Articles
Every read, by topic
40 short reads, about 152 minutes in all. Three to five minutes each. New to this? Follow the ladder walk first.
How much of my return does tax take, and what is worth doing about it?
-
What a dollar of gain keeps after tax
Sell a winner at the top California bracket and 35 cents of every dollar of gain goes to tax. Sell it inside a year and 52 cents go. Where those two numbers come from.
-
Eight tax levers, ranked by what they are worth
Never selling is worth the whole 35 cents, the one-year line 17, an ETF about 120 basis points a year. Loss harvesting, the lever with the marketing budget, 0 to 30 after year five. Each one, priced.
-
The lever worth the most is never selling
A $100,000 position with a $40,000 gain costs $14,040 of tax to sell. Held to the estate it costs nothing; given to charity it earns a deduction. The quiet lever at the top of the ranking.
-
The one-year line is worth 17 cents per dollar
Everyone knows the line. In 2015 individuals still realized $3.3 billion of stock gains within three months of it, at the short-term rate. Where a known rule gets crossed early, and what it costs.
-
Loss harvesting, in one example
Sell a stock at a $10,000 loss, buy a similar one the same day, and the loss cuts this year's tax bill by up to $5,210. What actually happened, step by step, and the catch.
-
Why the harvesting benefit fades after a few years
In AQR's simulation harvesting adds 155 basis points in year one and nothing by year five without new cash. Live accounts show the same curve. Why it fades, and the three things that keep it alive.
-
The wash-sale rule in four facts
Buy back the same stock within 30 days of selling it at a loss and the loss is disallowed. That part is well known. The four facts that catch households are about spouses, IRAs, brokers, and ETFs.
-
Two questions for any tax-savings pitch
Does the benefit last past year five? Does the long/short version make sense below $5 million? The published evidence says no to both as general claims. How to ask, and what a good answer sounds like.
-
One rule holds up the ETF tax advantage, and Treasury is looking at it
As of September 2026 nothing pending changes the ranking of the levers. One live item: Treasury has said it is examining the in-kind mechanism behind every ETF's tax edge. A dated watchlist.
Which of my accounts should hold which piece of the portfolio?
-
The rollover that never got invested
IRAs hold $19.2 trillion. Vanguard found that 28% of rollovers were still sitting in cash seven years later. The most under-managed retirement money is the money that never got invested at all.
-
Four accounts, three tax treatments
Taxable, traditional IRA or 401(k), Roth, HSA. Each taxes the same dollar of return at a different time and a different rate. One table, then the one rule that follows from it.
-
Turnover decides location, not asset class
Most households shelter index funds in the IRA and keep active holdings in taxable. A 3% return from monthly trading nets 1.4% in taxable and keeps its full value in an IRA. That setup is backwards.
-
Fill the sheltered slots before adding taxable money
A dollar inside a shelter never meets the 52% short-term rate or the yearly dividend tax. The 2026 limits, the two back doors for high earners, and one Californian catch.
-
What an IRA cannot do
It cannot borrow, so it cannot short. It destroys a loss harvested next door. And a long-only version of a long/short idea keeps about half the edge. Three limits of the best shelter most people have.
-
California taxes the HSA every year
Federally an HSA is deductible going in, untaxed while it grows, and untaxed coming out for medical costs. California recognizes none of it. What that costs, and the bill that would change it.
Which fund wrapper keeps the most after tax, and who keeps the fees?
-
Why ETFs pay less tax than mutual funds
The average stock mutual fund lost 1.9% a year to tax over the last decade; the average stock ETF lost 0.7%. The gap comes from one line in the tax code, not from smarter managers.
-
Direct indexing, in one page
Own the 500 stocks instead of the fund that owns them. A $1.16 trillion business built on that idea. What you get, what you pay, and who it fits.
-
Same harvesting, 9 basis points or 115
Retail direct indexing charges 9 basis points a year. Through an adviser and a platform the same logic costs about 115. Against a long-run benefit of 0 to 30, the price decides the sign.
-
The dividend tax no wrapper removes
An index ETF defers its capital gains almost completely. It cannot defer its dividends. At a 1.3% yield and a 35% rate, that is 46 basis points a year, and it is most of the ETF's remaining tax drag.
-
Where the basis points go on an adviser-sold account
An adviser-sold direct-indexing account charges about 115 basis points a year: 65 to the adviser, 30 to the platform, 20 to the harvesting engine. The engine is what a retail investor buys for 9.
-
ETF share classes: the patent that expired
One firm held a patent on attaching an ETF share class to a mutual fund until May 2023. Since November 2025 the SEC has let others do it, and the funds that get one stop paying out gains.
