Dividend Income vs. Selling Shares: Which Actually Funds Retirement Better?

Where does $40,000 a year come from when you have $1,000,000 invested, and how do you pull it without running dry? That is the question this settles.

Most people holding that balance and needing that income believe dividends are the safe route because they avoid selling. The belief is understandable and it is not quite right.

On the dividend fund's ex-date the price drops by the $3.50 payout. You hold $35,000 in cash and a $965,000 balance, which still adds up to $1,000,000. The dividend is a transfer from share price to cash, not extra money on top. Selling shares is not automatically eating principal either.

So the real question is narrower: which route funds the $40,000 while leaving the best mix of portfolio value, control, tax, and fees? Both routes deliver the same cash. They differ in who decides the timing, what a bad year costs you, and what the tax bill and the expense ratio take before the money reaches your checking account.

By the end you can run both routes on the same $1,000,000, check your account type first, and pick the method that matches either an income schedule or a withdrawal dial. Dividends and share sales both get you $40,000. The right choice depends on the wrapper and on you.

Point	Dividend income	Selling shares
Annual cash from $1,000,000	$35,000 dividend + $5,000 sale = $40,000	$40,000 sale = 400 shares at $100
Share count after a flat year	10,000 shares (50 sold to top up)	9,600 shares
Portfolio value after a flat year	$965,000 plus $40,000 cash	$960,000 plus $40,000 cash
In an -8% year	Price falls to $88.50; $885,000 in shares	Price falls to $92; 435 shares sold
Who sets the payout	Fund's dividend schedule	You, once a year
Tax in a taxable account	$35,000 dividend taxed when paid	$16,000 gain on a $60 cost basis
Annual fund fee on $1,000,000	$3,500 (0.35%)	$400 (0.04%)
Annual admin	Automatic deposit	One sale and one rebalance, about 15 minutes

Both routes deliver the same $40,000

Your $40,000 need gets met either way. Both a dividend fund yielding 3.5% and a plain total-market fund you sell from can hand you the same $40,000 a year before tax, and the differences are in the plumbing, not the total.

Start with the same $1,000,000, held as 10,000 shares at $100 each. At a 3.5% yield the dividend fund pays $3.50 a share, which is $35,000 over the year, or about $2,917 a month landing in your account on the fund's schedule. To reach $40,000 you sell 50 shares for the remaining $5,000. That is the whole annual job: collect the checks, sell a sliver.

The total-return fund does not care about yield. You sell $40,000 of it at $100 a share, which is 400 shares, and the $40,000 shows up in your cash account instead of as a distribution. Same number, different plumbing.

Notice what sets the target in both versions: not the fund, not the yield, not some rule about dividends being safer. The $40,000 does. If your need were $25,000, the dividend fund would cover it with cash left over and the selling fund would take 250 shares. If your need were $60,000, both routes would have to find more, one by selling more shares, the other by adding a larger sale on top of the payout.

The monthly math is where the two feel different in your hand. The dividend route deposits $2,917 and asks you to top it up once a year with a 50-share sale. The selling route asks you to move $3,333 a month yourself, or $40,000 in one annual sale, and the choice of when is entirely yours.

Same $40,000, then. What changes is who decides the timing and where the cash comes from.

Dividends are not extra money on top

The dividend is not a bonus that arrives on top of your portfolio. It is money moved out of the fund's own share price and into your cash account, which is why the price drops by the payout on the ex-date.

Put real numbers on it. Your dividend fund trades at $100 the day before the ex-date and pays $3.50 a share. On the ex-date the price opens around $96.50. You now hold $35,000 in cash and a balance of $965,000, and those two numbers add back to $1,000,000. Nothing was created. The transfer is the point.

The selling side does the same arithmetic in reverse. Take $40,000 out of the total-return fund at $100 a share and you are left with $960,000 in the fund plus $40,000 in cash. That is also $1,000,000 before any market move crawls in. The only difference is whether a fund committee decided the amount and the date or you did.

This is why share count is a bad scoreboard. The dividend investor still counts 10,000 shares after the ex-date and can feel like nothing was spent. Yet the shares are worth less than they were the day before, by exactly the payout. The seller counts 9,600 shares and feels poorer. Both hold the same $1,000,000 in the same mix of cash and stock.

