Tax strategy · Wealth transfer

Buy, Borrow, Die — how Larry Ellison uses Oracle stock as a tax-free ATM

Published 2026-09-03 · Anchor source: 24/7 Wall St. — "Buy, Borrow, Die: How Larry Ellison Can Borrow Against Oracle Stock Instead of Selling It" (Sept 3, 2026)

Ultra-wealthy holders of appreciated assets rarely sell. They buy and hold (unrealized gains aren't taxed), they borrow against the position as collateral (loan proceeds aren't taxed), and they die — at which point heirs inherit at a "stepped-up" cost basis under IRC §1014, and the lifetime capital gain evaporates from the income tax system entirely. Larry Ellison, co-founder of Oracle, is a textbook example. But the more useful part: the same mechanic is more accessible on real estate than on stocks, and many middle-class families with a long-held home have the ingredients without knowing it.

Noesis is not advocating this strategy.

Buy-Borrow-Die is a legal, factual pattern that ultra-wealthy individuals and long-tenured homeowners use to convert appreciated assets into spendable cash without triggering income tax. Whether it fits any particular reader depends on their assets, age, state, family situation, and risk tolerance. This is education, not tax, legal, or investment advice. Consult a CPA and estate attorney before acting on any of this. Any jargon you don't recognize — SBLOC, HELOC, LTCG, NIIT, QCD — is linked to the Noesis glossary.

Never sell what you can borrow against, and pass it to your heirs at death so the appreciation is never income-taxed.

That single sentence describes how a meaningful share of the top 0.1% of US household wealth funds its lifestyle. It rests on three provisions of the US tax code — one of which (step-up in basis) is the most-attacked provision in every Democratic reform proposal of the last decade.

The stock version (Ellison)

Step 1: Buy

Ellison co-founded Oracle in 1977. Whatever cost basis he has in his Oracle shares is negligible compared to today's market value. Every dollar of appreciation is an unrealized gain. Unrealized gains are not taxable — the IRS taxes realized events (sales, dividends, interest), not paper value.

Step 2: Borrow

Ellison pledges some fraction of his Oracle shares to a private-bank SBLOC or margin account. The bank hands him cash. Loan proceeds are not income under IRC §61, so the cash arrives untaxed. Typical ultra-wealthy SBLOC pricing: SOFR + 100–200 bps (~6.0–7.0% in 2026), 40–70% LTV, interest-only or accrue-to-principal. Provider: Morgan Stanley, Goldman Sachs private wealth, JPMorgan private bank, UBS, Bank of America Merrill.

Step 3: Die

At Ellison's death, his heirs inherit the Oracle shares. Under IRC §1014, their cost basis is not what Ellison paid in 1977 — it's the fair market value on the date of Ellison's death. If Oracle is at $200 on that date, heirs' basis is $200/share. Sell the next day at $200 → capital gain = $0. The lifetime capital gain — potentially $10 billion+ in Ellison's case — is never income-taxed by anyone, ever.

The math on a $1B slice

Suppose Ellison wants to fund $1B of spending. Three ways to do it:

Path	Federal tax	CA state tax	Net to Ellison	At death
Sell $1B of Oracle	23.8% ($238M)	13.3% ($133M)	~$629M	Nothing left to step up
Borrow $1B at 6.5% for 10 years	$0	$0	$1,000M	Owes ~$847M interest cumulative; heirs pay off from stepped-up estate; lifetime gain wiped
Traditional advisor plan — trim + diversify	Same as sell	Same as sell	Same	Reduced concentration risk but tax bill locked in

The borrow path costs ~$847M in interest over ten years but saves ~$371M in taxes immediately AND preserves the position for a full step-up at death. As long as Oracle compounds faster than 6.5% (net of dividend tax), the strategy wins by a wide margin.

IRC §1014 explained — the rule the whole strategy rests on

The whole strategy hangs on §1014. Understanding it is the difference between "how do rich people do this?" and "oh, I see."

First: what "cost basis" means

Cost basis is what you paid for an asset. When you sell:

Capital gain = Sale price − Cost basis

Tax owed = Capital gain × capital gains rate

Then: what §1014 does

When someone dies and passes an asset to their heirs, the heirs' cost basis resets to the fair market value on the date of death. The lifetime gain the deceased accumulated disappears from the income tax system.

