Salary Freeze

What a salary freeze is

A salary freeze is a decision to stop discretionary increases to the pay of current employees for a stated period. Salaries stay at their present amounts, nothing is deducted from them, and the review cycle is not cancelled: it still takes place, with nothing to award.

A salary freeze concerns money that has not yet been paid, not money already in the employee’s hands. Whether a particular increase may be withheld depends on the source of that increase, and the definition does not settle it. An increase to which an employee is already entitled under a contract or a regulation is a separate question with its own sources.

What a salary freeze saves in the first year

Take an organisation with 200 employees, a combined monthly payroll of SAR 2,000,000, and a usual annual review of 4%. The increase withheld comes to SAR 80,000 a month, and the saving in the first year is SAR 960,000.

The figures in this and the following examples are assumed, to show the effects of the decision. They are not a benchmark and are not taken from a source.

The SAR 960,000 is correct, and it is the figure a proposal for a freeze can put forward. It also understates the effect, because the decision did not stop one increase: it lowered the base on which every later increase is built.

The saving from a salary freeze outlasts the freeze

Assume the market moves 4% a year, and that after the freeze the organisation returns to its usual review with no compensation for the year it missed. Compare two paths after two years:

• No freeze: SAR 2,000,000 multiplied by 1.04 twice, which is SAR 2,163,200 a month.
• A freeze for one year, then a return to the usual review: SAR 2,000,000 multiplied by 1.04 once, which is SAR 2,080,000 a month.

The difference is SAR 83,200 a month, and it remains in every month after that unless the organisation addresses it. So the saving was not SAR 960,000 once. It is a recurring saving: the organisation has moved to a lower level of pay, and a decision about one year has created a permanent difference.

This side of the decision can be missed when the freeze is presented, because a freeze is put forward as a measure for one year, while its effect on the base does not end with that year. Nothing about it comes up for review the following year either, since by then nothing of the freeze is left to review.

What catching up after a salary freeze would take

If in the second year the organisation wanted to return to where its payroll would have been without the freeze, it would need to reach SAR 2,163,200 a month, which is an increase of 8.16% in a single year.

That increase costs no more cash than the path without a freeze, because it arrives at the same point. But it is a figure that is hard to approve in one year, being more than double the usual 4%. The obstacle a freeze creates is therefore not one of money but of the size of the number that has to be approved later. An organisation that does not catch up moves onto the second path without ever deciding so explicitly.

A salary freeze with no catching up: a gap fixed as a percentage, growing in riyals

If the organisation returns to 4% a year without catching up, the path without a freeze stays 4.0% above the path the organisation is on, neither more nor less, because both are moving at the same rate.

In riyals, the gap grows. After another five years it reaches SAR 101,225.52 a month, up from SAR 83,200. A review that looks at percentages sees a stable position that calls for nothing, and a review that looks at riyals sees a difference that is growing. Both describe the same thing, and a decision can end up resting on one of them alone.

A partial salary freeze: 60% of the people and 65% of the saving

A salary freeze does not have to apply to everyone. One limited form stops increases for employees whose pay is above the midpoint of their grade range and keeps them for those below it.

Assume that 120 of the 200 employees are above the midpoints of their ranges, with combined pay of SAR 1,300,000 a month, and that the other 80 are paid SAR 700,000. The increase withheld from the first group is SAR 52,000 a month, or SAR 624,000 a year, which is 65% of the saving from a full freeze. The remaining SAR 28,000 a month is paid to the others as before.

The decision touched 60% of the people and captured 65% of the money, because in this example the 120 employees above their midpoints carry 65% of the payroll. It has a second effect. Those whose increase is stopped are already better placed within their ranges, and those who keep it are placed lower, so in its effect the decision brings positions closer together instead of spreading them apart. That arrangement can be explained to each employee it affects, unlike a full freeze, which falls on everyone for a single reason that is specific to no one.

What a salary freeze does to pay indicators

If the range midpoints are moved with the market while salaries stay frozen, every employee’s compa ratio falls by the same proportion. With the midpoints moved by the 4% assumed above, the fall is 3.85% of each ratio’s value: an employee at 1.00 moves to 0.9615, and one at 0.96 moves to 0.9231.

