SYNTHOS RESEARCH

Synthos Think Pieces · macro & the dollar system · July 11, 2026

Stablecoins and the dollar system: a new Treasury buyer, or a bank run with better marketing?

A dollar stablecoin is a money-market fund with a chat-app wrapper: take a dollar, buy a T-bill, keep the yield, hand the holder a token that moves like a text message. Post the GENIUS Act, that structure is now US law — and it has quietly turned crypto's least glamorous product into one of the fastest-growing marginal buyers of US Treasuries on earth. The bull case is a two-parter our voices keep repeating: stablecoins are a new demand sink for short-term US debt and an extension of dollar hegemony onto smartphones the banking system never reached. The bear case is just as real: the same design drains bank deposits, imports SVB-style run risk onto instant rails, and rests — in Tether's case — on a private, historically opaque balance sheet. Our read: the Treasury-bid and hegemony arguments are strong and well-evidenced; the disintermediation and run risks are underpriced, not absent. This is a debate, not a verdict.

Synthos Research · synthosresearch.com · Think Piece · sources below · educational only, not investment advice

⚠ This topic is not one reading. "Stablecoins and the dollar system" spans several tracked Synthos topics that disagree with each other — we cite all three below rather than pretend they're one number.

Where the organism actually stands (three readings, not one)

Our expert-claim engine tracks these as separate topics, and they point in different directions. Net-stance scores run on a −100 (all bearish) to +100 (all bullish) scale, equal-weighted after dropping low-conviction claims. The bars below are decorative; the exact numbers are stated in words.

Stablecoins & tokenization
+81.8 held
net stance, delta −1 vs prior window · n=22 claimsNet stance plus 81.8 on a scale from minus 100 to plus 100. Held, essentially unchanged (down 1). Sample size 22 claims.
US dollar
+20.0 low sample
net stance, delta +50 · n=20 — small-sample, low confidenceNet stance plus 20.0 on a scale from minus 100 to plus 100, up 50 versus the prior window. LOW SAMPLE, n=20 — treat as low confidence and directional only.
Fiscal dominance / debasement
+62.5 held
net stance, delta −5 · n=16 claimsNet stance plus 62.5 on a scale from minus 100 to plus 100. Held, down 5. Sample size 16 claims.

Read together: our voices are strongly constructive on stablecoins as a technology and asset class (+81.8, and steady), mildly and unreliably constructive on the dollar itself (+20.0 — but on only 20 claims, a low-sample reading we will not lean on; it swung +50 in a window, which for n=20 is noise as much as signal), and convinced that fiscal dominance and debasement are the governing regime (+62.5). The tension is the whole story: the same panel that loves dollar stablecoins is largely bearish the long-run fiscal path of the dollar they are denominated in. Both can be true — a stablecoin can strengthen the dollar's network even as the debt behind it deteriorates. For the crypto-plumbing view, our crypto hub tracks On-chain finance / crypto infra at +77.3 (n=44, held); this piece is the macro / dollar-system layer above it.

The mechanism: how a token becomes a Treasury bid

Strip away the crypto vocabulary and a compliant dollar stablecoin is almost boring. A user sends $1. The issuer holds that dollar as reserves — under the GENIUS Act's framework for payment stablecoins, overwhelmingly in short-dated US Treasuries and cash-equivalents — and mints one token. The issuer earns the bill yield; the holder earns the convenience of a dollar that settles globally in seconds. Multiply by a float now measured in the hundreds of billions, and you have manufactured a captive, price-insensitive buyer of T-bills that grows with adoption rather than with the Fed's balance sheet.

A Forward Guidance panel put the structural version plainly on March 25, 2025: stablecoins are "the next great marginal buyer of Treasuries," with Tether already among the largest single holders, "selling T-bills directly to foreign citizens" — a mechanism that extends dollar hegemony rather than eroding it. On September 3, 2025 the same show reframed it as a "stealth financial-repression tool": Treasury-backed stablecoins near ~$900B and growing toward $2–3T create "a new captive buyer of government debt." Both framings describe the same plumbing — one calls it strength, one calls it repression.

The scale numbers, and their honest spread. Our voices don't agree on the level, and we won't paper over it: supply figures cited across 2025–26 range from Raoul Pal relaying Treasury Secretary Scott Bessent's ~$3T "within a few years" projection (Oct 23, 2025) and an "already ~$300B drawn from ~$17T of bank deposits" figure, to Forward Guidance's ~$900B → $2–3T, to Natalie Brunell's "$150B → $1.5T → $15T" adoption ladder (Nov 19, 2024). Treat these as competing estimates from named voices, not settled facts. What they agree on is direction: up, and structurally.

The bull case, in two moves

Move one — the Treasury sink. The All-In panel (Sept 18, 2025) put the maximal version on record: the GENIUS Act "unlocks $1–10T of stablecoins," and "the internet becomes the largest holder of US Treasuries within 5 years." Even discounted heavily, the logic is sound — every dollar of net-new stablecoin float that must be backed by bills is a dollar of net-new demand for US government paper, arriving exactly as ballooning deficits need buyers. Forward Guidance (June 30, 2025) called it "the policy-driven unlock for T-bill demand."

