Circle of Competence: How to Invest in What You Actually Know

Volatility is not the risk that actually blows up portfolios. Ignorance is.

Most new investors are taught to fear the chart that drops 20% in a month, the red day, the VIX on the evening news. A diversified index fund can fall 40% and, over a long enough horizon, recover — not because every constituent survives, but because the losers are replaced, the survivors keep earning, and the index keeps compounding while you wait. A single stock you bought because a colleague mentioned it, without understanding how the company makes money, what could break that model, or why the market price might already assume a perfect outcome, does not recover on its own. It recovers only if you got lucky, or it does not recover at all.

That distinction is the job of this piece.

This is the fourth article in the Investing 101 sequence we call The Foundations of Consistent Investing. The first three answered when you buy, what it costs to get in and out, and what size of company you are buying. This one is The Mindset: whether you should be buying the thing at all.

Warren Buffett and Charlie Munger built an investment philosophy around a deceptively simple filter for the ignorance problem: the Circle of Competence. Draw an honest boundary around the businesses and industries you can actually judge — and stay inside it. Everything outside the circle is not a bad investment by definition. It is an investment you are not equipped to judge, which is a different and more dangerous problem.

Buffett put the point cleanly in Berkshire Hathaway’s 1996 shareholder letter: you do not have to be an expert on every company, or even many. You only have to be able to evaluate companies within your circle of competence. The size of that circle is not very important. Knowing its boundaries is vital.

This is not a call to diversify less. A narrow circle plus a diversified core is a coherent plan. A narrow circle used as an excuse to own three familiar stocks and call it a philosophy is not. The circle does not replace dollar-cost averaging, liquidity awareness, or size-based risk sorting. It sits underneath all of them, as the filter that decides whether a name earns a place in the plan in the first place.

Two Doors: The Index Core and Everything You Add

The first three pieces were written for a reader building a plan — often a regular contribution into liquid, size-sorted instruments. This piece does not quietly replace that reader with a stock-picker. It splits the problem in two.

Door 1 — The index core. If the holding is a broad, diversified equity index fund or ETF — an S&P 500 vehicle, a developed-world fund, an ASX 200 fund — the competence test is narrower, and it should stay narrower. You need to understand that you own a slice of a market, that you will hold the losers with the winners, that the fee and the currency exposure are part of the cost, and that you are not required to write a research memo on every constituent. That is a circle. It is the circle Buffett has recommended for people who do not want to evaluate individual businesses. Do not pretend VOO or VGS requires a three-pillar paragraph on each holding. That is how a useful filter turns into homework theatre.

Door 2 — Everything you add on top. The filter switches on the moment a single stock, a concentrated sector fund, a thematic ETF, a covered-call product, or a “story” name is asked to sit beside that core. “I use the product,” “I heard it on a podcast,” and “the theme is obvious” are not substitutes for understanding the engine. If you cannot explain the payout machine, you do not own an investment. You own a narrative with a ticker.

Most damage happens at Door 2 while the holder still describes the whole portfolio as “just DCA.”

Defining the Circle

The single most common mistake is confusing a consumer relationship with business competence.

You can love your iPhone, use it twelve hours a day, and recommend it to everyone you know — and still have no real basis for judging Apple as an investment. Being a happy customer tells you the product works. It tells you nothing about services-margin trajectory, how App Store regulation in the EU and elsewhere is reshaping that revenue line, what on-device AI does to the replacement cycle, or how much of the current share price already assumes all of that goes Apple’s way. Knowing you like a product is not knowing the business behind it.

The same trap wears different clothes in 2026. Using ChatGPT does not mean you can underwrite Nvidia’s customer concentration, Palantir’s deployment cycle, or the difference between a picks-and-shovels semiconductor name and an unprofitable software wrapper with “AI” in the deck. Familiarity with the interface is still just familiarity with the interface.

A genuine circle of competence rests on three pillars. All three, not one or two. A business you understand on one axis and not the others is still outside the circle.

1. Operational understanding. Can you explain, in one paragraph and without marketing language, how the company actually makes what it sells and what physically drives its unit economics? Not “they’re a leader in cloud software” — what specifically do they build, and what happens operationally when demand doubles or halves?

2. Monetisation mechanics. How does revenue actually turn into profit? A subscription, an advertising business, a bank’s net interest margin, a royalty stream, and a one-off hardware sale all carry different risk profiles even if the top-line number looks the same. If you cannot say which lever the company pulls to make money, you cannot judge what breaks that lever.