Concrete Ellison example

Suppose Ellison bought 1M Oracle shares in 1986 at $0.10/share (split-adjusted, approximate).

Cost basis: $100,000
Current market value: $146,350,000 (1M × $146.35)
Unrealized gain sitting inside: $146,250,000

If Ellison sells today (age 82):

Federal LTCG 23.8%: $34.8M
California state 13.3%: $19.4M
Total tax: ~$54.2M
Nets ~$92M

If Ellison dies and shares pass to son David:

David inherits at $146.35/share
David's new basis: $146,350,000
Sell next morning at $146.35 → gain = $0, tax = $0
IRS never collects the $54M

Still owed at death — estate tax

Not literally free. Estate tax exemption is $15M individual / $30M married in 2026; above that, 40% flat. For Ellison specifically, estate tax on tens of billions is large — but it's still less than income-tax-on-sale + estate-tax-on-remainder would have been.

Does NOT apply to

Traditional IRA / 401(k) — heirs owe income tax on distributions
Roth IRA — no step-up needed (already tax-free)
Annuities — heirs still owe income tax on gains

This is why the strategy is useless if your wealth is trapped inside retirement accounts.

Three legitimate moves that rhyme with what Ellison does

Ellison's version needs tens of billions of dollars. But the underlying mechanics — never sell what you can borrow against, use step-up when possible, harvest the 0% LTCG bracket when eligible — are available at almost every wealth level. Three moves any tax-aware household can consider (with a fiduciary, not from an article):

1. Hold the low-basis lot until death

If you have a stock, fund, or property with decades of accumulated gain, leaving it to heirs preserves the step-up in basis. Selling it in your 70s to "simplify" may hand the IRS money your family would otherwise keep. The instinct to consolidate before you die often costs more in tax than the emotional simplicity is worth.

2. Use the 0% long-term capital-gains bracket

Retirees with taxable income below the 0% LTCG threshold (roughly $96,700 for a married couple in 2026, after deductions) can realize gains at a zero federal rate — and use the sale proceeds to buy back the same position with a stepped-up basis. Especially valuable for early retirees between 59½ and Social Security, when income is intentionally low.

3. Borrow, don't sell, for short-term needs

A HELOC or a modest SBLOC can bridge a cash gap without forcing a taxable sale of appreciated shares — provided you can service the interest and tolerate the collateral risk. Best for temporary needs (waiting for a house sale to close, funding a semester of tuition, covering an emergency medical bill). For permanent needs (funding ongoing retirement spending indefinitely), the math often flips back toward selling in low-tax years.

None of these is a strategy on its own — they're moves that fit into a coherent lifetime plan. The plan is what a fiduciary is paid to build.

Real estate application — where this pattern is actually accessible

For most readers, the Ellison version is theoretical (needs a $2M+ concentrated appreciated stock position in a taxable account). But the real estate version is much more accessible, and in some ways more powerful per dollar because of a second tax layer step-up also erases: depreciation recapture.

Three amplifiers stack on top of the plain step-up

Amplifier 1: Depreciation recapture is wiped out

If you own rental property, the IRS lets you deduct depreciation every year (residential = 27.5 years straight-line; commercial = 39 years). When you sell, all that depreciation is "recaptured" and taxed at up to 25% federal — higher than the LTCG rate. Step-up at death wipes out both the capital gain and the recapture liability. Heirs inherit as if newly purchased, and get to start depreciating all over again.

Amplifier 2: Real estate is natively borrowable and interest is often deductible

You don't need a private-bank SBLOC. Every homeowner has access to HELOC, cash-out refinance, DSCR loans for rental property, portfolio loans, or a reverse mortgage after age 62. Mortgage interest deductibility can drop effective borrowing cost by 20–37% depending on tax bracket. Investment property mortgage interest is fully deductible against rental income.

Amplifier 3: §1031 "swap till you drop"

§1031 lets real estate investors trade one investment property for another "like-kind" property and defer all capital gains + depreciation recapture indefinitely. Do it enough times, then die: heirs inherit at stepped-up basis. Every deferred gain from every prior exchange vanishes.

Concrete example — a middle-class California homeowner

The Ellison numbers are unrelatable. This one isn't:

Bought Bay Area house in 1985 for $300,000
Current market value (2026): $2,500,000
Unrealized gain: $2,200,000

A. Sell during retirement

LTCG on $2.2M − $500k §121 exclusion = $1.7M taxable. Fed 23.8% + CA 13.3% ≈ 37% blended → ~$629k tax. Net: ~$1.87M.