Move two — hegemony by smartphone. This is where the dollar-milkshake crowd and the crypto crowd converge. Brent Johnson, whose Dollar Milkshake thesis holds that offshore dollar-debt scarcity structurally bids the dollar up (Mar 30, 2025), is unambiguous (Nov 16, 2025): "USD stablecoins are an incredibly big deal and will keep growing, reinforcing rather than threatening dollar dominance." Raoul Pal (Oct 23, 2025) calls them "an obvious, hyperscaling use case — geopolitically US-aligned, backed by Treasuries"; and (Jun 5, 2025) sees "3bn unbanked" who are "choosing dollar stablecoins over failing local fiat via smartphones." The dollar wins savers the US State Department could never reach. Adoption is not theoretical: Jordi Visser (Feb 21, 2026) notes stablecoins are processing ~$11T a month — more than Mastercard did in all of the prior year (~$10T a year).

The bear case, in four moves — and it is not weak

One — deposit disintermediation. A yield-bearing dollar that any phone can hold is a direct competitor to a checking account, and the deposits it drains are the deposits that fund lending. Anthony Pompliano (Sept 16, 2025): the GENIUS Act "begins narrow banking; regional and super-regional banks face existential risk as deposits flow into yield-bearing stablecoins and DeFi." An Odd Lots guest (July 14, 2025) warned that tokenized money-market stablecoins "pull deposits from banks, starving small-business and farm credit." Brent Johnson — a bull on the coins — flags the same wound (Nov 2, 2025): stablecoin licenses for mega-banks "drain deposits from small regional banks, accelerating their failure."

Two — run risk on instant rails. Stablecoins remove the friction that historically slowed bank runs. Jordi Visser (Dec 10, 2025): stablecoins "enable instant capital flight from weak-currency emerging markets, setting up SVB-style 48-hour digital bank runs." Raoul Pal (Oct 23, 2025) turns the lens on the coins themselves: "SVB was effectively a bank run on a leveraged stablecoin" — the banking system is "inherently fragile, 13-to-1 leverage, long duration against instant liquidity." A fully-reserved coin is safer than a fractional bank, but a redemption panic against even a bill-backed issuer forces fire-sales into a thin market.

Three — the opacity problem, and it is specific. This is where private matters. Tether is privately held — it has no ticker and cannot be bought as a stock; it is context here, never a recommendation. Its float is reportedly among the largest holders of US Treasuries — estimates across our voices range from roughly seventh-largest buyer to top-ten to top-twenty depending on the source — run by a headcount Arthur Hayes (Oct 11, 2025) pegs at ~150 people against JPMorgan's 315,000. That efficiency is the bull case and the risk in one sentence. Lyn Alden has flagged since July 16, 2021 that "Tether reserves aren't pure cash and carry non-zero risk"; an Empire discussion (Oct 4, 2024) noted holders "bear real depeg risk yet capture little of the underlying yield." Attestations are not audits, and the entity setting global short-rates-adjacent flows is one no public shareholder can inspect.

Four — does it even entrench the dollar, or just paper over its decline? The sharpest dissent comes from Luke Gromen, who has argued since Jan 14, 2021 that "the USD is slowly losing reserve-currency status — a structural process, not a crash," as foreign central banks "stop growing Treasury holdings in favor of gold." By his read (Nov 3, 2025), "gold has flipped Treasuries as the dominant foreign central-bank reserve asset." In that frame, a stablecoin Treasury bid is real but marginal and captive — it finances the deficit for another cycle without fixing it. Gromen's own phrase for pushing eurodollar users into US-backed stablecoins and T-bills (Sept 5, 2025): "burning the furniture to stay warm." The counter, from Gromen's opposite number Brent Johnson (Nov 16, 2025): "de-dollarization is largely a myth" — it happens at the margins while "the dollar has never been more systemically important." Two of our highest-skill macro voices, diametrically opposed. We hold both.

The honest tension we won't resolve for you. The Treasury-bid and hegemony cases are strong, evidenced, and now backed by US law. But they and the bear case are describing the same machine from two ends: the thing that makes stablecoins a powerful dollar-extension — frictionless, yield-bearing, deposit-competitive, held by anyone — is the same thing that makes them a deposit-drain and a run vector. You cannot buy the bull case without importing the bear case's plumbing.

The regulatory turn: GENIUS as the pivot

The GENIUS Act is the hinge of the whole story — it converted stablecoins from a regulatory grey zone into sanctioned dollar infrastructure. Empire (June 20, 2025) marked the Senate's 68–30 passage, alongside JPMorgan's own JPMD token, as validation: "every incumbent entrant is validation, not a threat." All-In (Jan 23, 2026) treats the 100%-short-term-Treasury-reserve requirement as "now settled law," making regulated stablecoins "safer than fractional-reserve banks." But the same law has second-order effects our skeptics flag: Forward Guidance (June 11, 2025) warned that restricting collateral to "T-bills and reverse repos will distort those markets, like Basel capital treatment did before," and the interest ban simply pushed yield into "marketing rewards" workarounds (a guest on Pompliano's show, Sept 16, 2025) — banks' deposit franchises get competed anyway, just through a side door. Regulation didn't remove the disintermediation risk; it legalized and accelerated it.