3. Competitive moats. Why can’t a well-funded competitor copy this in eighteen months? Network effects, switching costs, a regulatory licence, brand, patents, or genuine scale economics are moats. “They were first” and “everyone’s heard of them” usually aren’t.

The price rider. Understanding a business is not the same as the current price being sane. Buffett’s circle was never “I understand See’s Candies.” It was “I understand See’s Candies and I can tell whether I am paying a reasonable price for it.” You do not need a spreadsheet. You do need one extra sentence: what you are paying for, and what would make that price a mistake. A great company bought at a price that already assumes a perfect decade is how “circle of competence” becomes a slogan for concentrating in familiar names at any multiple.

The internalisation test. After you have done the work — filings, results, the boring pages — close the document and write two or three honest sentences on each pillar, plus the price rider. If you still need the annual report open to remember how the company makes money, you do not own the understanding yet. Do the work first. Then close the tab and write. Research done after the purchase does not move the tier. The tier reflects what you understood before you owned it.

A circle without a kill-switch is a fan club. Before a name is allowed into Tier 1, write the thing that would prove you wrong.

What the Paragraph Actually Looks Like

A rubric without an example trains nobody. These are illustrations of the format, not recommendations, valuations, or invitations to buy anything.

Illustration A — engine visible (the shape of Tier 1 work).
Take a membership retailer of the Costco type, which this publication has already unpacked as a model.

• Operations: a limited-SKU warehouse that turns inventory fast and uses low prices to keep members walking through the door.
• Monetisation: the merchandise box runs at a thin margin; a large share of profit is the membership fee paid before the first trolley is filled; high renewal rates keep that fee base in place.
• Moat: the fee-funded price advantage is a flywheel that is hard to copy without eating years of margin.
• Price rider: a high multiple is not disqualifying by itself, but the thesis has to survive a slower membership add or a year of ordinary retail luck.
• Kill-switch: renewal rates break, or the merchandise box stops being a customer-acquisition engine and starts having to carry the profit.

If you can write that cold, you are doing the work. If you can only write “it’s a great shop and people love the hot dog,” you are still a customer.

Illustration B — model clear, visibility not (the shape of Tier 2).
Take a card network of the Visa type, which this publication has already unpacked as a toll road.

• Operations: it does not lend and it does not hold the deposit. It is pipes. Qualifying card transactions route through the network.
• Monetisation: a small take-rate on volume. Cross-border spend and value-added services usually pay more than a domestic swipe.
• Moat: a two-sided network of issuers and merchants that is slow and expensive to rebuild.
• Price rider: a quality compounder can still be priced for a decade of uninterrupted volume growth. Understanding the toll is not the same as the multiple being sane.
• Kill-switch: regulation or routing rules compress the take-rate, or the mix of volume shifts permanently away from the high-toll rails.

The mechanics are teachable. The five-year path may not be, because regulation, competition for routing, and the consumer-versus-commercial mix can move earnings power without breaking the idea of the business. That is honest ground. It is not ignorance. It is why the name, if it belongs at all, belongs smaller than a business whose next five years you can actually underwrite.

Illustration C — black box, local version (Tier 3).
“Pre-revenue explorer, one tenement, share price lives on the drilling update, and the company raises capital to keep exploring.”

There is no three-pillar paragraph here. There is a story about a place and a commodity. Stories can be sized as speculation. They do not get to borrow the language of investing.

Illustration D — black box, imported version (Tier 3).
“AI name everyone at the barbecue is discussing. I cannot explain the customer, the contract, or the margin without opening someone else’s thread.”

Same failure, different jacket. Using the consumer app, or watching the sector dominate an index, does not write the paragraph.

Thematic and packaged products belong in this test too. A semiconductor ETF, a uranium fund, a covered-call income vehicle, or a single-country thematic is not “diversification” if you cannot explain the payout engine and the concentration inside the wrapper. If the product is a black box, the wrapper does not make it Tier 1. It makes it Tier 3 with better branding.

The Three-Tier Sorting Filter

Before any purchase that sits outside the broad index core, sort the ticker honestly — not optimistically.

This mapping is a working rule of thumb, consistent with the core / satellite / speculative bands laid out in How to Size a Position. It is a ceiling-setting device, not a requirement to fill any sleeve, and not a recommendation to own any name.

Run the internalisation test from the previous section before the order. If the paragraph will not write without external help in the room, the ticker defaults to Tier 3. A rising price does not promote it.

Competence is personal. A geologist’s Tier 1 explorer can be a teacher’s Tier 3. A banker can underwrite net interest margin and still have no business in a clinical-stage biotech. “Outside my circle” is not a verdict on the company. It is a verdict on the match between the company and you.

