Phantom Saudization: SAR 20,000 Is the Floor

On 12 May 2026 the Ministry of Human Resources and Social Development warned private-sector establishments flagged with suspicion indicators for phantom Saudization, telling them to "verify that the Saudis registered in their records are performing actual and regular work," and noting that most establishments previously flagged this way "were subsequently proven to have committed the violation." Those who do not correct their status face "exclusion from the Nitaqat programme" and "application of the prescribed statutory penalties."

So you go looking for the size of the fine, and you find six different numbers. One page says SAR 10,000. Another says 20,000. A third says 50,000. An HR account on X gives you a ladder by establishment size starting at 5,000. All published confidently, and all sharing one mistake: they describe the figure as a ceiling.

It is not a ceiling. In its most important source it is a floor, and it grows with every month the registration stayed open.

The official definition, and two elements every summary drops

The ministry's dedicated page defines it as follows:

"An agreement between the Saudi worker and the employer, whereby the worker is registered in labour market and social insurance data and is paid a wage without any indication of actual work being performed at the establishment. The establishment's motive is usually to benefit from ministry services and raise its Saudization rate."

Two elements in that text fall out of almost everything written about it:

First: "an agreement." The violation has two parties. The Saudi worker who accepted the registration is not a witness, he or she is a participant. This is why reports so often arrive from the inside.

Second: "without any indication of actual work being performed." Note that the test is not "does the worker exist?" but "is there any indication of actual work?" The burden of showing work falls in practice on the establishment, and that indication is either produced by your systems or it is not. Absence of evidence here is itself the indicator, not a defence against one.

That distinction is not linguistic. An establishment with a genuine Saudi employee who genuinely works, but with no attendance record, no documented tasks, no job description and no trace of activity in any system, stands in front of the inspection committee in exactly the same place as an establishment selling a name.

The number everybody gets wrong

The text the General Organization for Social Insurance publishes on registering a worker with no employment relationship reads:

"A financial penalty of up to double the contributions due for that registered period or SAR 20,000, whichever is higher, per case."

Read the last clause before anything else: whichever is higher. And per case.

The Saudi Press Agency and the financial papers all carried the story under the headline "a fine of up to SAR 20,000." That phrasing is grammatically defensible and practically backwards: it presents 20,000 as the worst case, when in the text it is the best one. SAR 20,000 is what you pay when the registered period is short. Once it lengthens, the fine becomes double the contributions, and no cap is stated for it.

On the other side of the agreement, a separate violation applies to the beneficiary for benefits drawn without entitlement, capped at the amount of those benefits plus restitution, and the registered contribution periods are cancelled as invalid. The Saudi worker who thought he was accruing free pension time walks away with a cancelled contribution period.

When the fine passes SAR 20,000

The arithmetic is direct. Contributions are calculated on basic pay plus housing allowance, and the combined rate for a Saudi employee is 21.5 percent on the existing scheme (11.75 employer plus 9.75 employee), or 23.5 percent on the new scheme for the period 1 July 2026 to 30 June 2027. Why the rate differs between two employees on the same payroll is set out in the GOSI deduction rate explained.

Take a single case with a contributory wage of SAR 4,000, the common figure in nominal registrations because it is the threshold at which a Saudi employee counts in full under Nitaqat:

The crossover sits at roughly month twelve on the existing scheme and month eleven on the new one. Past that point, SAR 20,000 stops being relevant at all.

And the fine is per case. Five names registered over three years at that wage is SAR 309,600 from social insurance alone, before anything else. Raise the contributory wage to SAR 6,000 and the crossover moves forward to month eight, with a single three-year case reaching SAR 92,880.

This is what explains the contradictory sources. Anyone who read "up to SAR 20,000" as a ceiling built a risk estimate roughly five times smaller than reality.

The second penalty: the schedule nobody reads

The second track is the schedule of violations and penalties for the Labour Law and its Implementing Regulation. The operative schedule was issued by Ministerial Decision 112377 dated 21/08/1447H and published on 25 February 2026.

A practical note worth recording here: we downloaded the file from the ministry's site and inspected it. It is a scanned 24-page document, produced on a document scanner and passed through an OCR engine that failed on Arabic, so the text layer inside it is not machine-readable.

The consequence is that any page giving you a number from this schedule has not read it. It copied a press summary or an earlier edition. That is precisely why the circulating figures disagree: 5,000 and 8,000 and 10,000 and 20,000 and 50,000 are not one ladder, they are the residue of schedules from different years competing in the same search results.

