additions to Implement work team’s recommendations
As noted above, there are over 500 titles in use in the unclassified staff. There are hundreds more classified titles. To implement the recommendations of the compensation work team, at a minimum, the following operations must be implemented:
- Review all classification titles of the 15,000 job incumbents to ascertain if the title and position description are correct. OHR staff would conduct a statistically significant sample of job analyses within all large and selected benchmark classes to determine congruity of incumbent to class.
- Organize and train teams to develop job related performance evaluation instruments. Develop or contract to build internal capacity for training, development and evaluation of instruments including metrics. Instruments should be subject to tests of reliability and validity.
- Create a Center for Labor Market Analysis within OHR. Hire labor economists and compensation analysts to organize and maintain analytic system. Collect data from cohorts of empirically identified similarly-sized institutions on total compensation of all major occupational groups. Annually update data of selected groups and all classes in three-year cycles.
- Create an Office of Performance Evaluation within OHR. Hire job analysts, training specialists and survey research specialist to oversee program. Develop and implement training program in performance evaluation for all supervisors. This is an initial intensive training and followed by periodic reviews. The Office will oversee quality assurance program in areas such as studies of discriminatory impact and other statistical analyses of evaluations for anomalies. The Office will serve as the initial “step” in disputes related to validity of instruments and individual reviews.
- All supervisors with responsibility for performance reviews will participate and successfully complete training program in use of performance evaluation instruments. Performance evaluation will be conducted quarterly with an annual comprehensive review that is linked to pay.