Salary Negotiation · Deciding Pay Raise Rates
Salary Review and Pay Raise Decision
What is salary negotiation?
- Budget allocation rules come before negotiation
- A matrix that weighs rating and band position together
- Pay cuts need the employee’s consent; without agreement, prior terms continue
The phrase salary negotiation conjures up two people across a table trading numbers, but at most companies individual managers have little room to move. What actually happens is dividing a pre-set raise budget according to rules, then explaining the result and gaining agreement. So HR should focus less on interview technique and more on how large the budget is and what criteria it is split by. Negotiations are won or lost before anyone enters the meeting room.
Raise rates are usually built up in three layers: a base increase applied at the same rate to all employees, a performance increase that varies by rating, and an adjustment layer that covers promotions and market gaps. When allocating, don't look at the rating alone. Also look at where each person sits within their pay band. A raise matrix gives a higher rate to someone at the bottom of the band and a lower rate to someone at the top, even at the same rating, which narrows the internal gaps that widen year after year. Market pay, the benchmark for comparison, varies widely with the survey date and sample, so don't rely on figures from a single source.
It is a different matter from collective bargaining. At a workplace with a labor union, pay raises for union members are set through collective bargaining, and an individual cannot agree to terms less favorable than that. Individual negotiations for employees on annual salary contracts take place outside or on top of that framework. In either case, any change that cuts existing salary requires the employee's consent, and newly writing freeze or cut criteria into the rules of employment can trigger the procedure required for changes that disadvantage employees. Pay must not be withheld just because an agreement has not been reached. The principle is to keep paying on the previous terms while talks continue.
In practice, things most often go wrong in how the decision is communicated. If you hand over a contract right before payday and ask for a signature, it is an agreement on paper, but employees take it as an announcement. A bigger problem is pay inversion, where new hires earn more than existing employees. If you raise only starting salaries to match the hiring market and set aside no budget to adjust current employees’ pay, key people resign in a wave right after the raises are announced. Sharing only the results, without explaining how ratings link to raise rates, has the same effect.
Hwi KimCEO of HRDIST & Talent Development Systems Consultant
Inwoo LeeExpert in Leadership, Communication & Performance ManagementNo single person originated this concept; it is a procedure that became standard practice as annual salary systems spread. It combines a basic principle of labor contracts (pay is a core contract term, so changing it requires both parties’ agreement) with a compensation technique that divides the budget by performance rating and position within the pay band.
Let's look at a hypothetical case. Company A runs an e-commerce platform with 320 employees, all on annual salary contracts. After announcing a flat 4% raise last year, 14 developers resigned within two months, and 9 of them were people with 3 to 5 years of experience who earned less than recent hires.
- When the compensation manager charted the pay distribution by level, 23 of the 61 developers with 3 to 5 years of experience earned less than someone hired in the past year. The manager reported to the CEO that "the problem is inversion, not the raise rate."
- The HR head set the total raise budget at 5% of labor cost and split it into 2% base, 2% performance, and 1% inversion adjustment. The finance head asked to cut the adjustment portion but backed down after seeing the data on replacement hiring costs for people who had left.
- They built a raise matrix on two axes, rating and band position, so that even for an A rating the raise was 7% at the bottom of the band and 4% at the top. The table was shared with all employees one week before the interviews.
- The 38 team leads started their interviews three weeks before payday. One lead said, "This year, your number came from this cell in this table," and pointed to the matching cell of the matrix.
- Of the 11 employees who raised objections, HR corrected the amounts for the 2 whose band calculation errors were confirmed. The 3 who delayed signing kept getting paid at their previous salary while talks went on for 2 more weeks.
- In the two months after the announcement, developer resignations fell from 14 to 4, and 19 of the 23 cases of pay inversion were resolved. The satisfaction score on the interview survey rose from 2.6 to 3.9.