Set out an allocation, a spending requirement and a horizon, and the simulator returns what the portfolio is likely to be worth after fees, after tax and after the money you take out of it. One model across the whole range, because the arithmetic does not change at each round number — the options do.
Open the simulatorThe question is usually asked as though the answer were indexed to the number: one portfolio for $10 million, a different one for $50 million, a different one again for $100 million. That is not how it works, and knowing why saves a great deal of time.
Two things genuinely change with size. The first is access. Below roughly $10 million, private markets are not realistically allocable — fund minimums, the number of vintages needed to diversify properly, and the cost of administering commitments and capital calls make it a different problem, not a smaller one. Between $10 million and $25 million a partial allocation becomes possible. Above that, the full opportunity set is open, and the question stops being what you can buy and becomes what you should.
The second is what a basis point is worth in cash. A hundred basis points of avoidable drag is $100,000 a year at $10 million and $2.5 million a year at $250 million. Nothing in the arithmetic differs; what differs is whether the sum justifies the professional time, complexity and structural cost of doing something about it. That is the real reason advice looks different at different sizes.
What does not change is the shape of the answer. Above about $25 million, two portfolios with the same allocation return the same percentage, carry the same fee load and lose the same share to tax whether they hold $50 million or $250 million. The worked examples further down show exactly that: the percentages stop moving, and only the cash figures grow. This is why one simulator covers the whole range rather than six pages of the same model with a different number at the top.
Pick a wealth level and a risk profile to start, then change anything. Returns, fees and tax characteristics are illustrative starting assumptions, not forecasts and not drawn from any dataset — they are there to be replaced with yours.
The simulator runs on the same engine as the Portfolio Tax Drag Calculator. There is one model of after-tax compounding on this site, not two, and the tax figures the two instruments produce for the same portfolio are identical by construction.
Each asset class is described by nine numbers: its weight, its gross return, its all-in fee, how much of the return is paid out as income rather than left as appreciation, how much of that income is taxed at ordinary rates, how much of the accumulated gain is realised each year, how much of those realisations are short-term, and how quickly the sleeve can be turned into cash. The first eight decide the money; the ninth decides whether the spending plan is fundable.
fee = value × fee rate
return = value × (gross return − fee rate)
income = return × income share → taxed as it arises
apprec = return − income → untaxed until realised
realised = accumulated unrealised gain × turnover
Fees come off before anything is taxed, because a fund distributes and realises net of what it charges. Tax is paid out of the portfolio. After-tax income and realisation proceeds are reinvested in the same sleeve, so cost basis steps up by both. The portfolio is rebalanced to target weights at each year end with basis carried proportionally, so rebalancing neither creates nor destroys basis.
Spending is funded by selling assets, and selling realises a proportional slice of whatever unrealised gain has built up. The tax on that slice has itself to be funded, so the amount sold is grossed up until the after-tax proceeds equal the cash actually needed. A family drawing $1 million from a portfolio carrying substantial embedded gain is selling more than $1 million to get it, and the model says so.
Withdrawals are assumed to come from long-term positions and taxed at the preferential rate supplied. Spending grows at the rate entered, so the figure you type is in today’s money.
When you select a wealth level, the private sleeves are adjusted before anything is calculated: zero below $10 million, half the profile weight between $10 million and $25 million, full weight above it. The released capital returns to public equity and fixed income in a 60/40 split. This is a modelling judgement, not a rule, and every weight remains editable — but a simulator that offered a family with $5 million the same 25% private equity allocation it offers a family with $250 million would be producing a number that could not be acted on.
The upper line on the chart is the same allocation compounding at its gross return with no fees, no tax and no withdrawals. It is not an achievable alternative and nothing on this page suggests it is. It exists so the gap can be decomposed, and the table beneath it does exactly that: the distance between benchmark and outcome is withdrawals taken, plus fees paid, plus tax paid, plus the return all three would have earned had they stayed invested. Those four figures sum to the gap exactly.
A projection is an argument about assumptions. These are the ones being argued, and the places where a serious user should be careful.
Every sleeve earns its stated return every year. There is no volatility, no sequence-of-returns risk and no distribution of outcomes. This matters most for a portfolio funding withdrawals: a real portfolio can be forced to sell into a drawdown, and this one never is. Read the terminal figures as a central case, not a forecast, and treat the liquidity panel as the partial answer to the risk this omits.
You supply the two effective rates that apply to you. Nothing here asserts a rate, a threshold or a residence, and no default is presented as anyone’s actual position.
No loss harvesting, no carryforwards, no offsetting. A disciplined harvesting programme will reduce realised drag below what this model reports, particularly in the public equity sleeve. The figures are conservative in that specific direction.
The starting figures are plausible round numbers chosen to demonstrate the mechanism. They are not a forecast, not a survey and not drawn from any dataset. The instrument is only as good as the assumptions you replace them with.
Sleeves are tagged daily, quarterly or multi-year, and the panel reports what share sits in each and how many years of spending the daily-liquid portion covers. It does not model gates, side pockets, capital calls or a secondary sale at a discount.
