Portfolio Diversification: The Only Free Lunch in Investing

Look at September 2026. In the same country, in the same 30 days, the Dow Jones Industrial Average fell 4.3%, the S&P 500 slipped about half a per cent and the Nasdaq Composite rose 1.9%. The headline index hid a wider split. The equal-weighted S&P 500, which gives each company the same vote, fell about 5%. Most stocks finished the month lower. Technology carried the cap-weighted index.

Over the same stretch the US 10-year Treasury yield climbed about 54 basis points and finished the month near 5.3%. On simple duration arithmetic, that meant a price fall of around 4% for a ten-year bond (WSDU estimate). Spot gold fell about 6–8%, depending on the fix.

If your whole portfolio was one index fund, you got one answer. If it was one stock, you got a different one, and nobody could have told you in advance which. That is the problem this piece is about.

This is the first article in a new Investing 101 run on portfolio structure. The earlier pieces dealt with single holdings: when to buy, what it costs to get in and out, what size of company, and whether you understand what you are buying at all.

The question now is no longer “is this a good stock?” It is “what does the whole thing behave like?”

What the “Free Lunch” Actually Means

The phrase is usually attributed to Harry Markowitz, whose 1952 paper on portfolio selection earned him a share of the 1990 Nobel Prize in economics. The claim is narrow and mathematical. If you combine assets that do not move in lockstep, the risk of the combination is lower than the average risk of the parts, while the expected return is simply the average of the parts’ expected returns.

You give up nothing in expected return and you get less variability. That is as close to a free lunch as finance offers. It is not free of cost: fees, tax, complexity and a smaller chance of a spectacular one-stock win are the bill. And it is not a guarantee against loss. It is a way of being wrong about one thing without being wrong about everything.

Two Kinds of Risk, Only One of Which You Can Remove

Unsystematic risk, also called idiosyncratic risk, belongs to a single company or industry. A profit warning, a failed drug trial, a chief executive’s sudden exit. Nike fell sharply in after-hours trading on 1 October after a revenue miss and an announcement of job cuts. A 20-stock portfolio absorbs that as a small bruise. A one-stock portfolio wears it in full. This is the risk diversification is designed to eliminate. The market does not pay you, in expected return, for carrying it, because you could have avoided it without giving up expected return.

Systematic risk, or market risk, hits everything at once: interest rates, oil, recession, a global bond sell-off. Adding more stocks from the same market does not remove it. It is the reason a diversified portfolio still fell in 2022 and still wobbles on a day like 1 October.

How quickly does diversification remove the first kind? The table is a simple model, not a forecast. It assumes every stock has 30% annual volatility and an average pairwise correlation of 0.3, with equal weights.

The pattern matters more than the exact numbers. Most of the benefit arrives with the first 10–20 holdings, and the curve is nearly flat after 30. The 16.4% floor is systematic risk: the part you cannot diversify away within one market. Real correlations are not a constant 0.3. They are higher inside a concentrated index or a single sector, which is why twenty banks diversify far less than twenty companies from twenty industries.

Correlation Is the Engine

Correlation measures how closely two assets move together, from +1 (perfect lockstep) to −1 (perfect opposites). Zero means no relationship. The lower it is between the things you own, the more diversification you capture.

A simple case: two assets, each with 20% volatility and an expected return of 8%, held 50/50.

The expected return never moves. The risk falls with every step down in correlation. That column of constant 8% is the free lunch. The practical question is not how many assets you own, but how different their reasons for moving are.

Where the Lunch Stops Being Free

Correlations rise when it matters. Assets that look independent in calm markets can move together in a panic or in an inflation shock. 2022 was the textbook case: stocks and bonds fell together. September 2026 repeated a version of it. Bond prices fell, gold fell, and the one thing that helped inside US equities was owning the right slice of technology. A diversified portfolio is built for the average month, not guaranteed to work in the worst one.

Concentration hides inside “diversified” funds. A fund with 500 holdings is not necessarily a 500-way bet. At 30 June 2026 the ten largest holdings in an S&P 500 index fund made up about 38% of the fund, and Nvidia alone was about 7.5%. September’s split between the cap-weighted index and the equal-weighted index was the same fact, shown in prices. The ASX 200 has the same issue in a different place. At 30 June 2026, Financials were 33.3% of the index and Materials 25.5%, so two sectors accounted for roughly 59%. Neither is a reason to avoid index funds. It is a reason to know what you actually own, which is the lesson of the recent Investing 202 piece on how index construction works.

Overlap disguised as variety. Three funds can behave like one. An S&P 500 fund, a Nasdaq-100 fund and a semiconductor fund all hold Nvidia, Broadcom and Micron. Counting funds is not counting exposures. The test is not how many tickers are in your account, but how many different things drive them.

Cost and clutter. Every additional holding adds a bid–ask spread, a fee, a trade to record and, for Australians, a capital gains calculation in Australian dollars at the exchange rate on each trade date. Over-diversification — Peter Lynch called it “diworsification” — buys a tiny reduction in risk and a large increase in paperwork.

It does not protect against market risk. A diversified portfolio of shares will still fall when shares fall. Diversification across asset classes may cushion that. It cannot guarantee it.

The Aussie Angle: Your Portfolio Is Bigger Than Your Brokerage Account

Diversification for an Australian investor has several dimensions that textbooks written for Americans never mention.

