Compounding Explained: Why Time in the Market Beats Timing It
Money that grows produces more money that grows. That sentence is the whole idea, and it sounds so mild that most people file it away and move on. What they miss is that the sentence describes a process which does almost nothing for a decade and then does something extraordinary, and that the gap between those two phases is where nearly every investing mistake lives.
This guide works the idea through with real numbers rather than a parable. The market figures come from one series: 11,848 daily closes of the S&P/TSX Composite, the index that tracks the bulk of the Canadian stock market, from June 1979 to September 2026. The company figures come from Royal Bank’s own quarterly filing. Nothing here is a projection dressed up as history, and where a table does project, it says so and shows the assumption.
One thing to settle first, because it decides how you read every number below. The index used here is a price index: it tracks share prices and ignores dividends entirely. A Canadian investor holding those companies also collected their dividends, so every market return in this guide understates what a holder actually earned. That is deliberate. If the case for time in the market survives being told with a piece of the return removed, it does not need the help.
What compounding actually is
Interest, dividends and capital gains are what an investment pays you. You have two choices about that money: take it out, or leave it in. Compounding is what happens when you leave it in, because next year’s return is then calculated on a larger balance. The Ontario Securities Commission puts it in one line on its compound interest page: you make your money grow faster if you also invest the money you earn along with the money you started out with.
Put $10,000 into something that returns 6.75% a year, which is what the S&P/TSX Composite delivered on price alone over the past 30 years. Year one adds $675. Year two does not add $675. It adds 6.75% of $10,675, which is $720. Year three adds $769. The percentage never changes. The dollars behind it get larger every single year, because the base they are charged on keeps getting larger.
That is the entire mechanism. There is no second step. What makes it hard to believe is not the arithmetic, which is trivial, but the shape it produces over long stretches: a line that looks flat for years, then bends, then goes nearly vertical. Human intuition draws straight lines. Compounding does not.
How long it takes to double
The most useful number in compounding is not the annual return, it is the doubling time, because doubling is how you feel the effect. The shortcut is called the rule of 72: divide 72 by the return, and you get roughly the number of years to double your money. It is a piece of mental arithmetic, and it is more accurate than a shortcut has any right to be.
| Annual return | Rule of 72 says | Actual doubling time |
|---|---|---|
| 4% | 18.0 years | 17.7 years |
| 6% | 12.0 years | 11.9 years |
| 7% | 10.3 years | 10.2 years |
| 8% | 9.0 years | 9.0 years |
| 10% | 7.2 years | 7.3 years |
At 7%, money doubles about every decade. That framing is worth holding onto, because it turns an abstract percentage into something you can count on your fingers: a 25-year-old with $10,000 invested at 7% and nothing added has roughly $20,000 at 35, $40,000 at 45, $80,000 at 55 and $160,000 at 65. The first double took ten years and added $10,000. The last double took ten years and added $80,000. Same rate, same patience, eight times the reward, purely because the base was bigger.
Over the full 47 years of the S&P/TSX series, from 1,618.4 in June 1979 to 36,513.8 in September 2026, the index multiplied 22.56 times. That is 6.83% a year, and again, that is before a single dividend.
The line bends late
Here is the same idea run against the real market rather than a smooth rate. Someone puts $500 into the S&P/TSX Composite on the first trading day of every month for 30 years, starting in September 1996: 360 purchases, at whatever the index happened to cost that morning, through the dot-com bust, the financial crisis, the oil crash and the pandemic.
They contributed $180,000 and finished with $573,562. Growth accounted for $393,562, or 68.6% of the final balance. But the distribution of that growth across the three decades is the part worth staring at.
| Period | Contributed by the end | Balance | Growth added in that decade |
|---|---|---|---|
| Years 1 to 10 | $60,000 | $92,916 | $32,916 |
| Years 11 to 20 | $120,000 | $181,628 | $28,712 |
| Years 21 to 30 | $180,000 | $573,562 | $331,934 |
The final decade produced $331,934 of the $393,562 total, which is 84.3% of all the growth. The second decade produced less growth than the first, because it ended shortly after the 2008 crash. Ten years in, this investor had put in $60,000 and was sitting on $92,916, which is a perfectly nice result and nothing like a life-changing one. Twenty years in, having doubled the contributions, they had barely doubled the balance.
Everything that makes the exercise worth doing happened in the last third, and it happened to a balance that only existed because of contributions made twenty years earlier. That is not an argument that the market got generous after 2016. It is arithmetic: 6.75% of a large number is a large number, and you only get a large number by having been invested long enough to build one.
The practical consequence is uncomfortable. The decade in which compounding delivers almost nothing visible is the decade that decides how much it delivers later, and it is also the decade in which people give up.
What ten years of waiting costs
The most expensive decision in investing is not a bad stock. It is a delayed start, and it is expensive in a way that no later effort recovers.
