Portugal is marketed as a place to put operations. It is rarely marketed as a place to put shareholdings, and yet its participation exemption is as good as the ones people fly to Amsterdam and Valletta for — dividends and capital gains from qualifying subsidiaries exempt outright, on conditions that most real groups already meet by accident.
Put your subsidiaries in the panel and see which ones flow up clean.
The exemption in article 51 of the corporate tax code applies to dividends, and article 51-C extends the same logic to capital gains on selling the shareholding. Both rest on the same four tests.
Note what is absent. No special company form is needed — the SGPS exists as a legal category but carries no separate tax regime of its own, so an ordinary Lda takes the same exemption. No minimum capital beyond the general rules. No advance ruling. No application, no fee, no annual renewal.
The honest comparison is narrower than the marketing on either side suggests.
The Netherlands has the better technical regime and a deeper treaty network, and it costs several times more to maintain. Malta’s effective rate can be lower and it arrives through a refund mechanism that some counterparties and some tax authorities look at harder than they look at anything Portuguese.
Portugal’s advantage is not that its exemption is the best. It is that the exemption is good enough, the company underneath it is cheap to run, and nobody anywhere raises an eyebrow at a Portuguese parent. For a group whose holding company also needs to be somewhere the founders can live, that combination has no real competitor.
A holding company does two things: receives dividends, and eventually pays them onward. The exemption deals with the first. The second is a withholding question, and it is where structures that looked elegant fall apart — particularly where the ultimate owner sits outside the EU.
Portugal handles this better than its reputation suggests, because its domestic exemption on outbound dividends reaches beyond the EU to treaty states that exchange information and tax companies above a threshold. Which routes are open to your owners is set out in full in the piece on getting dividends out, and it should be read before, not after, deciding where the holding company goes.
The other half is substance. A holding company with no people, no premises and no decisions taken where it says they are is the easiest kind of structure to unpick, and the participation exemption is no defence against a finding that the company is resident somewhere else entirely. That is covered in the piece on substance and in the one on permanent establishment.
If the group’s subsidiaries are mostly in jurisdictions that tax at or near zero, the exemption does not apply to them and the holding company achieves nothing. If you need a treaty network that reaches places Portugal has no agreement with, another jurisdiction will serve better. And if the holding company will be a pure conduit with nothing underneath it, no regime anywhere will save it from the anti-abuse rules that exist precisely for that case.
Where it works well is the ordinary situation: a group with real operating subsidiaries in taxed countries, owners who want to be somewhere pleasant and defensible, and no appetite for explaining their structure to a bank.
Which subsidiaries qualify, what the exit costs when dividends leave Portugal for your owners, what substance the holding company has to actually carry, and whether moving existing shareholdings triggers tax where they are now. That last question sinks more restructurings than the destination ever does.
Talk about the structure