The Aussie Angle: The Resource Trap, the Imported Story, and the Boring Oligopoly

The filter works hardest at the intersection of the ASX and the US names Australians now buy as a matter of course.

The resource trap

Australia has an unusually large population of speculative micro-cap mining and energy explorers relative to other developed markets — a consequence of a listing regime that makes it comparatively straightforward to take a pre-revenue exploration company public. Many of these names are genuinely speculative: no JORC-compliant resource yet defined, one drill rig, a share price that moves on the quarterly announcement rather than on cash flow, and a structural need to keep raising capital — diluting existing holders — simply to keep exploring.

That is the Aussie Resource Trap: a Tier 3 holding that feels like Tier 1, because the investor recognises the commodity, the region, or even the town the tenement sits in.

Recognising a place is not the same as understanding drilling economics, resource-conversion probabilities, or how placements at this end of the market actually dilute existing shareholders. Geographic familiarity is not circle-of-competence. It is proximity dressed up as insight, and it is how “I know this area” becomes the justification for a Tier 3 bet sized like a Tier 1 one.

A JORC-compliant figure is a reporting standard. It is not a valuation, a timetable, or a promise that the next raise will be the last one.

The imported-narrative trap

The same mistake now walks in through a US ticker.

If you are an Australian buying US names, proximity to an American story is not competence any more than proximity to a Pilbara tenement is. Hearing the pitch on a podcast, using the consumer app, or watching a semiconductor name dominate the index does not put you inside the circle. It puts you next to the conversation.

The live version of the iPhone error is the AI stack. The live version of “I know the town” is “I know the narrative.” Sizing a US mid-cap you met on YouTube like a core holding in a major bank is the resource trap in a different jacket. If the paragraph will not write, the geography of the exchange does not save it.

The boring oligopoly — with a hard limit

The contrast sits one shelf up, in the local banks, grocers, and large diversified miners most Australians already bump into. The big four operate inside a regulatory and deposit-funding structure that is difficult for a new entrant to replicate at scale. The major grocery retailers run a duopoly most households navigate weekly. A diversified miner with decades of production and transparent reserve reporting is a different research object to a single-tenement explorer, even when a screener files both under “resources.”

None of that makes these companies risk-free, cheap, or a recommendation. It means the on-ramp to the three pillars is shorter, because the business model is one you can start describing from ordinary life rather than from first-year geology.

It does not complete the pillars. Having a mortgage at a big four bank is not understanding net interest margin, credit concentration, or a loan book tied to Australian house prices. Shopping at a major grocer is not understanding shelf-price architecture, supplier power, or what a mid-single-digit margin does in a volume war. Everyday exposure gets you to the door. It does not write the memo, and it does not make any price acceptable.

Use the shorter on-ramp. Do not confuse it with a finished judgement.

The Danger of Circle Creep

The circle is not static, and the way it usually expands is not through careful study. It is through FOMO.

The pattern is familiar. An investor genuinely understands banking, builds a sound core position in a major bank, and feels — reasonably — confident in that judgement. Then a speculative biotech, an unprofitable EV name, or the AI stock everyone at a barbecue is discussing starts running, and the confidence earned in one circle gets silently spent in a completely different one. “I’m a good investor, so this is probably fine too” is not analysis. It is borrowed confidence, and it is how a well-researched portfolio ends up with a Tier 3 position sized like a Tier 1 one.

Outside your circle of competence, you are not investing. You are speculating. That is not a moral judgement. Speculation has a legitimate place in a portfolio, sized deliberately, as capital you can lose completely without it changing your life. The danger is not speculating. The danger is speculating while telling yourself you are investing, because that is what produces position sizes that assume Tier 1 safety on a Tier 3 bet.

A rising share price is not evidence that your circle has expanded. It is evidence that the price went up. Conflating those facts — “I’ve made money on this, so I must understand it” — is how circle creep compounds across a bull market until a single bad quarter in an unfamiliar business does disproportionate damage to a portfolio that looked diversified on paper.

How the Circle Is Allowed to Grow

A strict boundary is not an argument for staying in the same two stocks until retirement. Competence can be earned. It cannot be borrowed.

Earned expansion looks like this:

• Time in the industry, or a long stretch of primary documents, before capital is committed.
• A written three-pillar memo that still makes sense when you reread it six months later.
• A paper portfolio, or a deliberately tiny speculative sleeve, used as tuition rather than as a back door into the core.
• A kill-switch you would actually honour.

Borrowed expansion looks like this:

• A friend’s circle treated as yours.
• An analyst note you could not recreate with the filing closed.
• A thread, a barbecue, a price chart.
• “I’ll learn it once I own it.”