Take $300 a month, held to age 65, growing at that same 6.75%.
| Start at | Contributed | Value at 65 |
|---|---|---|
| 25 | $144,000 | $693,705 |
| 35 | $108,000 | $334,786 |
| 45 | $72,000 | $147,927 |
| 55 | $36,000 | $50,646 |
The 25-year-old contributes $36,000 more than the 35-year-old and ends with $358,919 more. Those ten years of $300 a month bought roughly ten times their own value, and they did it for one reason: they are the ten years that spent the longest compounding. The 45-year-old contributes half of what the 25-year-old does and ends with 21% of the result.
This table is arithmetic at a steady rate, not a forecast. No market delivers 6.75% in a straight line, and the shape of the real path matters, as the previous section showed. What is not a forecast is the relationship: the earliest dollars are worth a multiple of the latest ones, and every year of delay removes a year from the end of the process, where the growth is largest.
Anyone in their forties reading that table should note the other thing it says. Starting at 45 still turns $72,000 into $147,927. The best time to start has passed for most people; the second best time is the argument this table actually makes.
The worst timer still beat the investor who waited
The most common reason people delay is not laziness. It is the belief that they should wait for a better price. So test it, at the extremes, on the real Canadian market.
Five investors each put $6,000 a year into the S&P/TSX Composite for the 25 calendar years from 2001 to 2025. Same $150,000 of contributions, five different rules about when to buy.
- The first has perfect foresight and buys at the exact lowest close of each year.
- The second buys on the first trading day of each year without thinking about it.
- The third splits the money into twelve and buys on the first trading day of every month.
- The fourth has perfectly bad luck and buys at the exact highest close of each year.
- The fifth waits for a better entry point that never feels right, and holds the cash.
| Investor | Ended with | Return on $150,000 contributed |
|---|---|---|
| Bought the low every year | $505,061 | +236.7% |
| Bought on the first day of every year | $439,781 | +193.2% |
| Bought a twelfth every month | $434,385 | +189.6% |
| Bought the high every year | $390,164 | +160.1% |
| Never bought, held the cash | $150,000 | 0% |
Two numbers carry the entire argument. Twenty-five consecutive years of flawless timing, which no person has ever achieved and which requires knowing each year’s low before it happens, was worth $65,279 more than buying blindly on the first trading day. That is 14.8%, spread across a quarter of a century.
Twenty-five consecutive years of the worst possible timing, buying at the single highest price of every year, produced $390,164 against $150,000 of contributions. It beat waiting on the sidelines by $240,164.
The gap between the best timer and the worst timer is $114,897. The gap between the worst timer and the person who never invested is $240,164, more than twice as large. Timing decides a slice. Time decides the outcome.
Note the third investor too. Splitting the money into monthly instalments finished within 1.2% of the investor who put the whole $6,000 in on day one. Buying monthly is not a way to beat the market and does not claim to be. It is a way to keep contributing without needing a view, which for most people is worth considerably more than the 1.2%.
None of this says a market decline is nothing. It says that acting on the fear of one is what costs money, which is the same conclusion reached from the other direction in our guide to corrections and bear markets: the market’s best single days cluster inside its worst weeks, so getting out of the way of the falls means being absent for the recoveries.
The dividend compounds too
For a Canadian investor, the point above about excluded dividends is not a footnote. Many of the largest companies on the TSX pay one, and for a holder those payments are part of a return the price index leaves out entirely. A dividend also compounds in two directions at once.
The first is obvious: reinvested, it buys more shares, and those shares pay their own dividends. The second is the one people miss, which is that the payment per share itself grows. Here is Royal Bank’s, nine quarters as the bank reported it in its own supplementary pack.
The quarterly dividend went from $1.42 in the third quarter of fiscal 2024 to $1.76 in the third quarter of fiscal 2026, a rise of 23.9% in two years. Across full fiscal years the bank declared $5.60 per share in 2024 and $6.04 in 2025, which is growth of 7.86%. Those are the company’s own reported figures, not an estimate.
Now put the two effects together. Take $10,000 in a holding paying the 2.5% dividend yield the bank reports on page 4 of its report to shareholders, with the dividend growing at that 7.86% and the yield staying where it is. Compare reinvesting every payment against banking the cash.
| Year | Income if reinvested | Position if reinvested | Income if spent | Position plus cash taken |
|---|---|---|---|---|
| 1 | $250 | $10,250 | $250 | $10,250 |
| 5 | $373 | $15,311 | $338 | $14,996 |
| 10 | $617 | $25,286 | $494 | $23,350 |
| 20 | $1,682 | $68,961 | $1,052 | $53,346 |
This table is arithmetic under stated assumptions, not a forecast. It holds the yield constant for 20 years, which no real stock does, and it assumes a dividend growth rate taken from a single year, which no company guarantees. Dividends get cut, and a company that has raised its payment for nine straight quarters is under no obligation to raise it for a tenth. Read the comparison, not the levels. The reinvesting holder ends with roughly $15,615 more and, more importantly, an income stream 60% larger, from the identical starting position and the identical company. The only difference is what happened to the cash.
