The Federal Pay Labyrinth

By on November 30, 2012 in News, Pay & Benefits

What started out as a simple change in the federal pay system in 1990 has become a statistical labyrinth that is incomprehensible. The original idea was to modify the General Schedule (GS) system to mirror the model in the private sector. Outside of government, employers maintain separate pay systems for exempt and nonexempt jobs – the distinction is the required overtime after 40 hours. The model for nonexempt jobs is similar to the Federal Wage System (FWS). Exempt jobs – in government jargon the two-grade interval jobs – are generally paid under a single salary system.

It seemed like a good idea when the planners started– but in the backroom discussions they decided local pay rates would be good for everyone.

The locality pay idea is a standard chapter in textbooks – identify the other employers competing for talent and base salaries on surveys of those employers. Hospitals pay what other local hospitals pay; law firms pay what other law firms pay. There are surveys for virtually every sector. It’s logical, easy to communicate, and widely accepted. Significantly the FWS stays under the political radar.

The idea got convoluted when the planners started worrying about congressional support. Maps were prepared showing congressional districts. A sage realized splitting districts would not be good politics.

Over time the locality areas were expanded in size. Someone decided that if enough employees commute from the boondocks, then the boondocks should be added to a locality area. That I assume is how Adams and York counties in Pennsylvania (think Gettysburg) ended up in the Washington DC locality area – after all its only an 85 mile commute to DC.

The New York locality area now boggles the mind. It’s typical, however. It extends from the very rural Pike County in PA, with its population of 57,000, to New Haven – a distance of roughly 150 miles – and from Putnam County in New York to Ocean County in New Jersey – roughly 160 miles. Somehow Fort Dix and McQuire AFB were included although they are closer to Philadelphia.

Not surprisingly there are significant differences in market pay levels across that region. The differentials are reflected in the pay levels for one of the most common jobs, Senior Accountant or Accountant III (the top of the career ladder). The job duties and qualifications for this job would be very similar in every sector. It compares with GS 11 accountants.

The federal salary in the New Your locality area at GS 11, step 5 is $73,359. According to survey data compiled by the Economic Research Institute (ERI), possibly the largest database of pay information in the country, the average salary for comparable accounting jobs in Manhattan is $72,192. The job is paid the same in Stamford, CN, $72,904 and slightly lower in White Plains, NY, $70,196, labor markets dominated by Fortune 500 companies. But in Newark the average is $66,963, in New Haven $63,843, and in Stroudsburg (Monroe County, PA) the average is $53,280.

The comparative pay levels for accountants stand in direct contrast to the latest pay gap for the New York locality area – 41.2%. Accountants are only one job family out of hundreds but there is no rationale that would reconcile the discrepancies in survey data. One of the very real and ongoing issues is that the ‘gap’ estimates mask significant occupational differences. High demand, knowledge jobs are often badly underpaid.

As an example, private sector software developers in Manhattan at a level equivalent to GS 11 are paid $104,220, based on ERI data. That is 42% above the GS level.

The urban-rural differentials are typical of those across the country. Center city jobs are consistently paid more than in the suburbs. Companies locate backroom jobs in the suburbs to reduce labor costs. There is no justification for paying federal employees in New Haven or Stroudsburg the same as those working in lower Manhattan.

.Now with the pay freeze, a popular strategy is to request creation of new locality areas. The Federal Salary Council report released recently lists 12 possible areas. Three are state capital cities, with labor markets dominated by state agencies, universities, local government, and hospitals – not high paying companies.

One proposed state capital area is Harrisburg. Across that area it is difficult to find large private employers, except for Hershey Foods. The existence of a significant pay gap is surprising in light of the pay freeze for non-union Pennsylvania state employees from 2008 until this past July when salaries were increased 1%.

It would be useful to see a direct comparison of the pay for similar state and federal jobs. More than a few have essentially the same job duties.

Rube Goldberg would be proud. What started as a straightforward proposal to adopt a simple and logical program management practice has grown into a bureaucratic labyrinth, with statistical estimates layered on top of statistical estimates.

Salary planning is normally routine; junior HR analysts assemble the survey data. It’s not brain surgery. Local committees would be a practical basis for managing the salaries of many white collar job series.

Sweden relies on an approach that is similar to FWS. Local managers in Swedish agencies are responsible for managing resources within budget, including salaries. They have access to market pay data. Yes, they are expected to consider performance. They use the word ‘individualized’ in discussions of salary management. Managers work closely with unions to insure pay decisions are fair.

