Monday, 7 September 2026 · Issue 029 · Economics & Currency
AboutMethodologyFrameworksRegionsSearchSubscribe
The Corridor
A weekly publication of record on African tourism and the world that shapes it · Nairobi
HomeEconomics & CurrencyIssue 029

Booking.com can't stop a French hotel undercutting it. It can still stop a Kenyan one.


When a foreign visitor books an African lodge through an international online travel agency, 15 to 30 percent of what they pay leaves the continent as commission before they arrive. That much is visible. The deeper leak is the one the operator cannot escape: the platform's contract has historically barred the property from contacting the guest directly, from taking payment directly, and from offering a lower price on its own website, so the operator can never convert that guest into a direct customer and never stop paying the commission on the next booking. In 2024 the European Union banned these clauses and named Booking.com a regulated gatekeeper. African operators sign the same contracts with no such protection. This issue names the mechanism, enforced disintermediation, and reads it as the digital layer of tourism value capture.

Aerial view of a beachfront resort with pool, palm trees and sandy beach
A beachfront property from the air. The land, the building, the beds and the staff are on the continent; the booking, the payment, the guest's data and a fifth of the price increasingly are not. Photograph: Diego F. Parra / Pexels (illustrative).

Since November 2024, a hotel in Paris has been free to do something a lodge in the Maasai Mara still, in practice, cannot. It can offer a lower price on its own website than it shows on Booking.com. It can be contacted by its guest and take that guest's payment directly. It has a legal right to the data that guest generates on the platform. These freedoms did not arrive by the platform's goodwill; they were imposed by the European Union, which in May 2024 designated Booking.com a gatekeeper under its Digital Markets Act and from November required it to drop the contractual clauses that had prevented exactly these things across the European Economic Area.1 The African operator signs a contract with the same company, offering the same global distribution, and receives none of those protections, because no African jurisdiction has passed the equivalent law. The single platform behaves in two ways on two continents, and the difference is the subject of this issue.

The Corridor Index framework reads tourism economies by asking who captures the value a destination generates. Across recent issues it has traced the leak through physical channels: the enclave resort, the foreign-owned airport, the currency the destination does not control. This issue follows the value into the digital layer, where the capture is newer, less visible, and in one respect more complete, because it operates not only on the transaction but on the relationship and the data beneath it. The broader phenomenon is not new: development economists have documented tourism "leakage" for decades, the share of a destination's tourism earnings that flows back out through foreign ownership, imported inputs and foreign intermediaries, with UNCTAD estimating it at 40 to 50 percent of gross earnings for many small developing economies, and studies of African tourism describing a sector whose value is captured disproportionately offshore.6 What follows is the digital reconfiguration of that long-standing leak, the point at which the intermediary moved from the travel agent in the tourist's home city to the platform that now owns the booking, the payment and the data. The commission is the part everyone sees. The lock-in is the part that matters, and it has a shape worth naming precisely.

What leaves, and how much

Begin with the visible number. International online travel agencies charge accommodation providers commissions that generally run between 15 and 30 percent of the booking value, with Booking.com and Expedia the dominant pair; Expedia's brands typically sit in an 18 to 22 percent band, and Booking.com's effective rate climbs higher once a property joins the "Preferred Partner" and sponsored-visibility programmes that are increasingly necessary to be seen at all.2 The two companies are not minor intermediaries. Together they control close to half of the worldwide online travel agency market, which itself handles more than $400 billion in bookings a year, and in 2025 the largest OTAs spent a combined $20 billion on marketing, an advertising budget no individual African operator, or indeed African tourism board, can approach.2 For a foreign-facing African property that depends on these platforms to reach the European or American visitor, a fifth to nearly a third of the headline room rate is captured off the continent at the moment of booking. That is the first leak, and it is the one the industry complains about.

It is also the least of the problem, because a commission on a single transaction is survivable if the operator can turn that guest into a direct, repeat customer next time. The entire architecture of the standard OTA contract has historically been built to prevent precisely that.

