A single holding company is the simplest structure. A cross-border entity stack adds layers: a top holdco in one jurisdiction, an intermediate holdco in another, operating entities at the bottom. It looks complicated. It usually has a purpose.
The most common reason for stacking is treaty optimization. If you want capital to flow from a UAE operating company to a European shareholder, routing it through a Dutch or Luxembourg intermediate can reduce or eliminate withholding tax at the UAE level, while the European parent receives dividends under a participation exemption. Direct UAE-to-Europe flows may face withholding tax depending on the specific countries involved. The intermediate layer captures the treaty benefit.
A second reason is investor compartmentalization. If you have investors in multiple jurisdictions with different tax treatments, US LPs, EU family offices, Gulf sovereign vehicles, structuring them through different feeder entities into a common intermediate gives each investor the right tax treatment for their specific situation without complicating the others.
A third reason is exit planning. If you eventually want to sell the business, having the right intermediate structure in place can determine whether the exit is taxed as a capital gain (often favourable) or a dividend (often less so), and in which jurisdiction that gain is recognized. This is much easier to set up before the exit than to engineer in the middle of a deal process.
Entity stacks require ongoing maintenance: transfer pricing documentation, substance requirements at each level, annual filings in every jurisdiction. They are not set-and-forget. But for investors moving meaningful capital across borders, the after-tax difference between a well-structured stack and a single holding company routinely runs to millions over a decade.