A Moroccan services company holding Casablanca Finance City status used to be a short paragraph in a structuring memorandum. The status carried a corporate rate visibly below the ordinary one, and part of the case for placing a European parent above it was a rate case. That paragraph is now wrong, and it has been wrong for longer than most files admit. The regime was reviewed under Action 5 of the BEPS project, and the consolidated peer review results of the Forum on Harmful Tax Practices record Casablanca finance city as not harmful in its amended form, with ring fencing removed, substance requirements in place, and a grandfathering period that ended on 31 December 2022. A file still built on the terms that applied before that date is built on terms that stopped being available.
The change did not originate in Rabat alone. It came with a reorganization of the regime and a redrawing of who qualifies, in the same period in which preferential regimes generally were under international review. The direction is the same everywhere that review has reached: a regime survives if it rewards activity, and is dismantled if it rewards registration. What remains of Casablanca Finance City is still worth having. It is worth having for different reasons than it was before the review, and those reasons change how a Dutch holding above it should be built.
Where the old memorandum is now simply wrong
The error we see most often is a memorandum that compares a headline rate in Casablanca with a headline rate somewhere else and stops there. The peer review records ring fencing as removed and substance requirements as in place, so whatever the status gives today is no longer a benefit fenced off from the ordinary domestic position. Any file whose economics depend on a gap the review was designed to close has no economics left. The figures that matter now come out of the Code Général des Impôts as it stands for the financial year in question, not out of a memorandum written when the status was granted.
The second error follows from the first. Because the rate argument was doing the work, the substance argument was never made properly, and the Moroccan company was allowed to stay thin. A registered office, a service agreement with a related party, little else. That was survivable while the regime was a rate concession. It is not survivable now that the substance requirements the peer review records are met only by the company actually performing the services its status describes.
The third error is treating the status as a permanent attribute of the company. It is granted on the footing of qualifying activity, and the peer review ties what remains to substance rather than to registration. A company that drifts outside the activities its status covers does not lose a discount. It loses the basis of the file.
What the status still gives on the corporate side
The honest answer is that whatever remains on the corporate side is narrower and more conditional than the file usually assumes, and that it has to be read out of the Code Général des Impôts as it stands rather than recalled. The provisions attached to the status have been amended in step with the peer review, and the conditions they carry, rather than any headline figure, are what decides the value.
Which means the first piece of work on any Moroccan file is not a rate comparison. It is to establish, from the code in force for the year concerned, exactly what the status gives the company in front of you, for how long, and from what date any period is counted. We have reviewed files where the benefit had largely run out before anyone checked, because the clock had not started where the group assumed it started. A group that intends to rely on it should confirm how the period is counted before it builds a projection on it.
The withholding layer, which is where a holding structure meets the regime
For anyone building a group above a Moroccan company, the withholding position matters more than the rate. The Moroccan charge on a distribution is set by the Moroccan code. What the treaty does is put a ceiling on that charge, and the ceiling is not where most files assume it is.
Article 10 of the convention between the Netherlands and Morocco for the avoidance of double taxation and the prevention of fiscal evasion with respect to taxes on income and on capital, in the text in force, allows the state of which the paying company is a resident to tax the dividends, but caps the tax so charged at 10 per cent of the gross amount where the recipient is a company whose capital is wholly or partly divided into shares and which holds directly at least 25 per cent of the capital of the paying company, subject to a condition set out below, and at 25 per cent of the gross amount in all other cases.
Three things follow, and they decide how the structure is drawn. The lower ceiling is not automatic: it is available only at or above a 25 per cent direct holding, which is five times what the Dutch participation exemption requires, so a shareholding sized only for the Dutch side may not reach it. The general ceiling is high enough that for a minority position the treaty may be doing no work at all, and a file that treats the treaty as the thing that solves the distribution is usually a file in which nobody opened it. And the lower ceiling carries its own condition, that the relationship between the two companies was not created or maintained primarily to obtain the benefit of that reduced rate, which is the same question an administration will ask about the holding structure as a whole.
