The number on the transaction is not the number that compounds.
The variable that separates them is not the tax rate. It is the order in which things happen, and how long the money is idle in between.
A sale is settled in weeks. What follows is settled over decades, and almost all of the difference between two otherwise identical outcomes is decided in the first eighteen months: whether anything was structured before the event or only after it, how much of the proceeds was taken off the table at closing, how much was left in cash, and how long it stayed there.
None of those is a market call. Each is a decision with a knowable cost, and each is usually made without one being calculated. A twelve-month delay before deploying $23,817,500 is not a twelve-month problem: on the assumptions in the second worked example below, it costs $2,429,924 by year thirty. That is not the interest foregone. It is what the interest would have earned.
This instrument therefore models a sequence rather than a rate. It starts at the gross proceeds, takes each deduction in the order it actually falls, reports what is left to invest as a share of the headline figure, and then compounds it — with the idle period inside the horizon rather than added to it.
It also models the one decision that cannot be made afterwards. A structure established before a taxable event and a structure established after it are different transactions with different consequences. The instrument does not assume that the first is better; it asks what it would have to defer, and what it would have to cost, for the answer to change.
Every figure the instrument reports is produced by the shared Wealth Intelligence engine from the assumptions on screen. Nothing is sampled, fitted, or drawn from transaction data.
Transaction costs come off the gross proceeds first, because selling expenses reduce the amount realised. Debt settled at closing comes off next; it is a use of proceeds rather than a reduction of gain, so it reduces what you keep without reducing what you owe. Tax is computed on the gain, which is the amount realised less basis. The charitable allocation is then paid, the spending taken and the cash reserve set aside. What remains is deployable capital. It sits in cash for the period entered, and is then invested.
gross proceeds
− transaction costs
− debt settled at closing
− tax on the event
− charitable allocation
− spending at closing
− cash reserve
= deployable capital → idle period → portfolio
The bars in the instrument partition the gross figure exactly: each deduction begins where the one above it ended, and nothing is netted against anything else.
Gain is the amount realised, net of selling expenses, less basis. It is then split into three characters using the shares entered: an ordinary portion, a short-term portion and a long-term remainder. The ordinary and short-term portions are taxed at the ordinary rate; the long-term portion at the long-term rate; state and local tax is added to both. The net investment income tax is applied separately, to the share of the gain the user says it reaches, because whether a business sale produces net investment income depends on facts the model cannot see.
Character is where most of the variation between otherwise similar transactions comes from, and it is decided in the transaction documents rather than afterwards. A consulting or non-compete allocation, depreciation recapture, and ordinary-income assets inside the business all move gain from the long-term column to the ordinary one.
A gift is cash out of the proceeds and a deduction against the gain, and the two are entered separately because they are rarely the same number. The deduction is applied against the most expensive character first — ordinary, then short-term, then long-term — which is the order a taxpayer with a choice would use. How much of a gift is currently deductible is limited by 26 U.S.C. §170(b) and depends on the recipient and the property, so the instrument takes the allowed amount as an input rather than computing it.
The idle period is entered in months. During it the deployable capital earns the cash yield and the interest is taxed as ordinary income each year, so what is deployed is the grossed-up balance less that tax. Horizons are measured from the event, not from deployment, so the delay is inside the period rather than added to it — which is the only way the cost of waiting shows up as a cost rather than as a shifted starting point.
The destination is one of four illustrative mixes. Each is run through the same portfolio engine used by the Portfolio Tax Drag Calculator and the Wealth Simulator: every sleeve produces income and appreciation in stated proportions, income is taxed as it arises, appreciation is taxed only to the extent turnover realises it, cost basis is tracked year by year, and the portfolio is rebalanced annually with basis carried proportionally. The expected gross return and the all-in fee are entered by the user; the selected mix is then scaled so that its weighted return and its weighted fee match those two figures. The mix therefore decides the character of the return and the user decides its size.
A reserve is not spending. It is held in cash for the whole period, earns the cash yield, is taxed as ordinary income each year, and is counted in terminal wealth. Reporting it any other way would either overstate wealth or understate what the liquidity is costing.
The portfolio engine runs on a whole-year grid. A delay of six or eighteen months leaves a part-year at the end of the horizon; it is compounded at the realised after-tax rate of the whole-year run, and that run’s tax and fees are extended in the same proportion. Forcing delays into whole years would have been simpler and would have hidden the effect this instrument exists to show.
