Diversification is described as the only free lunch in finance. It is worth understanding precisely what is being served, because it is a smaller and more specific meal than most people assume.
Diversification removes idiosyncratic risk — the risk specific to one company — which carries no expected return, so removing it is genuinely free. It cannot remove systematic risk, the market-wide risk that is the reason equities are expected to return anything at all. Roughly 20–30 stocks eliminates most single-name risk, though one broad ETF does it in a single purchase. Canada is about 3% of world equity markets, yet Canadian investors typically hold more than half their equities domestically.
Every investment carries risk from two distinct sources, and the difference between them is the entire justification for diversifying.
That asymmetry is the whole argument. Holding one stock exposes you to both. Holding a broad index exposes you only to the systematic part. You have removed the risk that pays nothing and kept the risk that pays. That is why it is called a free lunch — you improve the risk-adjusted outcome without giving up expected return.
The classic finding is that the benefit of adding holdings falls away quickly. Going from 1 stock to 10 removes a large share of specific risk; from 10 to 25 removes much of the rest; beyond about 30 the curve is nearly flat.
Two caveats that matter more than the number.
First, 25 holdings only works if they are genuinely different. Twenty-five Canadian bank and energy stocks is not a diversified portfolio; it is one bet on the Canadian economy, expressed twenty-five ways. What counts is exposure to different economic drivers, not the number of line items.
Second, the practical answer for almost everyone is not to count at all. A single broad-market ETF holds hundreds or thousands of companies across every sector, in one purchase, for a fraction of a percent a year. Assembling 25 stocks by hand means research, monitoring, rebalancing, and 25 opportunities to make a behavioural error — to buy a diversification that is available off the shelf.
Diversification works through correlation — the degree to which two holdings move together. Assets that always move identically (correlation of 1.0) provide no diversification no matter how many you hold. The benefit comes from assets that are imperfectly correlated: when one falls, the other does something different.
This is why a portfolio of stocks and bonds has historically been less volatile than the weighted average of its parts. It is also why adding an eleventh technology stock does far less than adding a bond fund.
The correct conclusion is not that diversification is useless. It is that diversification manages risk over the long run rather than eliminating it in the short run — and the only genuine protection against a bad year is not needing the money that year, which is Lesson 4.1's three-year rule and Lesson 7.1's emergency fund doing their jobs.
Canadian investors typically hold well over half their equity allocation in Canadian stocks. Canada represents roughly 3% of global equity market capitalisation. That is an overweight of fifteen to twenty times the country's global share — and it is not unique to Canada, but the consequences here are unusually sharp because of what the Canadian market contains.
The S&P/TSX Composite is heavily concentrated in financials and energy, with materials adding a third resource-linked block. Technology, healthcare and consumer sectors are thinly represented compared with global markets. So a Canadian-only equity portfolio is a leveraged bet on banks, oil and commodities — and on an economy already paying your salary and, quite possibly, holding your house.
There are real arguments for holding more than 3%: the dividend tax credit makes Canadian eligible dividends the most tax-efficient income in a taxable account (Lesson 6.1); there is no currency conversion cost and no FX risk on money you will spend in Canadian dollars; and your future liabilities — groceries, property tax, retirement — are denominated in CAD. A 20–30% Canadian weighting is a defensible balance that most Canadian one-ticket ETFs land close to. What is hard to defend on any of those grounds is 60–100%.
The concentration argument is worth making concrete. If you work for a Canadian bank, own a home in a Canadian city, and hold a portfolio of Canadian banks, then a Canadian financial-sector downturn hits your salary, your house and your savings simultaneously. Diversification is not only about the portfolio — it is about the portfolio relative to everything else you own, including your own earning power. That is the professional way to think about it, and it usually argues for owning more of the rest of the world than instinct suggests.
Around 20 to 30 genuinely different companies removes most single-company risk, and the benefit flattens quickly after that. But the count matters less than the variety — 25 Canadian banks and energy producers is one bet expressed 25 ways, not a diversified portfolio. In practice a single broad-market index ETF holds thousands of companies across every sector and country for a fraction of a percent a year, which is why most investors should not be counting at all.
Canada is roughly 3% of global equity markets, yet Canadian investors typically hold more than half their equities domestically. A home tilt of 20–30% is defensible: the dividend tax credit favours Canadian eligible dividends in taxable accounts, there is no currency conversion cost, and your future spending is in Canadian dollars. Beyond that you are making a concentrated bet on banks and energy — often on top of a Canadian salary and a Canadian home.
Educational purposes only — not financial, investment or tax advice. RiskStock is not a registered dealer, adviser, or tax professional. Nothing in this course is a recommendation to buy or sell any security or product. Tax and benefit figures are for the 2026 tax year, were verified against official sources on 2026-07-27, and change annually — confirm your own numbers with the CRA and a qualified professional before acting.