Week 36: Building an Income Portfolio — the L3 Capstone

---

1. Why This Is Important

Every other week in this tutorial has assumed you are accumulating — working, saving, and compounding. This week assumes the opposite. You have the pile. Now you need it to pay you. Whether that pile funds retirement at 65, supports a partner who stopped working, or simply covers the tuition cheque that arrives every September, the moment you flip from accumulation to distribution the optimisation problem changes shape entirely. You stop optimising for terminal wealth and start optimising for *cash that lands in the chequing account on schedule, after tax, with the smallest possible chance of running out*.

Four reasons this matters as the capstone for the L3 income block.

Income is mechanically harder than growth. A growth portfolio

only has to compound. An income portfolio has to compound and produce reliable cash and survive a bad first decade — the "sequence-of-returns risk" that the famous 4% rule was invented to answer. We will explain why the 4% rule is correct in spirit and misleading in detail in 2026.

Sources of yield are not interchangeable. A 5% Treasury coupon,

a 5% qualified dividend, a 5% REIT distribution, and a 5% covered-call premium look identical on a brokerage statement and are radically different products. They have different default risks, different inflation sensitivities, different drawdown profiles, and — most expensively — different tax treatments. A portfolio that ignores the difference can lose 30% of its spendable yield to the IRS for no good reason.

**The hierarchy of after-tax yield is the single biggest lever you

control.** Tax efficiency is the explicit anchor. In the 32% federal bracket, a qualified dividend keeps about 80 cents of every dollar. A bond coupon keeps about 63 cents. A short-term option premium keeps the same 63 cents — sometimes less, depending on state. The order in which you fill your taxable, IRA, and Roth accounts with these income streams is worth more than picking better individual tickers ever will be.

**The four-tranche framework and the barbell

need a distribution-side translation.** During accumulation, the four tranches govern how you allocate. During distribution they govern which sleeve you draw from this quarter. The income capstone is what stitches everything from Weeks 4-5 (60/40 and bonds), Weeks 14-15 (pair trades and barbell), and Weeks 26-28 (options as orders) into a single cash-producing machine.

This lesson lays out the four real income sources, the after-tax ladder, the canonical product menu (SCHD / VYM / DVY / SPYI for dividend equity, BND / VTEB / PFF for fixed income, VNQ for REITs, JEPI / JEPQ for premium-write), the 4% rule and its 2026 critique, and ends with two model portfolios — one taxable, one tax-advantaged — that target a 4-5% sustainable yield with bond-like volatility.

---

2. What You Need to Know

2.1 The Four Real Sources of Portfolio Income

Strip away every product wrapper and there are only four ways a US portfolio actually produces cash. Memorise this list. Every yield ETF ever launched is some packaging of these four ingredients.

Treasury coupons. The US government promises to pay. Default

risk is rounded to zero. State and local income tax do not apply. Federal tax applies at ordinary rates. The 2-year Treasury is the floor your safe sleeve cannot reasonably go below; the 10-year is the duration anchor for a balanced portfolio.

Investment-grade and high-yield corporate coupons. Companies

pay you to lend to them. IG (LQD, AGG's corporate slice, BND's corporate slice) yields 80-150 bps above Treasuries. HY (HYG, JNK) yields 250-450 bps above. Both pay ordinary income at the federal and state level, which is critical when you compute after-tax.

Qualified dividends. Long-held common stock dividends meeting

the IRS holding-period test are taxed at long-term capital-gains rates: 0/15/20% federal depending on bracket, plus state. The S&P 500 yields about 1.4% as of April 2026. Curated dividend ETFs (SCHD, VYM, DVY) yield 2.5-3.7%. The premium over the index is compensation for tilting toward "boring" sectors — staples, healthcare, financials, industrials — and away from the cap-light compounders that drove the post-2020 rally.

Option premium. Selling calls or puts (Weeks 26-28) creates

short-term income. With near-zero exception, equity-option premium is taxed as short-term capital gains at the federal and state level. The headline distribution yields on JEPI (~7-8%), QYLD, PUTW, SPYI, and the entire buy-write fund family are large precisely because the IRS gets a big bite first.

A fifth category you will see in marketing — REIT distributions — is partly category 2 (the REIT pays through interest and rent collected) and partly category 3 (qualified dividends from any equity holdings). REITs as a wrapper get a 20% Section 199A deduction at the investor level since 2018, which makes them somewhat tax-friendlier than raw bond coupons but still meaningfully worse than qualified dividends.

