That all depends on whether you think they are overpaid or underpaid! The association members may think they are underpaid, so they may want to use those organizations that pay more than you currently do. The elected officials, on the other hand, may think that employees are often overpaid, and may want to use only organizations that pay less than you do. In actuality, the answer is neither and both. (You already knew that, didn’t you?)
This reminds us of a client whose employee association (that happened to be unionized) wanted to survey only other departments whose members were unionized. Seemed logical to them, since they were under the assumption that unionized departments paid more. We advised them that in their given profession, that it might be wise to use unionized and non-unionized organizations since there were few unionized comparisons within a reasonable geographical area (about 300 miles). They insisted on only unionized organizations. As circumstances would have it, we collected comparative salary data on both unionized and non-unionized organizations.
When we summarized the data, we found that the non-unionized departments actually paid more, and it would have been to the employees’ advantage to use both unionized and non–unionized organizations, (See Table 1). Unfortunately for them, the City Council held them to their wishes and compared their salaries only to the unionized departments. Consequently, they did not receive an increase that year.
Table 1: Unionized vs. Non-unionized Survey Data
| Average Non-Union Salary
|| Average Union Salary
Which leads us to repeat the time worn phrase: Be careful what you ask for! The fact of the matter is that there is no simple answer, but there are guidelines that you should consider.
As pointed out in the above example, you need to accept the fact that your association may have a totally different perception as to what the labor market should be than the view held by management, and even possibly the elected board. In order to bridge that gap, you will need to workwith the association to develop a mutual understanding of the issues that affect salaries.
Traditionally, organizations use such phrases as “employers from whom we recruit and to whom we lose personnel.” In many cases, those would be adjacent agencies in urban areas or other agencies of similar size and character within a state or regional area. Unfortunately, it is not as simple as it may seem. For example, your active recruiting market is probably more local for administrative support and maintenance/trades jobs, as well as most other jobs that are considered non-exempt under the Fair Labor Standards Act (FLSA). That market could include other public, and even private, sector employers within your county or within a reasonable geographic radius of your community, (perhaps 50 miles, since we have yet to find a trades person leave a job to move across the state for another 10 cents an hour!) Conversely, you may need to recruit much more widely for professional and/or management personnel. The market for those positions could be narrower in terms of the type of organization (such as only other government organizations) but cover a substantially greater geographic area including your entire state or even a major region of the country, (see Table 2). In certain states, the state’s labor board may have already defined a statewide market where collective bargaining is the required process for determining wages and where binding arbitration is required to settle pay disputes.
Table 2: Labor Market Definition by Job Type and Level
This reminds us of another client that wanted to only look at other counties, since they claimed that they had never, ever (!) lost anyone to a city. Furthermore, their jobs were not similar to city jobs. To add to their argument, they said all the cities around them that might be comparable were in dire economic straits (some had been in receivership) so their wages could not be comparable. (We were beginning to question the validity of the argument, but…) After a good bit of discussion, we decided to eliminate the cities from the survey.
When the study was completed and the report was being presented to the employees, one department head asked why we did not include any cities. We responded that the County Board felt that they were not comparable and that they had never lost any employees to the cities mentioned. To which the department head responded that her last four employees left to go directly to the cities that were not surveyed, and they left for higher wages. When you hear comments like these, it reinforces the value of having good data!
Again, be careful what you ask for! Things get more complicated when you factor in such things as jobs that are unique to your organization (electric utilities, convention centers, sports arenas and airports). For example, public safety jobs pose a unique issue because typically, the only competition is for the entry level and the department head positions since most police and fire departments tend to promote from within rather conduct open and competitive recruitments for the middle level job classifications in these job families.
For public safety jobs, the only issue for comparing salaries is being competitive at the entry level. Once starting salaries are determined, the only other issue is determining how far apart each level or rank above the first level should be. Clearly, public safety employees do not necessarily agree with that philosophy but the reality is the second, third, fourth, and fifth levels in most public safety organizations are not open to them as outside applicants even though the other agencies may encourage lateral entry at the first level. Consequently, what others pay for those levels has little, if any, direct bearing on what you will pay those positions.
For example, one of our large agency clients was forced into eliminating the Police Lieutenant rank because of pay compression between the Sergeant rank and the Captain rank since the Chief of Police’s salary was fixed. That agency did not wantto pay command officers more than the Chief. Since Police Officer pay was increasing exponentially due to market conditions, and those employees receive overtime pay pursuant to the FLSA, Sergeant pay had to be increased to a level that made it unattractive for anyone to apply for the Lieutenant jobs. The result was a redistribution of the Lieutenant duties between the Sergeant and the Captain.
