Family Office  ·  Lemma 322 Holdings

Technology changes.
Human nature does not.

We invest in what has already been proven.

AUM 721.43M HKD · Central, Hong Kong
Philosophy

We build on what is already proven.

Forecasting which technology wins is speculation. Knowing what people will still need in fifty years is analysis. We study human behavior — psychologically, philosophically — and position capital early where that behavior makes demand permanent.

lemma/ˈlɛmə/ · noun

In mathematics, a proposition proven first, so larger truths can be built on it. We invest only on top of our lemmas — premises that hold in every version of the future.

  • Lemma IDistance will always demand to be compressed.
  • Lemma IICivilization runs on energy, and its appetite only compounds.
  • Lemma IIIEvery life requires ground to stand on.
  • Lemma IVAll that is produced leaves a residue to be managed.
Investments

Four sectors. Four constants.

Infrastructure

Infrastructure

Distance will always demand to be compressed.

Transport, logistics, and the networks that carry people and goods. Vehicles change; the demand for movement never has.

City infrastructure at night
Energy & Utilities

Energy & Utilities

Civilization runs on energy, and its appetite only compounds.

Electricity, gas, and the utilities beneath daily life. The source of energy is a variable; the consumption of it is a constant.

Energy generation at dusk
Real Estate

Real Estate

Every life requires ground to stand on.

Property in cities where people will still want to be decades from now. Shelter is the oldest asset class in existence, and the least likely to become obsolete.

Architecture
Circular Economy

Circular Economy

All that is produced leaves a residue to be managed.

Consumption always leaves residue. We invest in the waste management, recycling, and resource-recovery systems that turn that residue back into value.

Natural cycle
Principles

Seven working rules.

Essay

On Proof — The Philosophy of Lemma 322

Read the full essay

I. The Name

In mathematics, a lemma is a proposition that has been proven — not for its own sake, but as a stepping stone toward something larger. The word descends from the Greek lēmma: "that which is taken." A lemma is what you are permitted to take with you into the next argument, because the work of establishing it is already done.

The history of mathematics is full of lemmas that outgrew the theorems they were built to serve. Itô's Lemma, written to tame the calculus of random motion, became the foundation of modern finance. The Yoneda Lemma, a technical stepping stone in category theory, revealed that an object is entirely determined by its relations to everything else. Zorn's Lemma quietly holds up half of modern analysis. The label says "auxiliary." The record says otherwise.

We chose this name because it is a claim about method. We do not begin from opinion, narrative, or the temperature of the crowd. We begin from what has already been proven, and we build only on that.

II. What We Take as Given

Every deductive system rests on premises. Ours are four, and we did not invent them — we merely wrote them down.

People will demand that distance be compressed. Civilization will run on energy, and its appetite will compound. Every life will require ground to stand on. And everything produced will leave a residue that must be managed.

These are not forecasts. A forecast is a conjecture about the future; these are records of the past — ten thousand years of roads, granaries, hearths, cities, and middens, written into the archaeological and economic record without a single interruption. Empires, currencies, and technologies have replaced one another many times over. The four demands have never once lapsed. When a proposition has survived every regime change in recorded history, we stop calling it a prediction and start calling it a premise.

III. The Discipline of Admission

Here we must be honest about a distinction that finance often blurs. Markets are not deductive systems. In mathematics, a lemma once proven is true forever. In markets, most "proven" regularities are statistical patterns wearing the costume of law — and patterns break. The graveyard of quantitative finance is filled with firms that mistook a decade of backtests for a theorem.

So the central discipline of this firm is not analytical brilliance. It is severity at the point of admission: deciding what may be called a lemma at all. A pattern observed over one cycle is an anecdote. A pattern observed over one century is a hypothesis. We admit as lemmas only regularities proven at the scale of civilization itself — behaviors rooted not in market structure, which changes, but in human nature, which does not.

From these admitted lemmas, every investment thesis is derived — not imagined, derived — and the chain of assumptions between premise and conclusion is kept as short as we can make it. Every additional assumption is another link that can fail. The shorter the chain, the less there is to break.