B. HELOC $500k at 8% for retirement

Annual interest ~$40k. Zero tax on $500k cash received. After 15 years: ~$600k cumulative interest, $500k principal consumed. Still owns the house.

C. Die, pass to kids

Kids inherit at $2.5M stepped-up basis. Sell next month at $2.5M → tax: $0. Estate pays off $500k HELOC. Kids net $2M cash, tax-free.

A vs. C: family "saved" ~$629k in taxes AND got to keep the house appreciating for 15 more years.

Rental property — where the numbers get dramatic

Four-plex bought in 1995 for $500,000
Current market value: $2,000,000
Accumulated depreciation (27.5 years, fully depreciated): $470,000
Adjusted cost basis: $30,000

If sold today:

Federal LTCG on $1.5M appreciation: 23.8% = $357k
Federal recapture on $470k depreciation: 25% = $118k
California (13.3% on both): ~$261k
Total tax: ~$736,000

If owner cash-out-refinances $1M at 7% instead, then dies: heirs inherit at $2M basis. Step-up wipes out BOTH the $1.5M gain AND the $470k recapture liability. Heirs can start depreciating $2M all over again. Tax the IRS never collects on this one small property: ~$736,000.

Stocks vs. real estate through the same lens

	Stocks (Ellison-style)	Real estate
Borrowing rate	SBLOC ~SOFR + 1–2% (~6–7%)	HELOC/refi ~6–9%
Interest deductibility	Investment interest, capped at investment income	Mortgage interest deductible up to $750k principal; fully against rental income
Liquidity	High — sell any %, any day	Low — can't sell 10% of a house
Margin call risk	Yes — pledged shares get sold if price drops	No margin calls; foreclosure risk if you can't service debt
Depreciation recapture	N/A (stocks don't depreciate)	25% federal on all accumulated — step-up erases this
§1031 defer-forever	No stock equivalent	Yes — swap till you drop
Access	Needs $2M+ concentrated position in taxable account	Every homeowner with equity qualifies
State wrinkle	State income tax on sale (uniform)	Property tax reassessment on inheritance varies

The four risks

Risk 1: Margin call (or foreclosure equivalent)

Not hypothetical — as of Sept 2 2026, Oracle is down 36.79% over the past year. Pledged shares that drop through the LTV covenant get sold by the lender to restore the ratio. That forced sale is exactly the taxable event the strategy exists to avoid. Real estate version: no margin calls, but if you can't service debt out of rent or salary, foreclosure follows the same script.

Risk 2: Interest accumulation eats the win

2020 SBLOC at 2%: hurdle is trivial. 2026 SBLOC at 6.5%: collateral needs to grow >6.5% net of dividend tax to beat selling. Real estate cash-out refi at 7%: interest-deductibility makes the effective rate 4.4–5.6%. Model the break-even against realistic asset return, not a hopeful one.

Risk 3: Political — §1014 repeal proposals

Step-up in basis is the most-attacked provision in the US tax code in Democratic reform proposals of the last 15 years. Repeal shapes: (a) heirs inherit at original basis with $1-2.5M exemption (Biden 2021), or (b) deemed sale at death taxing accumulated gain (Canadian model). Ultra-wealthy families hedge with parallel structures — GRATs, family limited partnerships, charitable remainder trusts. Middle-class families with one primary residence generally can't afford these and are more exposed to a policy shift.

Risk 4: State-level wrinkles

California Prop 19 is the most impactful example. Inherited property is now reassessed at market value UNLESS the heir uses it as primary residence within a year AND market value doesn't exceed old assessment by more than $1M. For the Bay Area example above: kids get the income tax step-up (still valuable) but face property tax reassessment from ~$4k/year to ~$25k/year — a $21k/year running cost. Other states (TX, FL, NV, WA) have no equivalent.

Where you live is a tax decision — the state tax landscape

Everything so far has been about federal tax. But state and local tax often equals or exceeds federal tax for high earners and long-term homeowners, and it varies far more dramatically than the federal code. A Buy-Borrow-Die strategy in Nevada looks very different from the same strategy in California. Where you domicile matters as much as which asset you own.