-
Section 351: moving a portfolio into an ETF without selling
Since December 2024, 113 ETFs have launched by taking in investors' stock portfolios for shares, no gain recognized, about $21 billion so far. The two tests, and why the IRS has not said yes.
-
Exchange funds, for one big stock
Contribute one concentrated stock to a partnership, receive a slice of everyone's, pay no tax, wait seven years. The fee is 0.4% to 0.95% a year, and a fifth of the fund is real estate by law.
Are the 'tax-efficient income' ETFs actually tax-efficient?
-
Income ETFs are income products, not tax products
Option-income funds grew from about $6 billion to $175 billion in five years, sold on yield and tax words like 60/40 and return of capital. Against the index each one wraps, the tax case is thin.
-
Return of capital comes out of your own cost basis
A return-of-capital payout is not taxed when received. It lowers your cost basis, so the tax arrives when you sell. At a 14% payout the basis is gone in about seven years; then every payout is a gain.
-
Where cash from stocks costs least: 14 cents, 35, or 52
Take $14,000 from an income ETF and the tax is 52 cents per dollar now, or 35 cents later. Sell $14,000 of index-fund shares with a 40% gain and it is 14 cents, with the rest still compounding.
-
Every large income ETF trails its index
Since launch, the biggest covered-call funds have returned 4 to 11 points a year less than the index each one wraps. The premium is real; it is paid for with the rallies the fund sells away.
-
The label predicts nothing; the filing does
Two funds with the same 'covered call' label paid 99.6% return of capital and 74% ordinary income in the same year. The monthly notices said otherwise. Only the year-end filing counts.
-
The new ETFs built on one tax right, and the regulator watching
A $14 billion fund that turns T-bill interest into deferred gain. A fund that never receives a dividend. Both lean on the in-kind rule behind every index ETF, and Treasury has said it is looking.
Which book of stocks, Treasuries and gold fits the purpose of the money, and how does the house change the count?
-
The house is stocks, at a fifth of its value
A house moves with the stock market at a beta near 0.2, so a $1.5M house supplies $300,000 of stock exposure. Counted, the savings of a household with a house hold more stock than the plain book.
-
The stock-bond mix alone swings twenty-year returns by up to three points a year
Since 1985, moving from an all-equity portfolio to 60/40 has changed trailing 20-year annualized returns by as much as 3.1 points a year, before any strategy or wrapper choice.
-
60/40 is the floor, and three books clear it from both starts
Three books of stocks, Treasuries and gold returned more per unit of risk than 60/40 from a 2006 start and a 2010 start, and lost less in 2022. The return gap depends on the start; the edge does not.
-
Long Treasuries and listed real estate do not pay for their risk
From 2006 and from 2010, long Treasuries returned under 3% a year for stock-sized swings and listed real estate returned less than stocks for more risk. Neither improved 60/40.
-
Gold is a view on the decade, not a backtest result
Gold held in 2008 and 2022 and sits at 15 to 30% in the optimizer's answer from a 2006 start and a 2010 start. It also lost 62% from 1980 to 1999. Its weight is a view on the next decade.
-
Bitcoin is sized by the loss a household can hold, not by the optimizer
Three percent of bitcoin added two points a year to the Grow wealth book since 2016 for a worst fall a third of a point deeper. The optimizer wants 69%. Once a year cannot hold the cap.
-
The mortgage is leverage, a rate and an option, and it has no price
A fixed-rate mortgage cannot be sold, so it has no market value to the household. It is leverage on the house, a fixed rate to compare with today's, and a refinance option that pays when rates fall.
Which equity approach, and does the evidence support paying for it over the index?
-
A value tilt has earned 0.5 points a year since 1985, and none of it since 2006
A 0.20 HML tilt run at roughly 2% tracking error beat the market by 0.5 points a year since 1985. Nearly all of that edge arrived before 2007, and it has not returned since.
What else, beyond stocks, bonds and gold, actually diversifies the portfolio?
-
Trend following made money in all five of the worst US equity years since 1985
AQR's diversified trend strategy was positive in every one of the five worst US equity years since 1985, averaging +21.6% while stocks fell 6 to 39%.
What does the whole household look like after tax, and what needs attention?
-
Asset location is backwards
Household A shelters bonds and its 20%-turnover tilt, not the index fund sitting in taxable. Vanguard prices the mistake at up to 60 basis points a year.
-
Knowing when nothing needs to change
Household A's dashboard proposed a trade in two months out of twelve. Rebalancing bands, the wash-sale window, and the one-year line explain the other ten.