Carry trade H2 2026 update
Disclaimer: This is not financial, tax, or investment advice. It is a description of how I structure my own balance sheet, with risks I understand and accept for my situation. Borrowing against BTC, using DeFi protocols, and buying preferred stocks can all lose money or create tax consequences. Do your own work and talk to qualified professionals before copying any part of this.
It's been roughly six months since my original essay. Enough time has passed and enough details have changed that I decided to write an update to share where things stand compared to where they started - what changed and why.
Security
A lot has happened in the past few months due to the substantial increase in AI competence at basically everything, especially cybersecurity. I really felt nervous at times, worried about crypto exploits, especially given the recent Binance hack of over $6B, the largest ever. My main fear was Morpho smart contract safety and contagion due to other exploits and hacks, as has happened in the past with Aave.
Luckily, Morpho contracts have proven to be very resilient so far, with no hacks that I'm aware of. My positions (and Morpho in general) have remained intact through this shitstorm of hacks, CVEs, exploits and what have you. Stay safe out there. It's the wild west era of the internet right now. Being paranoid and extra vigilant is the bare minimum I need to survive in this space.
Price action
My base case is that BTC bottomed around June/July and my current planning and allocation rely on that thesis. Should things change in the near future, that would prompt me to re-assess these positions and adjust accordingly. For now, I expect BTC to make a new ATH in the next 12 months, despite the rather weak performance over the past few weeks.
Collateral and loan layers
The first two layers of this trade, the BTC collateral and the stablecoin loan, remain mostly unchanged structurally. However, I moved positions from Morpho variable rates to the (recently launched) fixed-rate markets.
Variable rates have fluctuated a lot lately (I presume due to the Binance hack, as liquidity fled DeFi and borrowing demand spiked), which gave me pause. You can see the history below, first in the 7-day rates as reported by Morpho (6h averages), and second the 30-day rates as tracked by my own dashboard, where I take snapshots every few minutes. There are multiple spikes, sometimes approaching 10%.
7-day rates (Morpho)
30-day rates (my dashboard)
Meanwhile, there was a fixed-rate loan maturing Dec 25th, 2026 with a rate of only 4.2%. That looked very attractive in comparison, so I made the switch. I believe this is some form of promotional rate that Morpho is offering temporarily to attract capital to its new product, and that it is likely to increase for 2027, so I might have to switch back to a variable rate later or roll into a slightly higher fixed rate. That's okay. Right now, any fixed rate close to or under 5% is attractive, since that's roughly where variable rates have averaged. That's my benchmark.
Unfortunately, there was no one-click solution to make the switch, so I had Claude build one on top of Tenor's contracts, which did most of the heavy lifting. This allows me to run a couple of commands and move my full position from variable to fixed. Tenor's UI didn't work for me, so I went the custom-built path. Your mileage may vary. I might end up building the reverse path later if needed.
Treasury preferreds layer
I've fully closed the STRC position in favor of SATA, which was paying more with less volatility. With the recent STRC switch to daily dividends just like SATA, the payout structure gap has closed. The remaining yield gap is 12% for STRC versus 13.3% for SATA at the time of this writing.
I'm planning to keep holding as is until the yield gap closes fully or flips in favor of STRC (in which case I'd rotate back), or there's another clear BTC bull run. In the bull run case, I might fully deallocate from preferreds and move to either IBIT or a combination of IBIT, ASST and MSTR. That will depend on where prices move through Q1 '27.
Closing thoughts
Volatility in treasury preferreds has been high over the past few months (roughly 22% annualized for SATA and 28% for STRC over the past six months, with lows of $83.53 and $74.57 respectively in late June, against their $100 par). They're still new and unproven products that need more time to mature and build a longer track record. These products rest on the assumption that BTC price appreciation will outpace the yield they offer, which is of course unknowable. Price action since the summer low is encouraging, even if the past few weeks have been weak, and I remain cautiously optimistic that the BTC power law floor will hold. Using the 200-week moving average as a rough proxy puts that floor at $68k as of now. I'll keep that number front and center and course correct, should the floor break. If BTC has a weekly close below the 200-week moving average, I'll fully liquidate my preferred stock positions (currently SATA) and any IBIT and MSTR longs, and stay in cash. After a week or two, I'll reassess and most likely pay back the Morpho loan and walk away from this trade.
Meta note on writing
One thing I've changed my mind about, substantially so, is the use of AI for writing. After posting the previous essay, I realized how little of it sounded like my writing; it didn't feel natural. In retrospect, it read like slop. It didn't age well. The content was solid; the format and delivery, not so great. Thus, I've decided to keep AI usage to a minimum, for grammar, clarity and obvious mistakes only. I want to keep delivering these in my own voice. Writing is thinking and it is simply way too important a skill to delegate. Use it or lose it.