None of them has had a single riyal of pay changed. A report reviewed at the end of the year would show a general fall in positions that can be taken for a placement problem, when it is the arithmetic effect of one known decision. The effect is greater at the range minimum: an employee paid exactly at the minimum falls below it as soon as the table is moved, and enters a different category without any decision about them.

So the decision to freeze has to settle at the same time whether the midpoints move or are frozen with salaries. Freezing salaries alone keeps the table accurate and shows the gap. Freezing both hides the gap for a year and then makes it larger when the midpoints are moved again.

The number that decides a salary freeze: how many leave

The saving is known. The cost lies elsewhere, in the people who leave because of the decision. The one comparison that holds up is to ask how many additional leavers would cancel out the saving.

Assume that replacing a leaver costs SAR 25,000, including everything counted in cost per hire. The saving of SAR 960,000 then equals the cost of 38.4 leavers, which means the 39th additional leaver takes the decision past the point where it pays for itself.

This number is a breakeven point, not a forecast of how many more people will leave, and we found no published measurement of a freeze’s effect on leaving for the Saudi market. Its value is that it turns the question into one that can be judged: is a year without an increase expected to add 39 or more leavers among 200 employees? Someone who knows the organisation can answer that with reasonable judgement, which cannot be said of the original question.

A further cost does not appear in the saving at all: the time a position stays empty after someone leaves, which is what the vacancy rate tracks.

Who leaves after a salary freeze is not a random sample

The calculation above assumes that all leavers cost the same, but they do not. The decision falls on everyone equally, but its effect does not: those with an alternative in the market can move first, and those without one stay.

It follows that a freeze can work as adverse selection in particular occupations, and that the visible saving may come with a change in the makeup of those who remain that shows in no figure. The retention rate of the group employed when the freeze began shows how many of them remain, but not which ones. For this reason the decision should be examined against the occupations in high demand in the market, not across the organisation as a whole.

How a salary freeze differs from neighbouring terms

The neighbouring terms differ on what each one stops or takes, whom it reaches, and how long it lasts.

• Hiring freeze. A hiring freeze stops the addition of new employees, and its effect is on headcount. A salary freeze stops increases for the people already employed, and its effect is on average pay. The two can be taken together, and the effect of each is calculated separately.
• Red circled pay. Red circling concerns an individual whose pay is above the maximum of their range. Its cause is the design of the pay structure, and it lasts until the maximum moves. A freeze, by contrast, applies to everyone the decision covers, its cause is money, and its length is announced.
• Wage deduction. A deduction takes something from pay already earned, for a stated reason, under conditions and within a ceiling, and the rules are covered in our guide to wage deductions. A freeze takes nothing away. It simply adds nothing.

What undermines a salary freeze

• No announced end. A freeze with no known end can be taken as a new policy rather than a temporary measure, and its effect on whether people stay is then the effect of a permanent measure. The definition sets no length for a freeze, because the length follows the reason for the decision and what is expected to happen before it ends.
• No stated condition for lifting it. If the organisation does not say what will lift the freeze, every improvement in its results becomes a reason to ask, and every postponement becomes evidence that it will never be lifted.
• Unannounced exceptions. A freeze from which individuals are exempted without a written rule can do more harm to internal pay equity than it saves. Who is exempted, and under what rule, should be decided before the announcement, not after the first request that is hard to turn down. An exception settled in advance under a published rule stays inside the decision, while exceptions decided one at a time empty the decision of its substance and leave it standing in the announcement only.
• Hiring at market rates during the freeze. A new hire joins at today’s price while existing staff stay on yesterday’s pay, and the gap between them narrows year after year until it reverses. How the organisation’s pay stands against the market as a whole is the subject of external pay equity.

Before a salary freeze is announced

Three numbers should be calculated before the announcement, not after it: the saving in the first year, the permanent difference in the base once the freeze ends, and the increase that would have to be approved later to catch up. The third shows whether the decision is truly temporary or permanent in its effect.

The announcement should then state three things alongside the decision: how long the freeze lasts, what will lift it, and whether the organisation intends to make up what was missed. Silence on the third can be taken as a no, and that is worse than an explicit no, because people can plan around an explicit no, whereas silence leaves them waiting.

This is an explanation of the concept, not legal advice.

Related terms

• Cost per Hire
• Retention Rate
• Wage Drift
• Geographic Pay Differentials
• Attrition