The yield-hierarchy chart shows the headline (pre-tax) numbers as of April 2026. Read it as the menu before tax.

2.2 The After-Tax Hierarchy and Why It Reverses the Menu

Pre-tax, JEPI's 7.8% looks like the obvious winner. After tax in a taxable account at the 32% federal bracket plus 5% state, the picture inverts. The next chart applies the right tax rate to each line item:

• Treasury coupons: 32% federal, 0% state ⇒ 32% effective.
• IG / HY corporate coupons: 32% + 5% = 37% effective.
• Qualified dividends (S&P 500, SCHD, VYM, DVY): 15% federal LTCG +

5% state = 20% effective.

• REIT distributions (VNQ): 37% combined, but the 199A deduction

reduces the taxable portion to 80%, giving roughly 29.6% effective.

• Option-premium ETFs (JEPI/QYLD/SPYI): 32% + 5% = 37% effective.

After-tax, SCHD's 3.6% becomes 2.9%. JEPI's 7.8% becomes 4.9%. The 11-bp gap on the y-axis is much smaller than it looked. And after you adjust for option-premium ETFs' historical underperformance versus their underlying index (Week 27, QYLD vs QQQ; Week 28, PUTW vs SPY), the apparent yield premium is mostly a tax leak you are paying *to yourself*.

The operational rule that falls out:

• Hold qualified-dividend equity in the taxable account — the

preferential rate is the whole point.

• Hold option-premium and HY-corporate sleeves in the IRA / 401(k) —

the IRS bite gets deferred (traditional) or eliminated (Roth).

• **Hold Treasuries in the taxable account if you live in a high

state-tax state** (CA/NY/HI) — the state exemption is a free 5%.

• Hold REITs in the IRA if your bracket is high enough that the

199A deduction does not fully offset your state rate.

This is location, not allocation. It is one of the few free lunches the US tax code gives you.

2.3 The Canonical Product Menu

Theory only matters if you can implement it with three or four tickers at a Vanguard, Fidelity, or Schwab account. Here is the April-2026 product menu by sleeve.

Dividend equity (qualified yield, growth-of-income).

• SCHD — Schwab US Dividend Equity. ~100 holdings, 10-year

dividend-growth screen, sector-balanced toward industrials, staples, healthcare. Yield ~3.6%, expense ratio 6 bps.

• VYM — Vanguard High Dividend Yield. ~440 holdings, market-cap

weighted within the high-yield half of the index. Yield ~2.9%, ER 6 bps.

• DVY — iShares Select Dividend. ~100 holdings, value-heavier than

SCHD. Yield ~3.4%, ER 38 bps.

• SPYI — NEOS S&P 500 High-Income. Index-tracking with a partial

call-write overlay; positions itself as a hybrid of dividend equity and option premium. Yield ~12% headline but the option-premium share inherits the option-premium tax problem.

The yield ladder is real. SCHD sits in the middle of the ladder by design — the dividend-growth screen filters out the highest-yielding names that are in distress. That is a feature, not a bug, and is why we treat SCHD as the default dividend sleeve for the model portfolio in §2.6.

Fixed income.

• BND — Vanguard Total Bond Market. Treasuries + IG corporates +

agency MBS, intermediate duration ~6 years. The default core bond.

• VTEB — Vanguard Tax-Exempt Bond. Investment-grade municipals.

Tax-equivalent yield rises sharply for high-bracket investors in taxable accounts; this is the right tool for the bond sleeve in a taxable account if your federal bracket is 32% or higher.

• PFF — iShares Preferred & Income Securities. Preferred stock,

yields 6-7%, but most preferreds are issued by financial firms so this sleeve is a bet on bank credit risk, not a pure bond.

• TLT / IEF — long and intermediate Treasury duration tools when

you want pure interest-rate exposure separated from credit.

Real assets / inflation hedge.

• VNQ — Vanguard Real Estate. The REIT sleeve. Yield ~3.8%,

inflation sensitivity comes through rent escalators and replacement- cost equity in real estate.

• SCHH — cheaper REIT alternative, similar profile.

Option premium (Weeks 26-28).

• JEPI — JPMorgan Equity Premium Income. ~80% defensive low-vol

equity + ~20% equity-linked notes that synthesise call-write premium. Yield ~7-8%, ER 35 bps. The single largest fund in the category.