Your biggest asset is probably not on your brokerage screen. For most Australian households it is the family home. Add your job and your superannuation, and the typical household balance sheet is already heavily exposed to one economy, one housing market and one central bank. The RBA has lifted the cash rate four times this year, by 100 basis points in total, to 4.60% after the 29 September decision. A mortgage holder, a bank shareholder and a person whose employer depends on consumer spending are feeling the same decision. Adding more ASX bank shares to that picture does not add much diversification.

Home bias has a tax reason, and it is not irrational. Franked dividends from Australian companies are worth more to Australian taxpayers than the headline yield suggests, which is a legitimate reason to hold local shares. The honest reading is that diversifying offshore costs something. US dividends carry no franking credits, and US withholding tax generally applies at 15% with a W-8BEN form on file. On a US index fund yielding about 1.1%, that withholding is roughly 0.17% of the portfolio a year before any offset (WSDU calculation). For many Australian taxpayers the foreign income tax offset credits that 15% against Australian tax, so the permanent cost is often nil and the real friction is cash timing and paperwork. Either way it is small relative to the risk reduction. It is still a cost, and it is why “free lunch” always comes with a footnote.

Currency is a diversifier, not just a risk. Unhedged offshore shares come with an Australian dollar exposure attached. Whether that exposure is hedged is its own decision, not a detail. Saturday’s Market Moves gave a live example: over three weeks the S&P 500 rose about 0.9% in US dollars, while a fall in the Australian dollar from about 0.717 to 0.6933 lifted the return for an unhedged Australian holder to roughly 4.3% (WSDU calculation, spot-to-spot, ignoring dividends and costs). The Australian dollar has historically tended to weaken in risk-off periods, which cushions an unhedged holding. It is a tendency, not a contract. In the 2001–07 period it ran the other way for years.

Check what you already own. Most default superannuation options are already diversified across Australian shares, international shares, property, infrastructure, bonds and cash, and many already choose a hedge ratio on the offshore equity. Before you add anything on top, work out what your super fund holds, and whether the international shares are hedged. A self-managed fund is a different matter: the diversification is whatever you build, and a single concentrated position is easy to create by accident.

The Four-Question Diversification Audit

A framework you can reuse once a year, or whenever you add a new holding.

Then size the result against the core, satellite and speculative bands in How to Size a Position. Diversification decides what you own and how the pieces relate. Position sizing decides how much of each.

⚠️ Diversification and Risk Warning

Diversification reduces the risk that any single holding hurts you badly. It does not eliminate market risk, it does not prevent loss in a broad downturn, and it can reduce the upside from a concentrated winner. Correlations are unstable and can rise sharply in stress. The volatility and correlation figures in this piece are illustrative models, not forecasts. Past relationships between asset classes may not repeat.

Putting It Into the Plan

Run the audit before adding. Do the four questions on what you already hold. The most diversifying thing you can buy may be nothing.

Count drivers, not holdings. Three funds with the same top ten is one bet.

Know your floor. Decide how much market risk you can live with, since diversification will never remove it.

Include the assets you do not see. Your home, your job and your super belong in the picture.

Treat it as ongoing. Weights drift as prices move. A winner quietly becomes a larger share of the portfolio, which is its own form of concentration.

The free lunch is real, but it is small, it is conditional and it requires attention. The investor who understands that will build a sturdier portfolio than the one who treats it as a slogan.

Data Sources:

• Markowitz, H., “Portfolio Selection,” The Journal of Finance, 1952; 1990 Nobel Memorial Prize in Economic Sciences (shared).
• September 2026 index moves: Reuters, 30 September 2026 (Dow −4.29%, S&P 500 −0.45%, Nasdaq Composite +1.86%). Equal-weighted S&P 500: NISA Investment Advisors, September 2026 overview (−4.8%).
• US 10-year yield: about +54 basis points in September, finishing near 5.29% (StreetStats, 30 September 2026). Bond price change is a WSDU duration estimate.
• Gold: NISA, September 2026 (−6.3%); BullionVault, 30 September 2026 (spot −8.5%). The piece uses the 6–8% range.
• S&P 500 concentration: Vanguard S&P 500 ETF fact sheet / N-PORT, period ended 30 June 2026 (top 10 about 37.9%, Nvidia about 7.5%).
• ASX 200 sector weights: State Street Global Advisors, as of 30 June 2026 (Financials 33.33%, Materials 25.54%).
• RBA cash rate: Monetary Policy Decision, 29 September 2026 (4.60%; fourth hike of 2026, +100 basis points year to date).
• Nike: after-hours and subsequent coverage of the FY2027 Q1 revenue miss and job cuts, 1–2 October 2026 (Yahoo Finance, Reuters).
• Currency example: WSDU Market Moves, week ended 2 October 2026.
• Withholding: 15% treaty rate with a W-8BEN on file; 0.17% is a WSDU calculation on a 1.1% yield, before any foreign income tax offset. ATO and US–Australia treaty guidance; confirm with your broker or adviser.
• Model maths (illustrative only): equal-weighted volatility (σₚ² = σ²[ρ + (1-ρ)/n]), with (σ = 30%) and (ρ = 0.3); two-asset volatility (σₚ = σ√(1+ρ)/2), with (σ = 20%).
• Prior WSDU pieces: Investing 101 series, including index construction, position sizing, and US withholding tax reclaim.

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.