Treasury Dealers Sit on a Two-Year High in Bills, With October 30 Circled in Red

Dealer balance sheets are fuller than they've been in two years, and there are exactly five business days between now and the moment that matters.

The Federal Home Loan Bank of New York pushed out its weekly update on October 23. Most people skimmed past it. They shouldn't have, because the read-through isn't about the bank itself, it's about how much room the firms in the middle of the Treasury market have left to absorb supply. October 30 is the checkpoint. Everything between now and then is just waiting.

Three Dates, One Squeeze

Start with the sequence, because the sequence is the story.

Treasury dealers have been building bill positions for months. They're now at a two-year high. That's not a rounding error or a seasonal blip. It's a structural shift in who's holding the shortest paper the US government issues, and it happened quietly while everyone was arguing about rate cuts.

Then came October 23, when the FHLBNY update landed and gave the market a fresh read on secured funding conditions. Granted, a single weekly release isn't a smoking gun. But it's one of the few public windows into how collateral is moving through the system, and it arrived at a genuinely awkward moment, with dealer books already heavy.

And then October 30. That's the test. Month-end rolls bunch up, balance sheet reporting gets finalized, and the repo market does its quarterly-ish stress test in miniature. If dealers are already at capacity going in, the exit gets tight.

Here's the part most crypto coverage misses. Bitcoin's spillover from all of this is conditional. Not automatic. Not mechanical. Conditional.

What Actually Shifted

When a dealer buys a Treasury bill, they don't pay for it out of a vault. They finance it, usually overnight, usually in repo. So every additional bill on the book is another piece of collateral pledged and another chunk of balance sheet consumed.

Crowded books mean less capacity. Less capacity means the marginal buyer of Treasury supply starts stepping back, and when the marginal buyer steps back, the cost of that financing goes up. That's the whole mechanism. It's plumbing, and it's boring, and it's exactly the kind of thing that moves markets before anyone writes a headline.

So who feels it first? Money market funds, mostly, then the repo desks, then anything that marks itself to short-term funding costs. Crypto sits way down the list.

Which brings me to a take I'll defend: Bitcoin doesn't trade off Treasury supply. It trades off the perception of Treasury supply. And perception is a much cheaper thing to move than actual collateral.

Anyone telling you a crowded dealer balance sheet is a bullish signal for Bitcoin is reading the plumbing upside down. Crowded dealers mean less intermediation capacity, and less intermediation capacity is a constraint, not a gift. The proponents of the liquidity thesis love to point at balance sheet expansion as fuel. Fine. But this isn't expansion. This is congestion.

Color me skeptical, but I've watched the "Bitcoin as a liquidity sponge" thesis get trotted out for four years now, and the track record is mixed at best. It works when the shock is large and obvious. It doesn't work when the shock is a repo rate ticking up a few basis points at month-end.

The question worth asking: does anyone actually believe a repo hiccup in late October transmits to Bitcoin within the same week? History suggests otherwise. The lag is real, and it's usually measured in weeks or months, not days.

What Comes After October 30

Here's what I'd watch, specifically.

First, the October 30 to 31 window. If repo rates spike into month-end and dealers show stress, that tells you capacity was already stretched on October 23. If it clears quietly, the two-year high in bill holdings is just a positioning story and nobody needs to panic.

Second, the early November borrowing estimates. If Treasury signals heavier bill issuance ahead, dealers have to absorb more of it with books that are already full. That's the setup that actually matters for funding markets into year-end, and year-end is a much bigger stress point than month-end. December 31 has broken more funding markets than any October in memory.

Third, watch whether the bill share of total issuance keeps climbing. Dealers at a two-year high in bills is one data point. Dealers at a two-year high in bills going into a heavy issuance quarter is a different situation entirely.

My honest thesis: this is a constraint story dressed up as a liquidity story, and the people trading it as a liquidity story will get the direction right for the wrong reason. Bitcoin probably grinds higher into year-end if funding markets stay calm. If they don't, crypto finds out how thin its own liquidity is before it finds out anything about the Treasury market.

To be fair, there's a bullish case. Quiet dealer congestion with no repo stress is basically a nothingburger, and the market has a long record of absorbing bill supply without drama. Admittedly, the plumbing has held up better than the skeptics predicted through most of this cycle.

But five business days isn't a lot of runway. And when the calendar compresses like this, the market stops pricing fundamentals and starts pricing capacity.

Time will tell, though I'd rather not wait until December to find out.