Scenario B is not a product. The user states three things: what share of the surviving capital sits inside a structure, what share of the gain on that portion is not taxed at the event, and what the structure costs — an up-front load, a one-off structuring cost, an annual charge on value and annual fixed charges. Inside the structure nothing is taxed as it arises; the charges are assessed on the account value at the start of each year and paid out of it, so they do not themselves earn a return. The personal remainder is taxed exactly as in Scenario A. Both sides face the same event, the same delay, the same portfolio and the same rates.
The deferral figure is the assumption that decides the answer, and the model supplies no default for it beyond a starting illustration. Whether a pre-event arrangement changes the tax on the event turns on when it was established, what was transferred, who owns it and what was done with the proceeds. Set it to zero and the comparison reduces to a clean question: are the structure’s charges cheaper than the annual tax drag they displace? That is the question the PPLI Economics and Break-Even instrument exists to answer, and it is answered there rather than here.
Three settings. Held compares values as they stand, with embedded gains unsettled on both sides. Liquidated settles the taxable account’s embedded gain at the long-term rate and treats gain above investment in the contract as ordinary income under 26 U.S.C. §72(e). At death the taxable account’s basis is adjusted under §1014 and a policy pays under §101(a)(1). The setting applies to both scenarios, so the comparison is always like for like.
Each is a simplification made on purpose. Where one of them matters to a particular transaction, it matters more than anything the instrument computes.
The portfolio earns its stated return every year. There is no volatility and no sequence of returns. For a question that is largely about timing this is a real limitation: an investor who deploys into a falling market and one who deploys into a rising one have very different experiences of the same delay. What the instrument measures is the expected cost of waiting, not the risk of it.
The provisions cited are United States federal. State and local tax is a single figure added to the federal rates; apportionment, residency changes, state-level treatment of instalment sales and states that do not follow federal characterisation are outside it. Nothing here is jurisdiction-specific beyond that, and nothing in it is tax advice.
The whole liability is settled at closing. Instalment reporting, escrow and earn-out arrangements, qualified opportunity fund deferral, exclusions for qualified small business stock, loss carryforwards and estimated-payment timing all change when the money leaves, and none of them is modelled. Where any applies, the honest use of this instrument is to reduce the taxable gain and label the assumption.
The model does not decide whether a pre-event structure defers anything. It applies the share you give it. That is deliberate: the answer depends on facts and documents, and a model that assumed a figure here would be asserting the very thing that has to be established.
The retained-holding scenario shows what happens when a third of the position is not sold: no realisation, no turnover, and therefore no annual tax on its appreciation. It says nothing about the risk of holding it. A single position that halves is not described by a deterministic return, and the tax advantage of not selling is the smaller half of that decision.
Ongoing withdrawals are not modelled here. Money taken at closing leaves and the rest compounds untouched. For the drawdown question — what a portfolio supports year after year — the Wealth Simulator is the instrument.
Nothing here assesses whether an allocation is appropriate, whether a structure is available, or whether a transaction should happen at all. It applies stated assumptions consistently and reports what they imply.
All three use the instrument’s own defaults, so they can be reproduced above: a sale of a private company with 3% transaction costs, 10% of the gain taxed as ordinary income and the rest long-term, a 37% effective ordinary rate, a 20% long-term rate, 5% state and local, no net investment income tax (material participation assumed), a diversified destination returning 7.30% gross with a 0.65% all-in fee, and 4.20% on cash. Every one of those is editable.
| $25m sale deployed at closing | $50m sale twelve months in cash | $100m event six months in cash | |
|---|---|---|---|
| Gross proceeds | $25,000,000 | $50,000,000 | $100,000,000 |
| Tax basis | $500,000 | $1,000,000 | $2,000,000 |
| Transaction costs | $750,000 | $1,500,000 | $3,000,000 |
| Debt settled at closing | $2,000,000 | $4,000,000 | $8,000,000 |
| Tax on the event | $6,341,250 | $12,682,500 | $25,365,000 |
| Spent at closing | $2,000,000 | $3,000,000 | $5,000,000 |
| Cash reserve | $2,000,000 | $5,000,000 | $10,000,000 |
| Investable capital | $11,908,750 | $23,817,500 | $48,635,000 |
| Share of gross deployed | 48% | 48% | 49% |
| Cost of waiting, 30 years | — | $2,429,924 | $2,398,820 |
| Wealth at 10 years | $22,345,879 | $45,028,992 | $92,733,399 |
| Wealth at 30 years | $56,442,487 | $112,867,831 | $232,531,060 |
The share of the headline figure that reaches a portfolio is 48% in both cases, and it is almost entirely a function of the assumptions rather than of the size: the model is scale-free, so a $25 million sale and a $50 million sale on the same terms lose the same proportion. What separates them is the twelve months. On $23,817,500 the delay earns $1,000,335 of interest and gives $420,141 of it back in tax, and the money deployed is $24,397,694 rather than $23,817,500 — which looks like a gain until it is compared with the portfolio it was not in. By year thirty the delay has cost $2,429,924, or 2.1% of the terminal figure, for a wait that was 3.3% of the period.
