Letter No. 2 · The Snake Eating Its Tail · Part 2 of 3
The Blast Radius: What a Mag-7 Reversion Does to Your Retirement
Seven stocks are a third of the index. Your “diversified” fund is the AI bet. And the people who can least afford the drawdown are the ones holding it at the worst possible moment.
Part 1 traced thirty years of capital overbuilds — fiber, mortgages, and now GPUs — and ended on a question about repricing. This part asks the question almost nobody asks out loud: when the repricing comes, who actually absorbs it?
The answer is not "tech investors." The answer is closer to everyone with a target-date fund — and the people least positioned to recover are the ones nearest to, or just inside, retirement.
You are more concentrated than you think
Seven companies — the so-called Magnificent 7 — now represent roughly 31.5% to 40% of the entire S&P 500's market capitalization, depending on the month you measure in 2026. That is up from about 13% in 2018, and it is the highest top-seven concentration since the Nifty Fifty era of the early 1970s.
Read that against Part 1's numbers. Those same seven companies are the ones funding the AI capex loop — a projected $725 billion in 2026 alone, up 77% in one year. So the person who "played it safe" in an S&P 500 index fund is not diversified against the AI infrastructure bet. To more than a third of their portfolio, they are the AI infrastructure bet — and most of them have no idea.
That is the blast radius. The exposure is not held by people who chose it. It is held by people who were told an index fund was the prudent choice — and it was, back when the index was actually five hundred stories instead of seven.
Sequence-of-returns risk: why the same crash is not the same crash
A 40% drawdown at age 45 is an inconvenience. You keep contributing, you buy the bottom, the recovery does the work. The identical drawdown in the first five years of retirement is a structurally different event, because you are selling depressed shares to pay for groceries. Every withdrawal at the bottom converts a paper loss into a permanent one, and the portfolio that "averages 7%" over thirty years can still fail if the bad years come first.
Financial planners call it sequence-of-returns risk. Most retirees have never heard the phrase — and the cohort at the top of the Gen X bracket, or freshly retired, investing today the way the last five years trained them to, is walking into it with record equity concentration.
The worked example nobody had to imagine: SPCX
In June 2026, SpaceX completed the largest IPO in history — $75 billion raised, a first-day close near $161 valuing the company around $2.1 trillion. This is not a story-stock shell: government launch contracts, Starlink subscription revenue, commercial space — arguably the most defensible revenue mix in the entire complex.
Seven weeks later it traded at $108.37 — down 33% from its debut close, and roughly 20% below its own IPO price. Call it $686 billion of market value gone in seven weeks, from the most credible name available.
Two lessons. First: if the market cannot hold the valuation of a company with real, contracted, diversified revenue, consider what supports the names that have none. Quality broke first because quality is what could actually be priced. Second, and this is the blast-radius point: a $2 trillion listing enters the major indices, and index funds must buy it — not because any manager judged it attractive, but because the rules say so. The 62-year-old with a target-date fund took a position in SPCX at debut-era prices without any human choosing it, rode it down 33%, and will learn all of this at statement time.
Where the pendulum starts from matters
Measure the run: from the post-GFC lows of 2009 to the pre-COVID highs, through the COVID blip, and then the steepest stretch of all — the Mag-7 "wind in the sails" years. If assets revert toward long-run means, the correction is proportional to how far the pendulum swung. It has rarely swung this far, this concentrated, this fast. Reversion is not a prediction with a date. It is arithmetic waiting for a catalyst — and Part 1 catalogued the candidates.
The part your broker will not tell you: a crash is a tax event — and some of it is opportunity
Here the conversation leaves Wall Street commentary and enters my actual profession. Four mechanisms decide whether a drawdown merely hurts or permanently damages a retirement — and every one of them is a tax mechanism, not an investment one.
1. RMDs are computed on a balance that no longer exists
Required minimum distributions force recognition of traditional retirement-account income beginning at age 73 (the Guide, Ch. 11, p. 184) — and the distribution is calculated on the account's December 31 balance of the prior year. A December peak followed by a spring crash means the IRS's number was set at the top while the sale happens at the bottom: a forced sale, of depressed shares, sized by a valuation that evaporated.
2. A drawdown is the best Roth conversion window most people will ever see
Converting traditional dollars to Roth while the account is down means paying tax on the depressed value and letting the recovery compound tax-free inside the Roth. The Guide frames the discipline as a question every advisor should have to answer (p. 121): when the conversion was recommended, what was the plan for the tax? In a drawdown, that question has its best possible answer — the tax is computed on the smallest number the account may ever show again. Almost nobody executes it, because the window opens precisely when everyone is panicking instead of planning.
3. Loss harvesting against decade-old gains
Concentrated Mag-7 positions in taxable accounts are sitting on enormous embedded gains with ancient basis. A reversion is the one moment when offsetting losses exist at scale (the Guide, §8.3 on gain and loss harvesting, p. 109) — and harvesting them correctly, without tripping wash-sale rules, is planning work, not brokerage work.
4. The gap years become priceless
Between retirement and RMD age sits a window where income is controllable — the years when conversions, bracket-fills, and asset-location moves (the Guide, p. 229) do their best work. A drawdown makes that flexibility more valuable, not less. This is Chapter 8's core argument: planning is permanent, and it is decided in years exactly like the one a reversion creates.
The market decides when the pendulum swings. The calendar decides when your RMDs start. Only planning decides what the collision costs you.
Part 3: Where's the Next Tail?
The repricing arrives — Oracle's debt, the canaries in the credit market, and where the loop breaks first. Plus the book.
One email when it publishes. No list-selling, ever.
The reference behind the Letter: THE TAX CUTTERY® Guide to Federal Income Taxation — 564 pages, nearly 2,000 primary-source citations. Join the launch waitlist →
The firm behind this analysis: THE TAX CUTTERY® — back taxes, IRS letters, planning, and returns. Start a conversation any time.