Chapter 2 of 6Portfolio Construction
How Your Portfolio Should Be Put Together
How a portfolio gets built when the balance sheet includes a business, real property, and assets that cannot be sold on a Tuesday.
7 min read
The short versionIf you read nothing else on this page, read these three.
- 1When an operating business dominates the balance sheet, the allocation question inside the brokerage account is the small one. The large one is what else you own that rises and falls with that business.
- 2Sort capital by when you will need it, not by which account holds it. An illiquid asset is not risky or safe in the abstract — only appropriate or inappropriate to a given horizon.
- 3A funded cash reserve is not a drag on returns. It is what makes a long horizon real, because it is the only thing standing between a bad market and a forced sale.
Why the stock-and-bond frame breaks
A closely held business cannot be sold in pieces or priced daily, so when it dominates net worth the allocation debate is the small question.
A portfolio is not the sum of your brokerage accounts. The largest asset is often not in a portfolio at all — a business, a building, or a concentrated position in employer stock. Construction starts from the whole balance sheet, because the risks that matter are already there.
The classic frame assumes your wealth is marketable, divisible, and repriced daily. A closely held business is none of the three — you cannot sell a quarter of it on a Tuesday, nobody quotes it, and its value depends on your presence. At roughly 60% of net worth it turns the brokerage allocation debate into a rounding error.
The exposure inside the business makes it worse. A construction company owner who also holds homebuilder stocks and a rental property in the same metro does not have three positions; he has one bet placed three times. Each sits in a different statement, so nothing flags it unless someone assembles the whole picture — Chapter 1, applied to allocation.
Why owning different things matters
Diversification is about whether holdings answer to different drivers, and because correlations converge under stress, a funded reserve protects better than an estimate.
Twelve funds that own the same two hundred large American companies are one position, and you are paying twelve managers to hold it. What matters is whether your holdings answer to different causes — rent payments, interest payments, a lease renewal — which is what lets one hold up while another does not.
Correlations also move. Assets that look independent in calm markets can converge in a liquidity event, when what gets sold is chosen not by asset class but by whatever can be sold. The answer is not to abandon diversification but to stop treating a correlation estimate as a fact and fund a reserve that means you are never the forced seller.
Nobody can identify holdings that reliably zig while others zag — if that were available it would already be priced. What is achievable is not holding the same risk in four places by accident.
Sorting money by when you need it
Sorting capital by when it is needed settles most allocation arguments, because an illiquid holding is only right or wrong relative to a tier.
Money you need in two years behaves nothing like money you need in twenty, so sort the balance sheet by when money is needed rather than by account. Most allocation arguments then answer themselves — the question is no longer "how much belongs in private real estate" but "which tier does it come out of."
| Tier | What it is for | What it can hold | What it must never do |
|---|---|---|---|
| Reserve | Spending and shocks inside roughly the next two years | Cash and equivalents; short, high-quality fixed income | Depend on a reasonable market price on the day |
| Transition | Known outlays a few years out — tuition, a tax bill | Short and intermediate fixed income sized to the date | Hold a lockup that outlasts the outlay it funds |
| Growth | A horizon measured in a decade or more | Public equities, real assets, diversified private strategies | Get raided to cover a reserve-tier failure |
| Legacy | Capital you do not expect to spend in your lifetime | Long-lived private strategies, closely held interests, low-basis appreciated positions | Be sized before the tiers above it are covered |
The reserve tier is not a drag to be minimized — it is what buys the ability to hold everything else through a bad stretch, which is the only way a long horizon stays long.
Where alternatives fit, and how big
Alternatives get sized last, out of tiers whose money is not needed, and the common error is measuring a commitment against the brokerage account rather than total liquidity.
Alternatives get sized last, after the concentration on the balance sheet is counted and out of tiers whose money is not needed. Their return drivers differ from public equity beta — private credit is paid by borrowers making interest payments, real assets by tenants and by replacement cost, and neither is primarily a bet on quarterly earnings sentiment. What alternatives are covers the categories; sizing and liquidity covers what most people get wrong.