The exposure map — public vs. private, rated vs. not

Horizons: Short = 0–6 months · Medium = 6–24 months · Long = 2+ years. Direction chips carry words, never color alone. A ticker links only where a Synthos deep dive exists; anything without one is shown unrated. Crypto tokens and private companies are not equities and carry no verdict.

NameShortMediumLongWhy
CRCLCircle · public · our verdict: Watch, FV $90 · rate-hostage· contested▲ structural The only pure-play, publicly-tradable, rated name here. ~96% of revenue is Fed-rate-dependent reserve income and roughly half of USDC economics is shared with Coinbase — a "regulated-stablecoin toll road whose toll rate is set by the Fed." Long-run adoption is the tailwind; near-term it is hostage to rate cuts and to the member-owned OUSD consortium that inverts the issuer-premium model. Verdict and fair value per our Circle deep dive — no change here.
USDTTether · private — no ticker · unrated · minimal· watch· contested Privately held; cannot be bought as a stock and carries no Synthos verdict. Context only: ~150 people, reportedly a top-20 US Treasury holder, ~95% margins on reserve yield (All-In, Nov 22, 2025). The efficiency case and the opacity risk are the same balance sheet. Offshore float is insulated short-term; the compliant, GENIUS-era end of its future is the contested zone.
OUSDOpen USD consortium · see think piece ▼ vs issuers· pivotal▲ if it ships Not a token you can price yet — a 140-company member-owned coin that pays reserve yield to members, not an issuer. Structurally bearish the pure-play issuer premium (CRCL), structurally bullish the rails. We cover it in full in the OUSD think piece rather than repeat it here.
Regional banksdeposit franchises · no deep dive · unrated · minimal▼ drag▼▼ structural The clearest bear-case exposure and deliberately unrated — no Synthos deep dive covers the group. Yield-bearing dollars compete directly for the low-cost deposits that fund small-business and farm credit; "narrow banking" (Pompliano) and "burning the furniture" (Gromen) both land here first.
US T-bill marketnot a security to trade · structural ▲ demand▲ demand· double-edged The real object of the thesis. A growing, captive, price-insensitive buyer supports bill demand and dollar reach — while concentrating a new, redemption-sensitive holder base at the front of the curve. Financial-repression tailwind and fragility vector at once.

What we're watching (the falsifiers)

  1. Does the Treasury bid actually show up in the data — verifiable stablecoin reserve holdings climbing into the trillions, or does supply stall near ~$300B–$900B and reveal the demand was narrower (capital-controlled / unbanked users, per Forward Guidance, Apr 1, 2026) than the maximalists claimed?
  2. A first real run. The bull case is "safer than a fractional bank." The bear case is "instant redemption into a thin bill market." The first genuine redemption panic — on any major issuer — reprices the entire thesis. We haven't seen one at scale post-GENIUS.
  3. Measured deposit flight. If regional-bank deposits visibly bleed into yield-bearing dollars, the disintermediation bear case moves from theory to balance sheet — watch small-bank deposit betas and credit availability, not just stablecoin supply.
  4. Tether transparency. A full, independent audit (not an attestation) would defuse the single largest opacity risk in the system. Its continued absence keeps the biggest holder a black box.
  5. The dollar-reserve scoreboard. Gromen vs. Johnson resolves in the data: does the gold-vs-Treasuries central-bank reserve share keep shifting (Gromen), or does the stablecoin network entrench dollar primacy faster than official reserves diversify away (Johnson)?

These get graded like everything else we publish — misses included. Our US-dollar topic reading (+20.0) sits on just 20 claims; we flag it as low-confidence and will not build a call on it until the sample deepens.

Sources

Expert claims above are drawn from the Synthos knowledge base — each carries a named speaker and a real date, spanning Forward Guidance, All-In, Empire, Odd Lots, Real Vision and named macro voices (Brent Johnson, Raoul Pal, Luke Gromen, Lyn Alden, Arthur Hayes, Anthony Pompliano, Jordi Visser, Natalie Brunell). Topic net-stance figures are from our July 11, 2026 sentiment baseline.
Forward Guidance — the stablecoin Treasury-demand thesis · Empire — GENIUS Act passage & issuer validation · Odd Lots — stablecoins as a banking charter · Real Vision — dollar hegemony & the deposit-drain · US Treasury — Bessent stablecoin remarks (as relayed by our voices)

Synthesis note: this piece is a synthesis of dated, attributed expert claims — every quote above carries a named speaker and a real date — distilled by the Synthos organism from its knowledge base; it is a model-generated signal, not a human recommendation. The three topic net-stance readings rest on n=22 (stablecoins & tokenization), n=20 (US dollar) and n=16 (fiscal dominance) claims respectively, and are directional; the US-dollar figure (+20.0, n=20) is explicitly low-sample. Nothing here is investment advice — it is educational analysis of a public debate, and the honest conclusion is that the debate is unresolved.