Dividend Reinvestment: What Actually Drives the Difference

The mechanism, stated exactly

Hold S shares at price P. A dividend at yield y pays S·y·P in cash. Reinvested at that same price P, it buys S·y·P ÷ P = S·y new shares. The price cancels out. The share count therefore grows by exactly the factor (1 + y) each year, and after T years you hold S·(1 + y)^T shares — a compound-growth calculation with the yield as the rate and the shares as the unit. Whatever the price does over those years, it does it to a larger and larger number of shares.

That gives a terminal value with a clean product structure: value = (starting value) × (price multiple) × (share multiple). On our case — $10,000, 3 percent yield, 4 percent price growth, 30 years — the price multiple is 1.04^30 = 3.2434 and the share multiple is 1.03^30 = 2.4273, so the position ends at $10,000 × 3.2434 × 2.4273 = $78,726. Spending the dividends leaves the share multiple at 1 and the position at $32,434. Reinvestment therefore multiplies the terminal value by exactly the share multiple, 2.4273 — not approximately, exactly, because the two effects are independent factors in the same product.

Decomposing the claim that dividends are most of the return

The claim is repeated constantly and almost never decomposed, so here is the decomposition. Growth compounds multiplicatively, so the honest way to split it is on logarithms: ln(3.2434) = 1.1766 for price and ln(2.4273) = 0.8866 for shares, out of a total of 2.0632. Price contributes 57.0 percent of the compound growth, the extra shares 43.0 percent. The annual figures agree: the reinvesting holder compounds at 7.12 percent a year against 4.00 percent for the price alone.

So in this case the price does slightly more of the work, not less — and the general rule falls straight out of the algebra. The share side contributes more than half whenever ln(1 + y) exceeds ln(1 + g), which is simply whenever the yield exceeds the price growth. The crossover is exact and the table shows it: at a 4 percent yield against 4 percent price growth the split is precisely 50-50. The received wisdom that reinvested dividends account for most of the total return is therefore a statement about a particular history — one in which yields were high and real price growth was modest — rather than a property of dividends. Run your own two numbers and the answer will be whatever those two numbers say.

Tax is taken before reinvestment, so it compounds negatively

This is the flaw in almost every quick version of the calculation. Where a dividend is taxable, the tax is charged when it is paid — often withheld before it reaches you — so only the net amount can buy shares. A 3 percent yield taxed at 30 percent reinvests at 2.1 percent, and the share multiple over 30 years collapses from 1.03^30 = 2.4273 to 1.021^30 = 1.8654. The terminal position falls from $78,726 to $60,502. The loss is $18,223, or 23.1 percent of the untaxed result.

The striking part is the comparison with the tax actually handed over. Running the schedule year by year, the total nominal tax paid across the 30 years is $7,644 — but the position ends $18,223 lower. The difference is everything those taxed euros or dollars would themselves have earned, and everything their earnings would have earned. That is what 'compounds negatively' means, and it is why a tax-sheltered wrapper is worth more on a dividend-heavy holding than on a growth one. Rates differ sharply by country and by wrapper — a flat withholding regime, a progressive one, a partial exemption and a fully sheltered account can all apply to the same share — and cross-border holdings can face a withholding tax at source on top. Put your own rate into the calculation; do not borrow one from an article written for a different country.

A dividend is not free money: the ex-date drop

When a company pays a dividend, cash leaves the company. On the ex-dividend date, buyers no longer acquire the right to that payment, and the price is marked down by roughly the dividend to reflect it. A $100 share paying $3 opens at about $97 with $3 on its way to you: you hold $100 either way. Nothing was created — the payout moved value from inside the company to your account. Actual moves are noisier than the theory, because tax treatment, index events and ordinary trading all interfere, but the direction is not in dispute and the mechanism is not a market opinion, it is accounting.

Two consequences follow, and both are practical. First, a high payout is not by itself a return: what you should judge is total return, price change plus dividends, and a company that pays out more will, all else equal, show less price growth because it is retaining less to grow with. Second, reinvesting is close to a no-op in the absence of tax and costs — you sell nothing and buy nothing, you simply decline to take the cash out, and the position tracks the total-return series instead of the price series. It is precisely tax and dealing costs that make the choice consequential.