The correct posture toward this track is not to quote a number but to treat it as a variable: classification and amount are read from the schedule in force at the time of the incident, from the ministry's own copy rather than from a report of it. The violation covering an employer's failure to meet Saudization ratios drew objections on the public consultation platform over being classified "non-grave," which is itself a sign that this schedule is under continuing review rather than settled text.

The third penalty, and the one that changes the category

This is the paragraph that makes the argument over 10,000 versus 20,000 an argument in the wrong place. The ministry's own penalties section reads:

"Where phantom Saudization practices are discovered at an establishment, in addition to intermediaries, the penalties prescribed in the Labour Law, its Implementing Regulation and the decisions issued by this ministry in execution thereof shall be applied, and referral shall be made to the competent authorities to apply the penalties set out in the Penal Law for Forgery Crimes."

The Penal Law for Forgery Crimes was issued on 18 Safar 1435H. Its Article 8, where the document is attributed to a public entity, carries imprisonment from one to five years and a fine of up to SAR 500,000; Article 9, for a private document, carries imprisonment of up to three years and a fine of up to SAR 300,000, or either penalty.

The substantive difference is not the amount but the category. An administrative violation is settled, paid and closed. A criminal referral leaves the space of things an HR manager resolves through an e-service. Note too that the text names intermediaries explicitly: the service office selling to "clear your Saudization flag" is not outside the circle.

Suspicion criteria: the ministry published its detection logic

The most actionable part of the official page is that the ministry published the suspicion criteria it works from:

"The ministry has defined suspicion criteria for phantom Saudization... These criteria include verification of the work location, the equipment and the schedules, and confirmation that actual activity is being carried out. Indicators and signs that may point to phantom Saudization can also be analysed, such as a repeating pattern in the hiring of Saudi labour and the absence of evidence that real job opportunities are being provided."

Read that as a technical specification, because in practice that is what it is. All five criteria are queries over data you already hold:

The fifth criterion is the dangerous one, because it does not examine an employee, it examines you. A repeating pattern is visible from outside the establishment, in the flow of registration and cancellation data, with no site visit required.

The defence is not a denial, it is a data trail

If a suspicion notice reaches you, the question you will be asked is not "are these employees?" but "show us the indication." Clean establishments fail this test routinely, not because they are in violation, but because the indication is spread across four systems that do not talk to each other:

• Qiwa holds the contract, the occupation and the Saudization rate.
• GOSI holds the contributory wage and the contribution period.
• WPS holds the actual transfer and its date.
• Attendance or the operational system holds the only evidence that anybody actually worked.

The contradictions among those four are exactly what the inspector is looking for. A GOSI wage that does not match the WPS transfer, a regular transfer with no attendance record against it, an occupation in Qiwa with no activity trace: each of these is a gap that reads against you. We covered wage-and-transfer reconciliation in detail in WPS violations and payroll data reconciliation, and you can check your file before uploading it with the WPS file validator.

The layer that solves this is not a new system. It is a periodic reconciliation report over the systems you already run, emitting the exceptions and nothing else. That report is your defence file, and building it before the notice is far cheaper than assembling it afterwards.

Before you assume the fine is SAR 20,000

Three steps this week:

1. Pull the list of registered Saudis and compare it against the attendance record for the last twelve months. Every name with no matching attendance record is an item in the suspicion file, whether it is genuine or not.
2. Calculate the real exposure for each doubtful case using the double-the-contributions formula rather than the 20,000 figure, and multiply by the number of cases. That result is what goes to management, not the number in the press headline.
3. Separate the band problem from the evidence problem. If your band is red the path is entirely different, and we set it out in Nitaqat Motawar: why your band turned red; you can measure where you currently stand with the Nitaqat calculator. If you are planning transfers to adjust your rate, the conditions and limits are in Qiwa employee transfer and Nitaqat conditions.

Phantom Saudization is a problem of intent for a minority, and a problem of a missing data trail for the majority. The first is not our business. The second is an engineering problem with a solution.

If you want a neutral read on where you stand, request a review of your Saudization data: we reconcile Qiwa, GOSI, WPS and the attendance record, and hand you back the list of cases with no indication of actual work, and the calculated exposure for each one.

Sources: Phantom Saudization page, Ministry of Human Resources and Social Development · Schedule of violations and penalties, Ministerial Decision 112377 dated 21/08/1447H · General Organization for Social Insurance, statement on registering a worker with no employment relationship · Penal Law for Forgery Crimes, 18 Safar 1435H · Sabq, 12 May 2026.