Yields First, Currencies Later: The Sequencing Question in U.S. Fiscal Deterioration

01 The Question Behind the Debt Discussion

The investor Howard Marks' latest discussion of U.S. debt, as reported, raises a question that is more analytically interesting than the headline debt number: at what point does fiscal deterioration affect currency preference rather than merely Treasury yields? The distinction matters because the two channels price different things. Treasury yields are the price of lending to the U.S. government in dollars — a spread over the expected path of short rates plus compensation for term and inflation risk. Currency preference is a portfolio decision about which unit of account to hold at all. A deterioration can raise yields while leaving currency preference untouched, which is the normal, well-documented case: the market demands compensation for holding more government paper but has not yet questioned the paper's denomination. The concern in Marks' argument, as a reported claim rather than an observed fact, is that the second channel is the one that ultimately binds — and that by the time it activates, policy has few good options left.

The framing for this analysis is a reserve-currency frame: how fiscal-deterioration stress transmits to dollar preference; what reserve-composition evidence (the IMF's Currency Composition of Official Foreign Exchange Reserves, COFER, being the standard dataset) can and cannot show; the sequencing hypothesis — yields first, currency later — as a causal claim to be tested; and Marks' argument treated throughout as a reported claim, not established fact. The discipline matters here more than usual, because the subject invites both alarmism and complacency, and both are forms of failing to distinguish evidence categories.

A note on sources, for the record: the Wikipedia reference material accompanying this analysis covers the United States federal budget — the budget being the financial representation of the priorities of the government, primarily spending on healthcare, retirement, and defense, with the non-partisan Congressional Budget Office providing extensive analysis of the budget and its economic effects (source: Wikipedia summary — United States federal budget). The budget process is the mechanism that generates the debt path, so the budget structure is where any debt analysis must ground itself.

02 The Transmission Sequence: From Deficits to Yields to Currencies

The causal chain runs in stages, and each stage has its own threshold. Stage one: persistent deficits — spending on healthcare, retirement, and defense exceeding revenue — require debt issuance year after year (source: Wikipedia summary — United States federal budget). The stock of debt grows, and with it the gross interest bill, which adds to spending, which adds to deficits: the familiar debt-arithmetic feedback that makes fiscal deterioration self-reinforcing above certain rate-growth differentials. Stage two: the market prices the growing supply. Yields incorporate a term premium for the volume of paper the Treasury must place — this is the observed, mechanical phase. Stage three, the hypothesis: at some point, the portfolio calculus of large official and private holders shifts from "what yield do I demand for holding dollars" to "do I want this much of my portfolio denominated in dollars at all." The question is not default — a sovereign issuing its own currency does not face a mechanical default in that currency — but the composition of global portfolios, the share of international transactions invoiced and settled in dollars, and the willingness of foreign official holders to keep absorbing Treasury supply.

The sequencing claim — yields first, currencies later — has a specific analytical content: it asserts that the price channel adjusts continuously and early, while the quantity channel (portfolio shares) adjusts discretely and late. Why would that be? Because changing currency preference is costly: reserve composition is sticky, invoicing habits are sticky, and switching involves operational rewiring across the entire international financial system. Yields, by contrast, re-price every day. So the system rationally does its adjusting in the cheap dimension first. The implication is uncomfortable: a long period of rising yields with stable currency preference is not evidence against the currency risk — it is exactly what the sequencing hypothesis predicts as the intermediate phase. The relevant test is not whether dollar preference is falling today but whether the marginal buyer of the growing Treasury supply is changing character — from foreign official accumulation to domestic price-insensitive holdings to, at the limit, central-bank accommodation of some form.

03 What the Reserve Evidence Can and Cannot Show

The reserve-composition evidence is the natural place to look for currency preference in action, and the IMF's COFER dataset is the standard source. What it can show: the share of allocated official foreign exchange reserves held in dollars versus other currencies, over time. What it cannot show, and this is the analytically important limitation: COFER records stocks at a coarse frequency and does not distinguish the reasons for changes — a shift in reserve shares could reflect currency preference, exchange-rate valuation effects (the dollar share falls mechanically when the dollar depreciates, with no behavioral change at all), or reserve growth differentials across holders. It is also a lagging indicator of private behavior: official reserves are the slowest-moving tranche of global portfolios, so stability in COFER is weak evidence against shifts in private currency preference.