• JEPQ — Nasdaq cousin of JEPI. Higher volatility, higher yield,

closer to QYLD's payoff shape but with active selection on the equity sleeve.

• QYLD / XYLD / RYLD — passive monthly at-the-money buy-write

on QQQ / SPY / IWM. Pure call-write exposure; the cap-on-upside is severe (Week 27).

• PUTW / WTPI — systematic put-write ETFs. Mirror image of

buy-writes; cap-on-upside replaced with cushion-on-downside (Week 28).

2.4 The 4% Rule and Why It Needs an Update for 2026

Bengen (1994) and the Trinity study (1998) gave retirees the most quoted rule of thumb in finance: with a 60/40 portfolio, you can safely withdraw 4% of the initial balance, increase that dollar amount by inflation each year, and have a 95%+ chance of not running out over a 30-year retirement. The rule is based on rolling 30-year windows of US data 1926-1995.

Three things have changed.

**Bond yields started Bengen's window above 7% and a typical

retirement window started above 5%.** The 2010-2021 zero-rate regime crushed the bond contribution to 4%-rule sustainability. Updates by Pfau and others put the safe withdrawal rate as low as 2.8% for portfolios starting in 2010-2020. The 2022-2024 yield reset has lifted that back to roughly 3.5-3.8%. April 2026 is the first time since 2008 that the math works close to the original number again, but it is not 4%.

Sequence-of-returns risk dominates. A retiree drawing 4%

through 1973-1974 (60/40 down a real ~30%) ran out of money in roughly year 22. A retiree drawing 4% through 1995-1999 (60/40 up a real ~140%) finished with roughly ten times the starting balance. The same rule, the same numerical sequence, the same asset mix — totally different lived experience.

Variable-spending rules dominate fixed-real spending. Guyton-

Klinger guard rails, the Bogleheads' "spend a percent of current balance," and Vanguard's "dynamic spending" all beat the rigid 4% rule by 50-100 bps of safe withdrawal in Monte-Carlo, with essentially no behavioural cost. This week's interactive uses a simplified version: target 4-5% of current balance, with a hard floor and ceiling.

The honest April-2026 number for a 60/40 portfolio with no discretionary-spending flexibility is roughly 3.7%. With the flexibility to cut 10-15% in bad years, 4.5%. The model income portfolio in §2.6 is engineered around that 4-5% target.

2.5 The Distribution-Side Four-Tranche Framework

The four tranches in accumulation translate cleanly to distribution. The same buckets exist; the direction of cash flow reverses.

Tranche	Accumulation purpose	Distribution purpose
1. Cash / T-bills	Emergency fund	1-2 years of spending, the "no-sell zone"
2. Bonds	Volatility ballast	5-7 years of spending; refilled from sleeve 3 in good years
3. Diversified equity	Compounding engine	The wealth machine; refills sleeve 2
4. Concentrated bets	Asymmetric upside	Optional yield overlays (JEPI, premium-write)

The barbell is the cross-section: short-duration safe income on one end (T-bills, 2-year Treasuries, VTEB if taxable), and long-duration real assets on the other (VNQ, equity dividend growers, optional premium overlay). The middle — long-duration intermediate Treasuries — is the least useful sleeve in a distribution portfolio relative to the two ends, because it has neither the nominal-rate sensitivity that long Treasuries give nor the inflation pass-through that equities and real estate give.

The interactive in §2.7 is the live version of this. You set the five-sleeve mix and get blended yield, after-tax yield (with bracket toggle), expected volatility, and expected drawdown.

2.6 Two Model Portfolios

Two reference points. Both target a 4-5% pre-tax distribution yield with bond-like volatility (8-10% annualised standard deviation). Both are buildable today at any major US broker for under 12 bps in weighted expense ratio.

Model A — Taxable, 32% federal bracket.

Sleeve	Ticker	Weight	Yield	Rationale
Dividend equity	SCHD	35%	3.6%	Qualified dividends, low ER, sector diversity
Dividend equity	VYM	10%	2.9%	Adds breadth, market-cap weight
Tax-exempt bonds	VTEB	25%	3.4% (≈5.0% TEY)	Federal exempt; lifts after-tax yield substantially
REITs	VNQ	10%	3.8%	Inflation pass-through; 199A partial relief
Treasuries	IEF (7-10yr)	15%	4.2%	State-tax-exempt anchor
T-bills	SGOV	5%	4.5%	Spending buffer
Blended	100%	3.7%	After-tax ≈ 3.0%

Volatility ~9% annualised; max-drawdown estimate -22% (1973-style inflation shock or 2022-style dual decline). After-tax distribution yield 3.0%; with a balance-of-portfolio variable-spending rule you can sustain 4.0-4.5% real spending.