The three main taxes: state income tax (on wages, dividends, gains), property tax (on real estate), and sales tax (on retail purchases). A handful of states have no income tax (NV, FL, TX, WA, WY, SD, AK, TN). Others have no sales tax (OR, MT, NH, DE, AK). None has zero of all three AND low property tax simultaneously — every state raises revenue somewhere.

US state tax landscape — all 50 states + DC · tap column headers to sort 51 rows

				Notes
Alabama (AL)	5.0% top	4% state + ~5.3% local ≈ 9.29%	0.4%	Among lowest property tax in US
Alaska (AK)	0%	0% state (local ~1.8%)	1.2%	No state income OR sales tax; Permanent Fund Dividend paid to residents
Arizona (AZ)	2.5% flat	5.6% + ~2.8% ≈ 8.4%	0.6%	Flat-tax state (moved from graduated 2023)
Arkansas (AR)	4.9% top	6.5% + ~2.94% ≈ 9.44%	0.6%	Near-highest combined sales tax
California (CA)	13.3% (14.4% w/ MHT >$1M)	7.25% + ~2.5% ≈ 9.75%	0.7% (low, thanks to Prop 13)	Highest top income tax in US; see Prop 13 + Prop 19 sections
Colorado (CO)	4.4% flat	2.9% + ~4.9% ≈ 7.8%	0.5%	Low property tax; TABOR limits on tax growth
Connecticut (CT)	6.99% top	6.35%	2.1%	High property tax (New England pattern)
Delaware (DE)	6.6% top	0% (no sales tax)	0.6%	No state sales tax; famous corporate incorporation home
District of Columbia (DC)	10.75% top	6%	0.6%	DC follows federal tax code closely; high income tax
Florida (FL)	0%	6% + ~1% ≈ 7%	0.9%	Retirement + hedge-fund destination; homestead exemption caps annual assessment growth at 3%
Georgia (GA)	5.49% (transitioning lower)	4% + ~3.4% ≈ 7.4%	0.9%	Rates cutting under multi-year plan; moving to 4.99% flat
Hawaii (HI)	11.0% top	4.0% GET (general excise, applies to services too)	0.3% (lowest in US)	GET taxes services + business receipts, not just retail; island lifestyle trade-off
Idaho (ID)	5.8% flat	6% + ~0.03% ≈ 6.03%	0.6%	Flat-tax state as of 2023
Illinois (IL)	4.95% flat	6.25% + ~2.6% ≈ 8.85%	2.3%	Second-highest property tax in US; constitutional flat-rate income tax
Indiana (IN)	3.05% flat	7% (no local sales tax)	0.8%	Low income tax; no local sales tax
Iowa (IA)	3.8% flat (2026)	6% + ~0.94% ≈ 6.94%	1.5%	Transitioning from graduated to 3.8% flat by 2026
Kansas (KS)	5.7% top	6.5% + ~2.2% ≈ 8.7%	1.4%	Recent income tax cuts under multi-year plan
Kentucky (KY)	4.5% flat	6% (no local sales tax)	0.8%	Flat-tax state; cutting toward 3-3.5%
Louisiana (LA)	4.25% top	4.45% + ~5.1% ≈ 9.55% (highest combined in US)	0.6%	Sky-high combined sales tax; low property tax; homestead exemption ($75k)
Maine (ME)	7.15% top	5.5%	1.4%	New England pattern: moderate-high income + moderate-high property
Maryland (MD)	5.75% state + local (up to 3.2%)	6%	1.1%	County-level income tax on top of state (unusual pattern)
Massachusetts (MA)	5% flat + 4% surtax >$1M	6.25%	1.1%	"Millionaire tax" surtax added 2023; Prop 2½ caps local property tax growth
Michigan (MI)	4.25% flat	6% (no local sales tax)	1.5%	Flat income tax + no local sales tax
Minnesota (MN)	9.85% top	6.875% + ~1% ≈ 7.9%	1.1%	Fourth-highest income tax in US
Mississippi (MS)	4.7% top (transitioning to 3% flat)	7% + ~0.06% ≈ 7.06%	0.8%	Cutting income tax under multi-year plan
Missouri (MO)	4.7% top	4.225% + ~4% ≈ 8.4%	1.0%	Recent income tax cuts
Montana (MT)	5.9% top	0% (no state sales tax)	0.8%	No state sales tax; scenic-value premium in Bozeman / Big Sky
Nebraska (NE)	5.84% top	5.5% + ~1.5% ≈ 7.0%	1.7%	High property tax relative to nearby states
