New Jersey's FY2027 Compensation Compendium took effect on July 11, 2026 for twelve-month employees and September 5, 2026 for ten-month employees. We rebuilt our salary data from it and compared the result against last year's, line by line. The headline is not a raise. Base rates on the largest schedule did not move at all. The real news sits in two places most people will never open a PDF to find: the list of titles you can be hired into above the bottom step grew from 106 codes to 232, and a handful of schedules turn out to be frozen at rates last set years or even decades ago.
The CWA base schedule did not move
The A-P-R-S schedule covers the administrative, professional, primary supervisory and higher supervisory units, which together account for 4,871 of the job titles in our index. It is the schedule most state employees are paid on. Comparing every published range and step against the prior year, the change is zero. Range 9 still starts at $39,357.93. Range 21 still starts at $66,592.01. Not a single common range moved by so much as a cent.
That is worth saying plainly because it is easy to assume a new fiscal year document means new money. For the schedules that make up the bulk of the workforce, the FY2027 tables print the same numbers as the year before.
Some schedules are frozen, a few for a very long time
Each schedule in the compendium carries its own effective date, and eight of them predate FY2027. The managerial schedule and the two related management schedules were last set on July 7, 2007. The exempt schedule dates to December 23, 2006. The State Police managerial schedule carries an effective date of July 3, 1999.
The practical effect is large. A title at range 35 on the managerial schedule shows a published range of $83,210 to $116,477. The same range number on the A-P-R-S schedule runs $126,854 to $187,691. The gap is not a pay cut, it is an artifact of one table standing still since 2007 while the other kept moving. Managerial and exempt titles are generally paid off-schedule, meaning the salary is set individually rather than by stepping up a published ladder, so the frozen figure is the official schedule rather than a good guess at what someone earns.
We now label these schedules as frozen wherever they appear on the site, with the date the rates were last set, because showing the number without that context invites the wrong conclusion.
Not every union is paid on the same table
This one is a correction to our own data rather than a change in the state's. Until this update we treated most schedules as if they shared the A-P-R-S numbers. The FY2027 document makes clear they do not. At range 9, step 1, the three largest schedules diverge:
- A-P-R-S, the CWA units: $39,357.93
- H, the AFSCME health care and rehabilitation unit: $38,927.63
- C-I-O, the IFPTE crafts, inspection and operations units: $38,640.95
Thirty-six of the thirty-seven schedules now carry their own transcribed values on this site instead of borrowing another unit's. For the roughly 300 titles in the IFPTE and AFSCME units, the salary shown on their pages changed as a result. The old figure was never the state's number, it was our approximation of it.
The hire-above-minimum list more than doubled
Section 3 of the compendium lets an agency appoint a qualified candidate at an advanced step on their first day, usually step 9, without getting prior sign-off from the Civil Service Commission chair and the director of the Office of Management and Budget. It exists for jobs the state has trouble filling.
We had been reading two of its lists, health care and information processing. The FY2027 document makes clear there are more, and they work the same way. Adding them takes the count of eligible title codes from 106 to 232. Nothing was dropped. The additions break down as 99 codes under miscellaneous titles, 14 under professional investment titles, and 13 under engineering and other technical titles.
The engineering entries are a useful reminder that "step 9" is shorthand rather than a rule. Several of those titles cap at step 5, and a few health care titles cap at step 6 or 7. The compendium sets a ceiling per title, not one ceiling for the whole program.
One title was renamed. Codes 53301 and 53316, listed for years as Computer Operator Assistant, are now Information Technology Assistant.
Fixed rates for jobs that never had a schedule
Appendix A of the compendium publishes rates for titles paid a flat amount or a trainee band rather than a step ladder. These are the codes ending in the nineties that used to show no salary at all on our pages, because there was no step table to read. There are now published figures for 114 of them. A Correctional Police Officer is a single rate of $55,081.11. A State Police Captain is $180,443.06.
Emergency and winter weather hourly rates
Section 2 sets hourly rates for emergency work beyond normal hours, and we had not captured it before. Seven codes run from $32.19 an hour for unskilled manual or clerical work up to $73.29 for supervisors running statewide or regional operations. They apply from July 11, 2026 through July 9, 2027.
Three additional codes exist only for winter weather work such as snow removal and ice control, running from $56.84 to $66.74 an hour, and only between September 5, 2026 and April 16, 2027.
One correction worth flagging
While rebuilding this data we found that our own step advancement note was wrong. We had been telling readers that reaching step 11 required 24 months at step 10. The compendium cites N.J.A.C. 4A:3-4.5 and sets it at 52 pay periods, and the same rule requires 39 pay periods at step 8 before step 9 and 52 pay periods at step 9 before step 10. The page now reads those figures straight from the regulation data rather than restating them.
Data and methodology
Figures come from the FY 2027 Compensation Compendium published by the New Jersey Civil Service Commission, effective July 11, 2026 for twelve-month employees and September 5, 2026 for ten-month employees, covering the fiscal year that ends June 30, 2027. Salary values were extracted from the source PDF by coordinate-based table reading, then checked two ways: an internal consistency test confirming every step interval matches the printed increment and that the schedules obey their laddering rule, and a cross-check against a second, independent parse of the same document. Those two methods agreed on 2,307 salary cells with no disagreements. Year-over-year comparisons were produced by diffing the committed FY2027 dataset against the FY2026 dataset in version control. Title counts come from our index of 5,958 job specifications.
We publish the extraction and verification scripts alongside the data, and the schedules on this site regenerate from that dataset rather than being typed in by hand. If you spot a number that looks wrong, please tell us through the contact page and we will check it against the source and correct it.