The common sizing error has a name: denominator blindness. An investor with a business, two properties, and a modest brokerage account sizes a commitment against the brokerage account — the number on the screen. Against net worth it is trivially small; against liquid assets it is enormous — and only the second decides whether a capital call in a bad year is an inconvenience or a crisis.
What has to be true first
The preconditions for an illiquid commitment are not about the investment at all — they are about whether your balance sheet can hold it when you cannot sell.
Before an illiquid commitment is on the table, a short list of conditions has to hold. None are exotic, and enthusiasm for an opportunity makes them feel negotiable.
- The reserve tier is funded and sized in years of spending, not a dollar figure that felt comfortable.
- No high-rate debt is sitting unpaid — paying it off is a known, certain outcome, and certain outcomes are scarce.
- The tax picture for this year and next is understood, so new income or a new state filing is not a surprise.
- You know how capital calls work in the structure, and where the cash for the later ones comes from.
- You could hold the position through a stretch where you cannot sell it or see a price — psychological as much as financial.
- The commitment is spread across more than one vintage year, so the outcome is not decided by the month you signed.
The quality of an opportunity is a separate question from whether your balance sheet can hold it — the opportunity carries a deadline and the balance sheet does not. A good investment in the wrong tier becomes a problem the moment you need money that is not there, so answer the balance sheet question away from any particular fund.
Rebalancing around what you cannot sell
When the largest asset cannot be traded you rebalance around it, building the liquid side to be unlike the business and redirecting its distributions for years.
When the largest asset cannot be traded, you rebalance around it rather than through it. The liquid side becomes the ballast — underweight whatever the business is exposed to, and holding what the business would suffer alongside. Less satisfying than trimming the concentration directly, but for an owner who is not selling it is the answer available.
The other mechanism is cash flow. A business that distributes cash funds the diversification of the balance sheet that owns it, without a taxable sale of the asset. For most owners the path out of concentration is not a transaction but a decade of directing distributions into what the business is not. A sale inverts the problem: what to do with the proceeds.
Rebalancing a taxable account creates a tax event, so sequencing matters as much as the target — which is why it belongs in the same annual conversation as harvesting and conversions, not a quarterly ritual. And unwinding a concentrated low-basis position costs enough that the plan is measured in tax years, not trades.
Common questions
- Should my business count as part of my portfolio?
- It should count as part of your balance sheet, which is the number that matters for sizing everything else. Treat it as a concentrated, illiquid holding with a specific set of economic exposures, and then build the liquid side to be genuinely different from it. Leaving it out of the picture is how people end up with three copies of the same bet.
- How do you decide how large an alternatives sleeve should be?
- It falls out of the liquidity tiering rather than from a target percentage. Once the reserve and transition tiers are funded and the concentration already on the balance sheet is measured, the capital genuinely available for a long lockup is usually a smaller number than people expect. We size to that, not to a model allocation.
- What if most of my wealth is in low-basis stock I cannot sell without a large tax bill?
- Then the plan is measured in tax years rather than trades. Staged sales across years, charitable structures for the most appreciated shares, and careful attention to what the rest of the portfolio holds are the usual tools, and the right combination depends on your bracket, your state, and your estate plan. That is a conversation to have with your CPA in the same room, not sequentially.
- Does adding alternatives mean selling my stocks and bonds?
- Not usually, and reflexively selling public holdings to fund a private commitment can create a tax bill that swallows the benefit. More often the funding comes from cash flow, from a concentrated position being unwound for other reasons, or from new capital after a liquidity event. The source of the funding is part of the decision, not an afterthought.
7 min for the whole chapter · 6 sections
Want to know whether this applies to you?
Reading about a strategy and knowing whether it fits your situation are two different things. Tell us what you are working with and we will be straight with you about whether this is the right tool.