This problem is going to continue to be a red flag until credible market data are assembled.

© 2016 Howard Risher. All rights reserved. This article may not be reproduced without express written consent from Howard Risher.

About the Author

Howard Risher is a private consultant who focuses on pay and performance. His career extends over 40 years and includes years managing consulting practices for two national firms. He recently became the editor of the journal Compensation and Benefits Review. He has written four books, including Aligning Pay and Results. He has an MBA and Ph.D from the Wharton School of the University of Pennsylvania.

13 Replies

Comments RSS

  1. John Edward says:

    Contrary to the author’s general tone that the original idea was “a good idea,” “a standard chapter in textbooks” and so on, the idea was fundamentally flawed from the beginning. There are vastly different differentials between federal and private sector pay between occupational groups and within occupational groups from lower graded to higher graded employees. The fundamental error has always been that the federal pay system should mirror private sector pay. It should not. The idea was doomed from the start and it was always obvious that it would lead to the “Rube Goldberg” system that surprises the author now. Federal pay should be higher at the low end and lower at the high end. Corporate pay is not something worth emulating at the federal government level. Cost of living was left out of the equation and should have played more of a role. I’m sure Dr. Risher will be right in there explaining why once the powers-that-be tell him that it is where they want to go.

  2. Howdy says:

    What would be interesting to see is what the locality pay would look like if the indices for these far out counties (ie York and Adams county relative to the WDC locality) were eliminated from the locality pay computation.  I have to wonder how much weight those distanced counties receive in the locality computation, and how much they drive down the relative wage for fed employees in big metro areas.   

  3. $15300432 says:

    Trying to compare private sector pay and skills with an unaccountable federal workforce is nothing but folly. The avg federal benefits package adds 60% to ther salaries and aren’t replicated in the private sector. Its like trying to compare a race horse (private sector) to a mule (the public sector)

    • Hello says:

      You’re right — mules exist to work and racehorses exist for show.  I guess they all go to glue factory in the end.

    • Andy2x says:

      Yet again, trolling sites with misinformation.  I hope you’ve found a way to make a living out of this because you do it so often there has to be something in it for you.

  4. JHN says:

    Another problem with locality pay is that it tries to mirror private sector pay, not the cost of living.  This doesn’t work for resort areas.  For example; Aspen, Colorado.  The average pay for a bartender, ski lift operator or carpenter may be $30,000, but a small home starts at $1 million.  It is wrong to base the pay on private sector salaries and not the cost of living. 

    • Andy2x says:

      If locality pay mirrored private sector pay, as you say, then I would be making more than a member of Congress as a GS-13.  But you are correct there should be some consideration of COL, which has been all but ignored the past 3 years.

  5. former fed comp specialist says:

    I think Dr. Rischer makes some valuable points.  One that he missed, in my opinion, is that each GS grade level homogenizes many occupations together, occupations that outside the Federal Government probably are paid at different rates.  Thus, another way to help to deal with this issue is to create a salary structure that is more sensitive to variations between occupations that one would see when surveying employers outside the Federal Government. 

  6. FormerFed says:

    The other major problem is when there are no pay increases (last two years) or minimal (as proposed in the President’s budget) there is no movement in locality pay. In other words, the gap should be narrowed over time but when pay is frozen or pay increases are minimal there is no money for changing the amount of locality pay.

  7. Fed says:

    It seems one major problem with the system is that the locality pay is based on where people live and not where poeple work.  In the NY example if two people lived in pike county, one worked in NY and the other worked in Pike county the one person who commutes to NY would get the locality pay for NY while the other would not. Their respective salaries would be competitive with jobs in their local job market. The problem here is because ‘some’ people commute from an area outside the city, all the poeple in that area get the same high locality pay.

    • CuriousFed says:

      Fed – This is a commonly help misconception re: locality pay areas.  Pay is based on the location of the job.  Your rightly pointed out that this would be better and, fortunately, that’s the way it works now.  As a worker, I could choose to live in the boondocks but work in Metropolis.  My salary is based on Metropolis.  A problem arises in the “fringe” communities.  Why would I seek a job in the boondocks when, with some commute added in, I could be paid significantly more?  And, depending on where the “line” is between pay areas, two communities sitting side-by-side might have significantly different pay rates for the very same work.  Like so many other things, what started as a simple fix has become a monster.

      • Andy2x says:

        You are both right.  While the locality is based on the location of the job, the area covered by the locality has been increased artificially to cover where people actually live.