Model B — Tax-advantaged (IRA / 401(k)), 32% bracket.

Sleeve	Ticker	Weight	Yield	Rationale
Dividend equity	SCHD	30%	3.6%	Qualified-dividend feature wasted; here for dividend-growth
Premium-write	JEPI	15%	7.8%	Highest pre-tax yield; tax irrelevant inside IRA
Total bond	BND	25%	4.6%	IG + Treasury blend
HY corporate	HYG	10%	7.5%	Credit premium; HY taxed punitively in taxable
REITs	VNQ	10%	3.8%	199A irrelevant inside IRA
Cash	SGOV	10%	4.5%	Buffer
Blended	100%	5.1%	After-tax = 5.1% (no tax inside)

Volatility ~10% annualised; max-drawdown estimate -25%. The IRA portfolio runs higher pre-tax yield because every income source is taxed identically — so you load up on the punitively-taxed sleeves exactly here, where the punishment does not apply.

The interactive lets you build either, and the bracket toggle shows the Model-A and Model-B after-tax numbers side by side.

2.7 The Live Builder

The interactive is a five-sleeve allocator: Treasuries, IG corporate, qualified dividends, REITs, premium-write. It computes pre-tax yield (weighted average), after-tax yield (weighted by tax category at the selected bracket), expected volatility (square-root of weighted covariance), and expected maximum drawdown (heuristic: 0.6 × peak historical loss for each sleeve, blended by weight). The tax-bracket toggle covers 12% / 22% / 32% / 37%, which is the four brackets where real allocation decisions get made. Use it to verify the §2.6 model portfolios; use it to test your own.

---

3. Common Misconceptions

"High yield is always better." No. Pre-tax yield without an

after-tax conversion and a default-risk adjustment is marketing-grade arithmetic. JEPI's 7.8% in a taxable account at the 37% bracket is 4.9%. SCHD's 3.6% in the same account is 2.9%. The gap is one-third of the headline.

"Treasuries are tax-free." Treasuries are state-tax-free. They

are fully federally taxable. In Texas, Florida, or Tennessee that distinction is meaningless; in California or New York it is worth roughly 5% of the coupon.

"Municipals are always better than Treasuries." Only if your

federal bracket × (1 - state rate adjustment) makes the tax-equivalent yield higher. At the 12% federal bracket, munis almost never win; at the 32% federal bracket, they almost always do.

"Covered-call ETFs have free yield." No. The yield is the

pre-paid sale of upside (Week 27). Over a full cycle, JEPI and QYLD lag their unhedged underlyings by 300-500 bps annualised.

"REIT dividends are qualified dividends." Mostly false. Most

REIT distributions are non-qualified ordinary income, partially sheltered by the 20% 199A deduction.

"Dividend stocks are safer than the index." Sometimes. The

dividend-paying half of the S&P 500 has a lower beta than the index, but in a credit/financial crisis (2008) banks and REITs — which dominate dividend ETFs — drop further than the index.

"You should reinvest dividends in retirement." If you actually

need the cash, no — that is buying with one hand and selling with the other, and creates avoidable tax in a taxable account.

"4% withdrawal works in any market regime." It works for the

1926-1995 sample of US data. It has failed in retro-tested non-US samples and in zero-rate windows. April 2026 has the math close to working again; for low-flexibility retirees, plan for 3.5-4.0%.

"Bonds are always negatively correlated with stocks." Week 4

covered this in detail. The 1970s and 2022 are existence proofs of the opposite.

**"You should pick the sleeve with the highest yield first and

fill in the others."** Backwards. Pick the sleeve with the highest *after-tax expected return per unit of drawdown contribution* first.

---

4. Q&A Section

**Q1: My broker is showing JEPI's "30-day SEC yield" as 8.4% but the "distribution yield" as 7.6%. Which is right?**

Distribution yield is what landed in your account over the trailing 12 months divided by current price. SEC yield is a regulator-defined forward-looking estimate based on the most recent month's cash flow, annualised. For a stable holding, the two should be within 50 bps. For premium-income funds where the option-premium varies with VIX, the SEC yield can be misleading in either direction. Use distribution yield for budgeting; use SEC yield for cross-fund comparisons.