Nevada (NV)	0%	6.85% + ~1.5% ≈ 8.35%	0.6%	No state income tax; Incline Village at Lake Tahoe is 15 miles from CA border
New Hampshire (NH)	0% wages (5% int+div phasing to 0 by 2027)	0% (no sales tax)	2.1%	"Live Free or Die" — but property tax funds nearly everything
New Jersey (NJ)	10.75% top	6.625%	2.5% (highest in US)	High everything; densest state; multi-tier public school funding
New Mexico (NM)	5.9% top	5% GRT + ~2.72% ≈ 7.72%	0.8%	Uses Gross Receipts Tax (like HI) instead of pure sales tax
New York (NY)	10.9% state + up to 3.876% NYC = 14.8% top	4% + ~4.5% ≈ 8.5%	1.7%	Highest COMBINED rate in US for NYC residents; STAR credit reduces property tax
North Carolina (NC)	4.5% flat (moving to 3.99% by 2027)	4.75% + ~2.25% ≈ 7%	0.8%	Cutting income tax on multi-year path to sub-4%
North Dakota (ND)	2.5% top	5% + ~2% ≈ 7%	1.0%	Very low top income tax; energy-industry revenue offset
Ohio (OH)	3.5% top	5.75% + ~1.5% ≈ 7.25%	1.5%	Recent income tax cuts; municipal income tax on top varies by city
Oklahoma (OK)	4.75% top	4.5% + ~4.5% ≈ 9%	0.9%	High combined sales tax; homestead exemption reduces property tax
Oregon (OR)	9.9% top	0% (no sales tax)	0.9%	No sales tax at all; high income tax; Measure 5 caps property tax growth
Pennsylvania (PA)	3.07% flat	6% + ~0.34% ≈ 6.34%	1.6%	Lowest flat-rate income tax in US; local income tax varies (Philly ~4%)
Rhode Island (RI)	5.99% top	7%	1.5%	Small state; high everything relative to size
South Carolina (SC)	6.4% top (transitioning lower)	6% + ~1.5% ≈ 7.5%	0.6%	Cutting income tax; low property tax attracts retirees
South Dakota (SD)	0%	4.5% + ~2% ≈ 6.5%	1.2%	Preferred domicile for perpetual dynasty trusts (no rule against perpetuities)
Tennessee (TN)	0% (fully removed 2021)	7% + ~2.5% ≈ 9.55%	0.7%	Removed remaining dividend tax 2021; highest combined sales tax bracket
Texas (TX)	0%	6.25% + ~2% ≈ 8.25%	1.7%	High property tax offsets no income tax; homestead exemption
Utah (UT)	4.65% flat	6.1% + ~1.2% ≈ 7.3%	0.6%	Flat income tax; low property tax; Truth-in-Taxation limits growth
Vermont (VT)	8.75% top	6% + ~0.3% ≈ 6.3%	1.9%	High everything; small state; New England pattern
Virginia (VA)	5.75% top	4.3% + ~1.45% ≈ 5.75%	0.8%	Moderate rates; low property tax
Washington (WA)	0% wages (7% state capital gains >$262k since 2022)	6.5% + ~2.9% ≈ 9.4%	0.9%	State income tax = 0 but has capital gains tax on high-earner realizations
West Virginia (WV)	5.12% top	6% + ~0.55% ≈ 6.55%	0.6%	Recent income tax cuts
Wisconsin (WI)	7.65% top	5% + ~0.4% ≈ 5.4%	1.7%	High income and property; low sales tax
Wyoming (WY)	0%	4% + ~1.4% ≈ 5.4%	0.6%	Lowest overall tax burden in US; mineral extraction revenue offset

Want this table as its own standalone reference? See /reference/us-state-tax-landscape.

Three patterns worth noticing

You can't escape all three. Every state finds a way to raise revenue. Even Alaska (both zero) has federal-lowest revenue per capita and thin services.
High-tax coastal + Northeast vs. low-tax South + Mountain West is the map. CA, NY, NJ, MA, IL, HI, OR at the top. NV, TN, FL, TX, WY at the bottom. Decades-stable pattern.
Property tax and income tax are inversely correlated. States with no income tax (TX, NH) tend to have high property tax; states with high income tax (CA, NY) often have moderate property tax. Both are essentially arbitrage-free — you pay somewhere.