Judge the portfolio by what it is worth plus what is in your checking account, not by how many shares sit in the account. Do that and the dividend stops looking like free money and starts looking like what it is: a scheduled transfer you did not have to arrange.

Selling shares is not eating principal

Selling shares moves value out of the fund and into your pocket, and that is all it does. It does not quietly burn your principal, and it does not leave you poorer than the dividend route with the same market return, which is the part that surprises people.

Run both funds through the same bad year. Say each earns a total return of -8%, dividends and price change combined. The dividend fund, which started at $100 and paid out $3.50, ends at $88.50 a share. The total-return fund, which paid nothing out, ends at $92. Now fund the $40,000.

On the dividend side you already have $35,000 in cash. You hold 10,000 shares worth $885,000, and the extra $5,000 comes from selling about 56 shares at $88.50. On the selling side you sell $40,000 at $92, which takes 435 shares, and you are left with 9,565 shares worth about $880,000.

Look at the two ending numbers: roughly $880,000 and $885,000 of stock left after $40,000 of spending, in a year when both funds lost the same 8%. The dividend investor still owns all the share certificates and feels untouched. The seller owns fewer shares and feels like something was spent. The portfolios are within a rounding error of each other.

Share count tells you what happened to the number of pieces. Portfolio value tells you what happened to your money. With equal total returns, only the second one decides how you come out.

Selling is a transfer. Treat it as one and the fear of shrinking to nothing loses most of its grip.

A bad year makes selling take more shares

When both funds earn the same -8% in a single year, the total-return fund's price falls to $92, so pulling out $40,000 takes about 435 shares instead of the 400 it took the year before at $100. The dollar amount hasn't changed a cent. The share cost of the withdrawal has.

The dividend side isn't sitting pretty either. The same -8% drags its price to $88.50, so the $35,000 payout arrives, but the 10,000 shares behind it are now worth $885,000 instead of $1,000,000. That $5,000 top-up needed to reach $40,000 costs more shares at $88.50 too, more than it would have at the old price. The cash shows up, but it's coming from a smaller pie on both sides.

Neither route dodges the loss.

And this is where the timing gets visible: a falling price makes every share you have to hand over buy less of the withdrawal. The retiree selling by hand watches the share count climb just as the portfolio shrinks. The retiree living on dividends watches the price fall and still needs the top-up sale. Both feel the same market, just from different chairs.

The thing a bad year exposes is that the dividend side's payout is fixed while the price isn't, and the selling side's withdrawal stays fixed while the shares needed grow. Run your own $40,000 need through a -10% or -15% year before you commit, and check how many shares each route would have asked you to give up.

Dividend cuts hit income and price together

A bad market doesn't just make selling more expensive in shares; it can shrink the dividend check itself. If the fund trims its payout from $3.50 a share to $2.50, the 10,000 shares that used to throw off $35,000 now deliver $25,000. That's a $10,000 hole in a $40,000 budget, and it arrives with no warning label.

Usually the cut doesn't come alone. A company or fund cutting its payout tends to do it when the price is under pressure, so income and portfolio value tend to fall together, not one after the other. The retiree who leaned on $35,000 of dividends to fund the year now faces a smaller check and a smaller balance at the same time.

The selling side looks safer here, and in one narrow sense it is. When the portfolio is worth less, the retiree can still decide to take $40,000, drop to $25,000, or take nothing at all and patch the gap from cash. That flexibility is real.

But don't hand the dial more power than it has.

Control doesn't bring back lost portfolio value. If the balance fell from $1,000,000 to $920,000, choosing $25,000 instead of $40,000 protects the remaining shares, but it doesn't restore a dollar of what the market took. The dial lets you pick the withdrawal; it can't rewind the loss.

The practical move: keep the fixed income need in cash or short-term savings if you can, so a cut or a bad year doesn't force a sale at the worst moment. Even a cushion covering three to six months of the $3,333 monthly need buys you room to decide instead of react.

Selling gives you the dial; dividends the schedule

On the dividend side you take the income the fund decides to send, on the fund's timeline, not yours.

The check shows up at $35,000 a year whether that month needs more or less. To reach $40,000 the retiree sells about 50 shares on top of the payout; to take less than $35,000, there is no dial at all, only the cash landing in the account whether it's wanted that quarter or not. That is a schedule someone else owns.