If writing the paragraph is harder than telling the story of why the stock is going up, the circle has not moved. Only the price has.

⚠️ Speculation & Risk Warning

The Circle of Competence is a qualitative filter, not a guarantee. Understanding a business well does not protect you from being wrong about it, from a genuinely unforeseeable industry shock, or from paying too high a price for a great company. What it protects you from is a narrower and more common failure: holding a meaningful position in something you never actually understood, and discovering that only after it has gone materially against you.

Position-size anything outside your circle as money you can afford to lose in full. Index recovery is not a property of individual names. Diversification across three stories you cannot explain is not diversification.

Putting It Into the Plan

The circle does not replace the first three pieces. It decides which tickers are allowed to use that machinery.

Before a position gets a DCA schedule, before you think about its bid-ask spread, before you classify it by market cap and give it a role as core, satellite, or speculative, run this list:

1. Mindset. Door 1 or Door 2? If Door 2, the three-pillar paragraph exists, written cold, with a price rider and a kill-switch.
2. Tier. Sort it as 1, 2, or 3. Size it to the band that tier is allowed to occupy — or do not buy it.
3. Metric. The market-cap role matches the tier. A Tier 3 micro-cap does not get to sit in the core sleeve because the story is vivid.
4. Mechanic + Strategy. The vehicle is liquid enough for the contribution schedule you already chose. A correctly understood illiquid name can still be a bad input to a fortnightly DCA plan.

The circle does not need to be large to work. Buffett’s stayed narrow for decades; the useful lesson is not “be Buffett,” it is that knowing the boundary matters more than widening it. Draw it around what you actually understand today — not what you are confident you could learn, not what a friend understands on your behalf, and not what a rising chart is currently persuading you that you understand.

Everything genuinely useful in investing happens inside that boundary. Almost everything that quietly destroys a portfolio happens just outside it, dressed up as a great opportunity.

Data Sources:

• Warren E. Buffett, Berkshire Hathaway Inc. Chairman’s Letter to Shareholders, 1996 — circle of competence, the primacy of knowing the boundary rather than widening it, and the two courses Buffett said an investment student actually needs: how to value a business, and how to think about market prices. Letters archive: berkshirehathaway.com/letters/letters.html
• Berkshire Hathaway Owner’s Manual / subsequent shareholder letters — restatement of the same qualitative filter across multiple years; standard primary source for the Buffett–Munger framing used here
• JORC Code, 2012 Edition (current mandatory edition), Australasian Joint Ore Reserves Committee — minimum standard for public reporting of Exploration Results, Mineral Resources and Ore Reserves on the ASX. A draft update was released for consultation in 2024; the 2012 Code remains the operative standard until a replacement is adopted. jorc.org
• ASX Listing Rules and Guidance Note 1 (Admission to the Official List), including the asset-test pathway commonly used by pre-revenue mining, energy, and early-stage issuers, plus free-float and spread requirements. asx.com.au
• ASX Listing Rule 7.1 and related guidance — placement capacity and the dilution mechanics that sit behind repeated capital raises at the explorer end of the market
• ASIC Moneysmart, Check before you invest — including the plain-language test that you should be able to explain how the investment makes money and where the returns come from. moneysmart.gov.au/check-and-report-scams/check-before-you-invest
• ASIC Moneysmart, How to buy and sell shares and Choose your investments — general investor guidance on direct shares versus funds and ETFs, risk, and time horizon. moneysmart.gov.au/shares/how-to-buy-and-sell-shares · moneysmart.gov.au/how-to-invest/choose-your-investments
• S&P/ASX index composition, ASX 20 through Small Ordinaries — see Market Cap Is a Risk Sorting Tool for the full breakdown used in this series
• How to Size a Position: A Practical Guide for Aussie Investors in US Markets — core / satellite / speculative bands that the three-tier filter in this piece is designed to feed
• Prior WSDU model write-ups used here only as format illustrations, not as current recommendations: Visa — The Toll Road, Costco — The Warehouse That Prints Money Before It Sells a Single Item, PLTR Deep Dive

Wall St. Down Under | Australia

Subscribe | wallstdownunder.com.au

Disclaimer: Wall St. Down Under is an independent financial newsletter for informational and educational purposes only. Nothing published here constitutes financial advice, a recommendation to buy or sell any security, or a solicitation of any investment decision. Always do your own research and consult a licensed financial adviser before making investment decisions. Australian investors should consider their own financial situation, objectives, and risk tolerance. Past performance is not indicative of future results.