Primary sources for state tax rates: Tax Foundation — State Individual Income Tax Rates · Tax Foundation — Property Taxes by State · each state's Department of Revenue for authoritative current-year rates.

Tax domicile arbitrage — the Incline Village case study

Incline Village is a small resort town on the north shore of Lake Tahoe. It is in Nevada. It is fifteen miles from the California state line. Same lake, same trees, same climate. But every dollar earned by a resident of Incline Village avoids California's up-to-14.4% state income tax — and pays 0% state income tax to Nevada.

At California's top rate, this is a fifteen-mile line that saves a serious earner $144,000 per million dollars of earned income per year, every year.

Notable people who use this arbitrage

Larry Ellison — primary Nevada domicile at Incline Village
Michael Milken, Charles Schwab, many other ultra-HNW Californians
Hedge fund managers moving NY → Miami (2020–2023): Ken Griffin (Citadel), Carl Icahn, David Tepper. NYC's ~14.8% combined top rate vs. Florida's 0% is a $10–100M annual gap at hedge-fund scale
Trump moved from NYC to Palm Beach (2019) — same reason
Elon Musk became a Texas resident in 2020 with Tesla / SpaceX moves

The mechanics — six things you must do

Physical presence — spend >183 days per calendar year in the new state. California specifically counts nights via credit-card records, cell-tower pings, EasyPass records, flight itineraries.
Register as a resident — Nevada driver's license, Nevada voter registration, Nevada vehicle registration.
File a Declaration of Domicile — useful evidence in a residency audit.
Home base proportionate to your life — a studio apartment on paper won't hold up for someone earning $10M/year.
Sever California ties — sell your CA primary residence OR convert to bona fide rental. Close CA bank accounts, cancel CA club memberships, use NV doctors and dentists.
Continue paying CA tax on CA-source income — rental from CA property, salary for work performed in CA, K-1 flow from CA-based partnership stays taxable to California as non-resident income.

Why California is especially aggressive

The Franchise Tax Board runs one of the country's most sophisticated residency audit programs. Their "closer connection" doctrine can rule you a California resident and pull back multiple years of income tax retroactively (with interest and penalties) even if you spend <183 days in state — if your closest ties are still there (CA house, CA schools, CA business HQ, CA mail address, CA cell carrier).

At the ultra-HNW scale, proper domicile change requires a tax attorney and 12–24 months of documented lifestyle changes.

Who this arbitrage actually pays for

Earner profile	Rough annual savings	Worth doing just for tax?
W2 income $500k or less	$10–40k	Rarely; move for actual life reasons
W2 income $1M	$50–100k	Marginal; often less than one-time moving costs
W2 income $10M+	$700k–$1M+ per year	Frequently yes, if you want to actually live in Reno / Miami
Retiree, no wage income	Near zero	No — move for climate, family, cost of living
Founder w/ $100M+ liquidity event coming	Can be $10M+ on a single event	Almost always yes, with 12+ months of pre-planning
Ellison-scale	$100M+ per year	Trivially yes; lifestyle costs are noise

The moral asymmetry to be aware of: the Incline Village pattern is legal, but it depends on services California funds. Oracle's employees, customers, and offices are largely in California; the CHP protects the roads Ellison drives on when visiting; UC Berkeley and Stanford produce the engineers Oracle hires. Nevada residency arbitrage is a way to consume California without paying for it. This is why California keeps trying to close it — and why proposals for "wealth tax" or "exit tax" surface every few years in Sacramento.

Primary sources for CA residency rules: CA Franchise Tax Board — Residency Status guide · FTB Publication 1031 · Cal. Rev. & Tax. Code §17014

California Prop 13 — the property tax cap that reshaped California

Passed 1978 by ballot initiative (Howard Jarvis, 65% voter approval), Prop 13 is arguably the single most consequential tax law in California history. It has kept property taxes low for the state's longest-tenured owners for 48 years, and has been essentially untouchable politically for that entire time.

The two rules that changed California

Property tax is capped at 1% of assessed value (plus small local overrides averaging 0.1–0.25%)
Assessed value can only grow at max 2% per year regardless of market. Full reassessment happens ONLY on sale, transfer, or substantial improvement.