The mechanics of this are covered properly on the best Canadian dividend stocks page, and if you want to see what a given yield pays on a given position, the dividend income calculator does that arithmetic. Most Canadian brokerages offer a dividend reinvestment plan that buys the shares for you automatically, which turns this from a decision you make every quarter into one you make once. The terms differ, so check what yours supports.
What compounds against you
Compounding is a process, not a force for good. Anything charged as a percentage of the balance compounds against you on exactly the same mathematics, and two things routinely are.
Fees. The Financial Consumer Agency of Canada groups investment costs into what you pay when buying, when selling, and when holding. The third one is the one that compounds, because it is levied on the balance every year. Here is the identical $500 a month over the identical 30 years, with an annual fee taken out of the balance.
| Annual fee | Ends with | Cost of the fee | Share of the growth it took |
|---|---|---|---|
| None | $573,562 | 0 | 0% |
| 0.05% | $568,323 | $5,239 | 1.3% |
| 0.25% | $547,910 | $25,651 | 6.5% |
| 1.00% | $478,568 | $94,993 | 24.1% |
| 2.00% | $401,480 | $172,082 | 43.7% |
A 2% annual fee looks like a rounding error next to a 6.75% return. Over 30 years it took $172,082, which is 43.7% of everything the portfolio grew by. The investor took all of the risk and kept 56% of the reward. It is also invisible on a statement, because a fund fee is deducted from the fund’s value rather than billed to you, so nothing ever arrives saying what it took. Our mutual fund fee calculator runs this on your own numbers, and it is worth ten minutes if you hold funds and have never checked what they charge.
Tax. In a taxable account, tax on dividends and interest is paid in the year the income is received, which means those dollars leave the balance before they can compound. Registered accounts stop the leak. The CRA describes the TFSA in exactly those terms on its opening a TFSA page: it is for individuals who want to save and invest their money tax-free. An RRSP defers the tax instead, so the full pre-tax amount compounds and the tax is settled on withdrawal. Which account suits you depends on income and timeline, and both are worked through in our guides to how a TFSA works and how an RRSP works. If you are choosing what to hold inside one, our best TFSA stocks page argues the case for long-hold positions in the account whose whole advantage is uninterrupted compounding.
The mistakes people actually make
Waiting for a better entry point. The five-investor table prices this precisely. Perfect timing was worth 14.8% over 25 years; not being invested cost 160%. A wait has to buy an advantage bigger than the one perfect foresight delivers, and it never does, because the thing being waited for is a price that is only identifiable afterwards.
Starting late because the amount available feels too small. $300 a month from 25 became $693,705. The instinct to wait until there is a serious sum to invest reverses the arithmetic: the sum becomes serious because it started, not before.
Interrupting it. Selling during a decline converts a temporary paper loss into a permanent one and restarts the clock on a smaller base. In a registered account it costs more than that, because withdrawn TFSA room does not come back until January 1 of the next year and an RRSP withdrawal is taxed on the way out. The fall was temporary and the exit charge was not. One withdrawal escapes the tax and not the arithmetic: money taken out under the Home Buyers’ Plan comes out untaxed and still sits outside the market for up to fifteen years while you repay it.
Taking the dividends as spending money by default. Over 20 years on the assumptions above, that choice cost $15,615 on a $10,000 position and left the holder with an annual income 37% smaller. If you need the income, take it. Most people under 50 do not need it and take it anyway, because it arrives as cash and cash gets spent.
Paying 2% without knowing it. $172,082 over 30 years, on a portfolio that grew by $393,562. Nothing else on this list is that expensive.
What to take from this
Compounding pays for patience rather than intelligence, which is why it is both the most reliable edge available to an ordinary investor and the hardest one to hold onto. The mechanism is dull, the first decade is unconvincing, and the reward is back-loaded into a stretch of time you cannot reach any other way than by waiting.
The numbers in this guide all say the same thing from different angles. 84.3% of the growth arrived in the final decade. Ten years of delay cost more than ten times the contributions skipped. The worst market timer in Canada beat the patient sideline-sitter by $240,164. A 2% fee quietly claimed 43.7% of the gain. In every case, the variable that mattered most was how long the money was left alone to work.
Where to go next
- What a stock is, and what you actually own is where this series begins, if the share itself is still fuzzy.
- How to buy your first stock covers order types and what a trade actually costs, which is the step between deciding to start and starting.
- How to read a stock quote explains the dividend yield and the other numbers this guide compounds, field by field.
- The investing guides index lists the whole series in order.