Accepting payments in Egypt and the Maghreb. 7 chapters and a final quiz.

Chapter 1. Choosing your entry route: four countries, four structures.

The first decision is legal, not technical. To accept payments in Egypt, Morocco, Tunisia, or Algeria, you must designate the entity that will sign the acquiring contract, and that entity almost always has to be local. The payment provider comes next, and that choice follows from the first. An entry plan that starts by comparing pricing grids has the order backward.

There are three possible structures. The first is to set up locally, with a subsidiary, a bank account, and an in-country acquiring contract. The second hands the sale to an operator already established in the market, which becomes the seller for tax purposes and collects payments on its own account. The third sells from abroad, in foreign currency, with no domestic acquiring, so it reaches only the share of customers allowed to pay outside the country. That share is small, and foreign exchange rules cap it.

The approach that works reverses the usual order and starts with the bank. The local bank tells you what it will let you repatriate, against which supporting documents, and how quickly. You then pick a provider from those the bank already works with in production. This sequence saves months, while the reverse order can cost you a full financial year.

Chapter 2. Egypt: wiring Meeza, InstaPay, and cash-in into one checkout.

An Egyptian checkout that offers a single payment method misses its market: four channels coexist, and each reaches different customers. Meeza cards cover the cardholder base fed by public-sector salaries, subsidies, and pensions. InstaPay serves banked customers who push a transfer. The cash-in network reaches customers who will pay cash 50 meters from home. Cash on delivery covers everyone else.

• Accept Meeza explicitly: confirm with the acquirer that the domestic scheme's BIN ranges are routed, and don't rely on the international networks alone
• Test declines before the happy path: limit exceeded, expired reference, insufficient funds, transaction rejected by the payer's bank
• Treat the server notification as the source of truth: the browser redirect may never arrive if the customer closes the tab
• Make notifications idempotent: a payment notified twice must never release two orders
• Assign a unique order ID in every channel, or three-way reconciliation becomes manual
• Set the validity period of a cash-in reference based on inventory turnover, not a default setting

One design decision must be made early. InstaPay is a push credit transfer, and therefore final. Chargebacks don't apply, which removes card dispute risk. The flip side: a refund means pushing a second transfer yourself, to a recipient you must be able to identify. Capture the payer's alias at the time of payment, or customer service will be refunding blind.

The legal framework points the same way. Law No. 18 of 2019 on non-cash payments requires government bodies and public-facing services to offer electronic payment. It penalizes cash settlement with a fine of 2% to 10% of the amount, capped at EGP 1 million. Egyptian public contracts are therefore paid electronically. To sell to a public entity, you need to be set up for it from the first invoice.

Chapter 3. Egyptian telecom wallets: who issues, who collects.

An Egyptian mobile wallet carries a telecom operator's brand, but the operator does not issue it. Banks licensed by the Central Bank of Egypt provide the mobile payment service; the operator provides the brand, the agent network, and the access channel. The distinction matters: your contractual counterparty is a licensed institution, never the operator. That institution sets the limits, runs the anti-money laundering controls, and determines the payout schedule.

• Use a provider licensed or registered with the Central Bank of Egypt: it handles the multi-wallet connection, and it pays you out
• Check the payout schedule wallet by wallet: it does not necessarily match the card schedule
• Plan for outbound refunds: on a push channel, a refund is handled as a payout, with its own controls
• Find out who bears the end customer's cash-out fees; that line item shows up in complaints, not in the connection contract
• Don't promise an operator brand in the interface until you have confirmed the provider actually routes it

Wallet interoperability is not a selling point, because it is mandatory. The mobile payment regulation approved by the Central Bank of Egypt on November 29, 2016, gave licensed banks six months to open it to their customers (CBE, via Daily News Egypt, 2016). A holder of one wallet can therefore pay a holder of another, and the consequence for merchants is direct: a single connection is enough to accept wallets from different brands.

Chapter 4. Morocco: contracting for acquiring after CMI's exit.

Morocco switched regimes on May 1, 2025. Until then, merchants had no acquirer to choose: the Centre Monétique Interbancaire (CMI) onboarded all of them. Since that date, 11 operators licensed by Bank Al-Maghrib can onboard merchants and deploy their own terminals. For the first time in 20 years, merchants have something to negotiate. The question is what the negotiation actually covers.

Merchant fee = interchange + scheme fees + acquirer margin

  domestic interchange     capped by Bank Al-Maghrib, not negotiable
                           1.20%  ->  0.65% excl. tax   (September 2024)
                           0.65%  ->  0.50% excl. tax   (October 1, 2026)
                           0.15% excl. tax  public bodies and neighborhood retail

  scheme fees              depend on the network used (domestic / international)

  acquirer margin          the ONLY line that is truly negotiable

Clause to demand: “Any downward revision of the interchange cap set
by Bank Al-Maghrib shall automatically apply to the merchant fee
as of its effective date, without an amendment.”

Two facts drive sizing in Morocco. The first is the instant interbank transfer launched on June 1, 2023, by Bank Al-Maghrib and GSIMT: execution in under 20 seconds, round-the-clock service, and a transitional limit of 20,000 dirhams. The channel has a low marginal cost and no chargeback exposure. The second rules out building e-commerce on the card base: nearly 86% of the value loaded onto Moroccan cards comes back out as cash at ATMs (Bank Al-Maghrib, 2024 annual report). In Morocco, the card is first and foremost a withdrawal tool.