A $100 million transaction produces $48,635,000 of investable capital. The compound rate from the headline figure to wealth is negative for the first several years and does not turn positive until well into the second decade: at five years the position is $73,679,542 against a headline of $100,000,000. That is not underperformance. It is what a 25.4% charge on the gross figure does to a starting point, and it is the single most useful thing to understand before a projection built on the headline number is taken seriously.
| No planning | Structure, 50% deferred 40% of capital inside | Same structure, nothing deferred charges only | |
|---|---|---|---|
| Tax at the event | $25,365,000 | $20,292,000 | $25,365,000 |
| Investable capital | $48,635,000 | $53,708,000 | $48,635,000 |
| Structural charges to 30 years | — | $14,143,528 | $13,201,454 |
| Wealth at 5 years | $73,679,542 | $79,469,023 | $72,941,301 |
| Wealth at 10 years | $92,733,399 | $101,035,099 | $92,523,491 |
| Wealth at 30 years | $232,531,060 | $268,822,983 | $244,468,079 |
| Difference at 30 years | — | $36,291,923 | $11,937,019 |
| Break-even year | — | Year 1 | Year 12 |
The middle column is the version usually presented: half the gain on 40% of the capital deferred, and the structure ahead from the first year. The right-hand column is the same structure with the deferral assumption removed, and it is the more instructive of the two. Without deferral the arrangement is behind by $738,241 at five years and by $209,908 at ten, and does not overtake until year twelve. Whether it is worth doing then depends entirely on how long the capital is actually going to be held — which is a question about the family rather than about the arithmetic.
A structure with heavier charges and no deferral does not break even at all within thirty years, and the instrument reports that as plainly as it reports the other two. The assumptions are the argument; the model only applies them.
Four provisions do the work in the tax layer. They are cited because the model applies them, not to characterise any particular transaction.
Subsection (a)(1) imposes a tax of 3.8 per cent on an individual, in addition to any other tax in the subtitle. Subsection (c)(1)(A)(iii) brings in net gain attributable to the disposition of property other than property held in a trade or business not described in paragraph (2); subsection (c)(2) describes a trade or business that is a passive activity within the meaning of §469, or one of trading in financial instruments or commodities. Whether a particular sale is reached therefore turns on material participation, which is why the instrument asks what share of the gain it reaches rather than assuming all or none.
Where one or more applicable partnership interests are held, the excess of the taxpayer’s net long-term capital gain with respect to those interests over the same figure computed by substituting three years for one year in §1222(3) and (4) is treated as short-term capital gain. This is why the carried-interest preset starts with part of the gain in the short-term column, and why that share is editable rather than fixed.
Property acquired from a decedent generally takes a basis equal to its fair market value at the date of death, which extinguishes the embedded gain in a taxable account. Amounts received under a life insurance contract by reason of the insured’s death are excluded from gross income. Both are applied when the horizon is measured at death, and both are applied to whichever side of the comparison they belong to.
Headed “Percentage limitations”, subsection (b) limits an individual’s deduction to a percentage of the contribution base — broadly 50, 30 or 20 per cent, depending on the recipient and on whether the gift is capital gain property. Because the applicable limit depends on facts outside the model, the deductible amount is an input rather than a calculation.
Nothing here is tax or legal advice, and none of it is jurisdiction-specific beyond the United States federal provisions cited. Rates, characterisation, the availability of any structure and the treatment of a particular transaction are questions for your own advisers, and the assumptions that decide the answer are labelled in the instrument for exactly that reason.
Bring the term sheet, the basis, the character allocation and the timetable. Read personally by a senior specialist, with 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.