Why the Dutch parent is not there for the rate either
The Dutch company above a Moroccan operating or holding company is not a rate instrument. Under article 22 of the Wet op de vennootschapsbelasting 1969, in the text in force from 1 January 2026, Dutch corporate income tax is 19 per cent on the first 200,000 euro of taxable profit and 25.8 per cent above that. Nobody moves profit to the Netherlands for those numbers.
The Dutch company is there for three things. It receives dividends and realizes gains on qualifying shareholdings without a second layer of tax, because the participation exemption applies where the company is a shareholder in respect of at least 5 per cent of the nominally paid up capital of the subsidiary and the further conditions of article 13 are met. It gives a group with subsidiaries in several African jurisdictions one balance sheet, one audited set of accounts and one counterparty that European banks and European buyers already know how to assess. And it gives a treaty position that is legible to a purchaser at exit, which is often the point of the exercise even when nobody says so.
None of that requires the Moroccan company to be anything other than what it is. It does require the Dutch company to be more than a name on a register.
Substance, tested by two administrations, on two different questions
The two tests are not the same test and they fail in different ways. The Moroccan question is whether the company performs the services its status covers, from Morocco, with people, which is the requirement the peer review records as being in place. The Dutch question is whether the holding company makes its own decisions, has the means to make them and bears the risks it reports.
Groups that treat these as one exercise usually satisfy neither. They staff the Moroccan company and leave the Dutch board meeting in name only, or they build a proper Dutch board and let the Moroccan entity subcontract everything to an affiliate abroad. Each administration looks at its own entity and finds it hollow.
The workable answer is unglamorous. Real employment in Casablanca, sized to the services actually invoiced. Intercompany agreements that describe what those people do rather than what the group wishes were true. A Dutch board that meets in the Netherlands, on matters that are genuinely open, with minutes that show a decision being taken rather than recorded. Transfer pricing documentation that would survive being read by both administrations at once, because eventually it will be.
The employee side, and where it belongs in the file
Groups routinely put the personal tax side in the wrong part of the memorandum. Whatever treatment Moroccan law attaches in the year a hire is made to people employed by companies with the status, it is a recruitment provision and should never carry the weight of a corporate argument.
It belongs in the file anyway, for a practical reason. The regime that remains is conditional on real activity, real activity means senior people in Casablanca, and the cost of those people is the largest single number in the decision. That cost has to be priced on the terms actually available when the offers are made, checked against the code for the year concerned, and carried into the model as a cost. What the employee rules can do is make the substance affordable, which is a different thing from being a reason to build the structure.
A sequence that works, and one that does not
In one mandate we reviewed a group that had held the status for years and had built its projections on the preferential position the status carried when it was granted. That position had already been amended out, and the grandfathering that protected the old terms had ended. The benefit the group was still counting on had largely run its course. What remained was a set of conditions the group was not meeting and a benefit it had never quantified. We recommended keeping the status and fixing the substance, because the group was on a profit trajectory that would eventually make the remaining treatment worth having, and because losing and regaining a status is not a neutral event.
The sequence that works runs the other way. Decide first what the Moroccan company will do and who will do it. Establish what the code in force gives a company in that position, and incorporate and apply for the status so that any period it sets lines up rather than expires unused. Put the Dutch parent in place before the first distribution, and size the holding for the treaty ceiling as well as for the Dutch side. Then check each expected flow separately, because the treatment of the company’s own profit is not the treatment of everything that passes through it.
What we would tell a group deciding today
Casablanca Finance City is now what the peer review left behind: a regime whose benefit is tied to activity conducted from Morocco rather than to registration in it. It is no longer a rate. A group with people in Casablanca, clients across the continent and a European parent that does real work has a defensible structure and a benefit worth sequencing the incorporation around. A group with a registered office and an invoice has an exposure with a Moroccan file and a Dutch file, both of which now get read.
The right question is not whether the status is still attractive. It is whether the group would put those people in Casablanca anyway. If the answer is yes, the remaining benefits are worth organizing around, and the first task is to establish from the current code what they are. If the answer is no, no rate schedule is going to rescue the arrangement, because the rate schedule is the part that has already gone.
This article is informational and does not constitute tax or legal advice. The treatment of any structure depends on its facts and on the law of each jurisdiction involved. Each engagement is subject to scope and applicable regulation.