The selling side runs the numbers the other way around. In a year that needs $40,000, sell 400 shares at $100. In a year that needs only $25,000, sell 250 and leave the other 150 shares working. In a year where a roof, a move, or a medical bill eats the budget down to nothing, sell nothing and let the full $1,000,000 sit.

The dollar amount is the retiree's to set, once a year, in a single decision.

That is the real split: automation against the dial, and neither is free.

Sitting back and letting the deposit appear means no math, no orders, and no decisions to second-guess in a bad month; it also means the income is whatever the fund pays, and topping it up still requires that extra sale. Keeping your hand on the dial means deciding an amount each year, checking the balance, and placing the trade, roughly 15 minutes of work, in exchange for the ability to flex between $0 and $40,000 or more as the year demands.

For the retiree who wants a predictable deposit and no involvement, the schedule is the point. For the retiree who wants to spend more in a good year, less in a bad one, or skip a withdrawal entirely, the dial wins, and it is worth the small yearly chore.

Control, though, is not the same as free money. Choosing $25,000 instead of $40,000 leaves more shares in place, but a smaller sale does not stop the portfolio from falling in a bad market; it only changes how much you pull out of it.

All of this is a decision about method, and it lands before the tax bill and the fund's expense ratio get a vote.

Dividends are taxed whether spent or not

The dividend route can cost more in tax, and only partly because the money shows up without you asking. A $35,000 distribution is taxable in the year it is paid, whether you spend a dollar of it or reinvest every cent. Reinvesting does not defer anything. The tax bill lands on the 1099 the following January regardless.

The sale side is not tax-free either, but less of the $40,000 is exposed. Sell $40,000 of shares bought at a $60 cost basis and $16,000 of that is gain; the other $24,000 is simply your own principal coming back to you. Only the gain portion is taxed.

So $35,000 of taxable dividend against $16,000 of taxable gain. That gap is real in a taxable account, and it flips once you start moving pieces around. The actual bill depends on your bracket, your holding period, and how much of each sale is gain rather than return of principal.

Which is why headline income is the wrong scoreboard. Two retirees can report the same $40,000 of cash and owe wildly different amounts, because the mix of dividend, gain, and principal is different under each route.

Run the numbers against your own cost basis before you decide. IRS Publication 550 spells out how dividends and capital gains are treated, and a tax adviser can price your specific situation; guessing at your bracket is how people get surprised in April.

The dividend fund charges more for the hand-holding

The dividend fund is the more expensive one, and not by a rounding error. On a $1,000,000 balance, a 0.35% expense ratio costs $3,500 a year. A broad total-market fund at 0.04% costs $400. The gap between them is 0.31 percentage points, or $3,100 a year, taken out whether the fund has a good year or a bad one.

Over ten years that is $31,000 of drag, before you count what the money might have earned if it had stayed invested instead. Expense ratios are quiet that way. Nobody sends you an invoice for $3,100, so it never feels like a purchase.

What you buy with the extra cost is fewer decisions. The dividend lands in the account on its own schedule. The selling side needs one sale and a rebalance a year, roughly fifteen minutes of work, plus the willingness to actually sit down and do it.

That is not nothing for someone who does not want to be making choices about money at 74. Hand-holding has a price and the price is $3,100 a year on this portfolio. Whether it is worth it depends on whether you would otherwise make the sale at all.

A useful way to frame it: you are paying about $3,100 annually to have the income arrive without you. If the fifteen minutes a year is genuinely something you will do, the math favors the cheap fund. If it is something you will avoid, the cheap fund is not actually cheaper for you.

Your account type can erase the tax difference

The wrapper decides whether the tax gap between the two routes exists at all. Hold the same $1,000,000 inside an IRA or a 401(k) and the $35,000 dividend is not taxed in the year it lands, and the $40,000 sale of shares triggers no capital-gains bill either, because nothing inside a tax-advantaged account is taxed as it moves. The dividend's built-in tax cost, the thing that made it the weaker route, simply isn't there.

Which means the case for selling over dividends loses its sharpest argument. In a taxable account the gap is real and worth naming: the $35,000 distribution shows up as taxable income whether or not you spend it, while the $40,000 sale carries a $16,000 gain on a $60 cost basis, with the other $24,000 just your own principal coming back.

That is the number to run through your own tax picture, and IRS Publication 550 is where the rules live. The point isn't that one route wins. It's that the winner changes with the wrapper.

