Box 3 in 2028: what Dutch entrepreneurs should arrange now

The shift itself, in one sentence, moves from taxing a fixed percentage of an asset's value regardless of what it actually earns, to taxing what each asset genuinely earns, with one critical wrinkle: for most financial assets that includes unrealised value increases within the year, while for real estate and qualifying startup shares it means only cash actually received or gains actually realised through sale.

Key takeaway: 2027 is still the old system. Nothing structural needs to happen before 1 January 2027. The genuinely useful window runs from now until the end of 2027, and the most consequential items on it are real estate, where no action is needed since the opening balance applies automatically, savings, where the current forfait is quietly protecting most savers rather than punishing them, and prior-year OWR claims, which are worth checking immediately rather than waiting for 2028 at all.

Since box 3 sits alongside every other part of a Dutch entrepreneur's tax position, how much tax you pay covers that fuller picture.

Savings: Where the New System Is Actually Worse

The instinctive assumption about the new system is that it rewards anyone earning less than the current forfait implies. That is broadly true for investments. It is not true for savings right now, and understanding why matters for the months remaining under the old rules.

The banktegoeden forfait for 2026 sits at 1.28%. Actual interest on competitive Dutch savings accounts in September 2026 runs closer to 2.5% to 3.5%. That gap means a saver earning 3% on 200,000 euros in deposits is currently taxed as though they earned only 2,560 euros, 1.28% of the total, paying 36% of that, roughly 922 euros in box 3 tax. Under actual-return taxation, that same saver would be taxed on the real 6,000 euros of interest earned, paying 36% of that instead, 2,160 euros, more than double what the forfait currently charges.

The situation ran in reverse not so long ago. In 2021 and 2022, savings rates sat near zero, and the forfait applied to banktegoeden was itself close to zero too, so savers were not especially punished then either. What actually hurt savers in those years was a separate quirk: cash sometimes got assessed under the higher overige bezittingen forfait rather than the lower banktegoeden rate, and anyone caught by that specific misclassification has a genuine reclaim opportunity through the opgaaf werkelijk rendement, the OWR, which lets a taxpayer elect actual-return taxation for any year where it produces a lower bill than the forfait, provided they actively claim it rather than assume it applies automatically.

Going forward, though, the direction is clear: with savings rates well above the current forfait, moving to actual-return taxation in 2028 will cost savers with competitive deposits more, not less. There is no particular reason to move cash elsewhere before 2028 purely for tax purposes; the sensible move is simply recognising that the current system, whatever its reputation, happens to be cheaper than what is coming for anyone earning a decent rate on their savings today. Checking whether an OWR claim remains available for 2022 through 2025, where actual returns genuinely fell below the applicable forfait, is worth doing now, since bezwaartermijnen, objection deadlines, do eventually close. Since the timing of any such claim connects to the broader aangifte calendar, provisional tax assessment Netherlands covers that filing rhythm in more depth.

Investments: The Unrealised Gains Problem

The new system's main rule, vermogensaanwasbelasting, taxes the annual change in a portfolio's value whether or not that change was ever actually realised through a sale. For an investor whose portfolio climbs 15% in a single year, this produces box 3 tax of 5.4% of the portfolio's value that year, meaningfully more than the current 2.16% the forfait charges regardless of actual performance. In a genuinely strong market year, the new system costs more, not less, a fact that gets lost in the general framing of the reform as a relief for smaller investors.

The reverse holds too, at least on paper. A year where the portfolio drops 20% produces zero tax, with the loss carried forward to offset future gains. That carry-forward genuinely softens the picture across several years, but it does not eliminate the underlying problem, and losses cannot be carried backward, which means a bad final year at the end of a long bull run does nothing to recover tax already paid on the gains that preceded it.

Consider a DGA holding 500,000 euros personally in investments. Under the current forfait, the annual bill sits at a predictable 10,800 euros regardless of what markets actually did that year. Under the new system, a year where the portfolio rises 20% produces a bill of 36,000 euros, while a year it drops 10% produces nothing and a carried-forward loss instead. Whether the total tax paid across a full market cycle ends up higher or lower than the forfait is genuinely uncertain; what is certain is that year-to-year variability increases substantially, and that variability itself becomes a real cash management problem.

