Side Lesson 10: Dividends — Qualified vs Ordinary, Dividend Growth, and the SCHD/VYM/DGRO Landscape

---

1. Why This Is Important

Dividends are one of those topics where the folk-wisdom and the math disagree, loudly. Over the long run roughly 30-40% of the S&P 500's total return has come from reinvested dividends; once you compound the reinvestment, the figure climbs above 80% of cumulative terminal wealth from the 1930 base. That is a large slice of equity returns to hand-wave through. At the same time, an academic in any decent finance department will tell you that — in a frictionless world — the choice between paying a dividend and retaining the cash is irrelevant, because you can replicate either one by selling shares or reinvesting. Both statements are simultaneously true. The job of this lesson is to live inside that tension without pretending it is not there.

Four reasons to take dividends seriously even after you have read Modigliani-Miller:

Tax treatment is not irrelevant. A qualified dividend taxed

at 15% is materially better than an ordinary dividend taxed at 32% plus 5% state. Modigliani-Miller assumed no taxes; you live in a world that has them. Where a payment lands on the IRS's two parallel rate schedules changes the after-tax cash flow by hundreds of basis points per year.

Dividend growth is a quality signal, not a yield signal. A

stock that has raised its payment 25 years in a row had to grow earnings to support it. The Aristocrats list and the funds tracking it (NOBL, SCHD, DGRO) lean toward profitable, low-leverage, low-beta names. The yield is the by-product; the discipline is the feature.

Behaviour beats theory. A 70-year-old whose paycheck stops on

Friday cannot live on "the math says you can sell 4% per year." A distribution that arrives in cash without an active sell decision is, for many retail investors, the difference between staying invested through a 30% drawdown and capitulating at the bottom. That preference is not irrational; it is just not in the textbook. The old line — *the market can stay irrational longer than you can stay solvent* — applies to the investor too.

Dividend funds are a real allocation, not a meme. SCHD alone

crossed $70B in AUM by 2025, with VYM, DGRO, and NOBL adding another ~$120B between them. This is one of the largest single strategy categories in the ETF market. Knowing how the four canonical funds differ — SCHD's quality screen, VYM's market-cap-weighted yield, DGRO's growth-rate filter, NOBL's 25-year streak — is part of being literate in the income sleeve of a four-tranche portfolio.

This lesson is the operating manual for the income tranche — the Income slot in the four-tranche stack. You will not learn whether dividends are "good" or "bad"; that is the wrong question. You will learn which payments qualify for the lower schedule, what dividend growth actually buys you, how to tell the canonical ETFs apart, and where the math really does say "irrelevant" versus where the IRS code says "actually, that costs you 17 points."

The interactive panel on the website lets you set a starting yield, a dividend-growth rate, a holding horizon, and a federal bracket. It returns the yield-on-cost at year N, the cumulative dividends collected, and the after-tax cash. Use it after you finish the prose.

---

2. What You Need to Know

2.1 Qualified vs Ordinary — The 60-Day Holding Window

A dividend is qualified — and therefore taxed at the preferential long-term capital-gains schedule (0%, 15%, or 20%, plus the 3.8% NIIT above $200k single / $250k MFJ) — when all three of the following hold:

The payer is a US C-corporation or a **qualified foreign

corporation** (treaty country, or trades on an established US exchange via ADR).

You held the underlying share for **more than 60 days during the

121-day window centred on the ex-dividend date** (so the stock must be in your account for at least 61 days inside that 121-day block).

You did not hedge the position with a deep-in-the-money short call

or a synthetic short during the holding window. (Constructive sale rules under IRC §1259; the IRS counts a hedged long as not "really" held.)

A dividend that fails any of those tests is ordinary and is taxed at your regular wage rate — up to 37% federal plus state plus NIIT, i.e. as much as 44.6%. The biggest ordinary-bucket sources are:

• REIT distributions (most of them; the 199A 20% qualified

business income deduction takes the effective rate down by 20% but does not move them onto the LTCG schedule).

• MLP distributions (mostly return-of-capital, with the rest

taxed as ordinary income on the K-1).

• Foreign companies not traded on a US exchange (if you bought

the local-listed shares directly via a foreign broker, the dividends are usually ordinary).

• Bond ETFs paying "dividends" (these are interest payments

dressed up as distributions; always ordinary).

• Money-market and HYSA "dividends" (interest).

