Project division is the tool that lets a sponsor separate a large investment project into distinct projects – to sell one component, finance another separately, or match ownership to different partners. Decree 96/2026/ND-CP governs it, and the rule on what happens to investment incentives after a division is unusually precise.

When project division makes sense
Four situations recur. A mixed-use development where the commercial and residential components attract different investors and different financing. An industrial site where one production line is sold while the sponsor retains the rest. A phased scheme where later phases should carry their own timetable rather than dragging the completed phase into a shared implementation schedule. And a joint venture unwinding, where each partner takes a defined component rather than sharing a single asset.
In each case the alternative – keeping one project and allocating economics contractually – creates a structure that lenders dislike and regulators struggle to supervise. A project division makes the separation real.
The incentive rule after a project division
Decree 96/2026 provides that an investment project formed on the basis of dividing or separating a project enjoys whatever level of investment incentive its own conditions qualify it for – and enjoys that level for the remaining incentive period of the project as it stood before the division.
Two consequences follow, and both matter commercially. The clock does not reset: a division does not create a fresh incentive term, only a continuation of the remaining one. And the level is re-tested against each resulting project’s own conditions, so a component that no longer meets the sector, location or scale criteria may qualify for less than the parent project did. Modelling that before dividing is the difference between a neutral restructuring and an expensive one, as our investment incentives guide sets out.

Land is the constraint
Where the project holds land, a project division requires the land position to be divisible – separate parcels, separate certificates, and boundaries that survive cadastral scrutiny. Sites that were allocated as a single parcel for a single project cannot simply be notionally split; the land instrument must follow, and our land use rights guide covers the analysis.
Sponsors should test this before designing the corporate structure. A division that works on paper and fails at the land registry produces two certificates over one indivisible parcel, which is worse than the position it replaced.
Sequence and consents
The steps run in order: confirm the land can be divided; adjust or apply for the certificates recording each resulting project; address investment policy approval where the original project was subject to it; and allocate the implementation security between the resulting projects, as our project security deposit guide explains.
Where a component is to be sold, the division is usually paired with a project transfer, and running the two in the right order – divide first, then transfer the resulting project – avoids transferring an interest that has no independent legal existence.
Project division FAQs
Does a division create new companies?
Not necessarily. Dividing a project and dividing the company holding it are separate exercises, and they are often combined – see our merger and demerger guide.
Can incentives improve after a division?
A resulting project qualifies for the level its own conditions support, which can differ from the parent project’s – but only for the remaining incentive period.
What about the opposite direction?
Combining projects follows different incentive rules, covered in our project merger guide. Texts are published via the Ministry of Finance.

Modelling a project division before you file
The exercise that decides whether a division is worth doing takes an afternoon. List the incentive conditions the current project satisfies – sector, location, capital scale, headcount, disbursement – and the remaining incentive period. Then apply those same conditions to each project as it would exist after the division, on its own footprint and its own capital.
Three outcomes are possible. Both resulting projects still qualify at the same level, in which case the division is incentive-neutral and the decision turns purely on commercial factors. One qualifies and one does not, which prices the division and may argue for a different split line. Or neither qualifies at the original level, which is the answer that should stop the exercise and send the sponsor back to a contractual allocation instead.
Sponsors who skip this modelling discover the outcome after the certificates are reissued, at which point reversing the project division is considerably harder than it was to execute.