Watch out: Under vermogensaanwasbelasting, tax falls due on unrealised gains, the increase you see reflected in a portfolio's value but have not converted into cash. An investor holding a million euros in equity ETFs that rise 20% in a single year owes 72,000 euros in box 3 tax without ever selling a single share. Holding enough liquidity elsewhere to cover this annual bill becomes a genuinely new financial planning requirement that simply does not exist under the current forfait.

For anyone whose portfolio underperformed the 6% forfait in 2023, 2024, or 2025, an OWR claim can reduce box 3 tax for those specific years, though it needs proper documentation of actual trading records and portfolio valuations to hold up. Getting that documentation in order is exactly the kind of task worth having proper support for; how to prepare a balance sheet covers how asset valuation and record-keeping fit into a wider financial picture.

Real Estate: Where the Old System Taxed the Hardest

Property owners in box 3 have arguably had the roughest deal under the current system. A rental apartment bought for 280,000 euros in 2015, now worth 520,000 euros, gets taxed on that current value regardless of what it actually earns; at the 6% overige bezittingen forfait and 36% box 3 rate, that produces annual tax of 11,232 euros whether the property brings in 18,000 euros in rent or 25,000 euros. The new system is a genuine structural improvement for property owners, and it contains one feature that is not really a planning opportunity so much as a straightforward gift: the opening balance.

Under the proposed system, the starting point for calculating any future capital gains tax on box 3 real estate is the WOZ-waarde, the municipal property valuation, as it stands on 1 January 2028, an assessment owners actually receive from their gemeente in early 2029. Any appreciation between the original purchase price and that 2028 opening value is, in effect, never taxed at all; it is simply grandfathered into the new starting base. The apartment bought at 280,000 euros and sitting at 520,000 euros on 1 January 2028 enters the new system with a 520,000-euro cost basis, and capital gains tax only ever applies to whatever appreciation happens above that figure from 2028 onward.

Rental income itself changes character too. Under the new system, huurinkomsten get taxed annually as genuine income at 36%, calculated as gross rental income minus directly attributable costs, maintenance, management fees, mortgage interest tied to the property. This is a change in method rather than necessarily a heavier burden; the actual outcome depends on how the true rental yield compares to the flat 6% forfait currently applied to the entire property value regardless of what it actually generates.

The one genuinely practical action for property owners ahead of 2028 is making sure records are in order for whenever the WOZ assessment for that date actually arrives. The valuation itself is set by the gemeente, not the taxpayer, but if it looks meaningfully too high or too low relative to the real market value at that reference date, a bezwaarperiode exists to challenge it, and a more accurate opening balance protects a more defensible tax position for the entire period of ownership that follows. Since acquiring property in the first place carries its own separate tax considerations, overdrachtsbelasting Netherlands covers that acquisition-side picture in full.

Inside or Outside the BV: The Question Worth Asking Now

Dutch DGAs with meaningful personal wealth are increasingly asking whether it makes more sense to hold investments inside their BV under box 2 rather than personally under box 3, and the new vermogensaanwasbelasting, with its annual tax on unrealised gains at 36%, makes that comparison genuinely worth running properly rather than assuming the current forfait logic still applies unchanged.

Inside a BV, investment returns face VPB at 19% on the first 200,000 euros and 25.8% above, and eventually extracting that after-tax profit as dividend adds box 2 at 24.5% on the first tranche and 33% above. Combine a 19% VPB rate with 33% box 2 on the bulk of an eventual dividend and the effective combined rate lands close to 46%, meaningfully above the 36% box 3 rate on its own. But this comparison only holds if distribution happens immediately, which for most DGAs it does not.

The real advantage of the BV route is timing rather than the headline rate. A DGA who leaves investment returns inside the BV and simply reinvests them, rather than distributing, pays VPB annually while box 2 stays deferred indefinitely. Under the new box 3 rules, by contrast, tax falls due annually on unrealised gains regardless of whether any distribution ever happens at all. For a long-horizon portfolio with no near-term plan to extract cash, the BV can genuinely compound with only VPB as a drag, while the equivalent personal holding faces a 36% bite on paper gains every single year, which over a couple of decades of compounding can matter considerably more than the eventual combined rate suggests on paper.