The rule that catches retail investors is the 60-day holding requirement. Buy a stock the day before its ex-date, sell two weeks later — the dividend is non-qualified, and you owe ordinary rates on something you only held 14 days. Schwab and Fidelity flag this on the 1099-DIV, but the system reports the dollar split (Box 1a vs Box 1b); it does not retro-fix your cost basis. The 60-day rule is why dividend-capture trading strategies almost never work after tax.

2.2 The 2026 Rate Stack

Two parallel schedules, applied to the same dollar of investment income depending on which bucket it falls in.

Ordinary schedule (wages, interest, REIT distributions, non-qualified dividends, short-term capital gains; 2026 single filer):

Bracket	Taxable income
12%	$12,150 - $50,400
22%	$50,400 - $107,825
24%	$107,825 - $206,000
32%	$206,000 - $262,000
35%	$262,000 - $657,000
37%	above $657,000

LTCG / qualified-dividend schedule:

Rate	Taxable income (single)
0%	up to $49,450
15%	$49,450 - $545,750
20%	above $545,750

Plus a flat 3.8% NIIT on top of either schedule for households above $200k single / $250k MFJ.

Two practical corollaries:

• A 32%-bracket retiree taking $4,000 of SCHD dividends keeps

$3,200 (15% LTCG + 5% state). The same retiree taking $4,000 of VNQ (REIT) distributions keeps $2,624 (32% × 0.80 from 199A + 5% state). Same headline yield, almost 18% gap in cash to the household.

• A 12%-bracket investor pays zero on qualified dividends.

This is the single largest legal free lunch in the US tax code for low- and middle-income retirees. Until you cross $49,450 of taxable income, your SCHD distribution is tax-free. Asset location (this dividend in which account) is the lever that lets you stay in that bracket.

2.3 Yield vs Dividend Growth — Two Different Questions

The single most common framing error in dividend investing is to read the trailing-twelve-month yield (TTM) as if it were the rate of return. It is not. Yield is one term of two:

$$ \text{Total return} \approx \text{Dividend yield} + \text{Capital growth} $$

For a slow-moving dividend payer the second term tracks the first through dividend growth: a company that grows its dividend 8% per year tends to compound its share price at roughly the same rate over long windows, because the payout policy is anchored to earnings.

The four canonical filters and what each one is actually optimising for:

ETF	Index	Yield (Apr 2026)	5y div CAGR	What it screens for
SCHD	Dow Jones US Dividend 100	~3.6%	~10%	Quality + 10y dividend track + cash-flow ROE
VYM	FTSE High Yield ex-REIT	~2.9%	~6%	Top half of S&P 500 by yield, cap-weighted
DGRO	Morningstar US Div Growth	~2.3%	~10%	5y consecutive dividend increases, payout < 75%
NOBL	S&P 500 Dividend Aristocrats	~2.0%	~7%	25 consecutive years of dividend increases
SPYI	S&P 500 + monthly call overwrite	~12.0%	~0%	Premium-write income, not dividend growth

The point of putting SPYI on the same bar chart is the contrast. SPYI is a covered-call ETF (course/week27_covered_calls.md); its 12% distribution comes from option premium, not from companies growing their cash flows. The dividend growth column for SPYI is roughly zero — the distribution rises and falls with implied volatility, not with corporate earnings. That is a different kind of income, taxed differently (mostly short-term capital gains and return-of-capital, none of it qualified), and it should occupy a different sleeve of the four-tranche stack than SCHD or VYM.

Yield-on-cost (YoC). A position with a 3.6% starting yield that grows the dividend 10% per year reaches a YoC of:

$$ \text{YoC}_N = y_0 \cdot (1+g)^N $$

After ten years that is 3.6% × 1.10^10 = 9.34%. After twenty years it is 24.2% on the original cost. This is the math that makes dividend-growth investing emotionally durable: a retiree in 2046 who bought SCHD in 2026 is — on paper — collecting nearly 25% of her original investment in cash dividends every year, even though the current yield of SCHD has not moved. YoC is a vanity metric for performance attribution, but a real one for lifestyle planning.