Q2: Should I buy individual bonds or BND?

For amounts under roughly $100k of bond exposure, BND. The fund has ~10,000 bonds and you cannot replicate that diversification at retail sizes. Above that, individual Treasuries on the auction market are attractive — you control the maturity, you can ladder, and you skip the 3 bp expense ratio. Individual corporate bonds at retail are a trap: spreads are wide, liquidity is thin, and a credit event sinks the position completely.

**Q3: What about a "60/40 of yield" — half qualified dividends, half bond coupons?**

That is essentially Model A in §2.6. The distribution yield works out to 3.7% pre-tax, ~3.0% after-tax in the 32% bracket. The volatility is bond-like (~9%). For a baseline retiree-style portfolio that is the right ballpark — but tilt the bonds toward VTEB if you are taxable, and prefer SCHD over VYM for the dividend-growth tilt.

Q4: Is the "yield-on-cost" metric useful?

For pep-talk purposes, yes. For decisions, no. Yield-on-cost rises because the price you paid is fixed and the dividend grows. It tells you nothing about whether your current capital is generating an adequate yield versus alternatives. Always compare current distribution yields, not cost-basis yields.

Q5: How do I know if a high yield is sustainable?

Three quick screens. (a) Payout ratio: dividend / earnings under 70% for industrials and staples, under 90% for utilities, under 95% for REITs. (b) Free-cash-flow coverage: dividends plus buybacks under free cash flow. (c) The fund's distribution history: did it cut in 2008, 2020, 2022? A buy-write fund cutting in 2022 is normal; an equity-dividend fund cutting is a signal.

Q6: What about preferred stock?

PFF yields ~6.5% and feels like a bond. It is not. Preferreds are junior to bonds, senior to common, and almost entirely issued by banks and insurers. The 2008 crisis was a near-total wipe of bank preferreds. Sized at 5-10% of a bond sleeve they are a fine yield enhancer; sized at 30%+ they are a concentrated bank-credit bet.

Q7: Should I use leverage on bonds (NTSX, RPAR, etc.)?

Levered "stocks-and-bonds in the same fund" products amplify the correlation assumption. They worked beautifully 1995-2021 and melted down in 2022 when stocks and bonds fell together. They are an accumulation tool, not a distribution tool. Do not use them for the income sleeve.

Q8: How often should I rebalance an income portfolio?

Annually, plus a calendar-driven cash-refill of the spending-buffer sleeve. The "smart" rule is: in a year where the equity sleeve is up sharply, refill 18 months of spending into cash and bonds; in a flat or down year, do not sell equities — draw from cash and bonds. This is the four-tranche framework working in reverse.

Q9: What about international dividend ETFs (VYMI, IDV)?

Our default is US-only investable. International dividend ETFs add concentration in financials and energy and introduce withholding-tax leakage that takes a bite out of the qualified-dividend rate. If you want non-US exposure, take it on the total-return side of the portfolio (the equity growth sleeve), not the income sleeve.

Q10: Is there a fund-of-funds that does all of this for me?

The closest is VTINX (Vanguard Target Retirement Income), AOK (iShares conservative allocation), or JAAA + JBBB (Janus Henderson AAA / BBB CLO funds for a credit-tilted income sleeve). All of them solve part of the problem and none solve the account-location problem in §2.2 — which is the largest source of after-tax yield uplift available to you. Build it sleeve by sleeve.

Q11: How much cash should I hold?

For a retiree, 1-2 years of spending. For a working accumulator, 3-6 months. The reason for the asymmetry is the tail. A retiree who is forced to sell equities at a 30% drawdown to fund this month's groceries is destroying compounding more than any 1% cash drag could cost. The cash sleeve is insurance against the bad-sequence tail.

Q12: What about annuities?

A single-premium immediate annuity (SPIA) at age 75-80 looks much better than the 4% rule for that portion of the portfolio you genuinely need to floor. The right framing is not "annuity vs no annuity" but "what fraction of base spending does Social Security cover, and do I need to top up the rest with a SPIA?" That is a financial-planning question more than an investment one — and a fine reason to talk to a fee-only fiduciary at the right age.