Dividend Investing

A dividend is a capital allocation decision, not a gift. Three numbers together tell you whether it can survive.

Dividend investing is often described as the conservative option, the choice you make when you want income rather than growth. That framing undersells it. The real argument for dividends is arithmetic: over long periods a substantial share of total equity return has come from cash actually paid out to shareholders, not from the multiple the market is willing to pay. A company that never distributes anything is asking you to trust that the cash will eventually be worth more inside the business than in your pocket. Sometimes that is true. Often it is not.

A dividend is a decision, not a gift

The board chooses to pay it. That choice tells you something, and the most useful thing it tells you is about capital allocation. If a company can reinvest at 18% and pays out nothing, that is rational. If a company reinvests at 6% while its cost of capital is 9%, every retained dollar destroys value -- and a dividend would have been the better decision.

So the first question is never "how much does it pay". It is "what does the business earn on the money it keeps". The DuPont ROE calculator decomposes return on equity into margin, turnover and leverage, which tells you whether high returns come from a genuinely good business or from borrowed money.

Three numbers, not one

A dividend yield on its own is close to meaningless. Three numbers together give you a picture.

MetricWhat it answersComfortable range
Dividend yieldWhat income am I getting for the price?Varies by sector; 3-6% for mature payers
Payout ratioWhat share of earnings is being paid out?30-60%; above 80% is a warning
Dividend growth rateWill this income keep up with inflation?At or above inflation, so 3%+

The three interact. A 7% yield with a 95% payout ratio and flat earnings is not generous, it is fragile -- there is almost nothing left to absorb a bad year. A 2.5% yield with a 40% payout ratio and 8% growth is a lower current income with a far better chance of doubling over a decade.

The dividend yield calculator handles the income side and the payout ratio calculator the sustainability side. Run both before you buy a high yielder.

Testing whether the dividend can survive

Earnings-based payout ratios are a starting point, but earnings are an accounting construct. The dividend is paid in cash, so test it against cash.

Dividend cover = Free cash flow / Total dividends paid

Cover above 1.5x means the business generates half again as much cash as it distributes -- reasonable room for a bad year. Cover around 1.0x means the dividend consumes essentially all available cash. Below 1.0x means the company is funding the dividend from the balance sheet or from borrowing, which cannot continue indefinitely.

Worked example. A utility reports $420m of free cash flow and pays $300m of dividends. Cover is 1.4x. Now assume a mild recession shaves 15% off free cash flow, to $357m. Cover falls to 1.19x -- still covered, but the cushion is thin and management will be reluctant to raise. If the same company had been paying $400m, cover would have been 1.05x before the recession and 0.89x after, at which point a cut becomes likely.

Two other checks worth running: net debt divided by EBITDA (above 3x is uncomfortable for a payer, because refinancing competes with the dividend), and whether the dividend has been held flat or cut in the last downturn. A dividend that survived 2008 or 2020 without a cut tells you more than any current ratio.

The arithmetic of dividend growth

Growth in the dividend is where income investors are rewarded over time, and the effect compounds in a way that is easy to underestimate. The table below tracks a $10,000 position bought at a 3.0% yield with the dividend growing 6% a year.

YearIncome receivedYield on original cost
1$3003.0%
5$4014.0%
10$5375.4%
15$7197.2%
20$9629.6%

After twenty years the position pays nearly 10% on the money originally invested, without the share price doing anything. That is the argument for growth over headline yield. Note also what the same table says about inflation: at 6% dividend growth, real income roughly doubles over two decades. At 0% growth, real income falls by about a third.

Be sceptical of sustained double-digit dividend growth. It requires earnings to grow at least as fast, which for a mature business rarely lasts more than five years. A useful sanity check: the sustainable dividend growth rate is roughly return on equity multiplied by the retention ratio (one minus the payout ratio). A company with 12% ROE paying out 60% can sustainably grow the dividend around 4.8%.

What actually happens on the ex-dividend date

The share price is reduced by roughly the dividend amount on the ex-dividend date. This is not the market punishing you; it is mechanical, because the cash has left the company. Total return is unaffected -- you have exchanged part of the share price for cash.

In practice this matters for three reasons. First, it gives you a scheduling decision: if you are buying, buying just after the ex-date avoids the small tax event of receiving a dividend you did not really earn. Second, it explains why a chart that "drops every quarter" is not broken. Third, the total return calculator confirms that a 4% price fall paired with a 4% dividend is a flat outcome, not a loss.