The payout ratio is the sustainability check

The payout ratio is the dividend per share divided by earnings per share: a company earning $5 and paying $3 has a 60 percent payout. It answers the only question that matters about a dividend you plan to compound for decades, which is whether it will still be there. A ratio comfortably below 100 percent leaves room for a bad year; a ratio above 100 percent means the payment is coming from cash reserves, asset sales or borrowing, and the arithmetic of that cannot run indefinitely. Reading it alongside free cash flow rather than accounting earnings is better still, because earnings can be depressed by non-cash charges that do not threaten the payment at all.

The ratio also tells you what growth to expect, and the two are linked mechanically. Whatever is not paid out is retained and reinvested inside the company, so a rough sustainable growth rate is return on equity multiplied by the retention ratio: 12 percent return on equity with a 60 percent payout gives 0.12 × 0.40 = 4.8 percent. Raise the payout and the yield rises while the growth of the dividend falls. That trade-off is the honest reason a high yield is not automatically better than a low one — it front-loads income at the cost of the compounding you were counting on.

A very high yield is usually a falling price, not a generous board

Yield is a fraction, and the market moves the denominator far faster than a board moves the numerator. A company paying $3 on a $100 share yields 3 percent. Let the price fall to $40 and the same unchanged $3 yields 7.5 percent — a headline yield two and a half times higher produced entirely by a 60 percent capital loss. Screening on yield alone therefore selects, quite reliably, for shares the market has just marked down, which is not the same thing as selecting for income.

The reinvestment arithmetic makes this worse rather than better, because it assumes the dividend keeps being paid. If a payment is cut after five years, the share multiple stops compounding at the higher rate and the projection you built collapses — and a cut usually arrives together with a further price fall, so both factors in the product move against you at once. Check the payout ratio, check whether the yield rose because the dividend was raised or because the price fell, and treat a yield far above the market's as a question to investigate rather than a number to type into a calculator.

Frequently asked questions

Does reinvesting always beat taking the cash?

In the arithmetic, reinvesting into the same holding always ends with more of that holding — that is what the share multiple says. Whether it ends with more money depends on what the alternative does. Cash taken out and used to repay a debt at 8 percent, or invested elsewhere at a better risk-adjusted return, can easily win. Reinvesting also quietly concentrates you: every payment buys more of a security you already own, so a position that started as one tenth of a portfolio grows relative to the rest, and rebalancing is exactly what taking the cash lets you do.

What tax rate should I put into the calculation?

Your own effective rate on dividend income, in the account the shares actually sit in — and that is genuinely different from one country to the next. Some systems apply a single flat rate to investment income; some tax dividends at the marginal income rate; some grant a partial exemption or a credit for tax already paid by the company; and most have at least one wrapper in which dividends are sheltered entirely. A cross-border holding may also suffer withholding tax at source that a treaty may or may not let you reclaim. Because the drag compounds, the difference between a sheltered and an unsheltered account over thirty years is not a rounding item — run it both ways, and check your own rules rather than a figure quoted for somewhere else.

Why did the share price drop on the day the dividend was paid?

Because the company is worth less by exactly the cash it just sent out. The mark-down happens on the ex-dividend date, the first day on which a buyer no longer acquires the right to the payment, and it is roughly the size of the dividend. You have not lost anything: the value simply moved from the share price into your account. This is also why a chart of the price alone understates the return of a dividend-paying share — the price series has every payout subtracted out of it, while a total-return series adds them back. Always check which series you are looking at before comparing two investments.

Is a 9 percent yield a bargain?

It is a question, not an answer. Ask first whether the numerator rose or the denominator fell: a dividend that was increased is a different animal from one that is unchanged while the price halved. Then check the payout ratio against earnings and against free cash flow — a payment consuming more than the company generates is being funded from somewhere that will run out. Then ask what the market is pricing in, because a yield far above the rest of the market is, most of the time, a collective forecast that the payment will be cut. Sometimes the market is wrong; the point is that you are taking the other side of a specific bet, not collecting an unclaimed windfall.

Does the calculation change if my broker buys fractional shares?

The model assumes fractional reinvestment, which is exactly what a reinvestment plan usually delivers. If your broker cannot buy fractions, each payment leaves a cash residue that sits idle until it is large enough to buy a whole share, and the realised result is slightly below the model — the more so on a small position, where a single share costs a large fraction of the payment. Dealing commissions have the same effect and are usually the larger of the two. Neither changes the shape of the answer; both argue for a plan that reinvests automatically and free, and for checking whether yours does.