The lock-in, named

Three contractual mechanisms, working together, convert a one-time commission into a permanent one. The first is the rate-parity clause, which requires the property to offer the platform a price no higher than it offers anywhere else, including on the property's own website, so the operator cannot use its own site to undercut the platform and draw the booking direct.3 The second is the anti-steering provision, which restricts the operator from using the platform's booking process to divert the guest to a direct channel, and in its documented forms has extended to preventing the property from obtaining the guest's own email address.1 The third is the handling of payment and data: when the platform intermediates the payment and owns the guest record, the operator finishes the transaction without the contact details, the payment relationship, or the booking history that would let it market to that guest again. The guest belongs to the platform. The bed belongs to the operator. Only one of those is a durable asset.

The combined effect deserves a name, because naming it clarifies what a commission figure obscures. Call it enforced disintermediation:5 the platform inserts itself between operator and guest, and then uses contract terms to make certain the operator can never remove it, so that the intermediary becomes permanent by design rather than by the ongoing consent of either party. The operator is not paying for a service it could choose to stop buying; it is paying rent on a customer relationship it is contractually forbidden to own. For an African tourism economy, this is the mechanism by which a sector can grow its arrivals, its occupancy and its reputation for years and still find that a rising share of the value routes permanently offshore.

15-30%
Commission an international OTA takes on an African accommodation booking, before visibility surcharges
~50%
Share of the worldwide OTA market held by Booking.com and Expedia together, of a $400bn+ annual market
9-15%
Travellers who find a property on an OTA and then book direct, down from 20-25 percent five years ago

Europe wrote the counter-example

The clearest proof that these terms are neither natural nor necessary is that a major jurisdiction has already abolished them. Rate parity in Europe had been under regulatory assault for a decade, with Germany, France, Italy, Austria and Belgium acting against wide parity clauses from 2015 onward, but the decisive step came with the Digital Markets Act. Having designated Booking.com a gatekeeper in May 2024, the European Commission required that from 14 November 2024 the company drop all parity clauses, narrow and wide, across the twenty-seven member states and the wider European Economic Area, and prohibited it from using measures with the same effect, such as raising commissions or downranking hotels that priced lower elsewhere.1 European hotels also gained a right of real-time access to the data they and their customers generate on the platform.1 The scale of the perceived past harm is visible in the response: in January 2026 more than 15,000 European hotels, backed by over thirty national hotel associations, filed a collective action in the Amsterdam District Court seeking compensation for the years the clauses were in force.4 Europe, in short, examined these contract terms, judged them anti-competitive, banned them, and is now suing over them.

African operators, selling to many of the same European tourists, sign contracts that in most of the continent still contain the terms Europe removed. Rate parity remains lawful and common across most of Africa; there is no continental gatekeeper regime, no guaranteed data-access right, and no collective-redress mechanism of the Amsterdam kind. The value that European regulation now keeps onshore for a French or Italian hotel continues to leak, unimpeded, from a Kenyan or Tanzanian or Moroccan one. The asymmetry is not a metaphor. It is the same company, the same clauses, and the presence or absence of a law.

The operator is not paying for a service it could choose to stop buying. It is paying rent on a customer relationship it is contractually forbidden to own.

The honest limit of the argument

The platforms are not villains in a morality tale, and the case against them can be overstated in two ways that a serious reading should concede. First, the OTAs deliver genuine value: they solve the discovery problem for a small African property that could never otherwise reach a traveller in Munich or Chicago, they handle payments and fraud and customer service, and the $20 billion they spend on marketing buys a demand-generation machine no single operator could build. A commission is, in part, payment for a real service, and an African lodge that vanished from these platforms tomorrow would lose bookings it has no other way to win. Second, even Europe's ban has not fully freed hotels, because the platform's ranking algorithm now does quietly what the parity clause once did openly: a room priced higher on Booking.com than on the hotel's own site simply converts worse and drifts down the search results, so the pressure toward parity persists without a clause to challenge.3 The Corridor's claim is therefore the narrower and sturdier one: the digital value chain in African tourism is structurally tilted toward offshore capture, the contractual mechanisms that tilt it have been judged unlawful where they have been tested, and the absence of any African regulatory response means the continent's operators bear the terms wealthy markets have already rejected, while also lacking the direct-booking infrastructure that would give them an alternative. The platform is worth paying. It is not worth being unable to stop paying.