One more clock runs inside the tax-advantaged accounts. IRAs and 401(k)s require withdrawals starting at the IRS required-beginning age, so a dividend income plan in an IRA still has to pull money out on that schedule regardless of what the fund pays. Current rules are at IRS.gov. Settle the account type first, then pick the fund: the tax math you're comparing depends on it.

Dividends fit hands-off income; selling fits flexibility

Start with one subtraction on a single sheet of paper: the $40,000 you need against the $35,000 a 3.5% yield throws off on a $1,000,000 portfolio. That gap of $5,000, filled by selling about 50 shares at $100, is the whole decision in miniature. If automatically covering most of the need sounds like relief, lean dividend. If that $5,000 shortfall makes your shoulders tighten, lean toward selling.

For the retiree who wants income to arrive without a decision, dividends do exactly that. The $35,000 lands on the fund's schedule, and the top-up sale of roughly 50 shares is small enough to do once and forget. What you are buying is fewer decisions. What you pay for is the higher expense ratio, 0.35% against 0.04%, which is $3,500 a year versus $400 on the same $1,000,000, plus the annual tax on the full distribution whether or not you spend it.

The other temperament wants the dial. One sale a year, about 15 minutes including the rebalance, and full say over whether the number is $40,000, $25,000, or nothing at all. Lower fees and control over when gains are realized are the reward. The cost is the attention: no check arrives, and in a year when prices are down you decide the withdrawal yourself, at $92 a share instead of $100.

None of this holds until the account type is settled. Inside an IRA or 401(k), the annual dividend tax goes away and the selling side loses its tax edge, which pushes the choice back toward pure convenience. In a taxable account the gap between a $35,000 distribution and a $16,000 realized gain is real money, and worth running past a tax adviser. Check the account before you pick the method.

Frequently Asked Questions

Do dividends count as income for retirement?

Yes. Dividends are income in the year they are paid, and they are reportable on your tax return even if you reinvest them. How much tax you owe depends on the account type and your bracket, so check IRS Publication 550 and your own return.

Is it better to live off dividends or sell shares in retirement?

It depends on your account type and how much you want to decide. Dividends suit a retiree who wants a predictable check and will top it up with a small sale. Selling suits a retiree who wants to choose the amount each year and keep fees and taxes lower. Run both against your $40,000 need before deciding.

How much do I need invested to earn $40,000 a year in dividends?

At a 3.5% yield, about $1,143,000 covers the full $40,000 with no share sales. On $1,000,000, the same yield pays $35,000, and the remaining $5,000 comes from selling about 50 shares at $100.

Will selling shares run out my portfolio?

Not automatically, as long as the portfolio earns a return and the withdrawal rate stays reasonable. A $40,000 withdrawal from $1,000,000 is 4%, which is at the edge of what most plans assume. Confirm your own numbers with a fee-only fiduciary before you rely on a specific rate.

Can my dividend be cut in retirement?

Yes. A payout cut from $3.50 to $2.50 a share would drop income from $35,000 to $25,000. A cut usually arrives with a price drop, so income and portfolio value fall together. Keep a cash cushion so a cut does not force a sale in a down market.

Should I hold dividend funds in my IRA or my brokerage account?

In an IRA, the dividend is not taxed annually, and a sale triggers no capital-gains tax, so the tax edge of selling largely disappears. In a taxable account, the difference remains. Settle the account type first, then choose the fund, and check the current IRS rules at IRS.gov.

The $40,000 need, the 3.5% yield, and the account type decide it

The verdict splits on temperament, so state it plainly. Dividends suit the retiree who wants a $35,000 check to arrive without a decision and will top it up with a small sale of 50 shares. Selling suits the retiree who wants the full $40,000 dial, the lower 0.04% fee, and control over the tax bill.

Before choosing, do two things: confirm which account holds the money, since an IRA can erase the tax gap entirely, then run your own $40,000 need against a 3.5% yield on your actual balance. If you are still years out, a five-year retirement blueprint can walk you through the accounts and spending plan before you give notice.

The fear that selling shrinks a portfolio while dividends leave it whole does not survive the arithmetic. Both routes spend the same $40,000; they differ in value, control, tax, and fees, and no route is free.

So return to the question you started with. Your $40,000 either arrives as a $35,000 transfer plus 50 shares sold, or as 400 shares sold on a day you pick. Which one you can live with is the answer.