I Put $28,000 Into Long Treasuries for Six Rate Cuts. Two Arrived.

Wednesday, Dec 18, 2024, 2:15 pm, living room couch, laptop open to my brokerage and the TV muted on the Fed press conference, because I could not stand hearing the words but could not stop reading the ticker at the bottom of the screen. The dot-plot graphic flashed up: two cuts projected for 2025, down from the four they had been signaling in September. My Treasury position sat in the corner of the laptop screen showing negative 11.9 percent since I snagged it, and I did the thing you do when a number refuses to change, which was close the laptop and look out the window at the neighbor's roof for a while.

The trade had been born a year earlier in pure consensus. January 2024, every desk on the street projecting six cuts, my bonds allocation sitting in cash, and me reading three separate strategy notes on a Sunday night that all agreed the direction of the next two years. So on Jan 9, 2024, I moved $28,000, money mentally labeled as the safe portion of a down payment we planned for 2027, into a long-duration Treasury ETF at $102.50 a share. Duration around 16 years. Yield to maturity around 3.9. I understood approximately one of those three numbers at the time of purchase, and it was the wrong one.

Six cuts were promised. Two arrived. Everything in between is a story about what the word safe actually protects you from, and the answer, it turns out, is almost nothin you were worried about and one thing you weren't.

the consensus that said six

It is worth remembering how unanimous it was, cuz the unanimity was the seduction. In January 2024 the futures market was pricing better than even odds on six cuts, a former Fed governor was on every panel saying inflation was beaten, and the strategy notes I printed at the office all used some version of the phrase duration is the trade. The logic felt less like a bet and more like reading a timetable. What I actually did was pay $28,000 for an instrument whose value moves inversely, and violently, to the one variable I was certain I could not forecast, while believing I had purchased the opposite of risk. Consensus is a wonderful thing to hold when it is right, and it is worth exactly its price, which is nothing, when the bond market decides to reprice the calendar.

duration, the word I underestimated

Here is the arithmetic I ultimately did in the spring of 2024, too late, on the back of a pharmacy receipt. The fund's modified duration was about 16, which means every one percentage point rise in long yields takes approximately 16 percent off its price, and long yields rose from approximately 3.9 to 4.8 by that autumn. Sixteen times point-nine is about 14 percent of price damage, softened a few points by the coupons the fund paid me along the way, and there is the whole of my negative 11.9 in one multiplication I could have done in the store parking lot before buying. Duration is leverage that wears a suit. It is the reason a fund backed by the full faith and credit of the United States Treasury, an instrument that cannot default in any way that matters, still managed to lose more money in 2024 than my stock index fund, which is backed by nothing but earnings, made in the same stretch.

what the boring cash did

The comparison that finished my education is almost too simple to publish. The identical $28,000, left in the high-yield savings account it came from, would have earned about 4.5 percent thru 2024 and approximately 4 percent in 2025, call it $2,300 over the stretch, with a zero drawdown, full liquidity, and FDIC insurance under it. The Treasury ETF, including every coupon and the price recovery, ended my whole adventure at approximately negative 1.8 percent total, after two and a half years of watching a press conference every quarter like a hostage. My safe allocation underperformed a savings account by somethin like $2,900 while I told myself I was being sophisticated. The word safe did all the damage. Safe from default, yeah. Safe from being wrong about the Fed, never, and being wrong about the Fed was the entire trade.

two cuts, two years late

The cuts ultimately came, and I want to be fair to the thesis: September and October of 2025, two of em, long yields fell from 4.6 toward 4.1, and the fund clawed from negative 15 at its worst back to almost flat. I sold the whole position on Jun 30, 2026 at $100.66 a share, a hair under my entry, plus the coupons along the way, netting out to that negative 1.8 percent on the ledger. If I had held it all the way to the down payment in 2027 mayb it would have printed a gain, and maybe yields go back to 5 and it prints a worse loss, and that word mayb is precisely the point. A safe allocation has no maybes in it. Mine was made of maybe the entire time, and I ultimately priced that correctly in June by taking the near-flat exit and putting the money where mayb is not part of the design.

the safe label did the damage

My stocks, the supposedly risky half, returned about 14 percent across the same two and a half years. The Treasury fund, the safe half, lost money. Roll that sentence around as long as I did, cuz it inverts every label on the shelf, and the inversion is not a paradox, it is a lesson about what the labels are measuring. Stocks are risky in the short run and generally rewarded in the long run, and long bonds are safe against default and brutally leveraged to a single macro variable that nobody, including the twelve people who set it, can predict two years forward. I now refuse to hear the word safe without asking safe from what, and I have found that the question ends most bond-fund conversations swiftly, sometimes rudely.

the 401k rollover question

The lesson got a live-fire test this spring. In April 2026 I ultimately did the 401k rollover from the employer I left in 2024, moving $61,000 from a plan with nine mediocre options into an IRA at a brokerage whose onboarding specialist, cheerful and reading from a script, advised their long-term bond fund as the core of it. I asked one question, what is the duration, and there was a pause long enough to hear her typing. Fourteen years, she said eventually. Fourteen. I built the IRA instead from the boring shelf: a total-market index fund as the spine, short Treasury bills for the part I might actually need predictably, and nothing whose return requires me to suspect the Fed's calendar correctly. The whole call took nineteen minutes and the specialist read relieved, which I have chosen to interpret as evidence that she asks her own accounts the same question now.

same couch, June 2026

I watched the June presser from the same couch on Wednesday, laptop open, TV muted out of habit, and the position list on screen now reads T-bills and an index fund and zero long-duration anything. The dot plot said what it said, three cuts projected, and I read the number with the mild interest of a person watching weather in a city she no longer lives in. The down payment fund is on schedule for 2027, sitting in instruments that cannot drop 12 percent because twelve people changed their minds, and the $2,900 my sophistication donated to the cause of my education is written on the first page of the notebook where I keep my market autopsies, under a title I am not proud of: the safe label. Two cuts arrived. I had planned my finances around six. The couch is the same, the mute habit is the same, and the only thing that changed is that I stopped mistaking a forecast for a plan, which is the most expensive synonym I have ever confused.