With those limitations stated, the honest reading of the reserve evidence as a category is mixed. The dollar remains the dominant reserve currency — no rival offers the combination of deep markets, convertibility, legal security, and network effects that reserve managers require. But the broad direction of the marginal change is what the sequencing hypothesis concerns: whether the newest marginal dollar of reserve accumulation is being placed in dollars at the same propensity as the existing stock, and whether the set of instruments counted as reserves is drifting — gold accumulation by official holders, swaps and bilateral arrangements, and other stores of value that do not show up as dollar-denominated reserve assets. The composition drift, not the level, is the observable that matters. A reserve system can be stable in level and shifting in composition for years before the shift becomes visible in headline shares.

This is where Marks' argument earns its analytical weight, as a reported claim: the concern is not that the dollar loses reserve status in a discrete event, but that the deterioration erodes the margins — the marginal buyer, the marginal invoicing decision, the marginal central bank's willingness to hold — while the headline aggregates remain calm enough to support complacency. The argument's structure is a threshold model: no visible change until a visible change, at which point the change is large. Threshold models are hard to test in real time and easy to dismiss, which is precisely why the debate persists.

04 The Counterfactual: What Deterioration Would Look Like Without the Currency Risk

The counterfactual discipline: what would the same fiscal path look like if currency preference were not a binding constraint at any threshold? The answer: it would look like Japan. A large industrial economy with an even larger government debt-to-GDP ratio than the United States has financed that debt for decades at very low nominal yields, because the debt is held overwhelmingly domestically, the domestic institutional base is a captive price-insensitive buyer pool, and — critically — the country's savings behavior is in domestic-currency assets. Yields absorbed the fiscal path; currency preference was never the binding channel, because the domestic pool never shrank. The counterfactual shows what the United States is not: a mostly-domestically-held debt stock. A meaningful share of U.S. federal debt is held by foreign investors, which means the currency-preference question cannot be neutralized by domestic savings, and the transmission chain has a stage three by construction. This comparison is illustrative and structural, not a prediction; it is used to establish that the sequencing question is not merely theoretical.

05 Competing Explanations for Calm Markets Under Rising Debt

The puzzle that motivates the debate: U.S. fiscal deterioration has been persistent and well documented — the budget's structural components, healthcare, retirement, and defense, being the primary spending categories (source: Wikipedia summary — United States federal budget) — and yet dollar asset markets have not produced the visible crisis the arithmetic seems to imply. Four competing explanations, with their distinguishing evidence. First, the safe-asset explanation: global demand for safe, liquid, dollar-denominated assets exceeds supply, so the growing debt stock is absorbed without a price penalty. Conflicting evidence: this holds only while the alternatives remain inferior; it is a description of the current hierarchy, not a law. Distinguishing test: watch whether the yield premium demanded at issuance auctions grows as supply grows.

Second, the sequencing explanation (the Marks-adjacent view): the price channel has been doing the adjusting, with yields higher than a fiscal-neutral counterfactual would produce, and the quantity channel has not yet been tested at threshold. Conflicting evidence: real yields have not been persistently extreme, which the strongest form of this hypothesis might predict. Distinguishing test: marginal-buyer composition, discussed above. Third, the fiscal-response explanation: markets price the political system's demonstrated capacity for eventual adjustment — periods of consolidation, however partial, keep the terminal-debt arithmetic from compounding unchecked. Conflicting evidence: the budget process has produced protracted standoffs and shutdown-adjacent episodes in recent years, which markets have mostly shrugged off, suggesting the fiscal-response explanation is not what is holding yields. Fourth, the complacency explanation: markets are simply underpricing a slow-moving risk, as they have before other threshold events. This is unfalsifiable in real time by construction, which is both its weakness and, its proponents note, the general property of threshold models before the threshold is crossed.