The Atherton example

A house bought in 1985 for $500,000 in Atherton:

Assessed value grows at 2%/year for 40 years: $500k × (1.02)^40 = $1.1M (2026 assessed value)
Actual market value in 2026: $15,000,000
Annual property tax bill: about $12,100 (1.1% of $1.1M)
A neighbor who bought last year for $15M pays about $180,000/year — 15× more property tax on essentially the same house

Consequences of the lock-in

Wealth preservation for holders — someone who bought a $500k house 40 years ago and pays $12k/year in property tax sits on a $15M asset with tiny running cost. Ideal collateral for a HELOC, ideal to pass to heirs.
Lock-in effect ("Prop 13 straitjacket") — retirees stay in 4-bedroom family homes long after kids leave, because a smaller condo across town would cost 10–15× more in annual property tax.
Massive intergenerational inequity — two identical houses on the same street can pay wildly different tax bills based purely on when purchased.
Housing shortage contribution — some studies attribute 10–15% of California's housing shortage to Prop 13 turnover suppression.
Fiscal starvation of local services — CA K-12 schools were once among the best-funded in the nation; post-Prop-13 they rely on state backfill and bond measures.
Elderly wealth concentration — 32%+ of California homeowners are 65 or older.

Political reality: every reform attempt has failed. Prop 15 (2020) tried to split-roll commercial property — lost 52-48 despite $70M in support and heavy Democratic backing. Neither party wants to be the one to raise property taxes on grandma.

Bottom line: Prop 13 makes California's effective property tax rate about 0.7% (one of the lowest in the US) despite the nominal 1% cap. The single largest de facto wealth transfer from young Californians to old Californians is not Social Security or Medicare — it is Prop 13.

Primary sources for Prop 13: CA Constitution Article XIII A (enacting text) · CA Board of Equalization — Prop 13 overview · Legislative Analyst's Office primer · Howard Jarvis Taxpayers Association history

California Prop 60 / Prop 90 / Prop 19 senior portability

Prop 13's lock-in is brutal on retirees who want to downsize or move within California — the tax basis doesn't follow you by default. Three subsequent propositions carved out age-based portability so seniors could move without triggering a full reassessment.

Prop 60 (1986)

Age 55+ homeowner could transfer old assessed value to a NEW home of equal or lesser value, within the same county, once per lifetime.

Prop 90 (1988)

Extended Prop 60 to cross-county transfers, but only if the receiving county opted in. Only about 10 counties did.

Prop 19 (2021)

Anywhere in CA · up to 3 lifetime uses · works even to more-expensive home (with partial upward adjustment) · 2-year window from sale of old home. Applies to age 55+, severely disabled, disaster victims.

Concrete example — retired Atherton couple moving to Palm Springs

Same setup: sell the $15M Atherton home with $1.1M assessed value and $12k/year property tax. Buy a $3M Palm Springs house.

New home ($3M) < old home ($15M) → couple keeps old assessed value of $1.1M
New Palm Springs property tax bill: still about $12,000/year
Young family buying the same $3M Palm Springs house would pay about $33,000/year
Couple can do this 3 times during their lifetimes

Concrete example — moving UP under Prop 19

Sell $15M Atherton home ($1.1M assessed). Buy a $20M Malibu house.

Delta between sale and new purchase: $20M − $15M = $5M
New assessed value = old assessed ($1.1M) + delta ($5M) = $6.1M — not $20M
New tax bill: about $67,000/year — much less than the ~$220,000 a fresh buyer at $20M would pay

Primary sources for Prop 19 / 60 / 90: CA BOE Prop 19 Portal · Cal. Rev. & Tax. Code §69.5 · CA Secretary of State — Prop 19 ballot text · CA Association of Realtors explainer

The wider arbitrage landscape — more drills coming

The differences between states, ages, account types, and asset classes create dozens of legal arbitrage opportunities. This publication is the base; future analyses drill into each category:

Real estate investment strategies (next major expansion)

§1031 deep-dive (45/180 day rules, reverse 1031, improvement 1031), cost segregation, bonus depreciation phase-out, Passive Activity Loss rules, Real Estate Professional Status (REPS), Short-Term Rental (STR) loophole, Opportunity Zones, Delaware Statutory Trust (DST), BRRRR + cash-out refi cycles, installment sales, §121 + rental conversion trap.