Three tests over the next year

The first test is regulatory: whether any African state or regional body, or the African Continental Free Trade Area's emerging digital-trade protocol, begins to treat the dominant travel platforms as the gatekeepers Europe has judged them to be, since the value at stake is a real and rising share of the continent's largest services export.1 The second is infrastructural: whether African tourism boards and operator associations build the shared direct-booking and payment capacity, and the guest-data ownership, that would let properties convert platform-acquired guests into direct ones, because a right to undercut the platform is worth little without the means to be found without it. The third is the data question that sits underneath the whole arrangement: whether operators and their governments come to see the guest record, the thing the platform captures most quietly, as the asset it is, since in tourism as in every other digital market, whoever owns the customer relationship owns the recurring value, and at present, for most of African tourism, that owner is offshore. A French hotel can now undercut the platform. Whether a Kenyan one ever can is a question of law and infrastructure that Africa has not yet chosen to answer.

Sources and notes
  1. The European Commission's designation of Booking Holdings as a gatekeeper under the Digital Markets Act (DMA) on 13 May 2024, and the requirement that from 14 November 2024 Booking.com cease applying "parity" clauses (narrow and wide) and refrain from measures with equivalent effect, such as raising commissions or de-listing partners who offer lower prices elsewhere, across the EU and EEA; and hoteliers' new right to real-time, continuous access to the data they and their customers generate on the platform. Per the European Commission ("Booking must now comply with the Digital Markets Act," 2024), Booking Holdings' DMA Compliance Reports (2024–2025), and analysis in the Kluwer Competition Law Blog, which documents the anti-steering dimension including access to the end user's email address.
  2. OTA commissions generally ranging 15–30 percent, with Expedia brands typically 18–22 percent and Booking.com's effective rate rising via "Preferred Partner" and sponsored-visibility programmes; Booking.com and Expedia together controlling close to half of the worldwide OTA market; the global OTA market exceeding $400 billion in annual bookings; Booking Holdings 2024 revenue ~$22 billion and Expedia ~$14 billion; and the largest OTAs spending a combined ~$20 billion on sales and marketing in 2025. Per EHL Hospitality Insights / Hospitality Net, Cloudbeds, BookingWhizz and StayFi (2025–2026).
  3. Rate parity defined and its regulatory status summarised (banned across most of the EU, UK, Switzerland and Australia; lawful in the US and much of the Asia-Pacific region as of 2026), per RevEvolve and industry summaries. The "billboard effect" of OTA-discovered guests booking direct declining to an estimated 9–15 percent in 2025 from 20–25 percent five years earlier, per BookingWhizz. The persistence of parity pressure through platform ranking algorithms after the clause's removal, per Hospitality.today ("Europe killed the parity clause. Parity didn't…," 2026) and the academic crawl studies of the French and Italian removals by Mantovani, Piga and Reggiani.
  4. The collective action filed at the Amsterdam District Court in late January 2026 by more than 15,000 hotels, supported by HOTREC and over thirty national hotel associations, seeking compensation for harm from Booking.com's historical parity clauses, with Booking.com contesting the legal basis. Per reporting summarised by The Hotel Blueprint and HOTREC materials (2026).
  5. The framing of "enforced disintermediation" and the characterisation of the guest record as the durable, recurring-value asset are the Corridor's own analysis, situated in the wider scholarship on platform intermediation and data value in digital markets. Figures throughout are indicative of order of magnitude; commission rates and market shares vary by property, region, contract and year.
  6. On tourism "leakage" as an established concept: UNCTAD's estimate that import-related leakage runs between 40 and 50 percent of gross tourism earnings for many small developing economies (and 10 to 20 percent for more diversified economies), per UNCTAD, and the widely cited finding that of every $100 spent by a tourist from a developed country, only around $5 remains in a developing destination; the analysis of African tourism leakage through foreign-owned airlines, tour operators, travel agencies and hotel chains and reliance on imported inputs, per ICTSD / Bridges Africa ("Tackling Leakages in Africa's Tourism Sector") and Africa Policy Research Institute ("Consuming Africa: The Impact of Tourism on Local African Economies"). This issue treats OTA-mediated capture as the digital form of that documented structural leakage.