2.4 The Modigliani-Miller View — Why Some People Hate Dividends

Here is the classical argument, in one paragraph. In a world without taxes or transaction costs, a $100 stock that pays a $4 dividend on the ex-date becomes a $96 stock plus $4 of cash in your account. You were holding $100 of value before; you are holding $100 of value after. If you preferred no payment, you reinvest the $4 and you are back to $100 in stock. If your neighbour holds $100 in a non-paying stock and wants $4 of cash, she sells $4 of stock and is also holding $96 plus $4. The two policies are identical in wealth terms; the only difference is the path. Modigliani and Miller (1961) proved this formally and won a Nobel Prize for it.

In the real world the assumption breaks in three places:

Taxes. A forced dividend is a forced taxable event. The

investor who sells $4 of stock chooses when to realise the gain; the investor who receives a $4 dividend has no such choice. At a 32% ordinary rate (non-qualified) that is a 1.28% friction per year imposed on you. Tax via the wrapper matters.

Transaction costs. Selling $4 of a $100 stock costs nothing

on a commission-free broker, but it costs attention — a real resource for many investors. The dividend is automatic. The "rip $4 off the top yourself" prescription is theoretically clean but behaviourally fragile.

Information signalling. A board that raises the dividend is

also signalling "we expect to keep earning this". A board that cuts is signalling the opposite. The market reads the change in payout as a forecast about future cash flows, and the empirical literature (Lintner 1956, Brav-Graham-Harvey-Michaely 2005) confirms that managers know it and manage accordingly. Pure irrelevance would imply boards do not care about dividend policy; they very obviously do.

The honest synthesis is: in a frictionless model, dividends are neutral; in the real world, the tax friction is real but quantifiable, and the behavioural value of the automatic payment is also real but not in the textbook. Both sides are right about something.

2.5 Foreign Withholding — The Canadian and European Cases

A US investor holding a foreign dividend payer will see foreign withholding tax deducted before the cash hits the brokerage account. The two cases retail investors actually meet:

• Canadian dividend payers (ENB, BCE, RY, TD). Canada withholds

15% at source under the US-Canada tax treaty (it would be 25% without the treaty). The 15% appears as "foreign tax paid" in box 7 of the 1099-DIV. You claim it back via the **Foreign Tax Credit (FTC)** on Form 1116.

• European payers (Nestlé, Novartis, Shell B-share via ADR).

Most treaty rates are 15%; Switzerland withholds 35% but allows reclaim via Form 88 (most retail investors never bother).

The FTC has two structural limits that catch beginners:

It is a credit, not a refund. You can offset it only against

your US tax bill. If your US tax owed in a given year is zero (e.g. you live primarily on Roth distributions), the foreign withholding is dead-weight loss. This is why holding foreign dividend payers in a Roth IRA is a worse outcome than holding them in a taxable account: in the taxable account you reclaim the 15%, in the Roth you eat it.

The credit is sleeved by income category. Passive-category

foreign tax credits offset passive-category US tax owed. Excess credits carry back one year and forward ten, but for most retail investors with modest foreign exposure the credit fully offsets in the year paid.

The general rule, restated: **foreign dividend payers belong in a taxable account so that the withholding is recoverable**; broad US dividend payers can live in any account; REITs and bond payers go into the IRA stack. (course/side04_tax_efficiency.md has the full asset-location matrix.)

---

3. Common Misconceptions

"High yield = high return." No. A 12% yield from a covered-call

ETF is not the same animal as a 3% yield from a dividend grower. Cover the difference with the dividend-growth column of the ETF table; the SPYI total return over 5 years has trailed SCHD by several percentage points despite the four-times-larger headline.

"Dividends are 'free money'." They come out of the share

price on the ex-date, dollar-for-dollar. Modigliani-Miller is right about that part.

"Reinvesting dividends in a taxable account is tax-free." No.

The dividend is taxed in the year paid, whether or not you reinvest it. DRIP is not a tax shelter; it is an autopilot for the reinvestment leg, nothing more.

"REIT dividends are qualified." Almost never. REIT

distributions are mostly ordinary income; the 199A 20% deduction is a partial offset, not a re-classification.

**"Foreign dividends should go in the Roth because they're

tax-advantaged."** Backwards. Putting foreign dividends in the Roth eats the foreign withholding with no offset. Foreign payers belong in taxable.

"Dividend stocks are safer than the S&P." Sometimes. Aristocrats

like KO, JNJ, PEP, PG do tend to have lower beta. But "safe" is regime-dependent: in 2022 the S&P 500 fell ~18% and many high-yield names (utilities, especially) fell more because they are long-duration bond proxies (course/week34_rate_shock_grid.png).