The yield trap

An unusually high yield is the market telling you something. Yields above roughly 8% on ordinary operating companies are usually one of four things:

Three of those four are traps. The way to tell them apart is the same test every time: does free cash flow cover the dividend with room to spare, and is the business's revenue growing in real terms over five years? If the answer to either is no, the yield is a warning rather than an opportunity.

Tax, and why the headline yield is not your yield

For many non-US investors, dividends from US-listed companies are subject to withholding tax at source. Under several bilateral treaties, including the one between China and the United States, the rate on portfolio dividends is 10% rather than the default 30%. That requires filing the correct form with your broker -- the W-8BEN or its successor -- otherwise the higher default rate applies and is difficult to recover.

The practical consequence: a quoted 4% yield delivers around 3.6% after a 10% withholding. Capital gains usually face no US withholding for non-resident aliens, which creates a real, if modest, tax preference for total return over pure income. Factor that in when comparing a high-yield strategy against a lower-yield one with faster growth.

How many payers do you need

Concentration is the main risk in dividend portfolios, because a single cut does disproportionate damage to both income and psychology. A workable structure for an individual investor:

Valuing a dividend stock

For a company whose dividend is stable and growing, the dividend discount model is the natural valuation approach, because the dividend is the cash flow you receive. The DDM calculator applies the Gordon growth form; for a company still growing quickly that will slow later, the two-stage DDM fits better.

Worked example. A stock pays a $2.00 dividend, expected to grow 5% a year, and you require a 9% return.

Value = 2.00 x 1.05 / (0.09 - 0.05) = 2.10 / 0.04 = $52.50

Note how sensitive this is. Require 10% instead of 9% and the value falls to $42.00. Assume 4% growth instead of 5% and it falls to $35.00. The model is not precise, which is exactly why the margin of safety matters -- you buy at a discount to the estimate rather than at it.

When to sell a dividend holding

The usual advice to "never sell a dividend stock" is wrong. Four situations justify an exit: the payout ratio has moved above 90% without a clear and temporary cause; the business has cut the dividend, which is management confirming that the cash is not there; the investment case was based on a growth rate that has structurally disappeared; or the position has grown so large that it dominates your income and your risk. A suspended dividend is information, and ignoring it because you liked the yield last year is how small losses become permanent ones.

Where to go next

Frequently asked questions

What is a good dividend yield?

For mature payers, roughly 3-6%, but the number on its own tells you almost nothing. The guide's two profiles make the point: a 7% yield with a 95% payout ratio and flat earnings is fragile, because almost nothing is left to absorb a bad year; a 2.5% yield with a 40% payout and 8% growth is lower income today with a far better chance of doubling over a decade. Treat anything above roughly 8% with suspicion -- an unusually high yield is usually the market pricing in a cut, not offering a gift.

How do I know if a dividend is safe?

Test it against cash, not earnings. Dividend cover is free cash flow divided by total dividends paid. The worked example: a utility with $420m of free cash flow paying $300m has cover of 1.4x. A mild recession shaving 15% off cash flow drops it to 1.19x -- still covered, but the cushion is thin. Had the company paid $400m instead, cover would have been 1.05x before the recession and 0.89x after, at which point a cut becomes likely. Two companion checks: net debt above 3x EBITDA is uncomfortable for a payer because refinancing competes with the dividend, and a dividend that survived the last downturn without a cut says more than any current ratio.

How much do I need to invest to make $500 a year in dividends?

Divide the income you want by the yield: $500 at a 3.0% starting yield needs about $16,700 invested. The guide's table tracks $10,000 bought at 3.0% paying $300 in year one -- but with the dividend growing 6% a year, that income reaches $401 by year five, $537 by year ten and $962 by year twenty, at which point the position yields 9.6% on its original cost without the share price moving. The starting yield sets the income; the growth rate determines what it becomes. A lower-yield, faster-growing payer can overtake a high-yield, flat one inside a decade.

What happens to the share price on the ex-dividend date?

It drops by roughly the dividend amount, mechanically, because the cash has left the company. This is not the market punishing you and total return is unaffected -- you have exchanged part of the share price for cash. It explains why a chart that "drops every quarter" is not broken, and why a 4% price fall paired with a 4% dividend is a flat outcome rather than a loss. One practical consequence for buyers: purchasing just after the ex-date avoids receiving a dividend you did not really earn, which is a small tax event in most jurisdictions.

Related guides

More guides

Free valuation calculators