Retirement account arbitrage

Backdoor Roth, Mega Backdoor Roth, HSA triple tax advantage, Solo 401(k), 529-to-Roth conversion (SECURE 2.0), QCD for charitable retirees, Net Unrealized Appreciation on employer stock, Roth conversion laddering.

Wealth transfer / trust arbitrage

GRAT, Intentionally Defective Grantor Trust (IDGT), Charitable Remainder Trust (CRT), Charitable Lead Trust (CLT), Donor-Advised Fund (DAF), Family Limited Partnership valuation discounts, dynasty trust in South Dakota / Nevada / Alaska, Spousal Lifetime Access Trust (SLAT).

Investing / portfolio arbitrage

Tax-loss harvesting at scale via direct indexing and 130/30 long-short, muni bond ladders for high-bracket investors, Master Limited Partnership return-of-capital, QSBS §1202 exclusion, wash-sale gap on cryptocurrency.

Domicile + geographic arbitrage

US expatriation + §877A exit tax, Foreign Earned Income Exclusion (§911), Puerto Rico Act 60 (4% corp tax + 0% capital gains for bona fide residents), US Virgin Islands EDC, multi-state residency for wage-splitting.

The unifying question: every one of these opportunities exists because the tax code differentiates based on some criterion — age, state, asset type, holding period, entity type, income level, family status. Wherever there's a line, there's arbitrage. The wealthy pay professionals to know where every line sits. This publication exists to give everyone else access to the same map.

Who this actually applies to

Fits well

Ultra-HNW with concentrated, low-basis appreciated stock position in a taxable brokerage
Long-tenured homeowners with low-basis primary residences (CA/NY/BOS coastal, bought 1980s-90s)
Rental property investors with fully-depreciated buildings
Family real estate businesses (multi-generational)
Concentrated founder positions post-IPO with lockup expired

Fits poorly

Wealth mainly in 401(k)/IRA — no step-up applies
Assets already diversified, low basis-low gain
No heirs, contentious heirs, or charitable-only intent
Cash flow can't service borrow interest through a full economic cycle
State rules erode benefit (CA + non-primary-residence heirs is a common gotcha)

Warning signs the strategy is being sold to you inappropriately

A wealth manager pitching SBLOCs to fund lifestyle when your position is already diversified and low-gain
A HELOC pitch to fund consumption spending when there's no coherent estate plan
Anyone claiming the strategy has "no downside" (§1014 political risk, margin call risk, and interest rate risk all exist)

One important caveat — if your wealth is in retirement accounts

Buy-Borrow-Die works because step-up in basis eliminates income tax on lifetime appreciation. Step-up does not apply to retirement accounts. A typical 401(k) rolled to an IRA is already tax-deferred, so there is no untaxed appreciation to preserve — heirs still owe income tax on withdrawals. And the IRA cannot be pledged as loan collateral.

If most of your wealth is in retirement accounts, the more useful conversation is:

Roth conversions in the gap years between retirement (or age 59½) and RMD start age (73 as of 2026). Pay tax at cheap rates now to lock in tax-free growth later.
QCDs after age 70½. Direct IRA-to-charity transfers up to $105k/year (2026 limit) satisfy RMD but are excluded from taxable income. Lowers AGI, cascades into lower Medicare premiums.
Withdrawal ordering. The default (taxable first, tax-deferred second, Roth last) is often wrong. Filling low brackets with intentional traditional withdrawals or Roth conversions before RMDs kick in can beat the default by 5–15% of retirement wealth over 30 years.

This is planning worth running with a fiduciary or CPA who does actual multi-year projections — not rules of thumb from a magazine article. Noesis is not a substitute for that math.

What Noesis is (and isn't) saying

Not saying

"You should do this."
"This is smart / clever / recommended."
"The wealthy are gaming the system."

Saying

This is how the game is legally played at the top of the wealth distribution
The rules are public code (IRC §1014, §1031, §121)
The same rules apply to middle-class homeowners with paid-off houses
Understanding the mechanic is a prerequisite to any adult conversation about how you fund retirement, plan a bequest, or weigh concentration vs. diversification cost

Every user of this framework deserves to know it exists, whether or not they choose to use it.