IV. Proof of Loss

A proposition that cannot in principle be refuted is not knowledge; it is faith. We therefore write the refutation before we write the investment. For every thesis, the conditions under which it would be wrong are stated in advance, in writing, while we are still calm.

This produces an unusual accounting. In our ledger, losses may take only two forms. The first is a falsification: a pre-stated refutation condition was triggered, the position was closed, and we learned — before the rest of the market — that one of our premises had failed. That is not a cost. That is the price of knowledge, and it is cheap at almost any size. The second is an error of execution, which is identified, attributed, and corrected.

What we do not permit is the third kind: the loss that cannot be explained. An unexplained profit is luck, and an unexplained loss is failure, and a firm that tolerates either has abandoned proof for superstition. We intend to prove our losses as rigorously as our gains. Over a long enough horizon, that discipline — not any single insight — is the asset.

V. The Corollary Harvest

Mathematics has a beautiful economy: when a theorem finally falls, its corollaries follow at almost no additional cost. The hard work is done once; the consequences arrive nearly free.

Investment has the same structure, though few firms are organized to exploit it. When we complete a thesis — when capital committed against a derived conclusion proves out in reality — the adjacent opportunities change state. What remains conjecture to the market is, to us, already half-proven: the land beside the substation, the logistics corridor implied by the port, the recovery stream implied by the waste. Others must begin their proof from the beginning. We need only write down the corollary.

One completed proof, harvested many times. This is where the mathematics stops being metaphor and becomes margin.

VI. The Same Nature, Twice

There is a symmetry at the center of our method. The four premises we invest behind are facts of human nature. But so is the mispricing we invest through. The same species that reliably demands movement, energy, shelter, and disposal also reliably panics, crowds, extrapolates, and forgets. Human constants create the demand; human constants create the discount.

Our resolution is simple to state and difficult to live: we buy human nature, and we refuse to trade with it. Every impulse we underwrite in the world — fear, urgency, appetite — is an impulse we work to remove from our own process. The systematic internal machinery, the written refutations, the derived theses: these are not bureaucracy. They are the mechanism by which we remain the one participant in the market whose behavior is not the product being sold.

VII. The Eminent Dead

Nothing in this document is self-made. The instruments of modern investment rest on lemmas we did not prove: Bachelier's random walk, Kolmogorov's axioms of probability, Itô's calculus, Markowitz's frontier, and behind them Euclid, Gauss, and the long line of minds who did the work centuries before there was any capital to compound on it. Charlie Munger advised going through life "making friends with the eminent dead who had the right ideas." We take the advice literally. Their proofs are our working capital, granted to us without interest, and a firm that forgets this has misunderstood its own balance sheet.

Gratitude of this kind is not sentiment. It is an accurate accounting of where our edge comes from — and it carries an obligation.

VIII. Becoming a Lemma

Euclid's Elements has been in continuous use for twenty-three centuries. It survived not because its theorems were the deepest ever found, but because they were recorded — stated, proven, and written down in a form the next generation could build on without repeating the work.

Capital compounds. But so does recorded proof, and over generations it compounds faster. Knowledge, capital, and experience are the three assets a firm can hold, and of the three, only proof committed to the page survives the people who produced it. So we keep the ledger the way a mathematician keeps a notebook: every thesis stated, every refutation condition written in advance, every outcome — profit or loss — proven and preserved.

Our debt to the eminent dead can be repaid in only one currency. Somewhere past the horizon of our own careers, an investor not yet born should be able to take what we have recorded and build on it without repeating the work — to treat this firm's record as we treat Itô's result: as something already established, something that may simply be taken.

That is the ambition contained in our name. Not to be remembered — to be used.

“If you go through life making friends with the eminent dead who had the right ideas, I think it will work better for you in life and work better in education.”
Charlie Munger
Strategy

How we operate.

Proprietary capital only

A single family office investing its own capital exclusively. We do not solicit, accept, or manage external capital.

Conviction over consensus

When a lemma is proven, we concentrate capital behind it — and hold across decades, not quarters.

First principles, first positions

Analysis of what people will still do, then early positioning in the assets that serve it. We buy the constant, not the trend.