Box 3 still wins in plenty of situations, though: short-horizon holdings, frequent distributions, or assets already sitting in the vermogenswinst category, real estate and qualifying startup shares, where both box 3 and the BV route defer tax to eventual sale in broadly similar ways. Moving existing personal assets into a BV is not free either; real estate transfers trigger overdrachtsbelasting, and other financial assets can trigger their own box 3 consequences on the way in. This is exactly the kind of comparison worth working through with a tax advisor well before 2028 actually arrives, since restructuring ahead of a new regime taking effect gives genuinely more room to optimise than trying to react once it is already running. Since a DGA's own extraction strategy sits right at the centre of this comparison, DGA salary vs dividend covers that decision in full, and key differences holding BV and operational BV covers how a BV functions as an investment vehicle more broadly.

Get Your Wealth Positioned for Whichever System Actually Lands

Nothing here requires an urgent decision before 1 January 2027, and the details of the new system could still shift meaningfully once the novelle actually reaches parliament, with a rate cut to 35%, a higher heffingsvrij rendement, and a one-year loss carry-back all reportedly among the options the cabinet is weighing before finalising its text. What is worth doing regardless of exactly how those details land is understanding, asset by asset, which parts of a personal balance sheet genuinely benefit from the change and which quietly do not.

If you want help mapping your own savings, investments, property, and any startup stakes against what is actually coming, book a demo and we will walk through your specific position. Our team can also help you incorporate your BV or get bookkeeping and payroll running correctly around whatever structure ends up making the most sense once the details are confirmed.

FAQs

When does the new box 3 system actually enter into force?

The stated goal remains 1 January 2028, but this is not yet guaranteed. The Eerste Kamer is waiting on a novelle expected around Prinsjesdag 2026, with formal treatment unlikely to start before January 2027, leaving open the possibility of a 2029 start if the process runs long.

What is the difference between vermogensaanwasbelasting and vermogenswinstbelasting?

Vermogensaanwasbelasting taxes the annual change in an asset's value, including unrealised gains, and applies as the main rule to savings, listed investments, and most financial assets. Vermogenswinstbelasting taxes only realised gains at the moment of actual sale, and applies specifically to real estate and qualifying startup shares.

Does real estate in box 3 benefit from the new system?

Generally yes. Rental income gets taxed annually going forward, but capital gains are only taxed at sale, and any appreciation accrued before 1 January 2028 is effectively grandfathered into the new opening balance and never taxed at all.

What is the WOZ opening balance and why does it matter?

It is the municipal property valuation as of 1 January 2028, which becomes the starting cost basis for future capital gains calculations on box 3 real estate under the new system, meaning all appreciation before that date is not subject to tax.

Are savings better or worse off under the new box 3 system?

Worse off for anyone currently earning above the banktegoeden forfait of roughly 1.28%, which describes most savers at today's interest rates. The current forfait is, somewhat counterintuitively, cheaper than actual-return taxation would be for most competitive savings accounts right now.

What is the opgaaf werkelijk rendement and can I use it for previous years?

It is an existing mechanism, already available for 2022 onward, that lets a taxpayer elect taxation on actual return instead of the forfait for any year where the actual figure is lower, provided it is actively claimed rather than assumed automatic.

Does holding investments in my BV become more attractive under the new box 3?

Potentially, particularly for long-horizon portfolios with no near-term distribution plans, since a BV defers box 2 tax on retained profit while the new box 3 system taxes unrealised gains annually regardless of distribution.

What happens to unrealised gains on my investment portfolio under vermogensaanwasbelasting?

They become taxable annually at 36%, even though no shares have actually been sold and no cash has changed hands, which creates a genuine need to hold separate liquidity specifically to cover this new annual tax obligation.

What is the heffingsvrij rendement and how is it different from the heffingsvrij vermogen?

The heffingsvrij vermogen is the current system's tax-free wealth threshold, roughly 59,357 euros per person for 2026. The heffingsvrij rendement is a smaller, different figure proposed for the new system, an annual tax-free return allowance of around 1,800 euros per person rather than a wealth threshold at all.

What are the box 3 rates and thresholds in 2026 and 2027?

For 2026, the banktegoeden forfait sits at 1.28%, overige bezittingen at 6.00%, with a box 3 rate of 36% and a heffingsvrij vermogen of roughly 59,357 euros per person. These figures are expected to be confirmed again for 2027, since the current system continues unchanged through that year regardless of the new system's eventual start date.