It answers what a stated allocation is likely to produce on stated assumptions. It does not recommend an allocation, and PPLI.com does not provide legal, tax or investment advice. See our editorial standards.
Both run on the balanced profile, spending 2% of starting capital a year growing at 3%, over thirty years, at an illustrative ordinary rate of 40% and preferential rate of 25%. The only difference between them is size, and what size makes available.
| $10,000,000 | $50,000,000 | |
|---|---|---|
| Public equities | 48.4% | 40.0% |
| Fixed income | 25.6% | 20.0% |
| Hedge funds | 5.0% | 10.0% |
| Private credit | 5.0% | 10.0% |
| Private equity | 4.0% | 8.0% |
| Real estate and REITs | 7.0% | 7.0% |
| Cash | 5.0% | 5.0% |
| Expected gross return | 6.56% | 7.05% |
| Fee drag | 61 bps | 90 bps |
| Tax drag | 110 bps | 119 bps |
| After-fee, after-tax return | 4.85% | 4.96% |
| Realisable within a year | 79% | 65% |
The larger portfolio earns 49 basis points more gross. It pays 29 basis points more in fees and 9 more in tax to get it. What survives is 11 basis points a year — and the price of those eleven points is that the share of the portfolio realisable within a year falls from 79% to 65%.
That is the private-markets decision stated honestly. It is not nothing: eleven basis points compounded over thirty years is real money, and the allocation may well be right. But it is a considerably smaller number than the gross-return comparison suggests, and it is bought with liquidity that a family funding withdrawals from the portfolio may find it wants back at exactly the wrong moment.
| $10,000,000 | $50,000,000 | |
|---|---|---|
| Untaxed benchmark | $67,270,763 | $385,984,756 |
| Projected wealth | $19,462,066 | $101,458,720 |
| Withdrawn over the period | $9,515,083 | $47,575,416 |
| Fees paid | $2,609,611 | $19,785,667 |
| Tax paid | $6,695,166 | $36,167,925 |
| Compounding forgone on all three | $28,988,837 | $180,997,028 |
The last line is the one worth sitting with. At $50 million, fees and tax together come to $55.9 million over thirty years — and the return those payments would have earned had they stayed invested comes to $181.0 million. The cost of a basis point is never the basis point. It is the basis point and everything it would have become.
Run the same profile at $250 million and the percentages are identical to the $50 million case: 7.05% gross, 90 basis points of fees, 119 of tax, 4.96% net, 65% realisable within a year. Every cash figure is five times larger and not one ratio moves.
This is worth stating plainly because a great deal of writing on the subject implies otherwise. Above roughly $25 million the portfolio question is scale-invariant. What changes with further size is not the allocation arithmetic but the range of things worth doing about it: at $250 million, 119 basis points of tax drag is $3 million a year, which pays for a quality of structural and professional attention that the same 119 basis points cannot justify at $10 million. The percentage is the same. The response to it is not.
Not the gross figure. This is the rate the capital actually compounds at, and it is the only return number that has any bearing on what the family ends up with. If it is materially below what you assumed, the gap is in the two drag lines beneath it.
Fees and tax are both costs, but they behave differently. Fees are disclosed, negotiable and payable whatever happens. Tax is none of those things but responds to how the portfolio is built and where it is held. Knowing which is the bigger line tells you which conversation to have first, and it is not always the one you expect.
A portfolio can be solvent on paper and awkward in practice. If daily-liquid assets cover fewer than about four years of withdrawals, the illiquid sleeves are financing near-term consumption, and that is the position from which secondary sales get made at a discount.
If tax drag is modest, the answer is no, and no amount of arithmetic will make it yes. If it is large, the question becomes whether the cost of a structure is smaller than the drag it removes over the family’s real holding period — a separate calculation with its own break-even, which produces a negative answer more often than the market that sells such structures admits. That calculation is here.
The model divides return into ordinary income, short-term gains and long-term gains because tax codes do. These establish those distinctions in United States federal law. Other jurisdictions draw the lines differently, and the simulator accepts whatever rates result.
Defines short-term and long-term capital gain by holding period: not more than one year, or more than one year. This is the distinction the short-term input represents.
uscode.house.gov — 26 U.S.C. § 1222Requires a holder to include original issue discount in income as it accrues, whether or not cash has been received. It is why a private credit sleeve can generate a liability ahead of the cash to pay it, and a reason to enter a high income share for such strategies.
uscode.house.gov — 26 U.S.C. § 1272Imposes an additional tax on net investment income above stated thresholds. It is one of the components to fold into the effective rates entered at the top of the simulator.
uscode.house.gov — 26 U.S.C. § 1411The administrative treatment of investment income, capital gains and losses, and fund distributions, as published for individual filers.
irs.gov — Publication 550Read personally by a senior specialist. Never routed into a sales funnel. A written reply, usually within one business day.
Confidential PPLI ReviewYour information is submitted over an encrypted connection and handled in accordance with our Privacy Policy. PPLI.com does not sell personal information. Any external introduction is made only with your permission.