"You should chase the highest-yielding stock in a sector."

The highest yield in a sector is usually the highest yield because the market is pricing a dividend cut. Both AT&T (2021) and General Electric (2018) screened "best yield in their sector" right before they cut. Yield is a result; quality is the input.

"NOBL is the best dividend ETF because Aristocrats." The 25-year

filter is mechanical, not predictive. NOBL excludes payers like Microsoft (started in 2003) and Apple (re-started in 2012) that have been the dominant capital-return stories of the last decade. Filtering rules have side effects.

---

4. Q&A Section

**Q1. If I bought a stock on its ex-dividend date, do I get the dividend?** No. The ex-date is the cut-off; the buyer on or after the ex-date is buying the stock without the right to the next payment. The price drops by the dividend amount on the ex-date specifically because of this.

Q2. What is the holding-period rule again, in plain English? You must hold the stock more than 60 days inside a 121-day window that straddles the ex-date (60 before, 60 after, plus the ex-date itself). Hold less, and the dividend is taxed at ordinary rates even if the company is a perfectly normal US C-corp.

Q3. Why did my SCHD distribution drop in Q2 of last year? Distributions vary quarter to quarter as the underlying companies declare on their own schedules. SCHD's annual total grows at roughly its 10% trailing CAGR, but a single quarter can be 30% above or below the smoothed line.

Q4. Are MLP distributions dividends? No. They are partnership distributions and arrive on a K-1, not a 1099-DIV. Most of the cash is technically a return-of-capital that reduces your basis until you sell, then is recaptured as ordinary income. Not friendly to retail.

**Q5. What is the difference between dividend growth and dividend yield investing?** Yield investing optimises today's cash; growth investing optimises the trajectory. SCHD blends both screens. VYM is more yield-tilted. DGRO is more growth-tilted. NOBL is the strictest growth filter (25-year streak) and consequently has a lower headline yield.

Q6. Will dividend ETFs underperform in a tech-led market? Often, yes. SCHD trailed VOO by 7-10 percentage points per year in the 2020-2024 window because the dominant winners (NVDA, MSFT, GOOGL, META, AMZN) either pay no dividend or do not pass the SCHD quality screen. This is a feature, not a bug; you do not buy SCHD to win when growth wins. You buy it for the income tranche.

Q7. Where should I hold SCHD? Either a taxable account (qualified dividends taxed at LTCG rates) or a Roth (tax-free forever). A Traditional IRA is fine but slightly suboptimal because you turn LTCG-rate cash flows into ordinary income on withdrawal. The location loss is roughly 5-10 bp/year.

Q8. Where should I hold REITs? Traditional IRA or Roth. Never a regular taxable account at a high bracket if you can avoid it. The 199A deduction helps but does not fully fix the ordinary-rate problem.

**Q9. What is "yield-on-cost" and why do dividend investors talk about it?** YoC is the current annual dividend divided by your original cost basis. After 20 years of 10% dividend growth on a 3.6% starter yield, your YoC is 24%. It is psychologically powerful for retirees because it makes the long-term compounding tangible — even if it has no implication for forward expected return.

Q10. Do dividend ETFs work outside the US? Yes, but US investors should stick to US-listed wrappers. The US is the only equity market with the liquidity, governance, and ETF infrastructure where retail can operate confidently. SCHD, VYM, DGRO, NOBL, and SPYI are all US-listed and US-domiciled. Do not buy the Irish-domiciled "international clones" — you trade foreign withholding for PFIC-status nightmares.

**Q11. What is the right size for the income sleeve in a four-tranche portfolio?** For a 50-year-old with a 15-year accumulation runway, ~25% of total financial assets in the income tranche is a defensible default. For a 65-year-old in early retirement, 35-45% is more typical. The split within the sleeve between dividend ETFs, bonds, and premium-write is what course/week36_income_portfolio.md walks through.

**Q12. Is there a dividend version of "the market is wrong about this stock"?** Yes — it's called the dividend trap. A 9% yield in a sector where peers yield 3% is the market saying "we are pricing a 67% chance of a cut." Sometimes the market is wrong; usually it is not. Verify the payout ratio, free-cash-flow coverage, and balance-sheet leverage before assuming the yield is real.