06 Scenarios: How the Sequencing Resolves

Scenario A — Fiscal response before threshold. The budget process produces a durable consolidation — revenue increases, entitlement adjustments, or spending restraint sufficient to stabilize the debt path relative to growth. Trigger: a bipartisan budget framework that the bond market treats as durable. Mechanism: the feedback loop of interest-bill-on-deficits is broken from the source. Indicators: Congressional Budget Office projections showing the debt path flattening; a decline in the term premium; auction statistics normalizing. Consequence: yields fall toward the fiscal-neutral path, the currency-preference question is deferred indefinitely, and the sequencing debate becomes moot for a cycle.

Scenario B — Muddling through, yields doing the work. No consolidation, no crisis: the price channel continues absorbing the supply at a premium, currency preference erodes at the margins but never crosses a visible threshold. This is the baseline continuation. Trigger: continued political gridlock without market disruption. Mechanism: stage two operating indefinitely, stage three advancing only at the composition margins. Indicators: COFER dollar share drifting without breaking; gold's official share of reserves continuing to rise; yields carrying a persistent fiscal premium; no acute auction failure. Consequence: a slow, expensive equilibrium — the debt service costs rise, crowding out other budget priorities (the budget being the representation of the government's priorities, per the reference record), and the deterioration is paid in opportunity cost rather than crisis.

Scenario C — Threshold crossing. The quantity channel activates: a visible shift in marginal and then stock currency preference — foreign official accumulation of Treasuries turning to diversification at scale, a meaningful rise in non-dollar invoicing, or a domestic accommodation of the debt that market participants read as fiscal dominance. Trigger: any combination of a political commitment to monetize debt, an acute failed-supply episode, or an exogenous shock that forces a large discretionary fiscal expansion on top of an already deteriorating path. Mechanism: the sticky quantity channel unsticking, which by its nature arrives late and all at once. Indicators: COFER dollar share falling across several consecutive periods beyond valuation effects; gold's official share accelerating; persistent auction weak-cover statistics; currency-swap arrangements displacing dollar settlement in trade corridors. Consequence: the uncomfortable phase — yields and currency adjust together, and policy loses the option of a gradual, price-mediated path.

07 Indicators to Watch

The indicator set for the sequencing question, each with its diagnostic purpose. First, Treasury auction statistics — bid-to-cover and indirect bidder share — as the direct read on marginal-buyer character. Second, the term premium embedded in long yields — the fiscal-risk price channel operating. Third, COFER dollar share changes decomposed from valuation effects — the quantity channel's slow observable. Fourth, official gold accumulation and its share of official reserve assets — the composition-drift observable that COFER's definition may understate. Fifth, debt-service costs as a share of budget outlays — the feedback loop's intensity measure, grounded in the budget structure where healthcare, retirement, and defense dominate spending (source: Wikipedia summary — United States federal budget). Sixth, CBO long-term debt projections and their revisions — the terminal-arithmetic monitor. Seventh, the share of global trade invoiced in dollars, where observable — the network-effect layer that currency preference ultimately rides on. Eighth, episodic political-risk events — debt-ceiling confrontations and shutdown episodes — as stress tests that reveal how the market prices fiscal dysfunction at the margin.

08 The Bottom Line

What we know: The U.S. budget's structure — healthcare, retirement, and defense as the primary spending programs, with CBO providing the non-partisan analysis (source: Wikipedia summary — United States federal budget) — generates a persistent deficit path; the debt-arithmetic feedback from interest costs is a mechanical property of that path; and reserve composition is sticky, adjusting slowly and mostly at the margins.

What we think we know: The yields-first, currency-later sequencing describes how the adjustment has actually been distributed so far — the price channel doing the work while the quantity channel drifts — and the calm in currency-preference aggregates is therefore weak evidence against the threshold concern. The marginal-buyer composition is the most informative single observable, and it is more informative than either headline yields or headline reserve shares.

What we do not know: Where the threshold is, if there is one; whether the current era's combination of high debt and high nominal rates produces the compounding dynamic faster than the political system's demonstrated response capacity; and whether the composition drift — the slow rise of non-dollar stores of value in official portfolios — is a leading indicator or a permanently sub-threshold rearrangement.

What to watch next: Auction marginal-buyer statistics; term-premium movements; COFER and gold-share trends with valuation effects separated; debt-service share of outlays; and any episode in which the market's reaction to fiscal-political dysfunction changes character — because in a sequencing model, the first sign that stage three has begun is that stage two's prices stop making sense.