Chapter 5. Pricing cash on delivery, then shrinking it.

Cash on delivery is hard to debate because it is hard to price. Many plans set an acquiring fee against a supposed “zero fee.” That comparison is wrong. The right metric is the full cost of a delivered, paid order, including returns and the cost of carrying the cash. It takes five lines to calculate, and once it is on paper, it often changes the decision.

Parameters to measure on YOUR data, not to assume:

  L   share of orders actually delivered AND paid           e.g. 0.75
  Fa  outbound shipping cost, per order shipped
  Fr  return shipping cost, per order refused
  Fc  carrier collection fee, as % of the amount collected
  P   average order value
  d   days between delivery and payout
  r   annual cost of capital

  cost = [ Fa + (1 - L) x Fr ] / L   +   Fc x P   +   P x r x d / 365
                  |                          |               |
        returns are paid for           carrier          cash carrying
        by the orders that             collection       cost
        go through

Compare, for the same order value, with the prepaid channel:

  cost = acquiring fee x P   +   P x r x d' / 365
                                          (d' = acquirer payout delay)

L is the dominant variable. At 0.60, even cheap round-trip shipping
is enough to make the channel more expensive than cards.

• Measure the return rate by segment before doing anything else: city, vertical, order value, customer tenure. Without that data, you can't weigh any lever
• Move the point of payment, don't remove it: collecting at shipment instead of on delivery eliminates wasted shipments without shutting out wary customers
• Reduce perceived risk: proof of delivery, a clear returns policy, delivery dates that are met. Distrust drives cash on delivery, not habit
• Offer the local instant rail (InstaPay in Egypt, the instant interbank transfer in Morocco): low marginal cost, immediate funds, no chargebacks
• Connect a cash-in network to turn “cash” customers into electronic payments without turning couriers into cash collectors
• Cap cash on delivery above a set amount, and require prepayment on orders where returns cost the most

Chapter 6. Getting the margin out: foreign exchange and repatriation.

Exchange controls determine the structure, yet they are addressed too late. The Moroccan dirham, the Tunisian dinar, and the Algerian dinar are not freely convertible. The Egyptian pound is subject to controls. Your margin doesn't leave the country because you earned it. It leaves because a foreign exchange filing authorizes it, within a deadline that runs whether or not your cash is ready.

• What supporting document will the bank require for each payment, channel by channel (card, instant transfer, cash-in, cash on delivery)?
• How much internal processing time does the bank add to the regulatory deadline, and from which document does it start counting?
• Can you open a foreign currency account, and what share of revenue can be credited to it?
• Which margin transfers (dividends, royalties, intragroup recharges) has the bank already processed for a comparable case?
• Where is the breaking point: above what monthly volume does the file go to enhanced review?

Exchange rules also define how much your customers can spend, and Morocco's Office des Changes is raising its limits from 2026. The e-commerce allowance for individuals rises from 15,000 to 20,000 dirhams a year. For companies with no foreign currency or convertible dirham account, it rises from 500,000 to 1 million dirhams. These amounts don't apply to sales settled in dirhams on the domestic market. They cap what a Moroccan customer can spend with a merchant based outside the country.

Chapter 7. Local invoicing and licenses: what blocks the sale.

Two administrative requirements can halt sales without ever touching the payment. The first is invoicing: three of the four countries require an electronic format validated by the tax authority before the invoice is issued. The second is tax: a non-resident seller may have to register locally. Both must be handled during scoping. If you discover them after launch, they block invoicing and therefore revenue recognition.

VAT on remotely sold services follows two approaches. Egypt applies a simplified registration regime to non-resident suppliers, triggered at EGP 500,000 of revenue over 12 months. The rate is 14%, reduced to 10% for professional and consulting services, which must register from the first transaction. The legal basis is Law No. 3 of 2022 amending VAT Law No. 67 of 2016, as detailed in Ministerial Decree No. 160 of 2023 (EY, 2023). When the customer is a resident VAT-registered business, the reverse charge applies and the seller collects nothing. Morocco followed suit in its 2024 Finance Act. Registration is done through an online service of the Direction Générale des Impôts, with a monthly return of revenue earned in Morocco and payment of the corresponding tax.

• Egypt, PSO/PSP regime published by the Central Bank of Egypt on June 19, 2025, under Articles 184 to 200 of Banking Law No. 194 of 2020, with a 12-month transition period for existing players
• Egyptian minimum capital: EGP 30 million for category A, which applies above an average monthly volume of EGP 750 million; EGP 10 million for category B; EGP 20 million for payment initiation and account information; EGP 500 million for a payment system operator (Matouk Bassiouny, 2025)
• Morocco, payment institution: a category created by Law No. 103-12 on credit institutions and similar entities, licensed by Bank Al-Maghrib, with activity limited to payment services and no deposit-taking
• What a license does not replace: the connection to the domestic switch, which is required to accept local cards
• What a licensed partner lets you avoid: tied-up capital and the licensing timeline, at the cost of a margin and a dependency you will need to renegotiate