When defining a labor market, there are certain criteria that can serve as a starting point for your discussion. Those include:
- A population that is neither less than one-half, nor more than double the population of your community.
- An organization with a similar number of employees in the occupational groups under study.
- A similar character (residential versus heavily industrialized).
- A similar tax base (sales and/or property tax revenues).
- A similar per capita income of its residents.
- Similar demographics.
- Located in a state that has a similar tax structure.
(It is difficult to compare compensation levels for jobs in other states if they have a different overall tax structure than that in your state. For example, the fact that Nevada and North Dakota do not have a state income tax makes it a bit more difficult to make an accurate comparison with compensation levels in their neighboring states of California or Minnesota. Oregon and Washington have the same issues although the fact that Oregon does not have a sales tax usually offsets the fact that Washington does not have an income tax. This comparison is difficult for organizations that are on the state border, since they may have employees that live in the no income tax state, but work in the state with income taxes).
- Provides a similar range of services (contract cities versus full service cities or water/wastewater utilities versus electric utilities).
- An organization looking for the same group of applicants (similar knowledge, skills and experience) as your organization in selected or all job categories.
- Located within the geographic parameters (by job level) of your recruiting market.
While other criteria may apply, these tend to provide the most defensible basis for comparing compensation levels. Remember these criteria will vary by the jobs being compared since different jobs have different labor markets. Clearly, if one organization has a considerably greater ability to generate revenue than another, it is difficult to justify comparing the two organizations. That is not to imply that organizations with limited resources need to treat their employees as second-class citizens. It simply means that the two organizations may not necessarily be looking for the same type of individual to fill their positions. That is the single most important criteria. Large agencies often look for individuals who are specialists while smaller organizations often look for generalists. Consequently, your applicant pool would be, and probably should be, different.
When all the data are collected and reported, you need to be ready to defend the job matches and the organizations used as comparables. We have rarely been in an organization where once the salary data was summarized, they accepted it at face value. We have had employees flat out deny that the data we collected was correct, because they know for a fact that people in those organizations are paid much more than the data we reported. This is usually anecdotal information gathered from a friend of a friend and is usually wrong. In fact, in one situation the department head claimed that we benchmarked the wrong job. Had we benchmarked the correct job, we would have found that the salaries were much higher. When we did further, very detailed investigations, we found that the job that he wanted to match was a much lower level position that did not have the same level of legal authority as his jobs. Further, the average salary was about $5,000 less than what they were currently paying. Wrong anecdotal information.
Finally, when all is said and done, you need to figure out what you are going to do with the data when you receive it. If you are going to use it to adjust each surveyed job in lock step with the market data, the need for care and precision in matching the right jobs and the right organizations is absolutely critical. There is very little room for error. (Also—look out if one year the market rate for a job goes down!) If, on the other hand, you just want to get a sense of the market and you will be adjusting the entire structure based on how your structure lines up with the market overall, the need for brain surgery precision is less critical. (Not that we encourage sloppiness, but the concept of benchmarking is just that…to get a sense of where you stand in relation to the market. It is not an exact science.) Once you superimpose internal equity considerations, organizational structure differences, and cultural differences into the mix, the market data is used to determine if you are generally leading, following or matching the market. In the latter case, we have found that a variance of plus or minus 5 percent of the overall market average basically means that you are matching the market. Anything beyond that and are you are either under the market or over the market.
Once you know that, then we often find that the discussion leads to how much it will cost, how much money is available, which job or job families are the worst off, or where is recruiting the most difficult because of salary levels. In short, the process of finding the right market comparables leaves the realm of compensation and enters the world of politics.
So, before you embark on a market survey, it might be a good idea to take stock of your own organization’s characteristics, the nature of the work force and how your organization is different or similar to others. Answers to those questions will help you to determine which organizations to survey.
The Comp Doctor™ is the team of Jim Fox and Bruce Lawson, Managing Directors in the Human Resources & Compensation Consulting practice of Arthur J. Gallagher & Co. They specialize in assisting governments in fixing their compensation and classification systems.
This article originally appeared in the IPMA-HR News in July 2004. It was revised and updated in February 2017.
If you have a question you would like to have them answer, please write to them at email@example.com or firstname.lastname@example.org. They will try to include it in a future issue of Comp Doctor™.